Electronic version 2026
originally by:
This version:
CONTENTS
8.9 CLASSICAL LIMIT AND ASYMPTOTIC EXPRESSIONS FOR THE CLEBSCH-GORDAN COEFFICIENTS
8.9.1 Asymptotic Expressions for a , c ≫ b a, c \gg b a , c ≫ b
8.9.2 Asymptotic Expressions for a , b , c , γ ≫ a + b − γ a, b, c, \gamma\gg a + b - \gamma a , b , c , γ ≫ a + b − γ
8.9.3 Semiclassical Formulas for a , b , c ≫ 1 a, b, c \gg1 a , b , c ≫ 1
8.9.4 Squares of the Clebsch-Gordan Coefficients in the Classical Limit
8.10 ZEROS OF THE VECTOR-ADDITION COEFFICIENTS
8.11 CONNECTION OF THE CLEBSCH-GORDAN COEFFICIENTS AND THE 3 jm SYMBOLS WITH ANALOGOUS FUNCTIONS OF OTHER AUTHORS
8.12 ALGEBRAIC TABLES OF THE CLEBSCH-GORDAN COEFFICIENTS
8.13 NUMERICAL TABLES OF THE CLEBSCH-GORDAN COEFFICIENTS
6j Symbols and the Racah Coefficients
Definition
6j Symbols
Racah Coefficients
R Symbols
General Expressions for the 6j Symbols. Relations Between the 6j Symbols and Other Functions
Expressions for the 6j Symbols in Terms of Finite Sums
Bargmann Formula [61]
Relations Between the 6j Symbols and the Generalized Hypergeometric Functions
Relations Between the 6j Symbols and the 3jm Symbols
Quasi-Binomial Representation of the 6j Symbols
Integral Representations of the 6j Symbols
Symmetries of the 6j Symbols and the Racah Coefficients
R-Symbols
6j Symbols
Racah Coefficients
"Mirror" Symmetry
Explicit Forms of the 6j Symbols for Certain Arguments
One of Arguments is Equal to Zero
One of Arguments is Equal to the Sum of Two Others
One of Arguments is Smaller by Unity than the Sum of Two Others
Arguments a , b , d , e a, b, d, e a , b , d , e are Equal in Pairs
Recursion Relations
Relations in Which Arguments are Changed by 1 / 2 1/2 1/2
Relations in Which Arguments are Changed by 1
Generating Function
Sums Involving the 6j Symbols
Asymptotics of the 6j Symbols for Large Angular Momenta
Asymptotic Relations Between the 6j Symbols and the Clebsch-Gordan Coefficients
Asymptotic Expressions for the 6j Symbols
Relations Between the Wigner 6j Symbols and Analogous Functions of Other Authors
Tables of Algebraic Expressions for the 6j Symbols
9j and 12j Symbols
Definition of the 9j Symbols
9j Symbols as Recoupling Coefficients
9j Symbol and r Symbol
Explicit Forms of the 9j Symbols and Their Relations to Other Functions
Expressions for the 9j Symbols in the Forms of Algebraic Sums
Wu Formulas [77]
9j Symbols as Sums of Products of the Clebsch-Gordan Coefficients or the 3jm Symbols [54]
Integral Representations of the 9j Symbols
Representation Involving the Generalised Hypergeometric Function
Representation Involving the Wigner D D D Functions
Integrals Involving Characters of Irreducible Representations of Rotation Group
Symmetry Properties of the 9j Symbols
Permutation Symmetry
Symmetries of the r Symbol
Originally published by
World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE
Open Access
This project is open access and is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/ ), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. It is based on the original work published by World Scientific Publishing Co. Pte. Ltd., Singapore, 1988, ISBN 978-9971-5-0107-5 (hardcover), ISBN 978-9971-5-0996-5 (paperback), ISBN 978-981-4415-49-1 (ebook for institutions), ISBN 978-981-4578-28-8 (ebook for individuals).
Open Access funded by SCOAP (Sponsoring Consortium for Open Access Publishing in Particle Physics)
The original ebook was converted to Open Access in 2021 through the sponsorship of SCOAP , licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/ ), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. Original copyright © 1988 by World Scientific Publishing Co. Pte. Ltd.
Preface
In the years since its publication, this book has become the standard reference on the quantum theory of angular momentum. The book provides a comprehensive treatment of the subject, covering both the theoretical foundations and practical applications of angular momentum in quantum mechanics. As an open access publication, it is freely available to researchers and students worldwide, ensuring that the knowledge contained within its pages can be accessed by anyone interested in the field. The book has been widely cited and used in research and teaching, and it continues to be a valuable resource for those studying quantum mechanics and related areas of physics.
However, there are some drawbacks to the form in which it was provided. The scanned version of the book is not searchable, and the quality of the scanned images is not always optimal. This can make it difficult to find specific information or to read the text clearly. Additionally, the scanned version may not be compatible with all devices or software, which can limit its accessibility. Also, there is no verification of the accuracy of the scanned text, which can lead to errors or misinterpretations of the content. Finally, the scanned version may not be easily editable or adaptable for different purposes, such as creating new editions or translations.
Thus this attempt to provide a more accessible and user-friendly version of the book, with improved searchability, image quality, and compatibility with different devices and software. The text has been carefully reviewed and verified for accuracy, and it is provided in a format that can be easily edited or adapted for different purposes. We hope that this version of the book will be a valuable resource for researchers, students, and anyone interested in the quantum theory of angular momentum. The author would like to thank the original publisher, World Scientific Publishing Co. Pte. Ltd., for their support in making this open access version available to the public.
Clearly in the era of AI, I have made substantial use of AI tools to assist in the preparation of this version of the book. I have used AI tools to help with tasks such as text recognition, formatting, and editing. These tools have helped to improve the accuracy and quality of the text, as well as to streamline the process generating code to verify equations. However, I have also carefully reviewed and verified the output of these tools to ensure that the final version of the book is as accurate and reliable. Overall, I believe that the use of AI tools has been a valuable aid in the preparation of this new open access version of the book.
Of course, the author would like to acknowledge that the original work was done by D. A. Varshalovich, A. N. Moskalev, and V. K. Khersonskii, and that this new version is based on their work. The author has made every effort to ensure that the content of this new version is faithful to the original work, while also improving its accessibility and usability for modern readers.
I do no see this as a static resource—there are undoubtedly errors in the text, and I would be grateful for any corrections or suggestions for improvement. Please post these at the GitHub repository for this project at . The intention is to improve the accessibility and usability of this important work, and I welcome any feedback, and contributions, that can help achieve that goal.
Manchester, Uk, September 2026
Original Preface
This book deals with one of the basic topics of quantum mechanics: the theory of angular momentum and irreducible tensors. Being rather versatile, the mathematical apparatus of this theory is widely used in atomic and molecular physics, in nuclear physics and elementary particle theory. It enables one to calculate atomic, molecular and nuclear structures, energies of ground and excited states, fine and hyperfine splittings, etc. The apparatus is also very handy for evaluating the probabilities of radiative transitions, cross sections of various processes such as elastic and nonelastic scattering, different decays and reactions (both chemical and nuclear) and for studying angular distributions and polarizations of particles.
Today this apparatus is finding ever increasing use in solving practical problems relating to quantum chemistry, kinetics, plasma physics, quantum optics, radiophysics and astrophysics.
The basic ideas of the theory of angular momentum were first put forward by M. Born, P. Dirac, W. Heisenberg and W. Pauli. However, the modern version of its mathematical apparatus was developed mainly in the works of E. Wigner, J. Racah, L. Biedenharn and others who applied group theoretical methods to problems in quantum mechanics.
At present a number of good books on the theory of angular momentum have been already published. The general principles and results of the theory may be found in the books by M. Rose [31], A. Edmonds [16], U. Fano and G. Racah [18], A. P. Yutsis, I. B. Levinson and V. V. Vanagas [44], A. P. Yutsis and A. A. Bandzaitis [45], D. Brink and G. Satcber [9]. Nevertheless, many formulas and relationships essential for practical calculations have escaped these books and are either scattered in various editions, or included as appendices in papers discussing somewhat disparate topics, making them generally inaccessible. Even greater difficulties arise when one tries to use the results, as each author employs his own phase conventions, initial definitions and symbols.
The authors of this book aimed at collecting and compiling ample material on the quantum theory of angular momentum within the framework of a single system of phases and definitions. This is why, in addition to the basic theoretical results, the book also includes a great number of formulas and relationships essential for practical applications.
This edition is the translated version of our book published in the USSR in 1975. In the course of its preparation we have tried to comply with a number of suggestions from our readers. For instance, each chapter opens with a comprehensive listing of its contents to ease the search for information needed. We also included some new results relating to different aspects of angular momentum theory which have recently appeared in journals. Unfortunately the limited volume of the present book prevented us from covering all the aforementioned results. We offer sincere apologies to the authors whose results we failed to include.
The monograph is a kind of handbook. Consequently the material is presented in concise form. Most of the formulas and relationships are given without proof. Their full derivation may be found in the literature listed at the end of the text. Some results which have become generally known are given without references, and for this we also apologize.
The sequence adopted is as follows: chapter, section and subsection. Many chapters are self-contained and can be read independently of the others. Sections have double numbering: the first figure denotes the number of the chapter, the second, the number of the section. Equations are numbered within the confines of the section they are included in. When referring to an equation from the same section only the number of the equation is given, e.g., (3), (27); when reference is made to an equation from another section the numbers of the chapter, section and equation are given, e.g., Eq. 4.2.(17). A similar system is adopted when referring to individual subsections, e.g., Sec. 1.2.5. For convenience the book also contains a glossary of all symbols used in the text with references to the pages where their corresponding definitions are given. The list of references is divided into parts: the first part lists books and reviews; the second, papers on different subjects; the third, tables; the fourth, references added during translation.
The authors hope that many specialists will find in the book some fresh and interesting information. The material is prepared and arranged so as to make it useful to those less familiar with theory and for students of physics. These readers can effectively use the monograph as a supplementary text to their main courses.
For those who wish to thoroughly familiarize themselves with the fundamentals of angular momentum theory we recommend the excellent new book by L. Biedenharn and J. Louck [132] Angular Momentum in Quantum Physics. Theory and Applications.
The authors wish to express their deep appreciation to D. G. Yakovlev who took the trouble of reading the English translation of the book and gave some valuable suggestions on its preparation.
Leningrad
Introduction: Basic Concepts
The evaluation of many physical quantities, such as expectation values of energy, electric and magnetic multipole moments, transition probabilities, etc., is considerably simplified by making use of the transformation properties of these quantities under coordinate rotation and inversion in three-dimensional space.
The transformation properties of physical quantities with respect to rotations reveal themselves either through rotations of the given physical system relative to some fixed reference frame or through rotations of the coordinate axes relative to the physical system.
Inversion characterizes the behavior of a physical quantity under transformation from a right-handed coordinate system to a left-handed one, or vice versa.
Many physical quantities, by their nature, are invariants under coordinate rotations. In particular, the properties of any closed physical system should be independent of rotations, as follows from the isotropy of space. As a consequence of this fundamental property of space, the total angular momentum of such a system is an integral of motion.
A similar situation occurs with regard to coordinate inversion. Excluding phenomena connected with the weak interaction, a wealth of atomic, molecular and nuclear processes look alike in right-handed and left-handed coordinate systems. As a result of this ”mirror” symmetry, quantum states of atoms, molecules, nuclei and elementary particles may be characterized by definite parity.
Strictly speaking, a quantum mechanical wave function Ψ ( r ) \Psi(\boldsymbol{r}) Ψ ( r ) of any closed physical system may be characterized by four quantum numbers ( ε , π , j , m ) (\varepsilon,\pi,j,m) ( ε , π , j , m ) which are the eigenvalues of four commuting operators: the Hamiltonian H ^ \hat{H} H ^ , the parity operator P ^ π \hat{P}_{\pi} P ^ π , the operator J ^ 2 \hat{J}^{2} J ^ 2 of the square of the angular momentum and the operator J ^ z \hat{J}_{z} J ^ z of the projection of this momentum onto a quantization axis. Thus Ψ ( r ) ≡ Ψ ε π α j m ( r ) \Psi(\boldsymbol{r}) \equiv\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r}) Ψ ( r ) ≡ Ψ ε π α j m ( r ) obeys the equations
H ^ Ψ ε π α j m ( r ) = ε Ψ ε π α j m ( r ) , \hat{H}\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r})=\varepsilon\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r}), H ^ Ψ ε π α j m ( r ) = ε Ψ ε π α j m ( r ) ,
P ^ π Ψ ε π α j m ( r ) = π Ψ ε π α j m ( r ) , \hat{P}_{\pi}\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r})=\pi\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r}), P ^ π Ψ ε π α j m ( r ) = π Ψ ε π α j m ( r ) ,
J ^ 2 Ψ ε π α j m ( r ) = j ( j + 1 ) Ψ ε π α j m ( r ) , \hat{J}^{2}\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r})=j(j+1)\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r}), J ^ 2 Ψ ε π α j m ( r ) = j ( j + 1 ) Ψ ε π α j m ( r ) ,
J ^ z Ψ ε π α j m ( r ) = m Ψ ε π α j m ( r ) , \hat{J}_{z}\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r})=m\Psi_{\varepsilon\pi\alpha jm}(\boldsymbol{r}), J ^ z Ψ ε π α j m ( r ) = m Ψ ε π α j m ( r ) ,
where α \alpha α denotes all other quantum numbers (if available) and r \boldsymbol{r} r represents a variety of arguments.
ε \varepsilon ε , π \pi π , j j j and m m m are the integrals of motion not only for a closed system, but also for any system exposed to the action of an external spherically-symmetric field. Moreover, the ( ε π j m ) (\varepsilon\pi jm) ( ε π j m ) -representation appears to be very convenient for practical calculation even if the external field is not spherically-symmetric.
For a given j j j , there exist 2 j + 1 2j+1 2 j + 1 wave functions which correspond to different m m m . These functions describe the quantum states of the system which differ only by the orientation of the angular momentum in space.
Under coordinate rotations these functions undergo linear mutual transformations which do not involve the functions of other quantum numbers ( ε , π , α , j ) (\varepsilon,\pi,\alpha,j) ( ε , π , α , j ) and may be written as
Ψ ε α j m ′ ( r ′ ) = ∑ m Ψ ε α j m ( r ) D m m ′ j ( α , β , γ ) \Psi_{\varepsilon\alpha jm'}(\boldsymbol{r}')=\sum_m\Psi_{\varepsilon\alpha jm}(\boldsymbol{r})D^j_{mm'}(\alpha,\beta,\gamma) Ψ ε α j m ′ ( r ′ ) = m ∑ Ψ ε α j m ( r ) D m m ′ j ( α , β , γ )
where the transformations coefficients D m m ′ j ( α , β , γ ) D^j_{mm'}(\alpha,\beta,\gamma) D m m ′ j ( α , β , γ ) are called Wigner D D D -functions. They are the elements of the finite rotation matrix in the j j j -representation and depend on the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ which determine rotation.
Let a quantum-mechanical system which possesses some fixed angular momentum j j j and its projection m m m consist of two subsystems, with certain angular momenta j 1 j_1 j 1 and j 2 j_2 j 2 , respectively. In this case the wave function Ψ j 1 j 2 j m ( r 1 , r 2 ) \Psi_{j_1j_2jm}(r_1,r_2) Ψ j 1 j 2 j m ( r 1 , r 2 ) may be constructed from the wave functions of subsystems according to the relation
Ψ j 1 j 2 j m ( r 1 , r 2 ) = ∑ m 1 m 2 C j 1 m 1 j 2 m 2 j m Ψ j 1 m 1 ( r 1 ) Ψ j 2 m 2 ( r 2 ) \Psi_{j_1j_2jm}(r_1,r_2)=\sum_{m_1m_2}C^{jm}_{j_1m_1j_2m_2}\Psi_{j_1m_1}(r_1)\Psi_{j_2m_2}(r_2) Ψ j 1 j 2 j m ( r 1 , r 2 ) = m 1 m 2 ∑ C j 1 m 1 j 2 m 2 j m Ψ j 1 m 1 ( r 1 ) Ψ j 2 m 2 ( r 2 )
The quantities C j 1 m 1 j 2 m 2 j m C^{jm}_{j_{1}m_{1}\,j_{2}m_{2}} C j 1 m 1 j 2 m 2 j m are called the Clebsch-Gordan coefficients. They are very important in quantum mechanics because they allow one to construct wave functions for various complex systems (nuclei, atoms, molecules, etc.). The addition of many angular momenta into some resultant total angular momentum may be performed in different ways, or, in other words, in accordance with different coupling schemes of the momenta in question. Each scheme may be associated with a certain representation of these angular momenta. The unitary transformations which relate various representations and describe the recoupling of angular momenta are realized by matrices expressed in terms of the 6 j 6j 6 j -9 j 9j 9 j -symbols and others 3 n j 3nj 3 nj -symbols of higher order.
One of the principal concepts of the quantum theory of angular momentum is that of an irreducible tensor. By definition, an irreducible tensor of rank λ \lambda λ has 2 λ + 1 2\lambda+1 2 λ + 1 components which transform under coordinate rotation as
M λ μ = ∑ μ ′ M λ μ ′ D μ ′ μ λ ( α , β , γ ) \mathcal{M}_{\lambda\mu}=\sum_{\mu'}\mathcal{M}_{\lambda\mu'}D^{\lambda}_{\mu'\mu}(\alpha,\beta,\gamma) M λ μ = μ ′ ∑ M λ μ ′ D μ ′ μ λ ( α , β , γ )
In particular, the wave functions Ψ j m \Psi_{jm} Ψ j m with fixed j j j but different m m m constitute an irreducible tensor of rank j j j . Moreover, irreducible tensors may be constructed from various physical quantities. For instance, energy is a tensor of zero rank (scalar), spin and magnetic moments are tensors of first rank (vectors), the quadrupole moment is a tensor of second rank, etc. Generally, any physical quantity or operator which corresponds to these quantities may be represented as a linear combination of irreducible tensors.
The introduction of irreducible tensors into the analysis of physical quantities or corresponding operators substantially simplifies the evaluation of matrix elements of irreducible tensor operators,
⟨ ϵ ′ π ′ α ′ j ′ m ′ ∣ M ^ λ μ ∣ ϵ π α j m ⟩ = ∫ Ψ ϵ ′ π ′ α ′ j ′ m ′ ∗ ( r ) M ^ λ μ Ψ ϵ π α j m ( r ) d r \langle\epsilon'\pi'\alpha'j'm'|\widehat{\mathcal{M}}_{\lambda\mu}|\epsilon\pi\alpha jm\rangle=\int\Psi^{*}_{\epsilon'\pi'\alpha'j'm'}(r)\widehat{\mathcal{M}}_{\lambda\mu}\Psi_{\epsilon\pi\alpha jm}(r)\,dr ⟨ ϵ ′ π ′ α ′ j ′ m ′ ∣ M λ μ ∣ ϵ π α j m ⟩ = ∫ Ψ ϵ ′ π ′ α ′ j ′ m ′ ∗ ( r ) M λ μ Ψ ϵ π α j m ( r ) d r
where ∫ … d r \int\ldots dr ∫ … d r denotes integration over continuous variables and summation over discrete ones. These matrix elements determine expectation values of physical quantities in definite quantum states, the probabilities of quantum transitions between various states, etc.
According to the Wigner-Eckart theorem, all the dependence of matrix elements on the orientation of coordinate axes, i.e., on quantum numbers m , m ′ , μ m,m',\mu m , m ′ , μ which determine the projections of angular momenta, is entirely contained in the Clebsch-Gordan coefficient
⟨ ϵ ′ π ′ α ′ j ′ m ′ ∣ M ^ λ μ ∣ ϵ π α j m ⟩ = C j m λ μ j ′ m ′ ⟨ ϵ ′ π ′ α ′ j ′ ∣ ∣ M ^ λ ∣ ∣ ϵ π α j ⟩ . \langle\epsilon'\pi'\alpha'j'm'|\widehat{\mathcal{M}}_{\lambda\mu}|\epsilon\pi\alpha jm\rangle=C^{j'm'}_{jm\lambda\mu}\langle\epsilon'\pi'\alpha'j'||\widehat{\mathcal{M}}_{\lambda}||\epsilon\pi\alpha j\rangle. ⟨ ϵ ′ π ′ α ′ j ′ m ′ ∣ M λ μ ∣ ϵ π α j m ⟩ = C j mλ μ j ′ m ′ ⟨ ϵ ′ π ′ α ′ j ′ ∣∣ M λ ∣∣ ϵ π α j ⟩ .
In this case ⟨ ϵ ′ π ′ α ′ j ′ ∣ ∣ M ^ λ ∣ ∣ ϵ π α j ⟩ \langle\epsilon'\pi'\alpha'j'||\widehat{\mathcal{M}}_{\lambda}||\epsilon\pi\alpha j\rangle ⟨ ϵ ′ π ′ α ′ j ′ ∣∣ M λ ∣∣ ϵ π α j ⟩ is an invariant factor called a reduced matrix element. The Wigner-Eckart theorem allows one to reduce evaluation of transition probabilities, angular distributions, polarizations, etc., to the calculation of standard sums of the vector addition coefficients and recoupling coefficients.
All the facts mentioned above reveal that the quantum theory of angular momentum provides the formalism which is universal and extremely convenient for various practical calculations.
Elements of Vector and Tensor Theory
The theory of angular momenta and irreducible tensors represents, in principle, a development of the classical theory of vectors and tensors. In this chapter only the basic definitions and relations of the vector and tensor theory are represented which will be used throughout. For more detailed analysis see corresponding monographs (e.g., Refs. [1, 2, 3]).
Coordinate Systems. Basis Vectors
In the quantum theory of angular momentum cartesian, polar and spherical coordinate systems are widely used.
Cartesian Coordinate System
In a rectangular cartesian coordinate system the position of a point is specified by three real numbers x x x , y y y , z z z which represent the distances between the point and coordinate planes (Figure 1.1 ). The position vector (radius vector) of a point r \mathbf{r} r may be written as
Figure 1.1. Cartesian coordinate system.
r = x e x + y e y + z e z . (1) \mathbf{r} = x\mathbf{e}_x + y\mathbf{e}_y + z\mathbf{e}_z. \tag*{(1)} r = x e x + y e y + z e z . (1)
The covariant cartesian basis (base) vectors e x , e y , e z \mathbf{e}_x,\mathbf{e}_y,\mathbf{e}_z e x , e y , e z form a real orthonormal basis
e i e k = δ i k , ( i , k = x , y , z ) , (2) \mathbf{e}_i\mathbf{e}_k = \delta_{ik},\ (i,k=x,y,z), \tag*{(2)} e i e k = δ ik , ( i , k = x , y , z ) , (2)
e i ∗ = e i , ( i = x , y , z ) . (3) \mathbf{e}_i^{*} = \mathbf{e}_i,\ (i=x,y,z). \tag*{(3)} e i ∗ = e i , ( i = x , y , z ) . (3)
The contravariant cartesian basis (base) vectors e i ( i = x , y , z ) \mathbf{e}^{i}\ (i=x,y,z) e i ( i = x , y , z ) coincide with the covariant ones
e i = e i . (4) \mathbf{e}^{i} = \mathbf{e}_i. \tag*{(4)} e i = e i . (4)
Throughout this book the right-handed coordinate system will be used. In this system
[ e i × e k ] = ε i k l e l , ( i , k , l = x , y , z ) , (5) [\mathbf{e}_i \times\mathbf{e}_k] = \varepsilon_{ikl}\mathbf{e}_l,\quad(i,k,l=x,y,z), \tag*{(5)} [ e i × e k ] = ε ik l e l , ( i , k , l = x , y , z ) , (5)
ε i k l = [ e i × e k ] e l . (6) \varepsilon_{ikl} = [\mathbf{e}_i \times\mathbf{e}_k]\mathbf{e}_l. \tag*{(6)} ε ik l = [ e i × e k ] e l . (6)
A detailed form of (1.1.5) is
[ e x × e y ] = e z , [ e y × e z ] = e x , [ e z × e x ] = e y . (7) [\mathbf{e}_x \times\mathbf{e}_y] = \mathbf{e}_z,\quad[\mathbf{e}_y \times\mathbf{e}_z] = \mathbf{e}_x,\quad[\mathbf{e}_z \times\mathbf{e}_x] = \mathbf{e}_y. \tag*{(7)} [ e x × e y ] = e z , [ e y × e z ] = e x , [ e z × e x ] = e y . (7)
Polar Coordinate System
In a polar coordinate system1 the position of a point is determined by r , ϑ , φ r,\vartheta,\varphi r , ϑ , φ , where r r r is the position vector length, ϑ \vartheta ϑ is the colatitude, and φ \varphi φ is the longitude (Figure 1.2 ). The angles ϑ \vartheta ϑ and φ \varphi φ are called the polar angles of vector r \mathbf{r} r . The relations between cartesian and polar coordinates are
Figure 1.2. Polar coordinate system.
x = r sin ϑ cos φ , r = x 2 + y 2 + z 2 , 0 ≤ r < ∞ , y = r sin ϑ sin φ , ϑ = arccos z x 2 + y 2 + z 2 , 0 ≤ ϑ ≤ π , z = r cos ϑ , φ = arccos x x 2 + y 2 , ( tan φ = y x ) , 0 ≤ φ < 2 π . (8) \begin{aligned}
x &= r\sin\vartheta\cos\varphi,\qquad&r &= \sqrt{x^2+y^2+z^2},\qquad&0 \le r < \infty,\\
y &= r\sin\vartheta\sin\varphi,\qquad&\vartheta&= \arccos\frac{z}{\sqrt{x^2+y^2+z^2}},\qquad&0 \le\vartheta\le\pi,\\
z &= r\cos\vartheta,\qquad&\varphi&= \arccos\frac{x}{\sqrt{x^2+y^2}},\quad\left(\tan\varphi=\frac{y}{x}\right),\qquad&0 \le\varphi< 2\pi. \tag*{(8)}
\end{aligned} x y z = r sin ϑ cos φ , = r sin ϑ sin φ , = r cos ϑ , r ϑ φ = x 2 + y 2 + z 2 , = arccos x 2 + y 2 + z 2 z , = arccos x 2 + y 2 x , ( tan φ = x y ) , 0 ≤ r < ∞ , 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π . (8)
The position vector r \mathbf{r} r may be written as
r = r e r . (9) \mathbf{r}=r\mathbf{e}_r. \tag*{(9)} r = r e r . (9)
The covariant polar basis vectors e r , e ϑ , e φ \mathbf{e}_r,\mathbf{e}_\vartheta,\mathbf{e}_\varphi e r , e ϑ , e φ are shown in Figure 1.2 . They form a real orthonormal basis
e α e β = δ α β , ( α , β = r , ϑ , φ ) , (10) \mathbf{e}_\alpha\mathbf{e}_\beta=\delta_{\alpha\beta},\quad(\alpha,\beta=r,\vartheta,\varphi), \tag*{(10)} e α e β = δ α β , ( α , β = r , ϑ , φ ) , (10)
e α ∗ = e α , ( α = r , ϑ , φ ) . (11) \mathbf{e}_\alpha^*=\mathbf{e}_\alpha,\quad(\alpha=r,\vartheta,\varphi). \tag*{(11)} e α ∗ = e α , ( α = r , ϑ , φ ) . (11)
The contravariant polar basis vectors e r , e ϑ , e φ \mathbf{e}^r,\mathbf{e}^\vartheta,\mathbf{e}^\varphi e r , e ϑ , e φ coincide with the covariant ones
e α = e α ( α = r , ϑ , φ ) . (12) \mathbf{e}^\alpha=\mathbf{e}_\alpha\quad(\alpha=r,\vartheta,\varphi). \tag*{(12)} e α = e α ( α = r , ϑ , φ ) . (12)
The unit vectors e r , e ϑ , e φ \mathbf{e}_r,\mathbf{e}_\vartheta,\mathbf{e}_\varphi e r , e ϑ , e φ form the right-handed basis
[ e r × e ϑ ] = e φ , [ e ϑ × e φ ] = e r , [ e φ × e r ] = e ϑ . (13) [\mathbf{e}_r \times\mathbf{e}_\vartheta]=\mathbf{e}_\varphi,\quad[\mathbf{e}_\vartheta\times\mathbf{e}_\varphi]=\mathbf{e}_r,\quad[\mathbf{e}_\varphi\times\mathbf{e}_r]=\mathbf{e}_\vartheta. \tag*{(13)} [ e r × e ϑ ] = e φ , [ e ϑ × e φ ] = e r , [ e φ × e r ] = e ϑ . (13)
The polar basis vectors e r \mathbf{e}_r e r , e ϑ \mathbf{e}_{\vartheta} e ϑ , e φ \mathbf{e}_{\varphi} e φ , contrary to the cartesian ones, depend on the angles ϑ \vartheta ϑ , φ \varphi φ . This should be taken into account when evaluating the derivatives
∂ ∂ r e r = 0 , ∂ ∂ r e ϑ = 0 , ∂ ∂ r e φ = 0 , ∂ ∂ ϑ e r = e ϑ , ∂ ∂ ϑ e ϑ = − e r , ∂ ∂ ϑ e φ = 0 , ∂ ∂ φ e r = e φ sin ϑ , ∂ ∂ φ e ϑ = e φ cos ϑ , ∂ ∂ φ e φ = − e r sin ϑ − e ϑ cos ϑ . (14) \begin{aligned}
\frac{\partial}{\partial r}\mathbf{e}_r &= 0, & \frac{\partial}{\partial r}\mathbf{e}_{\vartheta} &= 0, & \frac{\partial}{\partial r}\mathbf{e}_{\varphi} &= 0, \\
\frac{\partial}{\partial\vartheta}\mathbf{e}_r &= \mathbf{e}_{\vartheta}, & \frac{\partial}{\partial\vartheta}\mathbf{e}_{\vartheta} &= -\mathbf{e}_r, & \frac{\partial}{\partial\vartheta}\mathbf{e}_{\varphi} &= 0, \\
\frac{\partial}{\partial\varphi}\mathbf{e}_r &= \mathbf{e}_{\varphi}\sin\vartheta, & \frac{\partial}{\partial\varphi}\mathbf{e}_{\vartheta} &= \mathbf{e}_{\varphi}\cos\vartheta, & \frac{\partial}{\partial\varphi}\mathbf{e}_{\varphi} &= -\mathbf{e}_r\sin\vartheta-\mathbf{e}_{\vartheta}\cos\vartheta. \tag*{(14)}
\end{aligned} ∂ r ∂ e r ∂ ϑ ∂ e r ∂ φ ∂ e r = 0 , = e ϑ , = e φ sin ϑ , ∂ r ∂ e ϑ ∂ ϑ ∂ e ϑ ∂ φ ∂ e ϑ = 0 , = − e r , = e φ cos ϑ , ∂ r ∂ e φ ∂ ϑ ∂ e φ ∂ φ ∂ e φ = 0 , = 0 , = − e r sin ϑ − e ϑ cos ϑ . (14)
The results of applying the ∇ \nabla ∇ operator (see Sec. 1.3) to the polar basis vectors are presented in the form
( ∇ ⋅ e r ) = 2 r , ( ∇ ⋅ e ϑ ) = 1 r cot ϑ , ( ∇ ⋅ e φ ) = 0 , (15) (\nabla\cdot\mathbf{e}_r)=\frac{2}{r},\qquad(\nabla\cdot\mathbf{e}_{\vartheta})=\frac{1}{r}\cot\vartheta,\qquad(\nabla\cdot\mathbf{e}_{\varphi})=0, \tag*{(15)} ( ∇ ⋅ e r ) = r 2 , ( ∇ ⋅ e ϑ ) = r 1 cot ϑ , ( ∇ ⋅ e φ ) = 0 , (15)
[ ∇ × e r ] = 0 , [ ∇ × e ϑ ] = 1 r e φ , [ ∇ × e φ ] = 1 r cot ϑ e r − 1 r e ϑ . (16) [\nabla\times\mathbf{e}_r]=0,\qquad[\nabla\times\mathbf{e}_{\vartheta}]=\frac{1}{r}\mathbf{e}_{\varphi},\qquad[\nabla\times\mathbf{e}_{\varphi}]=\frac{1}{r}\cot\vartheta\mathbf{e}_r-\frac{1}{r}\mathbf{e}_{\vartheta}. \tag*{(16)} [ ∇ × e r ] = 0 , [ ∇ × e ϑ ] = r 1 e φ , [ ∇ × e φ ] = r 1 cot ϑ e r − r 1 e ϑ . (16)
Spherical Coordinate System
Spherical coordinates are widely used in the angular momentum theory.
The covariant spherical coordinates x μ x_{\mu} x μ (with μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ) are defined by the relations
x + 1 = − 1 2 ( x + i y ) = − 1 2 r sin ϑ e i φ , x 0 = z = r cos ϑ , x − 1 = 1 2 ( x − i y ) = 1 2 r sin ϑ e − i φ . (17) \begin{aligned}
x_{+1} &= -\frac{1}{\sqrt{2}}(x+iy)=-\frac{1}{\sqrt{2}}r\sin\vartheta e^{i\varphi},\\
x_0 &= z=r\cos\vartheta,\\
x_{-1} &= \frac{1}{\sqrt{2}}(x-iy)=\frac{1}{\sqrt{2}}r\sin\vartheta e^{-i\varphi}. \tag*{(17)}
\end{aligned} x + 1 x 0 x − 1 = − 2 1 ( x + i y ) = − 2 1 r sin ϑ e i φ , = z = r cos ϑ , = 2 1 ( x − i y ) = 2 1 r sin ϑ e − i φ . (17)
The contravariant spherical coordinates x μ x^{\mu} x μ (where μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ) are given by
x + 1 = − 1 2 ( x − i y ) = − 1 2 r sin ϑ e − i φ , x 0 = z = r cos ϑ , x − 1 = 1 2 ( x + i y ) = 1 2 r sin ϑ e i φ . (18) \begin{aligned}
x^{+1} &= -\frac{1}{\sqrt{2}}(x-iy)=-\frac{1}{\sqrt{2}}r\sin\vartheta e^{-i\varphi},\\
x^0 &= z=r\cos\vartheta,\\
x^{-1} &= \frac{1}{\sqrt{2}}(x+iy)=\frac{1}{\sqrt{2}}r\sin\vartheta e^{i\varphi}. \tag*{(18)}
\end{aligned} x + 1 x 0 x − 1 = − 2 1 ( x − i y ) = − 2 1 r sin ϑ e − i φ , = z = r cos ϑ , = 2 1 ( x + i y ) = 2 1 r sin ϑ e i φ . (18)
The relations between covariant and contravariant spherical coordinates are as follows
x μ = ( − 1 ) μ x − μ , x μ = ( − 1 ) μ x − μ , x μ = x μ ∗ , x μ = x μ ∗ , ( μ = ± 1 , 0 ) . (19) \begin{aligned}
x^{\mu} &= (-1)^{\mu}x_{-\mu}, & x_{\mu} &= (-1)^{\mu}x^{-\mu},\\
x^{\mu} &= x_{\mu}^{*}, & x_{\mu} &= x^{\mu*}, \qquad(\mu=\pm1,0). \tag*{(19)}
\end{aligned} x μ x μ = ( − 1 ) μ x − μ , = x μ ∗ , x μ x μ = ( − 1 ) μ x − μ , = x μ ∗ , ( μ = ± 1 , 0 ) . (19)
The covariant spherical basis vectors e μ ( μ = ± 1 , 0 ) \mathbf{e}_{\mu}(\mu=\pm1,0) e μ ( μ = ± 1 , 0 ) are defined by
e + 1 = − 1 2 ( e x + i e y ) , e 0 = e z , e − 1 = 1 2 ( e x − i e y ) . (20) \begin{aligned}
\mathbf{e}_{+1} &= -\frac{1}{\sqrt{2}}(\mathbf{e}_x+i\mathbf{e}_y),\\
\mathbf{e}_0 &= \mathbf{e}_z,\\
\mathbf{e}_{-1} &= \frac{1}{\sqrt{2}}(\mathbf{e}_x-i\mathbf{e}_y). \tag*{(20)}
\end{aligned} e + 1 e 0 e − 1 = − 2 1 ( e x + i e y ) , = e z , = 2 1 ( e x − i e y ) . (20)
The contravariant spherical basis vectors e μ ( μ = ± 1 , 0 ) \mathbf{e}^{\mu}(\mu=\pm1,0) e μ ( μ = ± 1 , 0 ) are given by
e + 1 = − 1 2 ( e x − i e y ) , e 0 = e z , e − 1 = 1 2 ( e x + i e y ) . (21) \begin{aligned}
\mathbf{e}^{+1} &= -\frac{1}{\sqrt{2}}(\mathbf{e}_x-i\mathbf{e}_y),\\
\mathbf{e}^{0} &= \mathbf{e}_z,\\
\mathbf{e}^{-1} &= \frac{1}{\sqrt{2}}(\mathbf{e}_x+i\mathbf{e}_y). \tag*{(21)}
\end{aligned} e + 1 e 0 e − 1 = − 2 1 ( e x − i e y ) , = e z , = 2 1 ( e x + i e y ) . (21)
Relations between the covariant and contravariant spherical basis vectors read
e μ = ( − 1 ) μ e − μ , e μ = ( − 1 ) μ e − μ , e μ = e μ ∗ , e μ = e μ ∗ , ( μ = ± 1 , 0 ) . (22) \begin{aligned}
\mathbf{e}^{\mu} &= (-1)^{\mu}\mathbf{e}_{-\mu}, & \mathbf{e}_{\mu} &= (-1)^{\mu}\mathbf{e}^{-\mu},\\
\mathbf{e}^{\mu} &= \mathbf{e}_{\mu}^{*}, & \mathbf{e}_{\mu} &= \mathbf{e}^{\mu*}, \qquad(\mu=\pm1,0). \tag*{(22)}
\end{aligned} e μ e μ = ( − 1 ) μ e − μ , = e μ ∗ , e μ e μ = ( − 1 ) μ e − μ , = e μ ∗ , ( μ = ± 1 , 0 ) . (22)
The spherical basis vectors form a complex orthonormal basis
e μ e ν ∗ = e μ e ν ∗ = δ μ ν , ( μ , ν = ± 1 , 0 ) . (23) \mathbf{e}_{\mu}\mathbf{e}^{\nu*}=\mathbf{e}_{\mu}\mathbf{e}_{\nu}^{*}=\delta_{\mu\nu},\qquad(\mu,\nu=\pm1,0). \tag*{(23)} e μ e ν ∗ = e μ e ν ∗ = δ μν , ( μ , ν = ± 1 , 0 ) . (23)
Vector products of spherical basis vectors may be written with the use of the Clebsch-Gordan coefficients (see Chap. 8) in the form
e μ × e ν = i 2 C 1 μ 1 ν 1 λ e λ , \mathbf{e}_{\mu} \times\mathbf{e}_{\nu}=i\sqrt{2}C^{1\lambda}_{1\mu1\nu}\mathbf{e}_{\lambda}, e μ × e ν = i 2 C 1 μ 1 ν 1 λ e λ ,
e μ × e ν = − i 2 C 1 μ 1 ν 1 λ e λ , ( μ , ν , λ = ± 1 , 0 ) . (24) \mathbf{e}^{\mu} \times\mathbf{e}^{\nu}=-i\sqrt{2}C^{1\lambda}_{1\mu1\nu}\mathbf{e}^{\lambda}, \qquad(\mu,\nu,\lambda=\pm1,0). \tag*{(24)} e μ × e ν = − i 2 C 1 μ 1 ν 1 λ e λ , ( μ , ν , λ = ± 1 , 0 ) . (24)
One may also rewrite these formulas in a form similar to (1.1.5)
e μ × e ν = − i ε μ ν λ e λ , ( μ , ν , λ = ± 1 , 0 ) \mathbf{e}_{\mu} \times\mathbf{e}_{\nu}=-i\varepsilon_{\mu\nu\lambda}\mathbf{e}^{\lambda}, \qquad(\mu,\nu,\lambda=\pm1,0) e μ × e ν = − i ε μν λ e λ , ( μ , ν , λ = ± 1 , 0 )
e μ × e ν = i ε μ ν λ e λ , ( μ , ν , λ = ± 1 , 0 ) . (25) \mathbf{e}^{\mu} \times\mathbf{e}^{\nu}=i\varepsilon_{\mu\nu\lambda}\mathbf{e}_{\lambda}, \qquad(\mu,\nu,\lambda=\pm1,0). \tag*{(25)} e μ × e ν = i ε μν λ e λ , ( μ , ν , λ = ± 1 , 0 ) . (25)
where ε μ ν λ = + 1 \varepsilon_{\mu\nu\lambda}=+1 ε μν λ = + 1 if the combination of indices μ , ν , λ \mu,\nu,\lambda μ , ν , λ is obtained by an even permutation of + 1 , 0 , − 1 +1,0,-1 + 1 , 0 , − 1 , ε μ ν λ = − 1 \varepsilon_{\mu\nu\lambda}=-1 ε μν λ = − 1 for an odd permutation, and ε μ ν λ = 0 \varepsilon_{\mu\nu\lambda}=0 ε μν λ = 0 if at least two indices among μ , ν , λ \mu,\nu,\lambda μ , ν , λ are equal.
A detailed form of (1.1.25) is as follows:
e + 1 × e 0 = i e + 1 , e 0 × e − 1 = i e − 1 , e 0 × e + 1 = i e + 1 = i e 0 + , e − 1 × e 0 = i e − 1 , e − 1 × e + 1 = i e 0 . (26) \begin{aligned}
\mathbf{e}_{+1}\times\mathbf{e}_{0}&=i\mathbf{e}_{+1}, & \mathbf{e}_{0}\times\mathbf{e}_{-1}&=i\mathbf{e}_{-1},\\
\mathbf{e}^{0}\times\mathbf{e}^{+1}&=i\mathbf{e}^{+1}=i\mathbf{e}_{0}^{+}, & \mathbf{e}^{-1}\times\mathbf{e}^{0}&=i\mathbf{e}^{-1},\\
\mathbf{e}^{-1}\times\mathbf{e}^{+1}&=i\mathbf{e}^{0}. \tag*{(26)}
\end{aligned} e + 1 × e 0 e 0 × e + 1 e − 1 × e + 1 = i e + 1 , = i e + 1 = i e 0 + , = i e 0 . e 0 × e − 1 e − 1 × e 0 = i e − 1 , = i e − 1 , (26)
Covariant and contravariant spherical components (see Sec. 1.2) of the basis vectors e μ \mathbf{e}_{\mu} e μ and e μ \mathbf{e}^{\mu} e μ are given by
[ e μ ] σ = ( − 1 ) σ δ σ − μ , [ e μ ] σ = δ σ μ , [ e μ ] σ = δ σ μ , [ e μ ] σ = ( − 1 ) σ δ σ − μ . (27) \begin{aligned}
[\mathbf{e}_{\mu}]_{\sigma}&=(-1)^{\sigma}\delta_{\sigma-\mu}, & [\mathbf{e}_{\mu}]^{\sigma}&=\delta_{\sigma\mu},\\
[\mathbf{e}^{\mu}]_{\sigma}&=\delta_{\sigma\mu}, & [\mathbf{e}^{\mu}]^{\sigma}&=(-1)^{\sigma}\delta_{\sigma-\mu}. \tag*{(27)}
\end{aligned} [ e μ ] σ [ e μ ] σ = ( − 1 ) σ δ σ − μ , = δ σ μ , [ e μ ] σ [ e μ ] σ = δ σ μ , = ( − 1 ) σ δ σ − μ . (27)
Helicity Basis Vectors
By analogy with Eqs. (1.1.20), (1.1.21) one may construct the following combinations of the polar basis vectors e r , e ϑ , e φ \mathbf{e}_{r},\mathbf{e}_{\vartheta},\mathbf{e}_{\varphi} e r , e ϑ , e φ
e + 1 ′ = − 1 2 ( e ϑ + i e φ ) , e + 1 = − 1 2 ( e ϑ − i e φ ) , e 0 ′ = e r , e ′ 0 = e r , e − 1 ′ = 1 2 ( e ϑ − i e φ ) , e ′ − 1 = 1 2 ( e ϑ + i e φ ) . (28) \begin{aligned}
\mathbf{e}'_{+1}&=-\frac{1}{\sqrt{2}}(\mathbf{e}_{\vartheta}+i\mathbf{e}_{\varphi}), &
\mathbf{e}^{+1}&=-\frac{1}{\sqrt{2}}(\mathbf{e}_{\vartheta}-i\mathbf{e}_{\varphi}),\\
\mathbf{e}'_{0}&=\mathbf{e}_{r}, &
\mathbf{e}'^{0}&=\mathbf{e}_{r},\\
\mathbf{e}'_{-1}&=\frac{1}{\sqrt{2}}(\mathbf{e}_{\vartheta}-i\mathbf{e}_{\varphi}), &
\mathbf{e}'^{-1}&=\frac{1}{\sqrt{2}}(\mathbf{e}_{\vartheta}+i\mathbf{e}_{\varphi}). \tag*{(28)}
\end{aligned} e + 1 ′ e 0 ′ e − 1 ′ = − 2 1 ( e ϑ + i e φ ) , = e r , = 2 1 ( e ϑ − i e φ ) , e + 1 e ′ 0 e ′ − 1 = − 2 1 ( e ϑ − i e φ ) , = e r , = 2 1 ( e ϑ + i e φ ) . (28)
The vectors e μ ′ \mathbf{e}'_{\mu} e μ ′ (μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ) are called the covariant helicity basis vectors and e ′ μ \mathbf{e}'^{\mu} e ′ μ (μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ) are called the contravariant helicity basis vectors (the explanation of the term “helicity” is given below in Sec. 6.3.6).
The helicity basis vectors e μ ′ \mathbf{e}'_{\mu} e μ ′ and e ′ μ \mathbf{e}'^{\mu} e ′ μ satisfy the same relations (1.1.22)–(1.1.26) as the spherical basis vectors e μ \mathbf{e}_{\mu} e μ and e μ \mathbf{e}^{\mu} e μ .
Relations Between Different Basis Vectors
Cartesian and Polar Basis Vectors
e x = e r sin ϑ cos φ + e ϑ cos ϑ cos φ − e φ sin φ , e y = e r sin ϑ sin φ + e ϑ cos ϑ sin φ + e φ cos φ , e z = e r cos ϑ − e ϑ sin ϑ . (29) \begin{aligned}
\mathbf{e}_{x}&=\mathbf{e}_{r}\sin\vartheta\cos\varphi+\mathbf{e}_{\vartheta}\cos\vartheta\cos\varphi-\mathbf{e}_{\varphi}\sin\varphi,\\
\mathbf{e}_{y}&=\mathbf{e}_{r}\sin\vartheta\sin\varphi+\mathbf{e}_{\vartheta}\cos\vartheta\sin\varphi+\mathbf{e}_{\varphi}\cos\varphi,\\
\mathbf{e}_{z}&=\mathbf{e}_{r}\cos\vartheta-\mathbf{e}_{\vartheta}\sin\vartheta. \tag*{(29)}
\end{aligned} e x e y e z = e r sin ϑ cos φ + e ϑ cos ϑ cos φ − e φ sin φ , = e r sin ϑ sin φ + e ϑ cos ϑ sin φ + e φ cos φ , = e r cos ϑ − e ϑ sin ϑ . (29)
Cartesian and Spherical Basis Vectors
e x = 1 2 ( e − 1 − e + 1 ) = 1 2 ( e − 1 − e + 1 ) , e y = i 2 ( e − 1 + e + 1 ) = − i 2 ( e − 1 + e + 1 ) , e z = e 0 = e 0 . (30) \begin{aligned}
\mathbf{e}_{x}&=\frac{1}{\sqrt{2}}(\mathbf{e}_{-1}-\mathbf{e}_{+1})=\frac{1}{\sqrt{2}}(\mathbf{e}^{-1}-\mathbf{e}^{+1}),\\
\mathbf{e}_{y}&=\frac{i}{\sqrt{2}}(\mathbf{e}_{-1}+\mathbf{e}_{+1})=-\frac{i}{\sqrt{2}}(\mathbf{e}^{-1}+\mathbf{e}^{+1}),\\
\mathbf{e}_{z}&=\mathbf{e}_{0}=\mathbf{e}^{0}. \tag*{(30)}
\end{aligned} e x e y e z = 2 1 ( e − 1 − e + 1 ) = 2 1 ( e − 1 − e + 1 ) , = 2 i ( e − 1 + e + 1 ) = − 2 i ( e − 1 + e + 1 ) , = e 0 = e 0 . (30)
Cartesian and Helicity Covariant Basis Vectors
e x = − e + 1 ′ 1 2 ( cos ϑ cos φ + i sin φ ) + e 0 ′ sin ϑ cos φ + e − 1 ′ 1 2 ( cos ϑ cos φ − i sin φ ) , e y = − e + 1 ′ 1 2 ( cos ϑ sin φ − i cos φ ) + e 0 ′ sin ϑ sin φ + e − 1 ′ 1 2 ( cos ϑ sin φ + i cos φ ) , e x = e + 1 ′ 1 2 sin ϑ + e 0 ′ cos ϑ − e − 1 ′ 1 2 sin ϑ . (31) \begin{aligned}
\mathbf{e}_{x}&=-\mathbf{e}'_{+1}\frac{1}{\sqrt{2}}(\cos\vartheta\cos\varphi+i\sin\varphi)+\mathbf{e}'_{0}\sin\vartheta\cos\varphi+\mathbf{e}'_{-1}\frac{1}{\sqrt{2}}(\cos\vartheta\cos\varphi-i\sin\varphi),\\
\mathbf{e}_{y}&=-\mathbf{e}'_{+1}\frac{1}{\sqrt{2}}(\cos\vartheta\sin\varphi-i\cos\varphi)+\mathbf{e}'_{0}\sin\vartheta\sin\varphi+\mathbf{e}'_{-1}\frac{1}{\sqrt{2}}(\cos\vartheta\sin\varphi+i\cos\varphi),\\
\mathbf{e}_{x}&=\mathbf{e}'_{+1}\frac{1}{\sqrt{2}}\sin\vartheta+\mathbf{e}'_{0}\cos\vartheta-\mathbf{e}'_{-1}\frac{1}{\sqrt{2}}\sin\vartheta. \tag*{(31)}
\end{aligned} e x e y e x = − e + 1 ′ 2 1 ( cos ϑ cos φ + i sin φ ) + e 0 ′ sin ϑ cos φ + e − 1 ′ 2 1 ( cos ϑ cos φ − i sin φ ) , = − e + 1 ′ 2 1 ( cos ϑ sin φ − i cos φ ) + e 0 ′ sin ϑ sin φ + e − 1 ′ 2 1 ( cos ϑ sin φ + i cos φ ) , = e + 1 ′ 2 1 sin ϑ + e 0 ′ cos ϑ − e − 1 ′ 2 1 sin ϑ . (31)
Cartesian and Helicity Contravariant Basis Vectors
e x = − e ′ + 1 1 2 ( cos ϑ cos φ − i sin φ ) + e ′ 0 sin ϑ cos φ + e ′ − 1 1 2 ( cos ϑ cos φ + i sin φ ) , e y = − e ′ + 1 1 2 ( cos ϑ sin φ + i cos φ ) + e ′ 0 sin ϑ sin φ + e ′ − 1 1 2 ( cos ϑ sin φ − i cos φ ) , e z = e ′ + 1 1 2 sin ϑ + e ′ 0 cos ϑ − e ′ − 1 1 2 sin ϑ . (32) \begin{aligned}
\mathbf{e}_x &= -\mathbf{e}^{\prime+1}\frac{1}{\sqrt{2}}\left(\cos\vartheta\cos\varphi-i\sin\varphi\right)+\mathbf{e}^{\prime0}\sin\vartheta\cos\varphi+\mathbf{e}^{\prime-1}\frac{1}{\sqrt{2}}\left(\cos\vartheta\cos\varphi+i\sin\varphi\right),\\
\mathbf{e}_y &= -\mathbf{e}^{\prime+1}\frac{1}{\sqrt{2}}\left(\cos\vartheta\sin\varphi+i\cos\varphi\right)+\mathbf{e}^{\prime0}\sin\vartheta\sin\varphi+\mathbf{e}^{\prime-1}\frac{1}{\sqrt{2}}\left(\cos\vartheta\sin\varphi-i\cos\varphi\right),\\
\mathbf{e}_z &= \mathbf{e}^{\prime+1}\frac{1}{\sqrt{2}}\sin\vartheta+\mathbf{e}^{\prime0}\cos\vartheta-\mathbf{e}^{\prime-1}\frac{1}{\sqrt{2}}\sin\vartheta. \tag*{(32)}
\end{aligned} e x e y e z = − e ′ + 1 2 1 ( cos ϑ cos φ − i sin φ ) + e ′0 sin ϑ cos φ + e ′ − 1 2 1 ( cos ϑ cos φ + i sin φ ) , = − e ′ + 1 2 1 ( cos ϑ sin φ + i cos φ ) + e ′0 sin ϑ sin φ + e ′ − 1 2 1 ( cos ϑ sin φ − i cos φ ) , = e ′ + 1 2 1 sin ϑ + e ′0 cos ϑ − e ′ − 1 2 1 sin ϑ . (32)
Polar and Cartesian Basis Vectors
e r = e x sin ϑ cos φ + e y sin ϑ sin φ + e z cos ϑ , e ϑ = e x cos ϑ cos φ + e y cos ϑ sin φ − e z sin ϑ , e φ = − e x sin φ + e y cos φ . (33) \begin{aligned}
\mathbf{e}_r &= \mathbf{e}_x\sin\vartheta\cos\varphi+\mathbf{e}_y\sin\vartheta\sin\varphi+\mathbf{e}_z\cos\vartheta,\\
\mathbf{e}_\vartheta&= \mathbf{e}_x\cos\vartheta\cos\varphi+\mathbf{e}_y\cos\vartheta\sin\varphi-\mathbf{e}_z\sin\vartheta,\\
\mathbf{e}_\varphi&= -\mathbf{e}_x\sin\varphi+\mathbf{e}_y\cos\varphi. \tag*{(33)}
\end{aligned} e r e ϑ e φ = e x sin ϑ cos φ + e y sin ϑ sin φ + e z cos ϑ , = e x cos ϑ cos φ + e y cos ϑ sin φ − e z sin ϑ , = − e x sin φ + e y cos φ . (33)
Polar and Spherical Covariant Basis Vectors
e r = − e + 1 1 2 sin ϑ e − i φ + e 0 cos ϑ + e − 1 1 2 sin ϑ e i φ , e ϑ = − e + 1 1 2 cos ϑ e − i φ − e 0 sin ϑ + e − 1 1 2 cos ϑ e i φ , e φ = e + 1 i 2 e − i φ + e − 1 i 2 e i φ . (34) \begin{aligned}
\mathbf{e}_r &= -\mathbf{e}_{+1}\frac{1}{\sqrt{2}}\sin\vartheta e^{-i\varphi}+\mathbf{e}_0\cos\vartheta+\mathbf{e}_{-1}\frac{1}{\sqrt{2}}\sin\vartheta e^{i\varphi},\\
\mathbf{e}_\vartheta&= -\mathbf{e}_{+1}\frac{1}{\sqrt{2}}\cos\vartheta e^{-i\varphi}-\mathbf{e}_0\sin\vartheta+\mathbf{e}_{-1}\frac{1}{\sqrt{2}}\cos\vartheta e^{i\varphi},\\
\mathbf{e}_\varphi&= \mathbf{e}_{+1}\frac{i}{\sqrt{2}}e^{-i\varphi}+\mathbf{e}_{-1}\frac{i}{\sqrt{2}}e^{i\varphi}. \tag*{(34)}
\end{aligned} e r e ϑ e φ = − e + 1 2 1 sin ϑ e − i φ + e 0 cos ϑ + e − 1 2 1 sin ϑ e i φ , = − e + 1 2 1 cos ϑ e − i φ − e 0 sin ϑ + e − 1 2 1 cos ϑ e i φ , = e + 1 2 i e − i φ + e − 1 2 i e i φ . (34)
Polar and Spherical Contravariant Basis Vectors
e r = − e + 1 1 2 sin ϑ e i φ + e 0 cos ϑ + e − 1 1 2 sin ϑ e − i φ , e ϑ = − e + 1 1 2 cos ϑ e i φ − e 0 sin ϑ + e − 1 1 2 cos ϑ e − i φ , e φ = − e + 1 i 2 e i φ − e − 1 i 2 e − i φ . (35) \begin{aligned}
\mathbf{e}_r &= -\mathbf{e}^{+1}\frac{1}{\sqrt{2}}\sin\vartheta e^{i\varphi}+\mathbf{e}^{0}\cos\vartheta+\mathbf{e}^{-1}\frac{1}{\sqrt{2}}\sin\vartheta e^{-i\varphi},\\
\mathbf{e}_\vartheta&= -\mathbf{e}^{+1}\frac{1}{\sqrt{2}}\cos\vartheta e^{i\varphi}-\mathbf{e}^{0}\sin\vartheta+\mathbf{e}^{-1}\frac{1}{\sqrt{2}}\cos\vartheta e^{-i\varphi},\\
\mathbf{e}_\varphi&= -\mathbf{e}^{+1}\frac{i}{\sqrt{2}}e^{i\varphi}-\mathbf{e}^{-1}\frac{i}{\sqrt{2}}e^{-i\varphi}. \tag*{(35)}
\end{aligned} e r e ϑ e φ = − e + 1 2 1 sin ϑ e i φ + e 0 cos ϑ + e − 1 2 1 sin ϑ e − i φ , = − e + 1 2 1 cos ϑ e i φ − e 0 sin ϑ + e − 1 2 1 cos ϑ e − i φ , = − e + 1 2 i e i φ − e − 1 2 i e − i φ . (35)
Polar and Helicity Basis Vectors
e r = e 0 ′ = e 0 , e ϑ = 1 2 ( e − 1 ′ − e + 1 ′ ) = 1 2 ( e ′ − 1 − e ′ + 1 ) , e φ = i 2 ( e − 1 ′ + e + 1 ′ ) = − i 2 ( e ′ − 1 + e ′ + 1 ) . (36) \begin{aligned}
\mathbf{e}_r &= \mathbf{e}'_0=\mathbf{e}^{0},\\
\mathbf{e}_\vartheta&= \frac{1}{\sqrt{2}}\left(\mathbf{e}'_{-1}-\mathbf{e}'_{+1}\right)=\frac{1}{\sqrt{2}}\left(\mathbf{e}^{\prime-1}-\mathbf{e}^{\prime+1}\right),\\
\mathbf{e}_\varphi&= \frac{i}{\sqrt{2}}\left(\mathbf{e}'_{-1}+\mathbf{e}'_{+1}\right)=-\frac{i}{\sqrt{2}}\left(\mathbf{e}^{\prime-1}+\mathbf{e}^{\prime+1}\right). \tag*{(36)}
\end{aligned} e r e ϑ e φ = e 0 ′ = e 0 , = 2 1 ( e − 1 ′ − e + 1 ′ ) = 2 1 ( e ′ − 1 − e ′ + 1 ) , = 2 i ( e − 1 ′ + e + 1 ′ ) = − 2 i ( e ′ − 1 + e ′ + 1 ) . (36)
Spherical and Cartesian Basis Vectors
e + 1 = − 1 2 ( e x + i e y ) , e + 1 = − 1 2 ( e x − i e y ) , e 0 = e z , e 0 = e z , e − 1 = 1 2 ( e x − i e y ) , e − 1 = 1 2 ( e x + i e y ) . (37) \begin{aligned}
\mathbf{e}_{+1} &= -\frac{1}{\sqrt{2}}\left(\mathbf{e}_x+i\mathbf{e}_y\right), & \mathbf{e}^{+1} &= -\frac{1}{\sqrt{2}}\left(\mathbf{e}_x-i\mathbf{e}_y\right),\\
\mathbf{e}_0 &= \mathbf{e}_z, & \mathbf{e}^{0} &= \mathbf{e}_z,\\
\mathbf{e}_{-1} &= \frac{1}{\sqrt{2}}\left(\mathbf{e}_x-i\mathbf{e}_y\right), & \mathbf{e}^{-1} &= \frac{1}{\sqrt{2}}\left(\mathbf{e}_x+i\mathbf{e}_y\right). \tag*{(37)}
\end{aligned} e + 1 e 0 e − 1 = − 2 1 ( e x + i e y ) , = e z , = 2 1 ( e x − i e y ) , e + 1 e 0 e − 1 = − 2 1 ( e x − i e y ) , = e z , = 2 1 ( e x + i e y ) . (37)
Spherical Covariant and Polar Basis Vectors
e + 1 = − e r 1 2 sin ϑ e i φ − e ϑ 1 2 cos ϑ e i φ − e φ i 2 e i φ , e 0 = e r cos ϑ − e ϑ sin ϑ , e − 1 = e r i 2 sin ϑ e − i φ + e ϑ 1 2 cos ϑ e − i φ − e φ i 2 e − i φ . (38) \begin{aligned}
\mathbf{e}_{+1} &= -\mathbf{e}_r\frac{1}{\sqrt{2}}\sin\vartheta e^{i\varphi}-\mathbf{e}_\vartheta\frac{1}{\sqrt{2}}\cos\vartheta e^{i\varphi}-\mathbf{e}_\varphi\frac{i}{\sqrt{2}}e^{i\varphi},\\
\mathbf{e}_0 &= \mathbf{e}_r\cos\vartheta-\mathbf{e}_\vartheta\sin\vartheta,\\
\mathbf{e}_{-1} &= \mathbf{e}_r\frac{i}{\sqrt{2}}\sin\vartheta e^{-i\varphi}+\mathbf{e}_\vartheta\frac{1}{\sqrt{2}}\cos\vartheta e^{-i\varphi}-\mathbf{e}_\varphi\frac{i}{\sqrt{2}}e^{-i\varphi}. \tag*{(38)}
\end{aligned} e + 1 e 0 e − 1 = − e r 2 1 sin ϑ e i φ − e ϑ 2 1 cos ϑ e i φ − e φ 2 i e i φ , = e r cos ϑ − e ϑ sin ϑ , = e r 2 i sin ϑ e − i φ + e ϑ 2 1 cos ϑ e − i φ − e φ 2 i e − i φ . (38)
Spherical Contravariant and Polar Basis Vectors
e + 1 = − e r 1 2 sin ϑ e − i φ − e ϑ 1 2 cos ϑ e − i φ + e φ i 2 e − i φ , e 0 = e r cos ϑ − e ϑ sin ϑ , e − 1 = e r 1 2 sin ϑ e i φ + e ϑ 1 2 cos ϑ e i φ + e φ i 2 e i φ . (39) \begin{aligned}
\mathbf{e}^{+1} &= -\mathbf{e}_r\frac{1}{\sqrt{2}}\sin\vartheta e^{-i\varphi}-\mathbf{e}_\vartheta\frac{1}{\sqrt{2}}\cos\vartheta e^{-i\varphi}+\mathbf{e}_\varphi\frac{i}{\sqrt{2}}e^{-i\varphi},\\
\mathbf{e}^{0} &= \mathbf{e}_r\cos\vartheta-\mathbf{e}_\vartheta\sin\vartheta,\\
\mathbf{e}^{-1} &= \mathbf{e}_r\frac{1}{\sqrt{2}}\sin\vartheta e^{i\varphi}+\mathbf{e}_\vartheta\frac{1}{\sqrt{2}}\cos\vartheta e^{i\varphi}+\mathbf{e}_\varphi\frac{i}{\sqrt{2}}e^{i\varphi}. \tag*{(39)}
\end{aligned} e + 1 e 0 e − 1 = − e r 2 1 sin ϑ e − i φ − e ϑ 2 1 cos ϑ e − i φ + e φ 2 i e − i φ , = e r cos ϑ − e ϑ sin ϑ , = e r 2 1 sin ϑ e i φ + e ϑ 2 1 cos ϑ e i φ + e φ 2 i e i φ . (39)
Spherical Covariant and Spherical Contravariant Basis Vectors
e + 1 = − e − 1 , e + 1 = − e − 1 , e 0 = e 0 , e 0 = e 0 , e − 1 = − e + 1 , e − 1 = − e + 1 . (40) \begin{aligned}
\mathbf{e}_{+1}&=-\mathbf{e}^{-1}, & \mathbf{e}^{+1}&=-\mathbf{e}_{-1},\\
\mathbf{e}_{0}&=\mathbf{e}^{0}, & \mathbf{e}^{0}&=\mathbf{e}_{0},\\
\mathbf{e}_{-1}&=-\mathbf{e}^{+1}, & \mathbf{e}^{-1}&=-\mathbf{e}_{+1}. \tag*{(40)}
\end{aligned} e + 1 e 0 e − 1 = − e − 1 , = e 0 , = − e + 1 , e + 1 e 0 e − 1 = − e − 1 , = e 0 , = − e + 1 . (40)
Spherical Covariant and Helicity Covariant Basis Vectors
e + 1 = e + 1 ′ 1 + cos ϑ 2 e i φ − e 0 ′ sin ϑ 2 e i φ + e − 1 ′ 1 − cos ϑ 2 e i φ , e 0 = e + 1 ′ sin ϑ 2 + e 0 ′ cos ϑ − e − 1 ′ sin ϑ 2 , e − 1 = e + 1 ′ 1 − cos ϑ 2 e − i φ + e 0 ′ sin ϑ 2 e − i φ + e − 1 ′ 1 + cos ϑ 2 e − i φ . (41) \begin{aligned}
\mathbf{e}_{+1}&=\mathbf{e}'_{+1}\frac{1+\cos\vartheta}{2}e^{i\varphi}-\mathbf{e}'_{0}\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}+\mathbf{e}'_{-1}\frac{1-\cos\vartheta}{2}e^{i\varphi},\\
\mathbf{e}_{0}&=\mathbf{e}'_{+1}\frac{\sin\vartheta}{\sqrt{2}}+\mathbf{e}'_{0}\cos\vartheta-\mathbf{e}'_{-1}\frac{\sin\vartheta}{\sqrt{2}},\\
\mathbf{e}_{-1}&=\mathbf{e}'_{+1}\frac{1-\cos\vartheta}{2}e^{-i\varphi}+\mathbf{e}'_{0}\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}+\mathbf{e}'_{-1}\frac{1+\cos\vartheta}{2}e^{-i\varphi}. \tag*{(41)}
\end{aligned} e + 1 e 0 e − 1 = e + 1 ′ 2 1 + cos ϑ e i φ − e 0 ′ 2 sin ϑ e i φ + e − 1 ′ 2 1 − cos ϑ e i φ , = e + 1 ′ 2 sin ϑ + e 0 ′ cos ϑ − e − 1 ′ 2 sin ϑ , = e + 1 ′ 2 1 − cos ϑ e − i φ + e 0 ′ 2 sin ϑ e − i φ + e − 1 ′ 2 1 + cos ϑ e − i φ . (41)
Spherical Contravariant and Helicity Covariant Basis Vectors
e + 1 = − e + 1 ′ 1 − cos ϑ 2 e − i φ − e 0 ′ sin ϑ 2 e − i φ − e − 1 ′ 1 + cos ϑ 2 e − i φ , e 0 = e + 1 ′ sin ϑ 2 + e 0 ′ cos ϑ − e − 1 ′ sin ϑ 2 , e − 1 = − e + 1 ′ 1 + cos ϑ 2 e i φ + e 0 ′ sin ϑ 2 e i φ − e − 1 ′ 1 − cos ϑ 2 e i φ . (42) \begin{aligned}
\mathbf{e}^{+1}&=-\mathbf{e}'_{+1}\frac{1-\cos\vartheta}{2}e^{-i\varphi}-\mathbf{e}'_{0}\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}-\mathbf{e}'_{-1}\frac{1+\cos\vartheta}{2}e^{-i\varphi},\\
\mathbf{e}^{0}&=\mathbf{e}'_{+1}\frac{\sin\vartheta}{\sqrt{2}}+\mathbf{e}'_{0}\cos\vartheta-\mathbf{e}'_{-1}\frac{\sin\vartheta}{\sqrt{2}},\\
\mathbf{e}^{-1}&=-\mathbf{e}'_{+1}\frac{1+\cos\vartheta}{2}e^{i\varphi}+\mathbf{e}'_{0}\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}-\mathbf{e}'_{-1}\frac{1-\cos\vartheta}{2}e^{i\varphi}. \tag*{(42)}
\end{aligned} e + 1 e 0 e − 1 = − e + 1 ′ 2 1 − cos ϑ e − i φ − e 0 ′ 2 sin ϑ e − i φ − e − 1 ′ 2 1 + cos ϑ e − i φ , = e + 1 ′ 2 sin ϑ + e 0 ′ cos ϑ − e − 1 ′ 2 sin ϑ , = − e + 1 ′ 2 1 + cos ϑ e i φ + e 0 ′ 2 sin ϑ e i φ − e − 1 ′ 2 1 − cos ϑ e i φ . (42)
Spherical Covariant and Helicity Contravariant Basis Vectors
e + 1 = − e ′ + 1 1 − cos ϑ 2 e i φ − e ′ 0 sin ϑ 2 e i φ − e ′ − 1 1 + cos ϑ 2 e i φ , e 0 = e ′ + 1 sin ϑ 2 + e ′ 0 cos ϑ − e ′ − 1 sin ϑ 2 , e − 1 = − e ′ + 1 1 + cos ϑ 2 e − i φ + e ′ 0 sin ϑ 2 e − i φ − e ′ − 1 1 − cos ϑ 2 e − i φ . (43) \begin{aligned}
\mathbf{e}_{+1}&=-\mathbf{e}'^{+1}\frac{1-\cos\vartheta}{2}e^{i\varphi}-\mathbf{e}'^{0}\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}-\mathbf{e}'^{-1}\frac{1+\cos\vartheta}{2}e^{i\varphi},\\
\mathbf{e}_{0}&=\mathbf{e}'^{+1}\frac{\sin\vartheta}{\sqrt{2}}+\mathbf{e}'^{0}\cos\vartheta-\mathbf{e}'^{-1}\frac{\sin\vartheta}{\sqrt{2}},\\
\mathbf{e}_{-1}&=-\mathbf{e}'^{+1}\frac{1+\cos\vartheta}{2}e^{-i\varphi}+\mathbf{e}'^{0}\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}-\mathbf{e}'^{-1}\frac{1-\cos\vartheta}{2}e^{-i\varphi}. \tag*{(43)}
\end{aligned} e + 1 e 0 e − 1 = − e ′ + 1 2 1 − cos ϑ e i φ − e ′ 0 2 sin ϑ e i φ − e ′ − 1 2 1 + cos ϑ e i φ , = e ′ + 1 2 sin ϑ + e ′ 0 cos ϑ − e ′ − 1 2 sin ϑ , = − e ′ + 1 2 1 + cos ϑ e − i φ + e ′ 0 2 sin ϑ e − i φ − e ′ − 1 2 1 − cos ϑ e − i φ . (43)
Spherical Contravariant and Helicity Contravariant Basis Vectors
e + 1 = e ′ + 1 1 + cos ϑ 2 e − i φ − e ′ 0 sin ϑ 2 e − i φ + e ′ − 1 1 − cos ϑ 2 e − i φ , e 0 = e ′ + 1 sin ϑ 2 + e ′ 0 cos ϑ − e ′ − 1 sin ϑ 2 , e − 1 = e ′ + 1 1 − cos ϑ 2 e i φ + e ′ 0 sin ϑ 2 e i φ + e ′ − 1 1 + cos ϑ 2 e i φ . (44) \begin{aligned}
\mathbf{e}^{+1}&=\mathbf{e}'^{+1}\frac{1+\cos\vartheta}{2}e^{-i\varphi}-\mathbf{e}'^{0}\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}+\mathbf{e}'^{-1}\frac{1-\cos\vartheta}{2}e^{-i\varphi},\\
\mathbf{e}^{0}&=\mathbf{e}'^{+1}\frac{\sin\vartheta}{\sqrt{2}}+\mathbf{e}'^{0}\cos\vartheta-\mathbf{e}'^{-1}\frac{\sin\vartheta}{\sqrt{2}},\\
\mathbf{e}^{-1}&=\mathbf{e}'^{+1}\frac{1-\cos\vartheta}{2}e^{i\varphi}+\mathbf{e}'^{0}\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}+\mathbf{e}'^{-1}\frac{1+\cos\vartheta}{2}e^{i\varphi}. \tag*{(44)}
\end{aligned} e + 1 e 0 e − 1 = e ′ + 1 2 1 + cos ϑ e − i φ − e ′ 0 2 sin ϑ e − i φ + e ′ − 1 2 1 − cos ϑ e − i φ , = e ′ + 1 2 sin ϑ + e ′ 0 cos ϑ − e ′ − 1 2 sin ϑ , = e ′ + 1 2 1 − cos ϑ e i φ + e ′ 0 2 sin ϑ e i φ + e ′ − 1 2 1 + cos ϑ e i φ . (44)
Equations (1.1.41)–(1.1.44) may be written in a more compact form using the Wigner D D D -functions (see Chap. 4).
e μ = ∑ ν D − ν − μ 1 ( 0 , ϑ , φ ) e ν ′ = ∑ ν ( − 1 ) ν D ν − μ 1 ( 0 , ϑ , φ ) e ′ ν , e μ = ( − 1 ) μ ∑ ν D − ν μ 1 ( 0 , ϑ , φ ) e ν ′ = ∑ ν ( − 1 ) μ + ν D ν μ 1 ( 0 , ϑ , φ ) e ′ ν , (45) \begin{aligned}
\mathbf{e}_{\mu}&=\sum_{\nu}D^{1}_{-\nu-\mu}(0,\vartheta,\varphi)\mathbf{e}'_{\nu}=\sum_{\nu}(-1)^{\nu}D^{1}_{\nu-\mu}(0,\vartheta,\varphi)\mathbf{e}'^{\nu},\\
\mathbf{e}^{\mu}&=(-1)^{\mu}\sum_{\nu}D^{1}_{-\nu\mu}(0,\vartheta,\varphi)\mathbf{e}'_{\nu}=\sum_{\nu}(-1)^{\mu+\nu}D^{1}_{\nu\mu}(0,\vartheta,\varphi)\mathbf{e}'^{\nu}, \tag*{(45)}
\end{aligned} e μ e μ = ν ∑ D − ν − μ 1 ( 0 , ϑ , φ ) e ν ′ = ν ∑ ( − 1 ) ν D ν − μ 1 ( 0 , ϑ , φ ) e ′ ν , = ( − 1 ) μ ν ∑ D − ν μ 1 ( 0 , ϑ , φ ) e ν ′ = ν ∑ ( − 1 ) μ + ν D ν μ 1 ( 0 , ϑ , φ ) e ′ ν , (45)
( μ , ν = ± 1 , 0 ) (\mu,\nu=\pm1,0) ( μ , ν = ± 1 , 0 ) .
Helicity Covariant and Cartesian Basis Vectors
e + 1 ′ = − e x 1 2 ( cos ϑ cos φ − i sin φ ) − e y 1 2 ( cos ϑ sin φ + i cos φ ) + e z 1 2 sin ϑ , e 0 ′ = e x sin ϑ cos φ + e y sin ϑ sin φ + e z cos ϑ , e − 1 ′ = e x 1 2 ( cos ϑ cos φ + i sin φ ) + e y 1 2 ( cos ϑ sin φ − i cos φ ) − e z 1 2 sin ϑ . (46) \begin{aligned}
\mathbf{e}'_{+1}&=-\mathbf{e}_{x}\frac{1}{\sqrt{2}}(\cos\vartheta\cos\varphi-i\sin\varphi)-\mathbf{e}_{y}\frac{1}{\sqrt{2}}(\cos\vartheta\sin\varphi+i\cos\varphi)+\mathbf{e}_{z}\frac{1}{\sqrt{2}}\sin\vartheta,\\
\mathbf{e}'_{0}&=\mathbf{e}_{x}\sin\vartheta\cos\varphi+\mathbf{e}_{y}\sin\vartheta\sin\varphi+\mathbf{e}_{z}\cos\vartheta,\\
\mathbf{e}'_{-1}&=\mathbf{e}_{x}\frac{1}{\sqrt{2}}(\cos\vartheta\cos\varphi+i\sin\varphi)+\mathbf{e}_{y}\frac{1}{\sqrt{2}}(\cos\vartheta\sin\varphi-i\cos\varphi)-\mathbf{e}_{z}\frac{1}{\sqrt{2}}\sin\vartheta. \tag*{(46)}
\end{aligned} e + 1 ′ e 0 ′ e − 1 ′ = − e x 2 1 ( cos ϑ cos φ − i sin φ ) − e y 2 1 ( cos ϑ sin φ + i cos φ ) + e z 2 1 sin ϑ , = e x sin ϑ cos φ + e y sin ϑ sin φ + e z cos ϑ , = e x 2 1 ( cos ϑ cos φ + i sin φ ) + e y 2 1 ( cos ϑ sin φ − i cos φ ) − e z 2 1 sin ϑ . (46)
Vectors. Tensors
Helicity Contravariant and Cartesian Basis Vectors
e ′ + 1 = − e x 1 2 ( cos ϑ cos φ + i sin φ ) − e y 1 2 ( cos ϑ sin φ − i cos φ ) + e z 1 2 sin ϑ , \mathbf{e}^{\prime+1}=-\mathbf{e}_x\frac{1}{\sqrt{2}}(\cos\vartheta\cos\varphi+i\sin\varphi)-\mathbf{e}_y\frac{1}{\sqrt{2}}(\cos\vartheta\sin\varphi-i\cos\varphi)+\mathbf{e}_z\frac{1}{\sqrt{2}}\sin\vartheta, e ′ + 1 = − e x 2 1 ( cos ϑ cos φ + i sin φ ) − e y 2 1 ( cos ϑ sin φ − i cos φ ) + e z 2 1 sin ϑ ,
e ′ 0 = e x sin ϑ cos φ + e y sin ϑ sin φ + e z cos ϑ , \mathbf{e}^{\prime0}=\mathbf{e}_x\sin\vartheta\cos\varphi+\mathbf{e}_y\sin\vartheta\sin\varphi+\mathbf{e}_z\cos\vartheta, e ′0 = e x sin ϑ cos φ + e y sin ϑ sin φ + e z cos ϑ ,
e ′ − 1 = e x 1 2 ( cos ϑ cos φ − i sin φ ) + e y 1 2 ( cos ϑ sin φ + i cos φ ) − e z 1 2 sin ϑ . (47) \mathbf{e}^{\prime-1}=\mathbf{e}_x\frac{1}{\sqrt{2}}(\cos\vartheta\cos\varphi-i\sin\varphi)+\mathbf{e}_y\frac{1}{\sqrt{2}}(\cos\vartheta\sin\varphi+i\cos\varphi)-\mathbf{e}_z\frac{1}{\sqrt{2}}\sin\vartheta. \tag*{(47)} e ′ − 1 = e x 2 1 ( cos ϑ cos φ − i sin φ ) + e y 2 1 ( cos ϑ sin φ + i cos φ ) − e z 2 1 sin ϑ . (47)
Helicity and Polar Basis Vectors
e + 1 ′ = − 1 2 ( e ϑ + i e φ ) , e ′ + 1 = − 1 2 ( e ϑ − i e φ ) , \mathbf{e}^{\prime}_{+1}=-\frac{1}{\sqrt{2} }(\mathbf{e}_{\vartheta}+i\mathbf{e}_{\varphi}),\qquad\mathbf{e}^{\prime+1}=-\frac{1}{\sqrt{2}}(\mathbf{e}_{\vartheta}-i\mathbf{e}_{\varphi}), e + 1 ′ = − 2 1 ( e ϑ + i e φ ) , e ′ + 1 = − 2 1 ( e ϑ − i e φ ) ,
e 0 ′ = e r , e ′ 0 = e r , \mathbf{e}^{\prime}_{0}=\mathbf{e}_{r},\qquad\mathbf{e}^{\prime0}=\mathbf{e}_{r}, e 0 ′ = e r , e ′0 = e r ,
e − 1 ′ = 1 2 ( e ϑ − i e φ ) , e ′ − 1 = 1 2 ( e ϑ + i e φ ) . (48) \mathbf{e}^{\prime}_{-1}=\frac{1}{\sqrt{2}}(\mathbf{e}_{\vartheta}-i\mathbf{e}_{\varphi}),\qquad\mathbf{e}^{\prime-1}=\frac{1}{\sqrt{2}}(\mathbf{e}_{\vartheta}+i\mathbf{e}_{\varphi}). \tag*{(48)} e − 1 ′ = 2 1 ( e ϑ − i e φ ) , e ′ − 1 = 2 1 ( e ϑ + i e φ ) . (48)
Helicity Covariant and Spherical Covariant Basis Vectors
e + 1 ′ = e + 1 1 + cos ϑ 2 e − i φ + e 0 sin ϑ 2 + e − 1 1 − cos ϑ 2 e i φ , \mathbf{e}^{\prime}_{+1}=\mathbf{e}_{+1}\frac{1+\cos\vartheta}{2}e^{-i\varphi}+\mathbf{e}_{0}\frac{\sin\vartheta}{\sqrt{2}}+\mathbf{e}_{-1}\frac{1-\cos\vartheta}{2}e^{i\varphi}, e + 1 ′ = e + 1 2 1 + cos ϑ e − i φ + e 0 2 sin ϑ + e − 1 2 1 − cos ϑ e i φ ,
e 0 ′ = − e + 1 sin ϑ 2 e − i φ + e 0 cos ϑ + e − 1 sin ϑ 2 e i φ , \mathbf{e}^{\prime}_{0}=-\mathbf{e}_{+1}\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}+\mathbf{e}_{0}\cos\vartheta+\mathbf{e}_{-1}\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}, e 0 ′ = − e + 1 2 sin ϑ e − i φ + e 0 cos ϑ + e − 1 2 sin ϑ e i φ ,
e − 1 ′ = e + 1 1 − cos ϑ 2 e − i φ − e 0 sin ϑ 2 + e − 1 1 + cos ϑ 2 e i φ . (49) \mathbf{e}^{\prime}_{-1}=\mathbf{e}_{+1}\frac{1-\cos\vartheta}{2}e^{-i\varphi}-\mathbf{e}_{0}\frac{\sin\vartheta}{\sqrt{2}}+\mathbf{e}_{-1}\frac{1+\cos\vartheta}{2}e^{i\varphi}. \tag*{(49)} e − 1 ′ = e + 1 2 1 − cos ϑ e − i φ − e 0 2 sin ϑ + e − 1 2 1 + cos ϑ e i φ . (49)
Helicity Contravariant and Spherical Covariant Basis Vectors
e ′ + 1 = − e + 1 1 − cos ϑ 2 e − i φ + e 0 sin ϑ 2 − e − 1 1 + cos ϑ 2 e i φ , \mathbf{e}^{\prime+1}=-\mathbf{e}_{+1}\frac{1-\cos\vartheta}{2}e^{-i\varphi}+\mathbf{e}_{0}\frac{\sin\vartheta}{\sqrt{2}}-\mathbf{e}_{-1}\frac{1+\cos\vartheta}{2}e^{i\varphi}, e ′ + 1 = − e + 1 2 1 − cos ϑ e − i φ + e 0 2 sin ϑ − e − 1 2 1 + cos ϑ e i φ ,
e ′ 0 = − e + 1 sin ϑ 2 e − i φ + e 0 cos ϑ + e − 1 sin ϑ 2 e i φ , \mathbf{e}^{\prime0}=-\mathbf{e}_{+1}\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}+\mathbf{e}_{0}\cos\vartheta+\mathbf{e}_{-1}\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}, e ′0 = − e + 1 2 sin ϑ e − i φ + e 0 cos ϑ + e − 1 2 sin ϑ e i φ ,
e ′ − 1 = − e + 1 1 + cos ϑ 2 e − i φ − e 0 sin ϑ 2 − e − 1 1 − cos ϑ 2 e i φ . (50) \mathbf{e}^{\prime-1}=-\mathbf{e}_{+1}\frac{1+\cos\vartheta}{2}e^{-i\varphi}-\mathbf{e}_{0}\frac{\sin\vartheta}{\sqrt{2}}-\mathbf{e}_{-1}\frac{1-\cos\vartheta}{2}e^{i\varphi}. \tag*{(50)} e ′ − 1 = − e + 1 2 1 + cos ϑ e − i φ − e 0 2 sin ϑ − e − 1 2 1 − cos ϑ e i φ . (50)
Helicity Covariant and Spherical Contravariant Basis Vectors
e + 1 ′ = − e + 1 1 − cos ϑ 2 e i φ + e 0 sin ϑ 2 − e − 1 1 + cos ϑ 2 e − i φ , \mathbf{e}^{\prime}_{+1}=-\mathbf{e}^{+1}\frac{1-\cos\vartheta}{2}e^{i\varphi}+\mathbf{e}^{0}\frac{\sin\vartheta}{\sqrt{2}}-\mathbf{e}^{-1}\frac{1+\cos\vartheta}{2}e^{-i\varphi}, e + 1 ′ = − e + 1 2 1 − cos ϑ e i φ + e 0 2 sin ϑ − e − 1 2 1 + cos ϑ e − i φ ,
e 0 ′ = − e + 1 sin ϑ 2 e i φ + e 0 cos ϑ + e − 1 sin ϑ 2 e − i φ , \mathbf{e}^{\prime}_{0}=-\mathbf{e}^{+1}\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}+\mathbf{e}^{0}\cos\vartheta+\mathbf{e}^{-1}\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}, e 0 ′ = − e + 1 2 sin ϑ e i φ + e 0 cos ϑ + e − 1 2 sin ϑ e − i φ ,
e − 1 ′ = − e + 1 1 + cos ϑ 2 e i φ − e 0 sin ϑ 2 − e − 1 1 − cos ϑ 2 e − i φ . (51) \mathbf{e}^{\prime}_{-1}=-\mathbf{e}^{+1}\frac{1+\cos\vartheta}{2}e^{i\varphi}-\mathbf{e}^{0}\frac{\sin\vartheta}{\sqrt{2}}-\mathbf{e}^{-1}\frac{1-\cos\vartheta}{2}e^{-i\varphi}. \tag*{(51)} e − 1 ′ = − e + 1 2 1 + cos ϑ e i φ − e 0 2 sin ϑ − e − 1 2 1 − cos ϑ e − i φ . (51)
Helicity Contravariant and Spherical Contravariant Basis Vectors
e ′ + 1 = e + 1 1 + cos ϑ 2 e i φ + e 0 sin ϑ 2 + e − 1 1 − cos ϑ 2 e − i φ , \mathbf{e}^{\prime+1}=\mathbf{e}^{+1}\frac{1+\cos\vartheta}{2}e^{i\varphi}+\mathbf{e}^{0}\frac{\sin\vartheta}{\sqrt{2}}+\mathbf{e}^{-1}\frac{1-\cos\vartheta}{2}e^{-i\varphi}, e ′ + 1 = e + 1 2 1 + cos ϑ e i φ + e 0 2 sin ϑ + e − 1 2 1 − cos ϑ e − i φ ,
e ′ 0 = − e + 1 sin ϑ 2 e i φ + e 0 cos ϑ + e − 1 sin ϑ 2 e − i φ , \mathbf{e}^{\prime0}=-\mathbf{e}^{+1}\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}+\mathbf{e}^{0}\cos\vartheta+\mathbf{e}^{-1}\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}, e ′0 = − e + 1 2 sin ϑ e i φ + e 0 cos ϑ + e − 1 2 sin ϑ e − i φ ,
e ′ − 1 = e + 1 1 − cos ϑ 2 e i φ − e 0 sin ϑ 2 + e − 1 1 + cos ϑ 2 e − i φ . (52) \mathbf{e}^{\prime-1}=\mathbf{e}^{+1}\frac{1-\cos\vartheta}{2}e^{i\varphi}-\mathbf{e}^{0}\frac{\sin\vartheta}{\sqrt{2}}+\mathbf{e}^{-1}\frac{1+\cos\vartheta}{2}e^{-i\varphi}. \tag*{(52)} e ′ − 1 = e + 1 2 1 − cos ϑ e i φ − e 0 2 sin ϑ + e − 1 2 1 + cos ϑ e − i φ . (52)
Equations (1.1.49)–(1.1.52) may be written in a more compact form using the Wigner D D D -functions (see Chap. 4).
e μ ′ = ∑ ν D ν μ 1 ( φ , ϑ , 0 ) e ν = ∑ ν ( − 1 ) ν D − ν μ 1 ( φ , ϑ , 0 ) e ′ , \mathbf{e}^{\prime}_{\mu}=\sum_{\nu}D^{1}_{\nu\mu}(\varphi,\vartheta,0)\mathbf{e}_{\nu}=\sum_{\nu}(-1)^{\nu}D^{1}_{-\nu\mu}(\varphi,\vartheta,0)\mathbf{e}^{\prime}, e μ ′ = ν ∑ D ν μ 1 ( φ , ϑ , 0 ) e ν = ν ∑ ( − 1 ) ν D − ν μ 1 ( φ , ϑ , 0 ) e ′ ,
e ′ μ = ∑ ν ( − 1 ) μ D ν − μ 1 ( φ , ϑ , 0 ) e ν = ∑ ν ( − 1 ) μ + ν D − ν − μ 1 ( φ , ϑ , 0 ) e ′ . \mathbf{e}^{\prime\mu}=\sum_{\nu}(-1)^{\mu}D^{1}_{\nu-\mu}(\varphi,\vartheta,0)\mathbf{e}_{\nu}=\sum_{\nu}(-1)^{\mu+\nu}D^{1}_{-\nu-\mu}(\varphi,\vartheta,0)\mathbf{e}^{\prime}. e ′ μ = ν ∑ ( − 1 ) μ D ν − μ 1 ( φ , ϑ , 0 ) e ν = ν ∑ ( − 1 ) μ + ν D − ν − μ 1 ( φ , ϑ , 0 ) e ′ .
( μ , ν = ± 1 , 0 ) . (53) (\mu,\nu=\pm1,0). \tag*{(53)} ( μ , ν = ± 1 , 0 ) . (53)
Helicity Covariant and Helicity Contravariant Basis Vectors
e + 1 ′ = − e ′ − 1 , e + 1 = − e − 1 , \mathbf{e}^{\prime}_{+1}=-\mathbf{e}^{\prime-1},\qquad\mathbf{e}_{+1}=-\mathbf{e}^{-1}, e + 1 ′ = − e ′ − 1 , e + 1 = − e − 1 ,
e 0 ′ = e ′ 0 , e 0 = e 0 , \mathbf{e}^{\prime}_{0}=\mathbf{e}^{\prime0},\qquad\mathbf{e}_{0}=\mathbf{e}^{0}, e 0 ′ = e ′0 , e 0 = e 0 ,
e − 1 ′ = − e ′ + 1 , e − 1 = − e + 1 . (54) \mathbf{e}^{\prime}_{-1}=-\mathbf{e}^{\prime+1},\qquad\mathbf{e}_{-1}=-\mathbf{e}^{+1}. \tag*{(54)} e − 1 ′ = − e ′ + 1 , e − 1 = − e + 1 . (54)
Vectors. Tensors
Vectors and tensors are usually defined by transformation properties of their components under rotations of coordinate systems. The transformation rule for cartesian components of vectors and tensors is given below in § 9.9 (Eqs. (175) –(1.4.51)). The transformation properties of spherical components of vectors and irreducible tensors are discussed in Chap. 3.
Vector Components
Any vector can be expanded in terms of basis vectors, i.e., written as
A = ∑ α A α e α = ∑ α A α e α . (55) \mathbf{A} = \sum_{\alpha} A^{\alpha}\mathbf{e}_{\alpha} = \sum_{\alpha} A_{\alpha}\mathbf{e}^{\alpha}. \tag*{(55)} A = α ∑ A α e α = α ∑ A α e α . (55)
The expansion coefficients A α A_{\alpha} A α are called the covariant components of the vector, and A α A^{\alpha} A α are the contravariant vector components
A α = A ⋅ e α , A α = A ⋅ e α . (56) A_{\alpha} = \mathbf{A} \cdot\mathbf{e}_{\alpha}, \qquad A^{\alpha} = \mathbf{A} \cdot\mathbf{e}^{\alpha}. \tag*{(56)} A α = A ⋅ e α , A α = A ⋅ e α . (56)
In a cartesian coordinate system one has
A = A x e x + A y e y + A z e z = A x e x + A y e y + A z e z . (57) \mathbf{A} = A_x\mathbf{e}_x + A_y\mathbf{e}_y + A_z\mathbf{e}_z = A^x\mathbf{e}_x + A^y\mathbf{e}_y + A^z\mathbf{e}_z. \tag*{(57)} A = A x e x + A y e y + A z e z = A x e x + A y e y + A z e z . (57)
The covariant cartesian components of a vector coincide with the contravariant ones. In a polar coordinate system
A = A r e r + A ϑ e ϑ + A φ e φ = A r e r + A ϑ e ϑ + A φ e φ . (58) \mathbf{A} = A_r\mathbf{e}_r + A_{\vartheta}\mathbf{e}_{\vartheta} + A_{\varphi}\mathbf{e}_{\varphi} = A^r\mathbf{e}_r + A^{\vartheta}\mathbf{e}_{\vartheta} + A^{\varphi}\mathbf{e}_{\varphi}. \tag*{(58)} A = A r e r + A ϑ e ϑ + A φ e φ = A r e r + A ϑ e ϑ + A φ e φ . (58)
The covariant polar components coincide with the contravariant ones. For a spherical coordinate system
A = A + 1 e + 1 + A 0 e 0 + A − 1 e − 1 = A + 1 e + 1 + A 0 e 0 + A − 1 e − 1 . (59) \mathbf{A} = A^{+1}\mathbf{e}_{+1} + A^0\mathbf{e}_0 + A^{-1}\mathbf{e}_{-1} = A_{+1}\mathbf{e}^{+1} + A_0\mathbf{e}^0 + A_{-1}\mathbf{e}^{-1}. \tag*{(59)} A = A + 1 e + 1 + A 0 e 0 + A − 1 e − 1 = A + 1 e + 1 + A 0 e 0 + A − 1 e − 1 . (59)
The relations between covariant and contravariant spherical components are given by
A μ = ( − 1 ) μ A − μ , A μ = ( − 1 ) μ A − μ , ( μ = ± 1 , 0 ) . (60) A_{\mu} = (-1)^{\mu} A^{-\mu}, \qquad A^{\mu} = (-1)^{\mu} A_{-\mu}, \qquad(\mu= \pm1,0). \tag*{(60)} A μ = ( − 1 ) μ A − μ , A μ = ( − 1 ) μ A − μ , ( μ = ± 1 , 0 ) . (60)
If A \mathbf{A} A is a real vector, i.e., if A ∗ = A \mathbf{A}^{*} = \mathbf{A} A ∗ = A , then
A μ ∗ = A μ , A μ ∗ = A μ , ( μ = ± 1 , 0 ) . (61) A_{\mu}^{*} = A^{\mu}, \qquad A^{\mu*} = A_{\mu}, \qquad(\mu= \pm1,0). \tag*{(61)} A μ ∗ = A μ , A μ ∗ = A μ , ( μ = ± 1 , 0 ) . (61)
If A \mathbf{A} A is a complex vector, then
A μ ∗ = ( A ∗ ) μ , A μ ∗ = ( A ∗ ) μ , ( μ = ± 1 , 0 ) . (62) A_{\mu}^{*} = (A^{*})^{\mu}, \qquad A^{\mu*} = (A^{*})_{\mu}, \qquad(\mu= \pm1,0). \tag*{(62)} A μ ∗ = ( A ∗ ) μ , A μ ∗ = ( A ∗ ) μ , ( μ = ± 1 , 0 ) . (62)
An expansion of a real vector A \mathbf{A} A in terms of spherical basis vectors is written as
A = ∑ μ A μ e μ = ∑ μ A μ e μ = ∑ μ A μ ∗ e μ ∗ = ∑ μ A μ ∗ e μ ∗ = ∑ μ A μ e μ ∗ = ∑ μ A μ ∗ e μ = ∑ μ A μ e μ ∗ = ∑ μ A μ ∗ e μ = ∑ μ ( − 1 ) μ A − μ e μ = ∑ μ ( − 1 ) μ A − μ e μ . (63) \begin{aligned}
\mathbf{A} &= \sum_{\mu} A_{\mu}\mathbf{e}^{\mu} = \sum_{\mu} A^{\mu}\mathbf{e}_{\mu} = \sum_{\mu} A_{\mu}^{*}\mathbf{e}_{\mu}^{*} = \sum_{\mu} A^{\mu*}\mathbf{e}_{\mu}^{*} \\
&= \sum_{\mu} A_{\mu}\mathbf{e}_{\mu}^{*} = \sum_{\mu} A_{\mu}^{*}\mathbf{e}_{\mu} = \sum_{\mu} A^{\mu}\mathbf{e}_{\mu}^{*} = \sum_{\mu} A^{\mu*}\mathbf{e}^{\mu} \\
&= \sum_{\mu} (-1)^{\mu} A_{-\mu}\mathbf{e}_{\mu} = \sum_{\mu} (-1)^{\mu} A^{-\mu}\mathbf{e}^{\mu}. \tag*{(63)}
\end{aligned} A = μ ∑ A μ e μ = μ ∑ A μ e μ = μ ∑ A μ ∗ e μ ∗ = μ ∑ A μ ∗ e μ ∗ = μ ∑ A μ e μ ∗ = μ ∑ A μ ∗ e μ = μ ∑ A μ e μ ∗ = μ ∑ A μ ∗ e μ = μ ∑ ( − 1 ) μ A − μ e μ = μ ∑ ( − 1 ) μ A − μ e μ . (63)
An expansion of an arbitrary vector A \mathbf{A} A in terms of helicity basis vectors is given by
A = A ′ + 1 e + 1 ′ + A ′ 0 e 0 ′ + A ′ − 1 e − 1 ′ = A + 1 ′ e ′ + 1 + A 0 ′ e ′ 0 + A − 1 ′ e ′ − 1 . (64) \mathbf{A} = A^{\prime+1}\mathbf{e}^{\prime}_{+1} + A^{\prime0}\mathbf{e}^{\prime}_{0} + A^{\prime-1}\mathbf{e}^{\prime}_{-1} = A^{\prime}_{+1}\mathbf{e}^{\prime+1} + A^{\prime}_{0}\mathbf{e}^{\prime0} + A^{\prime}_{-1}\mathbf{e}^{\prime-1}. \tag*{(64)} A = A ′ + 1 e + 1 ′ + A ′0 e 0 ′ + A ′ − 1 e − 1 ′ = A + 1 ′ e ′ + 1 + A 0 ′ e ′0 + A − 1 ′ e ′ − 1 . (64)
The helicity components of a vector satisfy the same relations (1.2.6)–(1.2.9) as the spherical components. The relations between vector components in different bases are the same as the relations between basis vectors. These relations are given by Eqs. (29) –(1.1.54) in which one should replace e α → A α \mathbf{e}_{\alpha} \to A_{\alpha} e α → A α and e α → A α \mathbf{e}^{\alpha} \to A^{\alpha} e α → A α . In particular,
A + 1 = − A − 1 = − 1 2 ( A x + i A y ) , A x = 1 2 ( A − 1 − A + 1 ) = 1 2 ( A − 1 − A + 1 ) , A 0 = A 0 = A z , A y = i 2 ( A − 1 + A + 1 ) = − i 2 ( A − 1 + A + 1 ) , A − 1 = − A + 1 = 1 2 ( A x − i A y ) , A z = A 0 = A 0 . (65) \begin{aligned}
A_{+1} &= -A^{-1} = -\frac{1}{\sqrt{2}}(A_x + iA_y), \qquad
A_x = \frac{1}{\sqrt{2}}(A_{-1} - A_{+1}) = \frac{1}{\sqrt{2}}(A^{-1} - A^{+1}), \\
A_0 &= A^0 = A_z, \qquad
A_y = \frac{i}{\sqrt{2}}(A_{-1} + A_{+1}) = -\frac{i}{\sqrt{2}}(A^{-1} + A^{+1}), \\
A_{-1} &= -A^{+1} = \frac{1}{\sqrt{2}}(A_x - iA_y), \qquad
A_z = A_0 = A^0. \tag*{(65)}
\end{aligned} A + 1 A 0 A − 1 = − A − 1 = − 2 1 ( A x + i A y ) , A x = 2 1 ( A − 1 − A + 1 ) = 2 1 ( A − 1 − A + 1 ) , = A 0 = A z , A y = 2 i ( A − 1 + A + 1 ) = − 2 i ( A − 1 + A + 1 ) , = − A + 1 = 2 1 ( A x − i A y ) , A z = A 0 = A 0 . (65)
The matrices of transformations between cartesian, contravariant spherical and polar components of vectors are given in Tables 1.1 and 1.2.
Spherical components of a real vector A \mathbf{A} A which contains no derivatives and is independent of spin variables are
A ± 1 = ∓ ∣ A ∣ sin ϑ 2 e ± i φ , A ± 1 = ∓ ∣ A ∣ sin ϑ 2 e ∓ i φ , A_{\pm1}=\mp\lvert\mathbf{A}\rvert\frac{\sin\vartheta}{\sqrt{2}}e^{\pm i\varphi},\qquad A^{\pm1}=\mp\lvert\mathbf{A}\rvert\frac{\sin\vartheta}{\sqrt{2}}e^{\mp i\varphi}, A ± 1 = ∓ ∣ A ∣ 2 sin ϑ e ± i φ , A ± 1 = ∓ ∣ A ∣ 2 sin ϑ e ∓ i φ ,
A 0 = ∣ A ∣ cos ϑ , A 0 = ∣ A ∣ cos ϑ . (66) A_0=\lvert\mathbf{A}\rvert\cos\vartheta,\qquad A^0=\lvert\mathbf{A}\rvert\cos\vartheta. \tag*{(66)} A 0 = ∣ A ∣ cos ϑ , A 0 = ∣ A ∣ cos ϑ . (66)
where ϑ , φ \vartheta,\varphi ϑ , φ are the polar angles of the vector A \mathbf{A} A .
Equation (1.2.12) may be written in terms of spherical harmonics (see Chap. 5) as
A μ = 4 π 3 ∣ A ∣ Y 1 μ ( ϑ , φ ) A_\mu=\sqrt{\frac{4\pi}{3}}\lvert\mathbf{A}\rvert Y_{1\mu}(\vartheta,\varphi) A μ = 3 4 π ∣ A ∣ Y 1 μ ( ϑ , φ )
A μ = 4 π 3 ∣ A ∣ Y 1 μ ∗ ( ϑ , φ ) , ( μ = ± 1 , 0 ) . (67) A^\mu=\sqrt{\frac{4\pi}{3}}\lvert\mathbf{A}\rvert Y^*_{1\mu}(\vartheta,\varphi),\qquad(\mu=\pm1,0). \tag*{(67)} A μ = 3 4 π ∣ A ∣ Y 1 μ ∗ ( ϑ , φ ) , ( μ = ± 1 , 0 ) . (67)
The expressions for cartesian components of A \mathbf{A} A in terms of spherical harmonics read
A x = 2 π 3 ∣ A ∣ { Y 1 − 1 ( ϑ , φ ) − Y 1 + 1 ( ϑ , φ ) } , A_x=\sqrt{\frac{2\pi}{3}}\lvert\mathbf{A}\rvert\{Y_{1-1}(\vartheta,\varphi)-Y_{1+1}(\vartheta,\varphi)\}, A x = 3 2 π ∣ A ∣ { Y 1 − 1 ( ϑ , φ ) − Y 1 + 1 ( ϑ , φ )} ,
A y = i 2 π 3 ∣ A ∣ { Y 1 − 1 ( ϑ , φ ) + Y 1 + 1 ( ϑ , φ ) } , A_y=i\sqrt{\frac{2\pi}{3}}\lvert\mathbf{A}\rvert\{Y_{1-1}(\vartheta,\varphi)+Y_{1+1}(\vartheta,\varphi)\}, A y = i 3 2 π ∣ A ∣ { Y 1 − 1 ( ϑ , φ ) + Y 1 + 1 ( ϑ , φ )} ,
A z = 4 π 3 ∣ A ∣ Y 10 ( ϑ , φ ) . (68) A_z=\sqrt{\frac{4\pi}{3}}\lvert\mathbf{A}\rvert Y_{10}(\vartheta,\varphi). \tag*{(68)} A z = 3 4 π ∣ A ∣ Y 10 ( ϑ , φ ) . (68)
Table 1.1. Matrix form of the transformations for vector components in different bases.
Scalar Product of Vectors
The scalar product of vectors A \mathbf{A} A and B \mathbf{B} B in an arbitrary orthonormal basis is defined by
A ⋅ B = ∑ α A α B α = ∑ α A α B α . (69) \mathbf{A}\cdot\mathbf{B}=\sum_\alpha A_\alpha B^\alpha=\sum_\alpha A^\alpha B_\alpha. \tag*{(69)} A ⋅ B = α ∑ A α B α = α ∑ A α B α . (69)
In a cartesian coordinate system we have
A ⋅ B = A x B x + A y B y + A z B z . (70) \mathbf{A}\cdot\mathbf{B}=A_xB_x+A_yB_y+A_zB_z. \tag*{(70)} A ⋅ B = A x B x + A y B y + A z B z . (70)
For polar coordinates the scalar product is given by
A ⋅ B = A r B r + A ϑ B ϑ + A φ B φ . (71) \mathbf{A}\cdot\mathbf{B}=A_rB_r+A_\vartheta B_\vartheta+A_\varphi B_\varphi. \tag*{(71)} A ⋅ B = A r B r + A ϑ B ϑ + A φ B φ . (71)
Equation (1.2.17) is valid only if A \mathbf{A} A and B \mathbf{B} B do not contain derivatives because the polar basis vectors depend on the polar angles ϑ , φ \vartheta,\varphi ϑ , φ (see § 9.1 ). In spherical coordinates we have
A ⋅ B = ∑ μ A μ B μ = ∑ μ A μ B μ = ∑ μ ( − 1 ) μ A μ B − μ = ∑ μ ( − 1 ) μ A μ B − μ , ( μ = ± 1 , 0 ) . (72) \mathbf{A}\cdot\mathbf{B}=\sum_\mu A^\mu B_\mu=\sum_\mu A_\mu B^\mu=\sum_\mu(-1)^\mu A_\mu B_{-\mu}=\sum_\mu(-1)^\mu A^\mu B^{-\mu},\qquad(\mu=\pm1,0). \tag*{(72)} A ⋅ B = μ ∑ A μ B μ = μ ∑ A μ B μ = μ ∑ ( − 1 ) μ A μ B − μ = μ ∑ ( − 1 ) μ A μ B − μ , ( μ = ± 1 , 0 ) . (72)
or, in a more detailed form,
A ⋅ B = − A + 1 B − 1 + A 0 B 0 − A − 1 B + 1 . (73) \mathbf{A}\cdot\mathbf{B}=-A_{+1}B_{-1}+A_0B_0-A_{-1}B_{+1}. \tag*{(73)} A ⋅ B = − A + 1 B − 1 + A 0 B 0 − A − 1 B + 1 . (73)
The scalar product in terms of helicity components of vectors is similar to (1.2.18)–(1.2.19). The scalar product of vectors is invariant with respect to rotations of the coordinate system.
Vector Product of Vectors
In a cartesian coordinate system the vector product of vectors A \mathbf{A} A and B \mathbf{B} B is defined by
A × B = ∣ e x e y e z A x A y A z B x B y B z ∣ = ∑ i = x , y , z [ A × B ] i e i , (74) \mathbf{A}\times\mathbf{B}=
\begin{vmatrix}
\mathbf{e}_x & \mathbf{e}_y & \mathbf{e}_z\\
A_x & A_y & A_z\\
B_x & B_y & B_z
\end{vmatrix}
=\sum_{i=x,y,z}[\mathbf{A}\times\mathbf{B}]_i\mathbf{e}_i, \tag*{(74)} A × B = e x A x B x e y A y B y e z A z B z = i = x , y , z ∑ [ A × B ] i e i , (74)
where
[ A × B ] x = A y B z − A z B y , (75) [\mathbf{A}\times\mathbf{B}]_x=A_yB_z-A_zB_y, \tag*{(75)} [ A × B ] x = A y B z − A z B y , (75)
[ A × B ] y = A z B x − A x B z , [\mathbf{A}\times\mathbf{B}]_y=A_zB_x-A_xB_z, [ A × B ] y = A z B x − A x B z ,
[ A × B ] z = A x B y − A y B x . [\mathbf{A}\times\mathbf{B}]_z=A_xB_y-A_yB_x. [ A × B ] z = A x B y − A y B x .
Equation (1.2.21) may be written in a more compact form as
[ A × B ] i = ∑ k l ε i k l A k B l , ( i , k , l = x , y , z ) . (76) [\mathbf{A}\times\mathbf{B}]_i=\sum_{kl}\varepsilon_{ikl}A_kB_l,\qquad(i,k,l=x,y,z). \tag*{(76)} [ A × B ] i = k l ∑ ε ik l A k B l , ( i , k , l = x , y , z ) . (76)
In the polar coordinate system
A × B = ∣ e r e ϑ e φ A r A ϑ A φ B r B ϑ B φ ∣ = ∑ α = r , ϑ , φ [ A × B ] α e α , (77) \mathbf{A}\times\mathbf{B}=
\begin{vmatrix}
\mathbf{e}_r & \mathbf{e}_\vartheta& \mathbf{e}_\varphi\\
A_r & A_\vartheta& A_\varphi\\
B_r & B_\vartheta& B_\varphi
\end{vmatrix}
=\sum_{\alpha=r,\vartheta,\varphi}[\mathbf{A}\times\mathbf{B}]_\alpha\mathbf{e}_\alpha, \tag*{(77)} A × B = e r A r B r e ϑ A ϑ B ϑ e φ A φ B φ = α = r , ϑ , φ ∑ [ A × B ] α e α , (77)
where
[ A × B ] r = A ϑ B φ − A φ B ϑ , (78) [\mathbf{A}\times\mathbf{B}]_r=A_\vartheta B_\varphi-A_\varphi B_\vartheta, \tag*{(78)} [ A × B ] r = A ϑ B φ − A φ B ϑ , (78)
[ A × B ] ϑ = A φ B r − A r B φ , [\mathbf{A}\times\mathbf{B}]_\vartheta=A_\varphi B_r-A_rB_\varphi, [ A × B ] ϑ = A φ B r − A r B φ ,
[ A × B ] φ = A r B ϑ − A ϑ B r . [\mathbf{A}\times\mathbf{B}]_\varphi=A_rB_\vartheta-A_\vartheta B_r. [ A × B ] φ = A r B ϑ − A ϑ B r .
Equation (1.2.24) is valid only if A \mathbf{A} A and B \mathbf{B} B are not differential operators.
In the spherical coordinate system
A × B = i ∣ e + 1 e 0 e − 1 A + 1 A 0 A − 1 B + 1 B 0 B − 1 ∣ = − i ∣ e + 1 e 0 e − 1 A + 1 A 0 A − 1 B + 1 B 0 B − 1 ∣ = ∑ μ = ± 1 , 0 [ A × B ] μ e μ = ∑ μ = ± 1 , 0 [ A × B ] μ e μ , (79) \mathbf{A}\times\mathbf{B}
=i\begin{vmatrix}
\mathbf{e}_{+1} & \mathbf{e}_0 & \mathbf{e}_{-1}\\
A_{+1} & A_0 & A_{-1}\\
B_{+1} & B_0 & B_{-1}
\end{vmatrix}
=-i\begin{vmatrix}
\mathbf{e}^{+1} & \mathbf{e}^{0} & \mathbf{e}^{-1}\\
A^{+1} & A^0 & A^{-1}\\
B^{+1} & B^0 & B^{-1}
\end{vmatrix}
=\sum_{\mu=\pm1,0}[\mathbf{A}\times\mathbf{B}]_\mu\mathbf{e}^\mu
=\sum_{\mu=\pm1,0}[\mathbf{A}\times\mathbf{B}]^\mu\mathbf{e}_\mu, \tag*{(79)} A × B = i e + 1 A + 1 B + 1 e 0 A 0 B 0 e − 1 A − 1 B − 1 = − i e + 1 A + 1 B + 1 e 0 A 0 B 0 e − 1 A − 1 B − 1 = μ = ± 1 , 0 ∑ [ A × B ] μ e μ = μ = ± 1 , 0 ∑ [ A × B ] μ e μ , (79)
where
[ A × B ] + 1 = i ( A 0 B + 1 − A + 1 B 0 ) = i ( A − 1 B 0 − A 0 B − 1 ) , [\mathbf{A} \times\mathbf{B}]_{+1} = i(A_0B_{+1} - A_{+1}B_0) = i(A^{-1}B^0 - A^0B^{-1}), [ A × B ] + 1 = i ( A 0 B + 1 − A + 1 B 0 ) = i ( A − 1 B 0 − A 0 B − 1 ) ,
[ A × B ] 0 = i ( A − 1 B + 1 − A + 1 B − 1 ) = i ( A + 1 B − 1 − A − 1 B + 1 ) , [\mathbf{A} \times\mathbf{B}]_0 = i(A_{-1}B_{+1} - A_{+1}B_{-1}) = i(A^{+1}B^{-1} - A^{-1}B^{+1}), [ A × B ] 0 = i ( A − 1 B + 1 − A + 1 B − 1 ) = i ( A + 1 B − 1 − A − 1 B + 1 ) ,
[ A × B ] − 1 = i ( A − 1 B 0 − A 0 B − 1 ) = i ( A 0 B + 1 − A + 1 B 0 ) . (80) [\mathbf{A} \times\mathbf{B}]_{-1} = i(A_{-1}B_0 - A_0B_{-1}) = i(A^0B^{+1} - A^{+1}B^0). \tag*{(80)} [ A × B ] − 1 = i ( A − 1 B 0 − A 0 B − 1 ) = i ( A 0 B + 1 − A + 1 B 0 ) . (80)
[ A × B ] + 1 = i ( A 0 B − 1 − A − 1 B 0 ) = i ( A + 1 B 0 − A 0 B + 1 ) , [\mathbf{A} \times\mathbf{B}]^{+1} = i(A_0B_{-1} - A_{-1}B_0) = i(A^{+1}B^0 - A^0B^{+1}), [ A × B ] + 1 = i ( A 0 B − 1 − A − 1 B 0 ) = i ( A + 1 B 0 − A 0 B + 1 ) ,
[ A × B ] 0 = i ( A − 1 B + 1 − A + 1 B − 1 ) = i ( A + 1 B − 1 − A − 1 B + 1 ) , [\mathbf{A} \times\mathbf{B}]^0 = i(A_{-1}B_{+1} - A_{+1}B_{-1}) = i(A^{+1}B^{-1} - A^{-1}B^{+1}), [ A × B ] 0 = i ( A − 1 B + 1 − A + 1 B − 1 ) = i ( A + 1 B − 1 − A − 1 B + 1 ) ,
[ A × B ] − 1 = i ( A + 1 B 0 − A 0 B + 1 ) = i ( A 0 B − 1 − A − 1 B 0 ) . (81) [\mathbf{A} \times\mathbf{B}]^{-1} = i(A_{+1}B_0 - A_0B_{+1}) = i(A^0B^{-1} - A^{-1}B^0). \tag*{(81)} [ A × B ] − 1 = i ( A + 1 B 0 − A 0 B + 1 ) = i ( A 0 B − 1 − A − 1 B 0 ) . (81)
Equations (1.2.26)–(1.2.27) may be written in a more compact form using the Clebsch-Gordan coefficients (see Chap. 8)
[ A × B ] μ = − i 2 ∑ ν λ C 1 ν 1 λ 1 μ A ν B λ , [\mathbf{A} \times\mathbf{B}]_\mu= -i\sqrt{2}\sum_{\nu\lambda} C^{1\mu}_{1\nu\,1\lambda} A_\nu B_\lambda, [ A × B ] μ = − i 2 ν λ ∑ C 1 ν 1 λ 1 μ A ν B λ ,
[ A × B ] μ = i 2 ∑ ν λ C 1 ν 1 λ 1 μ A ν B λ , ( μ , ν , λ = ± 1 , 0 ) . (82) [\mathbf{A} \times\mathbf{B}]^\mu= i\sqrt{2}\sum_{\nu\lambda} C^{1\mu}_{1\nu\,1\lambda} A^\nu B^\lambda,
\qquad(\mu,\nu,\lambda= \pm1,0). \tag*{(82)} [ A × B ] μ = i 2 ν λ ∑ C 1 ν 1 λ 1 μ A ν B λ , ( μ , ν , λ = ± 1 , 0 ) . (82)
Helicity components of the vector product are given by equations analogous to (1.2.25)–(1.2.28).
Products Involving Three or More Vectors
A ⋅ [ B × C ] = B ⋅ [ C × A ] = C ⋅ [ A × B ] = − A ⋅ [ C × B ] = − B ⋅ [ A × C ] = − C ⋅ [ B × A ] , (83) \mathbf{A} \cdot[\mathbf{B} \times\mathbf{C}] = \mathbf{B} \cdot[\mathbf{C} \times\mathbf{A}] = \mathbf{C} \cdot[\mathbf{A} \times\mathbf{B}] = -\mathbf{A} \cdot[\mathbf{C} \times\mathbf{B}] = -\mathbf{B} \cdot[\mathbf{A} \times\mathbf{C}] = -\mathbf{C} \cdot[\mathbf{B} \times\mathbf{A}], \tag*{(83)} A ⋅ [ B × C ] = B ⋅ [ C × A ] = C ⋅ [ A × B ] = − A ⋅ [ C × B ] = − B ⋅ [ A × C ] = − C ⋅ [ B × A ] , (83)
A × [ B × C ] = B ( A ⋅ C ) − C ( A ⋅ B ) , (84) \mathbf{A} \times[\mathbf{B} \times\mathbf{C}] = \mathbf{B}(\mathbf{A} \cdot\mathbf{C}) - \mathbf{C}(\mathbf{A} \cdot\mathbf{B}), \tag*{(84)} A × [ B × C ] = B ( A ⋅ C ) − C ( A ⋅ B ) , (84)
[ A × B ] ⋅ [ C × D ] = ( A ⋅ C ) ( B ⋅ D ) − ( A ⋅ D ) ( B ⋅ C ) , (85) [\mathbf{A} \times\mathbf{B}] \cdot[\mathbf{C} \times\mathbf{D}] = (\mathbf{A} \cdot\mathbf{C})(\mathbf{B} \cdot\mathbf{D}) - (\mathbf{A} \cdot\mathbf{D})(\mathbf{B} \cdot\mathbf{C}), \tag*{(85)} [ A × B ] ⋅ [ C × D ] = ( A ⋅ C ) ( B ⋅ D ) − ( A ⋅ D ) ( B ⋅ C ) , (85)
[ A × B ] × [ C × D ] = B ( A ⋅ [ C × D ] ) − A ( B ⋅ [ C × D ] ) = C ( A ⋅ [ B × D ] ) − D ( A ⋅ [ B × C ] ) . (86) [\mathbf{A} \times\mathbf{B}] \times[\mathbf{C} \times\mathbf{D}] = \mathbf{B}(\mathbf{A} \cdot[\mathbf{C} \times\mathbf{D}]) - \mathbf{A}(\mathbf{B} \cdot[\mathbf{C} \times\mathbf{D}]) = \mathbf{C}(\mathbf{A} \cdot[\mathbf{B} \times\mathbf{D}]) - \mathbf{D}(\mathbf{A} \cdot[\mathbf{B} \times\mathbf{C}]). \tag*{(86)} [ A × B ] × [ C × D ] = B ( A ⋅ [ C × D ]) − A ( B ⋅ [ C × D ]) = C ( A ⋅ [ B × D ]) − D ( A ⋅ [ B × C ]) . (86)
( A ⋅ [ B × C ] ) ( a ⋅ [ b × c ] ) = ∣ A ⋅ a A ⋅ b A ⋅ c B ⋅ a B ⋅ b B ⋅ c C ⋅ a C ⋅ b C ⋅ c ∣ = ( A ⋅ a ) ( B ⋅ b ) ( C ⋅ c ) − ( A ⋅ a ) ( B ⋅ c ) ( C ⋅ b ) − ( B ⋅ b ) ( A ⋅ c ) ( C ⋅ a ) − ( C ⋅ c ) ( A ⋅ b ) ( B ⋅ a ) + ( A ⋅ b ) ( B ⋅ c ) ( C ⋅ a ) + ( A ⋅ c ) ( B ⋅ a ) ( C ⋅ b ) . (87) \begin{aligned}
(\mathbf{A} \cdot[\mathbf{B} \times\mathbf{C}])(\mathbf{a} \cdot[\mathbf{b} \times\mathbf{c}]) &=
\begin{vmatrix}
\mathbf{A} \cdot\mathbf{a} & \mathbf{A} \cdot\mathbf{b} & \mathbf{A} \cdot\mathbf{c} \\
\mathbf{B} \cdot\mathbf{a} & \mathbf{B} \cdot\mathbf{b} & \mathbf{B} \cdot\mathbf{c} \\
\mathbf{C} \cdot\mathbf{a} & \mathbf{C} \cdot\mathbf{b} & \mathbf{C} \cdot\mathbf{c}
\end{vmatrix} \\
&= (\mathbf{A} \cdot\mathbf{a})(\mathbf{B} \cdot\mathbf{b})(\mathbf{C} \cdot\mathbf{c}) - (\mathbf{A} \cdot\mathbf{a})(\mathbf{B} \cdot\mathbf{c})(\mathbf{C} \cdot\mathbf{b}) - (\mathbf{B} \cdot\mathbf{b})(\mathbf{A} \cdot\mathbf{c})(\mathbf{C} \cdot\mathbf{a}) \\
&\quad-(\mathbf{C} \cdot\mathbf{c})(\mathbf{A} \cdot\mathbf{b})(\mathbf{B} \cdot\mathbf{a}) + (\mathbf{A} \cdot\mathbf{b})(\mathbf{B} \cdot\mathbf{c})(\mathbf{C} \cdot\mathbf{a}) + (\mathbf{A} \cdot\mathbf{c})(\mathbf{B} \cdot\mathbf{a})(\mathbf{C} \cdot\mathbf{b}). \tag*{(87)}
\end{aligned} ( A ⋅ [ B × C ]) ( a ⋅ [ b × c ]) = A ⋅ a B ⋅ a C ⋅ a A ⋅ b B ⋅ b C ⋅ b A ⋅ c B ⋅ c C ⋅ c = ( A ⋅ a ) ( B ⋅ b ) ( C ⋅ c ) − ( A ⋅ a ) ( B ⋅ c ) ( C ⋅ b ) − ( B ⋅ b ) ( A ⋅ c ) ( C ⋅ a ) − ( C ⋅ c ) ( A ⋅ b ) ( B ⋅ a ) + ( A ⋅ b ) ( B ⋅ c ) ( C ⋅ a ) + ( A ⋅ c ) ( B ⋅ a ) ( C ⋅ b ) . (87)
Tensors δ i k \delta_{ik} δ ik and ϵ i k l \epsilon_{ikl} ϵ ik l
In a cartesian basis two basic tensors δ i k \delta_{ik} δ ik and ϵ i k l \epsilon_{ikl} ϵ ik l are widely used. The first tensor, δ i k \delta_{ik} δ ik , is the symmetric unit tensor of rank 2. The second tensor, ϵ i k l \epsilon_{ikl} ϵ ik l , is the totally antisymmetric unit tensor of rank 3.
The tensor δ i k \delta_{ik} δ ik is called the Kronecker δ \delta δ -symbol and is defined by
δ i k = { 1 , i = k , 0 , i ≠ k , ( i , k = x , y , z ) (88) \delta_{ik} =
\begin{cases}
1, & i = k, \\
0, & i \ne k,
\end{cases}
\qquad(i,k = x,y,z) \tag*{(88)} δ ik = { 1 , 0 , i = k , i = k , ( i , k = x , y , z ) (88)
The components δ i k \delta_{ik} δ ik are invariant with respect to rotations and inversion of coordinate systems.
The tensor (or, more precisely, the pseudo-tensor) ϵ i k l \epsilon_{ikl} ϵ ik l is often called the Levi-Civita tensor. It is antisymmetric with respect to permutations of any pair of indices. Thus, ϵ i k l = 0 \epsilon_{ikl} = 0 ϵ ik l = 0 if at least two of the indices i , k , l i,k,l i , k , l are equal, and ϵ i k l ≠ 0 \epsilon_{ikl} \ne0 ϵ ik l = 0 only if all indices i , k , l i,k,l i , k , l are different. The components ϵ i k l \epsilon_{ikl} ϵ ik l are given by
ϵ i i i = 0 , ( i = x , y , z ) ( 3 components ) , \epsilon_{iii} = 0,\qquad(i = x,y,z)\quad(3\ \text{components}), ϵ iii = 0 , ( i = x , y , z ) ( 3 components ) ,
ϵ i i k = ϵ i k i = ϵ k i i = 0 , ( i , k = x , y , z ) ( 18 components ) , \epsilon_{iik} = \epsilon_{iki} = \epsilon_{kii} = 0,\qquad(i,k = x,y,z)\quad(18\ \text{components}), ϵ iik = ϵ ik i = ϵ k ii = 0 , ( i , k = x , y , z ) ( 18 components ) ,
ϵ x y z = ϵ y z x = ϵ z x y = − ϵ x z y = − ϵ y x z = − ϵ z y x = 1 ( 6 components ) . (89) \epsilon_{xyz} = \epsilon_{yzx} = \epsilon_{zxy} = -\epsilon_{xzy} = -\epsilon_{yxz} = -\epsilon_{zyx} = 1\quad(6\ \text{components}). \tag*{(89)} ϵ x y z = ϵ y z x = ϵ z x y = − ϵ x z y = − ϵ y x z = − ϵ z y x = 1 ( 6 components ) . (89)
The components ϵ i k l \epsilon_{ikl} ϵ ik l are invariant with respect to rotations and inversion of coordinate systems.
The tensor ε i k l \varepsilon_{ikl} ε ik l has the following properties: The product of two tensors ε i k l \varepsilon_{ikl} ε ik l and ε r g t \varepsilon_{rgt} ε r g t may be written in the form of a determinant
ε i k l ε r s t = ∣ δ i r δ i s δ i t δ k r δ k s δ k t δ l r δ l s δ l t ∣ = δ i r δ k s δ l t + δ i s δ k t δ l r + δ i t δ k r δ l s − δ i r δ k t δ l s − δ i s δ k r δ l t − δ i t δ k s δ l r . (90) \varepsilon_{ikl}\varepsilon_{rst} =
\begin{vmatrix}
\delta_{ir} & \delta_{is} & \delta_{it} \\
\delta_{kr} & \delta_{ks} & \delta_{kt} \\
\delta_{lr} & \delta_{ls} & \delta_{lt}
\end{vmatrix}
= \delta_{ir}\delta_{ks}\delta_{lt} + \delta_{is}\delta_{kt}\delta_{lr} + \delta_{it}\delta_{kr}\delta_{ls} - \delta_{ir}\delta_{kt}\delta_{ls} - \delta_{is}\delta_{kr}\delta_{lt} - \delta_{it}\delta_{ks}\delta_{lr}. \tag*{(90)} ε ik l ε r s t = δ i r δ k r δ l r δ i s δ k s δ l s δ i t δ k t δ l t = δ i r δ k s δ l t + δ i s δ k t δ l r + δ i t δ k r δ l s − δ i r δ k t δ l s − δ i s δ k r δ l t − δ i t δ k s δ l r . (90)
By summing over a pair of indices, one obtains
∑ i ε i k l ε i s t = ∣ δ k s δ k t δ l s δ l t ∣ = δ k s δ l t − δ k t δ l s . (91) \sum_i \varepsilon_{ikl}\varepsilon_{ist} =
\begin{vmatrix}
\delta_{ks} & \delta_{kt} \\
\delta_{ls} & \delta_{lt}
\end{vmatrix}
= \delta_{ks}\delta_{lt} - \delta_{kt}\delta_{ls}. \tag*{(91)} i ∑ ε ik l ε i s t = δ k s δ l s δ k t δ l t = δ k s δ l t − δ k t δ l s . (91)
Summation over two pairs of indices yields
∑ i , k ε i k l ε i k t = 2 δ l t . (92) \sum_{i,k} \varepsilon_{ikl}\varepsilon_{ikt} = 2\delta_{lt}. \tag*{(92)} i , k ∑ ε ik l ε ik t = 2 δ l t . (92)
Finally, the summation over three pairs of indices gives
∑ i , k , l ε i k l ε i k l = 6. (93) \sum_{i,k,l} \varepsilon_{ikl}\varepsilon_{ikl} = 6. \tag*{(93)} i , k , l ∑ ε ik l ε ik l = 6. (93)
For an arbitrary 3 × 3 3 \times3 3 × 3 matrix ∥ A i k ∥ \lVert A_{ik}\rVert ∥ A ik ∥ ( i , k = x , y , z ) (i,k=x,y,z) ( i , k = x , y , z ) the following relation holds
∑ i , k , l A x i A y k A z l ε i k l = det ∥ A i k ∥ = ∣ A x x A x y A x z A y x A y y A y z A z x A z y A z z ∣ . (94) \sum_{i,k,l} A_{xi}A_{yk}A_{zl}\varepsilon_{ikl} = \det\lVert A_{ik}\rVert=
\begin{vmatrix}
A_{xx} & A_{xy} & A_{xz} \\
A_{yx} & A_{yy} & A_{yz} \\
A_{zx} & A_{zy} & A_{zz}
\end{vmatrix}. \tag*{(94)} i , k , l ∑ A x i A y k A z l ε ik l = det ∥ A ik ∥ = A xx A y x A z x A x y A y y A z y A x z A y z A z z . (94)
Differential Operations
Operator
The operator ∇ \nabla ∇ (nabla) is the basic vector differential operator. Cartesian components of ∇ \nabla ∇ are given by
∇ x = ∂ ∂ x , ∇ y = ∂ ∂ y , ∇ z = ∂ ∂ z . (95) \nabla_x = \frac{\partial}{\partial x}, \qquad
\nabla_y = \frac{\partial}{\partial y}, \qquad
\nabla_z = \frac{\partial}{\partial z}. \tag*{(95)} ∇ x = ∂ x ∂ , ∇ y = ∂ y ∂ , ∇ z = ∂ z ∂ . (95)
These components may be expressed in terms of polar coordinates as
∇ x = sin ϑ cos φ ∂ ∂ r + cos ϑ cos φ r ∂ ∂ ϑ − sin φ r sin ϑ ∂ ∂ φ , \nabla_x = \sin\vartheta\cos\varphi\frac{\partial}{\partial r}
+ \frac{\cos\vartheta\cos\varphi}{r}\frac{\partial}{\partial\vartheta}
- \frac{\sin\varphi}{r\sin\vartheta}\frac{\partial}{\partial\varphi}, ∇ x = sin ϑ cos φ ∂ r ∂ + r cos ϑ cos φ ∂ ϑ ∂ − r sin ϑ sin φ ∂ φ ∂ ,
∇ y = sin ϑ sin φ ∂ ∂ r + cos ϑ sin φ r ∂ ∂ ϑ + cos φ r sin ϑ ∂ ∂ φ , \nabla_y = \sin\vartheta\sin\varphi\frac{\partial}{\partial r}
+ \frac{\cos\vartheta\sin\varphi}{r}\frac{\partial}{\partial\vartheta}
+ \frac{\cos\varphi}{r\sin\vartheta}\frac{\partial}{\partial\varphi}, ∇ y = sin ϑ sin φ ∂ r ∂ + r cos ϑ sin φ ∂ ϑ ∂ + r sin ϑ cos φ ∂ φ ∂ ,
∇ z = cos ϑ ∂ ∂ r − sin ϑ r ∂ ∂ ϑ . (96) \nabla_z = \cos\vartheta\frac{\partial}{\partial r}
- \frac{\sin\vartheta}{r}\frac{\partial}{\partial\vartheta}. \tag*{(96)} ∇ z = cos ϑ ∂ r ∂ − r sin ϑ ∂ ϑ ∂ . (96)
An expansion of the operator ∇ \nabla ∇ in terms of spherical basis vectors reads
∇ = ∑ μ ( − 1 ) μ e μ ∇ − μ = − e + 1 ∇ − 1 + e 0 ∇ 0 − e − 1 ∇ + 1 . (97) \nabla= \sum_{\mu}(-1)^{\mu}e_{\mu}\nabla_{-\mu}
= -e_{+1}\nabla_{-1} + e_0\nabla_0 - e_{-1}\nabla_{+1}. \tag*{(97)} ∇ = μ ∑ ( − 1 ) μ e μ ∇ − μ = − e + 1 ∇ − 1 + e 0 ∇ 0 − e − 1 ∇ + 1 . (97)
where spherical components of ∇ \nabla ∇ are given by
∇ + 1 = − 1 2 ( ∂ ∂ x + i ∂ ∂ y ) , \nabla_{+1} = -\frac{1}{\sqrt{2}}\left(\frac{\partial}{\partial x} + i\frac{\partial}{\partial y}\right), ∇ + 1 = − 2 1 ( ∂ x ∂ + i ∂ y ∂ ) ,
∇ 0 = ∂ ∂ z , \nabla_0 = \frac{\partial}{\partial z}, ∇ 0 = ∂ z ∂ ,
∇ − 1 = 1 2 ( ∂ ∂ x − i ∂ ∂ y ) . (98) \nabla_{-1} = \frac{1}{\sqrt{2}}\left(\frac{\partial}{\partial x} - i\frac{\partial}{\partial y}\right). \tag*{(98)} ∇ − 1 = 2 1 ( ∂ x ∂ − i ∂ y ∂ ) . (98)
Spherical components of ∇ \nabla ∇ in a polar coordinate system have the form
∇ + 1 = − e i φ 2 { sin ϑ ∂ ∂ r + cos ϑ r ∂ ∂ ϑ + i r sin ϑ ∂ ∂ φ } , \nabla_{+1} = -\frac{e^{i\varphi}}{\sqrt{2}}\left\{
\sin\vartheta\frac{\partial}{\partial r}
+ \frac{\cos\vartheta}{r}\frac{\partial}{\partial\vartheta}
+ \frac{i}{r\sin\vartheta}\frac{\partial}{\partial\varphi}
\right\}, ∇ + 1 = − 2 e i φ { sin ϑ ∂ r ∂ + r cos ϑ ∂ ϑ ∂ + r sin ϑ i ∂ φ ∂ } ,
∇ 0 = cos ϑ ∂ ∂ r − sin ϑ r ∂ ∂ ϑ , \nabla_0 = \cos\vartheta\frac{\partial}{\partial r}
- \frac{\sin\vartheta}{r}\frac{\partial}{\partial\vartheta}, ∇ 0 = cos ϑ ∂ r ∂ − r sin ϑ ∂ ϑ ∂ ,
∇ − 1 = e − i φ 2 { sin ϑ ∂ ∂ r + cos ϑ r ∂ ∂ ϑ − i r sin ϑ ∂ ∂ φ } . (99) \nabla_{-1} = \frac{e^{-i\varphi}}{\sqrt{2}}\left\{
\sin\vartheta\frac{\partial}{\partial r}
+ \frac{\cos\vartheta}{r}\frac{\partial}{\partial\vartheta}
- \frac{i}{r\sin\vartheta}\frac{\partial}{\partial\varphi}
\right\}. \tag*{(99)} ∇ − 1 = 2 e − i φ { sin ϑ ∂ r ∂ + r cos ϑ ∂ ϑ ∂ − r sin ϑ i ∂ φ ∂ } . (99)
An expansion of ∇ \nabla ∇ in terms of polar basis vectors may be written as
∇ = e r ∇ r + e ϑ ∇ ϑ + e φ ∇ φ , (100) \nabla= \mathbf{e}_r \nabla_r + \mathbf{e}_{\vartheta} \nabla_{\vartheta} + \mathbf{e}_{\varphi} \nabla_{\varphi}, \tag*{(100)} ∇ = e r ∇ r + e ϑ ∇ ϑ + e φ ∇ φ , (100)
where
∇ r = ∂ ∂ r , ∇ ϑ = 1 r ∂ ∂ ϑ , ∇ φ = 1 r sin ϑ ∂ ∂ φ . (101) \nabla_r = \frac{\partial}{\partial r}, \qquad\nabla_{\vartheta} = \frac{1}{r}\frac{\partial}{\partial\vartheta}, \qquad\nabla_{\varphi} = \frac{1}{r\sin\vartheta}\frac{\partial}{\partial\varphi}. \tag*{(101)} ∇ r = ∂ r ∂ , ∇ ϑ = r 1 ∂ ϑ ∂ , ∇ φ = r sin ϑ 1 ∂ φ ∂ . (101)
The order of operator components relative to the basis vectors in Eq. (100) is essential because e r \mathbf{e}_r e r , e ϑ \mathbf{e}_{\vartheta} e ϑ , e φ \mathbf{e}_{\varphi} e φ depend on ϑ \vartheta ϑ , φ \varphi φ .
The operator ∇ \nabla ∇ may be written in the form
∇ = n ∂ ∂ r + 1 r ∇ Ω , (102) \nabla= \mathbf{n}\frac{\partial}{\partial r} + \frac{1}{r}\nabla_{\Omega}, \tag*{(102)} ∇ = n ∂ r ∂ + r 1 ∇ Ω , (102)
where ∇ Ω \nabla_{\Omega} ∇ Ω is the angular part of ∇ \nabla ∇ , and n = r / r \mathbf{n} = \mathbf{r}/r n = r / r is the unit vector determined by angles ϑ \vartheta ϑ and φ \varphi φ . The operator ∇ Ω \nabla_{\Omega} ∇ Ω acts only on variables ϑ \vartheta ϑ and φ \varphi φ . In the polar coordinate system it has only two components
( ∇ Ω ) ϑ = ∂ ∂ ϑ , ( ∇ Ω ) φ = 1 sin ϑ ∂ ∂ φ . (103) (\nabla_{\Omega})_{\vartheta} = \frac{\partial}{\partial\vartheta}, \qquad(\nabla_{\Omega})_{\varphi} = \frac{1}{\sin\vartheta}\frac{\partial}{\partial\varphi}. \tag*{(103)} ( ∇ Ω ) ϑ = ∂ ϑ ∂ , ( ∇ Ω ) φ = sin ϑ 1 ∂ φ ∂ . (103)
The operator ∇ Ω \nabla_{\Omega} ∇ Ω may be written as
∇ Ω = − i n × L ^ , (104) \nabla_{\Omega} = -i\mathbf{n} \times\hat{\mathbf{L}}, \tag*{(104)} ∇ Ω = − i n × L ^ , (104)
where L ^ \hat{\mathbf{L}} L ^ is the orbital angular momentum operator (see § 10.2 ).
Laplace Operator
The Laplace operator (Laplacian) Δ \Delta Δ is a scalar differential operator
Δ = ∇ 2 . (105) \Delta= \nabla^2. \tag*{(105)} Δ = ∇ 2 . (105)
In the cartesian coordinate system Δ \Delta Δ has the form
Δ = ∂ 2 ∂ x 2 + ∂ 2 ∂ y 2 + ∂ 2 ∂ z 2 . (106) \Delta= \frac{\partial^2}{\partial x^2} + \frac{\partial^2}{\partial y^2} + \frac{\partial^2}{\partial z^2}. \tag*{(106)} Δ = ∂ x 2 ∂ 2 + ∂ y 2 ∂ 2 + ∂ z 2 ∂ 2 . (106)
In the polar coordinate system it is given by
Δ = 1 r 2 ∂ ∂ r { r 2 ∂ ∂ r } + 1 r 2 sin ϑ ∂ ∂ ϑ { sin ϑ ∂ ∂ ϑ } + 1 r 2 sin 2 ϑ ∂ 2 ∂ φ 2 . (107) \Delta= \frac{1}{r^2}\frac{\partial}{\partial r}\left\{r^2\frac{\partial}{\partial r}\right\} + \frac{1}{r^2\sin\vartheta}\frac{\partial}{\partial\vartheta}\left\{\sin\vartheta\frac{\partial}{\partial\vartheta}\right\} + \frac{1}{r^2\sin^2\vartheta}\frac{\partial^2}{\partial\varphi^2}. \tag*{(107)} Δ = r 2 1 ∂ r ∂ { r 2 ∂ r ∂ } + r 2 sin ϑ 1 ∂ ϑ ∂ { sin ϑ ∂ ϑ ∂ } + r 2 sin 2 ϑ 1 ∂ φ 2 ∂ 2 . (107)
The operator Δ \Delta Δ may also be written as
Δ = 1 r 2 ∂ ∂ r { r 2 ∂ ∂ r } + 1 r 2 Δ Ω , (108) \Delta= \frac{1}{r^2}\frac{\partial}{\partial r}\left\{r^2\frac{\partial}{\partial r}\right\} + \frac{1}{r^2}\Delta_{\Omega}, \tag*{(108)} Δ = r 2 1 ∂ r ∂ { r 2 ∂ r ∂ } + r 2 1 Δ Ω , (108)
where Δ Ω \Delta_{\Omega} Δ Ω is the angular part of Δ \Delta Δ
Δ Ω = ∇ Ω 2 = 1 sin ϑ ∂ ∂ ϑ { sin ϑ ∂ ∂ ϑ } + 1 sin 2 ϑ ∂ 2 ∂ φ 2 (109) \Delta_{\Omega} = \nabla_{\Omega}^2 = \frac{1}{\sin\vartheta}\frac{\partial}{\partial\vartheta}\left\{\sin\vartheta\frac{\partial}{\partial\vartheta}\right\} + \frac{1}{\sin^2\vartheta}\frac{\partial^2}{\partial\varphi^2} \tag*{(109)} Δ Ω = ∇ Ω 2 = sin ϑ 1 ∂ ϑ ∂ { sin ϑ ∂ ϑ ∂ } + sin 2 ϑ 1 ∂ φ 2 ∂ 2 (109)
expressed in terms of the orbital angular momentum operator L ^ \hat{\mathbf{L}} L ^ as
Δ Ω = − L ^ 2 (110) \Delta_{\Omega} = -\hat{\mathbf{L}}^2 \tag*{(110)} Δ Ω = − L ^ 2 (110)
Differential Operations on Scalars and Vectors
The gradient of a scalar function Φ ( r ) \Phi(\mathbf{r}) Φ ( r ) is the vector defined in terms of the operator abla as
grad Φ ( r ) = ∇ Φ ( r ) (111) \operatorname{grad}\Phi(\mathbf{r}) = \nabla\Phi(\mathbf{r}) \tag*{(111)} grad Φ ( r ) = ∇Φ ( r ) (111)
The components of grad Φ \operatorname{grad}\Phi grad Φ may be obtained by use of Eqs. (95) –(101) for the components of ∇ \nabla ∇ . If Φ \Phi Φ depends only on r = ∣ r ∣ r = \lvert\mathbf{r}\rvert r = ∣ r ∣ (spherically-symmetric field), then
∇ Φ ( r ) = n d Φ ( r ) d r (112) \nabla\Phi(r) = \mathbf{n}\frac{d\Phi(r)}{dr} \tag*{(112)} ∇Φ ( r ) = n d r d Φ ( r ) (112)
where n = r / r \mathbf{n}=\mathbf{r}/r n = r / r .
The directional derivative of a scalar function Φ ( r ) \Phi(\mathbf{r}) Φ ( r ) in the direction specified by a unit vector u \mathbf{u} u is the scalar defined by
d d s Φ ( r ) = ( u ⋅ ∇ ) Φ ( r ) . (113) \frac{d}{ds}\Phi(\mathbf{r})=(\mathbf{u}\cdot\nabla)\Phi(\mathbf{r}). \tag*{(113)} d s d Φ ( r ) = ( u ⋅ ∇ ) Φ ( r ) . (113)
The divergence of a vector field A ( r ) \mathbf{A}(\mathbf{r}) A ( r ) is the scalar product of ∇ \nabla ∇ and A \mathbf{A} A
div A = ∇ ⋅ A . (114) \operatorname{div}\mathbf{A}=\nabla\cdot\mathbf{A}. \tag*{(114)} div A = ∇ ⋅ A . (114)
The expression for div A \operatorname{div}\mathbf{A} div A in a cartesian coordinate system is
div A = ∂ A x ∂ x + ∂ A y ∂ y + ∂ A z ∂ z = ∑ i = x , y , z ∂ A i ∂ x i . (115) \operatorname{div}\mathbf{A}=\frac{\partial A_x}{\partial x}+\frac{\partial A_y}{\partial y}+\frac{\partial A_z}{\partial z}=\sum_{i=x,y,z}\frac{\partial A_i}{\partial x_i}. \tag*{(115)} div A = ∂ x ∂ A x + ∂ y ∂ A y + ∂ z ∂ A z = i = x , y , z ∑ ∂ x i ∂ A i . (115)
and in a spherical coordinate system it has the form
div A = − ∇ + 1 A − 1 + ∇ 0 A 0 − ∇ − 1 A + 1 = ∑ μ = ± 1 , 0 ( − 1 ) μ ∇ μ A − μ . (116) \operatorname{div}\mathbf{A}=-\nabla_{+1}A_{-1}+\nabla_0A_0-\nabla_{-1}A_{+1}=\sum_{\mu=\pm1,0}(-1)^\mu\nabla_\mu A_{-\mu}. \tag*{(116)} div A = − ∇ + 1 A − 1 + ∇ 0 A 0 − ∇ − 1 A + 1 = μ = ± 1 , 0 ∑ ( − 1 ) μ ∇ μ A − μ . (116)
where the spherical components ∇ μ \nabla_\mu ∇ μ are given by Eqs. (98) –(99) . In the polar coordinate system we have
div A = 1 r 2 ∂ ∂ r ( r 2 A r ) + 1 r sin ϑ ∂ ∂ ϑ ( sin ϑ A ϑ ) + 1 r sin ϑ ∂ A φ ∂ φ . (117) \operatorname{div}\mathbf{A}=\frac{1}{r^2}\frac{\partial}{\partial r}(r^2A_r)+\frac{1}{r\sin\vartheta}\frac{\partial}{\partial\vartheta}(\sin\vartheta A_\vartheta)+\frac{1}{r\sin\vartheta}\frac{\partial A_\varphi}{\partial\varphi}. \tag*{(117)} div A = r 2 1 ∂ r ∂ ( r 2 A r ) + r sin ϑ 1 ∂ ϑ ∂ ( sin ϑ A ϑ ) + r sin ϑ 1 ∂ φ ∂ A φ . (117)
The curl of a vector field A ( r ) \mathbf{A}(\mathbf{r}) A ( r ) is the vector product of ∇ \nabla ∇ and A \mathbf{A} A
curl A = ∇ × A . (118) \operatorname{curl}\mathbf{A}=\nabla\times\mathbf{A}. \tag*{(118)} curl A = ∇ × A . (118)
The cartesian components of curl A \operatorname{curl}\mathbf{A} curl A are given by
[ curl A ] x = ∂ A z ∂ y − ∂ A y ∂ z , [ curl A ] y = ∂ A x ∂ z − ∂ A z ∂ x , [ curl A ] z = ∂ A y ∂ x − ∂ A x ∂ y , (119) \begin{aligned}
[\operatorname{curl}\mathbf{A}]_x&=\frac{\partial A_z}{\partial y}-\frac{\partial A_y}{\partial z},\\
[\operatorname{curl}\mathbf{A}]_y&=\frac{\partial A_x}{\partial z}-\frac{\partial A_z}{\partial x},\\
[\operatorname{curl}\mathbf{A}]_z&=\frac{\partial A_y}{\partial x}-\frac{\partial A_x}{\partial y}, \tag*{(119)}
\end{aligned} [ curl A ] x [ curl A ] y [ curl A ] z = ∂ y ∂ A z − ∂ z ∂ A y , = ∂ z ∂ A x − ∂ x ∂ A z , = ∂ x ∂ A y − ∂ y ∂ A x , (119)
or, in a more compact form,
[ curl A ] i = ∑ k l ε i k l ∂ A l ∂ x k . (120) [\operatorname{curl}\mathbf{A}]_i=\sum_{kl}\varepsilon_{ikl}\frac{\partial A_l}{\partial x_k}. \tag*{(120)} [ curl A ] i = k l ∑ ε ik l ∂ x k ∂ A l . (120)
Moreover, curl A \operatorname{curl}\mathbf{A} curl A may also be written in the form
curl A = ∣ e x e y e z ∂ ∂ x ∂ ∂ y ∂ ∂ z A x A y A z ∣ . (121) \operatorname{curl}\mathbf{A}=
\begin{vmatrix}
\mathbf{e}_x & \mathbf{e}_y & \mathbf{e}_z\\
\frac{\partial}{\partial x} & \frac{\partial}{\partial y} & \frac{\partial}{\partial z}\\
A_x & A_y & A_z
\end{vmatrix}. \tag*{(121)} curl A = e x ∂ x ∂ A x e y ∂ y ∂ A y e z ∂ z ∂ A z . (121)
The spherical components of curl A \operatorname{curl}\mathbf{A} curl A are given by
[ curl A ] + 1 = i ( ∇ 0 A + 1 − ∇ + 1 A 0 ) , [ curl A ] 0 = i ( ∇ − 1 A + 1 − ∇ + 1 A − 1 ) , [ curl A ] − 1 = i ( ∇ − 1 A 0 − ∇ 0 A − 1 ) , (122) \begin{aligned}
[\operatorname{curl}\mathbf{A}]_{+1}&=i(\nabla_0A_{+1}-\nabla_{+1}A_0),\\
[\operatorname{curl}\mathbf{A}]_0&=i(\nabla_{-1}A_{+1}-\nabla_{+1}A_{-1}),\\
[\operatorname{curl}\mathbf{A}]_{-1}&=i(\nabla_{-1}A_0-\nabla_0A_{-1}), \tag*{(122)}
\end{aligned} [ curl A ] + 1 [ curl A ] 0 [ curl A ] − 1 = i ( ∇ 0 A + 1 − ∇ + 1 A 0 ) , = i ( ∇ − 1 A + 1 − ∇ + 1 A − 1 ) , = i ( ∇ − 1 A 0 − ∇ 0 A − 1 ) , (122)
or in a more compact form involving the Clebsch-Gordan coefficients
[ curl A ] μ = − i 2 ∑ ν λ C 1 ν 1 λ 1 μ ∇ ν A λ ( μ , ν , λ = ± 1 , 0 ) (123) [\operatorname{curl}\mathbf{A}]_\mu=-i\sqrt{2}\sum_{\nu\lambda}C^{1\mu}_{1\nu\,1\lambda}\nabla_\nu A_\lambda\qquad(\mu,\nu,\lambda=\pm1,0) \tag*{(123)} [ curl A ] μ = − i 2 ν λ ∑ C 1 ν 1 λ 1 μ ∇ ν A λ ( μ , ν , λ = ± 1 , 0 ) (123)
The spherical components of curl A \operatorname{curl}\mathbf{A} curl A may also be written as
curl A = i ∣ e + 1 e 0 e − 1 ∇ + 1 ∇ 0 ∇ − 1 A + 1 A 0 A − 1 ∣ . (124) \operatorname{curl}\mathbf{A}=i
\begin{vmatrix}
\mathbf{e}_{+1} & \mathbf{e}_0 & \mathbf{e}_{-1}\\
\nabla_{+1} & \nabla_0 & \nabla_{-1}\\
A_{+1} & A_0 & A_{-1}
\end{vmatrix}. \tag*{(124)} curl A = i e + 1 ∇ + 1 A + 1 e 0 ∇ 0 A 0 e − 1 ∇ − 1 A − 1 . (124)
The polar components of curl A \mathbf{A} A read
[ curl A ] r = 1 r sin ϑ ∂ ∂ ϑ ( sin ϑ A φ ) − 1 r sin ϑ ∂ A ϑ ∂ φ , [\operatorname{curl}\mathbf{A}]_{r}=\frac{1}{r\sin\vartheta}\frac{\partial}{\partial\vartheta}(\sin\vartheta A_{\varphi})-\frac{1}{r\sin\vartheta}\frac{\partial A_{\vartheta}}{\partial\varphi}, [ curl A ] r = r sin ϑ 1 ∂ ϑ ∂ ( sin ϑ A φ ) − r sin ϑ 1 ∂ φ ∂ A ϑ ,
[ curl A ] ϑ = 1 r sin ϑ ∂ A r ∂ φ − 1 r ∂ ∂ r ( r A φ ) , [\operatorname{curl}\mathbf{A}]_{\vartheta}=\frac{1}{r\sin\vartheta}\frac{\partial A_{r}}{\partial\varphi}-\frac{1}{r}\frac{\partial}{\partial r}(rA_{\varphi}), [ curl A ] ϑ = r sin ϑ 1 ∂ φ ∂ A r − r 1 ∂ r ∂ ( r A φ ) ,
[ curl A ] φ = 1 r ∂ ∂ r ( r A ϑ ) − 1 r ∂ A r ∂ ϑ . (125) [\operatorname{curl}\mathbf{A}]_{\varphi}=\frac{1}{r}\frac{\partial}{\partial r}(rA_{\vartheta})-\frac{1}{r}\frac{\partial A_{r}}{\partial\vartheta}. \tag*{(125)} [ curl A ] φ = r 1 ∂ r ∂ ( r A ϑ ) − r 1 ∂ ϑ ∂ A r . (125)
The above equations are summarized in Table 1.3 . Note also the following differential operations of the second order
Table 1.3. Differential operations.
div grad Φ = ∇ ⋅ ( ∇ Φ ) = Δ Φ . (126) \operatorname{div}\operatorname{grad}\Phi=\nabla\cdot(\nabla\Phi)=\Delta\Phi. \tag*{(126)} div grad Φ = ∇ ⋅ ( ∇Φ ) = ΔΦ. (126)
curl grad Φ = ∇ × ( ∇ Φ ) = 0 , (127) \operatorname{curl}\operatorname{grad}\Phi=\nabla\times(\nabla\Phi)=0, \tag*{(127)} curl grad Φ = ∇ × ( ∇Φ ) = 0 , (127)
div curl A = ∇ ⋅ [ ∇ × A ] = 0 , (128) \operatorname{div}\operatorname{curl}\mathbf{A}=\nabla\cdot[\nabla\times\mathbf{A}]=0, \tag*{(128)} div curl A = ∇ ⋅ [ ∇ × A ] = 0 , (128)
curl curl A = ∇ × [ ∇ × A ] = ∇ ( ∇ ⋅ A ) − Δ A = grad div A − Δ A . (129) \operatorname{curl}\operatorname{curl}\mathbf{A}=\nabla\times[\nabla\times\mathbf{A}]=\nabla(\nabla\cdot\mathbf{A})-\Delta\mathbf{A}
=\operatorname{grad}\operatorname{div}\mathbf{A}-\Delta\mathbf{A}. \tag*{(129)} curl curl A = ∇ × [ ∇ × A ] = ∇ ( ∇ ⋅ A ) − Δ A = grad div A − Δ A . (129)
Rotations of Coordinate System
An arbitrary rotation of a coordinate system about the origin is completely specified by three real parameters. The most useful description of rotation is that in terms of the Euler angles α \alpha α , β \beta β , γ \gamma γ . Note that two other sets of parameters are also widely used to describe rotations:
direction of the rotation axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) (2 parameters) and the rotation angle ω \omega ω (1 parameter);
the Cayley-Klein parameters.
Description of Rotations in Terms of the Euler Angles
Any rotation of the coordinate system S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } may be performed by three successive rotations about the coordinate axes (Figure 1.3 )
Figure 1.3. Succession of rotations of a coordinate system according to scheme A.
(a) rotation about the z z z -axis through an angle α ( 0 ≤ α < 2 π ) \alpha(0\leq\alpha<2\pi) α ( 0 ≤ α < 2 π ) ,
(b) rotation about the new y 1 y_{1} y 1 -axis through an angle β ( 0 ≤ β ≤ π ) \beta(0\leq\beta\leq\pi) β ( 0 ≤ β ≤ π ) ,
(c) rotation about the new axis z 2 = z ′ z_{2}=z' z 2 = z ′ through an angle γ ( 0 ≤ γ < 2 π ) \gamma(0\leq\gamma<2\pi) γ ( 0 ≤ γ < 2 π ) .
The same rotation S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } may also be performed by another succession of rotations Figure 1.4 .
Figure 1.4. Succession of rotations of a coordinate system according to scheme B.
B
(a) rotation about the z z z -axis through an angle γ ( 0 ≤ γ < 2 π ) \gamma(0\leq\gamma<2\pi) γ ( 0 ≤ γ < 2 π ) ,
(b) rotation about the initial y y y -axis through an angle β ( 0 ≤ β ≤ π ) \beta(0\leq\beta\leq\pi) β ( 0 ≤ β ≤ π ) ,
(c) rotation about the initial z z z -axis through an angle α ( 0 ≤ α < 2 π ) \alpha(0\leq\alpha<2\pi) α ( 0 ≤ α < 2 π ) .
Here the angles α , β , γ \alpha,\beta,\gamma α , β , γ are the same as those in the first case. The relative orientations of initial and final coordinate axes S { x , y , z } S\{x,y,z\} S { x , y , z } and S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } , obtained in both cases, A and B, are shown in Figure 1.5 .
Figure 1.5. The Euler angles α \alpha α , β \beta β , γ \gamma γ .
The angles α , β , γ \alpha,\beta,\gamma α , β , γ are called the Euler angles. They completely define the rotation of the coordinate system. The inverse rotation which returns the coordinate system S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } back into S { x , y , z } S\{x,y,z\} S { x , y , z } is specified by the Euler angles − γ , − β , − α -\gamma,-\beta,-\alpha − γ , − β , − α , or, equivalently, by the angles π − γ , β , − π − α \pi-\gamma,\beta,-\pi-\alpha π − γ , β , − π − α .
Sometimes the following successive rotations are used to obtain the general rotation of the coordinate system:
(a) rotation about the z z z -axis through an angle α ′ \alpha' α ′ ;
(b) rotation about the new x 1 x_1 x 1 -axis through an angle β ′ \beta' β ′ ;
(c) rotation about the new axis z 2 = z ′ z_2=z' z 2 = z ′ through an angle γ ′ \gamma' γ ′ .
The angles α ′ , β ′ , γ ′ \alpha',\beta',\gamma' α ′ , β ′ , γ ′ describe the same rotation of the coordinate system as the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ if
α ′ = α + π 2 , β ′ = β , γ ′ = γ − π 2 . (130) \alpha'=\alpha+\frac{\pi}{2},\qquad\beta'=\beta,\qquad\gamma'=\gamma-\frac{\pi}{2}. \tag*{(130)} α ′ = α + 2 π , β ′ = β , γ ′ = γ − 2 π . (130)
The absolute value of a vector is invariant with respect to rotations, but the polar angles ϑ , φ \vartheta,\varphi ϑ , φ , which determine the vector direction, change. The relations between angles ϑ , φ \vartheta,\varphi ϑ , φ and ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ which specify vector directions in the initial and final coordinate systems, S { x , y , z } S\{x,y,z\} S { x , y , z } and S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } , are given by
cos ϑ ′ = cos ϑ cos β + sin ϑ sin β cos ( φ − α ) , cot ( φ ′ + γ ) = cot ( φ − α ) cos β − cot ϑ sin β sin ( φ − α ) . (131) \begin{aligned}
\cos\vartheta' &= \cos\vartheta\cos\beta+\sin\vartheta\sin\beta\cos(\varphi-\alpha),\\
\cot(\varphi'+\gamma) &= \cot(\varphi-\alpha)\cos\beta-\frac{\cot\vartheta\sin\beta}{\sin(\varphi-\alpha)}. \tag*{(131)}
\end{aligned} cos ϑ ′ cot ( φ ′ + γ ) = cos ϑ cos β + sin ϑ sin β cos ( φ − α ) , = cot ( φ − α ) cos β − sin ( φ − α ) cot ϑ sin β . (131)
The inverse relations are
cos ϑ = cos ϑ ′ cos β − sin ϑ ′ sin β cos ( φ ′ + γ ) , cot ( φ − α ) = cot ( φ ′ + γ ) cos β + cot ϑ ′ sin β sin ( φ ′ + γ ) . (132) \begin{aligned}
\cos\vartheta&= \cos\vartheta'\cos\beta-\sin\vartheta'\sin\beta\cos(\varphi'+\gamma),\\
\cot(\varphi-\alpha) &= \cot(\varphi'+\gamma)\cos\beta+\frac{\cot\vartheta'\sin\beta}{\sin(\varphi'+\gamma)}. \tag*{(132)}
\end{aligned} cos ϑ cot ( φ − α ) = cos ϑ ′ cos β − sin ϑ ′ sin β cos ( φ ′ + γ ) , = cot ( φ ′ + γ ) cos β + sin ( φ ′ + γ ) cot ϑ ′ sin β . (132)
Description of Rotations in Terms of Rotation Axis and Rotation Angle
Any rotation of a coordinate system S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } may be treated as one rotation through an angle ω ( 0 ≤ ω ≤ π ) \omega(0\leq\omega\leq\pi) ω ( 0 ≤ ω ≤ π ) about some axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) . The direction of this rotation axis n \mathbf{n} n is defined by the polar angles Θ , Φ ( 0 ≤ Θ ≤ π , 0 ≤ Φ < 2 π ) \Theta,\Phi(0\leq\Theta\leq\pi,0\leq\Phi<2\pi) Θ , Φ ( 0 ≤ Θ ≤ π , 0 ≤ Φ < 2 π ) which are the same in the initial coordinate system S { x , y , z } S\{x,y,z\} S { x , y , z } and in the final one S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } (Figure 1.6 ).
Figure 1.6. Rotation of a coordinate system through an angle ω \omega ω about an axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) .
The angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ completely determine the relative orientation of the initial and final coordinate axes. The rotation defined by the angles − ω -\omega − ω , π − Θ \pi-\Theta π − Θ , π + Φ \pi+\Phi π + Φ is identical to the rotation defined by the angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ . The inverse rotation S ′ { x ′ , y ′ , z ′ } → S { x , y , z } S'\{x',y',z'\}\to S\{x,y,z\} S ′ { x ′ , y ′ , z ′ } → S { x , y , z } is specified by the angles − ω -\omega − ω , Θ \Theta Θ , Φ \Phi Φ or, equivalently, by the angles ω \omega ω , π − Θ \pi-\Theta π − Θ , π + Φ \pi+\Phi π + Φ .
The direction cosines of the unit vector n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) in the initial coordinate system S { x , y , z } S\{x,y,z\} S { x , y , z } coincide with those in the final coordinate system S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } .
n ⋅ e x = n ⋅ e x ′ = sin Θ cos Φ , n ⋅ e y = n ⋅ e y ′ = sin Θ sin Φ , n ⋅ e z = n ⋅ e z ′ = cos Θ . (133) \begin{aligned}
\mathbf{n}\cdot\mathbf{e}_x &= \mathbf{n}\cdot\mathbf{e}'_x=\sin\Theta\cos\Phi,\\
\mathbf{n}\cdot\mathbf{e}_y &= \mathbf{n}\cdot\mathbf{e}'_y=\sin\Theta\sin\Phi,\\
\mathbf{n}\cdot\mathbf{e}_z &= \mathbf{n}\cdot\mathbf{e}'_z=\cos\Theta. \tag*{(133)}
\end{aligned} n ⋅ e x n ⋅ e y n ⋅ e z = n ⋅ e x ′ = sin Θ cos Φ , = n ⋅ e y ′ = sin Θ sin Φ , = n ⋅ e z ′ = cos Θ. (133)
The polar angles ϑ , φ \vartheta,\varphi ϑ , φ of vectors not parallel to the n \mathbf{n} n -axis vary under coordinate rotations. The relations between the polar angles ϑ , φ \vartheta,\varphi ϑ , φ and ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ which specify the direction of a vector with respect to S { x , y , z } S\{x,y,z\} S { x , y , z } and S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } , respectively, are given by
cos ϑ ′ = cos ϑ ( cos ω sin 2 Θ + cos 2 Θ ) + sin ϑ sin Θ [ ( 1 − cos ω ) cos Θ cos ( φ − Φ ) − sin ω sin ( φ − Φ ) ] , cot ( φ ′ − Φ ) = cos ( φ − Φ ) [ cos ω cos 2 Θ + sin 2 Θ ] + sin ( φ − Φ ) sin ω cos Θ − cot ϑ ( cos ω − 1 ) sin Θ cos Θ − cos ( φ − Φ ) sin ω cos Θ + sin ( φ − Φ ) cos ω + cot ϑ sin ω sin Θ . (134) \begin{aligned}
\cos\vartheta'={}&\cos\vartheta\left(\cos\omega\sin^{2}\Theta+\cos^{2}\Theta\right)\\
&+\sin\vartheta\sin\Theta\left[(1-\cos\omega)\cos\Theta\cos(\varphi-\Phi)-\sin\omega\sin(\varphi-\Phi)\right],\\
\cot(\varphi'-\Phi)={}&\frac{\cos(\varphi-\Phi)\left[\cos\omega\cos^{2}\Theta+\sin^{2}\Theta\right]+\sin(\varphi-\Phi)\sin\omega\cos\Theta-\cot\vartheta(\cos\omega-1)\sin\Theta\cos\Theta}{-\cos(\varphi-\Phi)\sin\omega\cos\Theta+\sin(\varphi-\Phi)\cos\omega+\cot\vartheta\sin\omega\sin\Theta}. \tag*{(134)}
\end{aligned} cos ϑ ′ = cot ( φ ′ − Φ ) = cos ϑ ( cos ω sin 2 Θ + cos 2 Θ ) + sin ϑ sin Θ [ ( 1 − cos ω ) cos Θ cos ( φ − Φ ) − sin ω sin ( φ − Φ ) ] , − cos ( φ − Φ ) sin ω cos Θ + sin ( φ − Φ ) cos ω + cot ϑ sin ω sin Θ cos ( φ − Φ ) [ cos ω cos 2 Θ + sin 2 Θ ] + sin ( φ − Φ ) sin ω cos Θ − cot ϑ ( cos ω − 1 ) sin Θ cos Θ . (134)
The inverse relations are
cos ϑ = cos ϑ ′ ( cos ω sin 2 Θ + cos 2 Θ ) + sin ϑ ′ sin Θ [ ( 1 − cos ω ) cos Θ cos ( φ ′ − Φ ) + sin ω sin ( φ ′ − Φ ) ] . \cos\vartheta=\cos\vartheta' \left(\cos\omega\sin^{2}\Theta+\cos^{2}\Theta\right)+\sin\vartheta'\sin\Theta\left[(1-\cos\omega)\cos\Theta\cos(\varphi'-\Phi)+\sin\omega\sin(\varphi'-\Phi)\right]. cos ϑ = cos ϑ ′ ( cos ω sin 2 Θ + cos 2 Θ ) + sin ϑ ′ sin Θ [ ( 1 − cos ω ) cos Θ cos ( φ ′ − Φ ) + sin ω sin ( φ ′ − Φ ) ] .
cot ( φ − Φ ) = cos ( φ ′ − Φ ) [ cos ω cos 2 Θ + sin 2 Θ ] − sin ( φ ′ − Φ ) sin ω cos Θ − cot ϑ ′ ( cos ω − 1 ) sin Θ cos Θ cos ( φ ′ − Φ ) sin ω cos Θ + sin ( φ ′ − Φ ) cos ω − cot ϑ ′ sin ω sin Θ . (135) \cot(\varphi-\Phi)=\frac{\cos(\varphi'-\Phi)\left[\cos\omega\cos^{2}\Theta+\sin^{2}\Theta\right]-\sin(\varphi'-\Phi)\sin\omega\cos\Theta-\cot\vartheta'(\cos\omega-1)\sin\Theta\cos\Theta}{\cos(\varphi'-\Phi)\sin\omega\cos\Theta+\sin(\varphi'-\Phi)\cos\omega-\cot\vartheta'\sin\omega\sin\Theta}. \tag*{(135)} cot ( φ − Φ ) = cos ( φ ′ − Φ ) sin ω cos Θ + sin ( φ ′ − Φ ) cos ω − cot ϑ ′ sin ω sin Θ cos ( φ ′ − Φ ) [ cos ω cos 2 Θ + sin 2 Θ ] − sin ( φ ′ − Φ ) sin ω cos Θ − cot ϑ ′ ( cos ω − 1 ) sin Θ cos Θ . (135)
Introducing parameters ω , n ( θ , Φ ) \omega,\mathbf{n}(\theta,\Phi) ω , n ( θ , Φ ) to describe rotations, we are able to write the transformation properties of components of the position vector r \mathbf{r} r in compact vector form:
r ′ = r cos ω + n ( n r ) ( 1 − cos ω ) + [ n × r ] sin ω , \mathbf{r}'=\mathbf{r}\cos\omega+\mathbf{n}(\mathbf{n}\mathbf{r})(1-\cos\omega)+[\mathbf{n}\times\mathbf{r}]\sin\omega, r ′ = r cos ω + n ( nr ) ( 1 − cos ω ) + [ n × r ] sin ω ,
r = r ′ cos ω + n ( n r ′ ) ( 1 − cos ω ) − [ n × r ′ ] sin ω . (136) \mathbf{r}=\mathbf{r}'\cos\omega+\mathbf{n}(\mathbf{n}\mathbf{r}')(1-\cos\omega)-[\mathbf{n}\times\mathbf{r}']\sin\omega. \tag*{(136)} r = r ′ cos ω + n ( n r ′ ) ( 1 − cos ω ) − [ n × r ′ ] sin ω . (136)
Equations (1.4.5)–(1.4.6) may be derived by projecting Eq. (136) onto the coordinate axes.
Description of Rotations in Terms of Unitary 2 × 2 Matrices. Cayley-Klein Parameters.
The position vector of an arbitrary point r = x e x + y e y + z e z \mathbf{r}=x\mathbf{e}_{x}+y\mathbf{e}_{y}+z\mathbf{e}_{z} r = x e x + y e y + z e z may be represented by the following Hermitian 2 × 2 2\times2 2 × 2 matrix X X X :
X = X + = ( z x + i y x − i y − z ) = ∑ i = x , y , z x i σ ~ i , (137) X=X^{+}=\begin{pmatrix} z & x+iy \\ x-iy & -z \end{pmatrix}=\sum_{i=x,y,z}x_i\widetilde{\sigma}_i, \tag*{(137)} X = X + = ( z x − i y x + i y − z ) = i = x , y , z ∑ x i σ i , (137)
where σ ~ i ( i = x , y , z ) \widetilde{\sigma}_i(i=x,y,z) σ i ( i = x , y , z ) are the transposed Pauli matrices (Eq. (298) ). Note that
− det X = r 2 = x 2 + y 2 + z 2 -\det X=r^{2}=x^{2}+y^{2}+z^{2} − det X = r 2 = x 2 + y 2 + z 2
Each rotation S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } may be represented by a unitary transformation U U U of matrix X X X into X ′ X' X ′
X ′ = U X U − 1 . (138) X'=UXU^{-1}. \tag*{(138)} X ′ = U X U − 1 . (138)
Here U U U is the unitary unimodular 2 × 2 2\times2 2 × 2 matrix
U + = U − 1 , det U = 1. (139) U^{+}=U^{-1},\qquad\det U=1. \tag*{(139)} U + = U − 1 , det U = 1. (139)
Bearing in mind that r \mathbf{r} r is real, one can easily prove that U U U , indeed, is unitary. The relation det U = 1 \det U=1 det U = 1 ensures the invariance of the absolute value of the position vector under coordinate rotations.
Equations (1.4.10) imply the following form of U U U
U = ( a b − b ∗ a ∗ ) (140) U = \begin{pmatrix} a & b \\ -b^{*} & a^{*} \end{pmatrix} \tag*{(140)} U = ( a − b ∗ b a ∗ ) (140)
where a a a and b b b are complex numbers which satisfy the condition
∣ a ∣ 2 + ∣ b ∣ 2 = 1 (141) \lvert a\rvert^{2} + \lvert b\rvert^{2} = 1 \tag*{(141)} ∣ a ∣ 2 + ∣ b ∣ 2 = 1 (141)
Thus, the matrix U U U depends on three real independent parameters. The numbers a a a and b b b are called the Cayley-Klein parameters. They uniquely determine rotation of the coordinate system. The inverse statement is not true because the parameters − a -a − a and − b -b − b describe the same rotation as a a a and b b b .
The inverse rotation S ′ { x ′ , y ′ , z ′ } → S { x , y , z } S^{\prime}\{x^{\prime},y^{\prime},z^{\prime}\} \to S\{x,y,z\} S ′ { x ′ , y ′ , z ′ } → S { x , y , z } corresponds to the matrix
U − 1 = U + = ( a ∗ − b b ∗ a ) . (142) U^{-1} = U^{+} = \begin{pmatrix} a^{*} & -b \\ b^{*} & a \end{pmatrix}. \tag*{(142)} U − 1 = U + = ( a ∗ b ∗ − b a ) . (142)
The Cayley-Klein parameters permit us to propose an alternative interpretation of coordinate rotations. Let us consider a sphere (of unit diameter) about the origin. Each point of the sphere with coordinates x , y , z x,y,z x , y , z (x 2 + y 2 + z 2 = 1 / 4 x^{2}+y^{2}+z^{2}=1/4 x 2 + y 2 + z 2 = 1/4 ) corresponds to the point ζ = ξ + i η \zeta=\xi+i\eta ζ = ξ + i η on the complex plane which is called the stereographic projection of the point x , y , z x,y,z x , y , z (Figure 1.7 ). The complex number ζ \zeta ζ is related to x , y , z x,y,z x , y , z by
Figure 1.7. Stereographic projection of a point on a sphere.
ζ = x + i y 1 2 − z = 1 2 + z x − i y (143) \zeta= \frac{x+iy}{\frac{1}{2}-z} = \frac{\frac{1}{2}+z}{x-iy} \tag*{(143)} ζ = 2 1 − z x + i y = x − i y 2 1 + z (143)
The rotation of the coordinate system which transforms the coordinates x , y , z x,y,z x , y , z of the point on the sphere into x ′ , y ′ , z ′ x^{\prime},y^{\prime},z^{\prime} x ′ , y ′ , z ′ generates the following complex-plane bilinear transformation
ζ ′ = a ζ + b − b ∗ ζ + a ∗ . (144) \zeta^{\prime} = \frac{a\zeta+b}{-b^{*}\zeta+a^{*}}. \tag*{(144)} ζ ′ = − b ∗ ζ + a ∗ a ζ + b . (144)
The coefficients of this bilinear transformation are just the Cayley-Klein parameters a a a and b b b .
Relations Between Different Descriptions of Rotations
Relations Between Angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ and Euler Angles α \alpha α , β \beta β , γ \gamma γ
The angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ are expressed in terms of the Euler angles α \alpha α , β \beta β , γ \gamma γ by
cos ω 2 = cos β 2 cos α + γ 2 , tan Θ = tan β 2 sin α + γ 2 , Φ = π 2 + α − γ 2 . (145) \begin{aligned}
\cos\frac{\omega}{2} &= \cos\frac{\beta}{2}\cos\frac{\alpha+\gamma}{2},\\
\tan\Theta&= \frac{\tan\frac{\beta}{2}}{\sin\frac{\alpha+\gamma}{2}},\\
\Phi&= \frac{\pi}{2}+\frac{\alpha-\gamma}{2}. \tag*{(145)}
\end{aligned} cos 2 ω tan Θ Φ = cos 2 β cos 2 α + γ , = sin 2 α + γ tan 2 β , = 2 π + 2 α − γ . (145)
The inverse relations are written as
sin β 2 = sin Θ sin ω 2 , tan α + γ 2 = cos Θ tan ω 2 , α − γ 2 = Φ − π 2 . (146) \begin{aligned}
\sin\frac{\beta}{2} &= \sin\Theta\sin\frac{\omega}{2},\\
\tan\frac{\alpha+\gamma}{2} &= \cos\Theta\tan\frac{\omega}{2},\\
\frac{\alpha-\gamma}{2} &= \Phi-\frac{\pi}{2}. \tag*{(146)}
\end{aligned} sin 2 β tan 2 α + γ 2 α − γ = sin Θ sin 2 ω , = cos Θ tan 2 ω , = Φ − 2 π . (146)
Note also the following useful relations between the angles α \alpha α , β \beta β , γ \gamma γ and ω \omega ω , Θ \Theta Θ , Φ \Phi Φ .
∂ α ∂ ω = ∂ γ ∂ ω = cos Θ 2 cos 2 β 2 , ∂ β ∂ ω = sin 2 Θ sin ω sin β , ∂ α ∂ Θ = ∂ γ ∂ Θ = − 1 2 tan Θ sin ( α + γ ) , ∂ β ∂ Θ = 2 sin 2 Θ sin 2 ω 2 sin β , ∂ α ∂ Φ = − ∂ γ ∂ Φ = 1 , ∂ β ∂ Φ = 0. (147) \begin{aligned}
\frac{\partial\alpha}{\partial\omega}=\frac{\partial\gamma}{\partial\omega}&=\frac{\cos\Theta}{2\cos^2\frac{\beta}{2}},&
\frac{\partial\beta}{\partial\omega}&=\frac{\sin^2\Theta\sin\omega}{\sin\beta},\\
\frac{\partial\alpha}{\partial\Theta}=\frac{\partial\gamma}{\partial\Theta}&=-\frac{1}{2}\tan\Theta\sin(\alpha+\gamma),&
\frac{\partial\beta}{\partial\Theta}&=\frac{2\sin2\Theta\sin^2\frac{\omega}{2}}{\sin\beta},\\
\frac{\partial\alpha}{\partial\Phi}=-\frac{\partial\gamma}{\partial\Phi}&=1,&
\frac{\partial\beta}{\partial\Phi}&=0. \tag*{(147)}
\end{aligned} ∂ ω ∂ α = ∂ ω ∂ γ ∂ Θ ∂ α = ∂ Θ ∂ γ ∂ Φ ∂ α = − ∂ Φ ∂ γ = 2 cos 2 2 β cos Θ , = − 2 1 tan Θ sin ( α + γ ) , = 1 , ∂ ω ∂ β ∂ Θ ∂ β ∂ Φ ∂ β = sin β sin 2 Θ sin ω , = sin β 2 sin 2Θ sin 2 2 ω , = 0. (147)
∂ ω ∂ α = ∂ ω ∂ γ = cos Θ , ∂ ω ∂ β = tan β 2 tan ω 2 , ∂ Θ ∂ α = ∂ Θ ∂ γ = − sin Θ 2 tan ω 2 , ∂ Θ ∂ β = cos Θ sin Θ sin β , ∂ Φ ∂ α = − ∂ Φ ∂ γ = 1 2 , ∂ Φ ∂ β = 0. (148) \begin{aligned}
\frac{\partial\omega}{\partial\alpha}=\frac{\partial\omega}{\partial\gamma}&=\cos\Theta,&
\frac{\partial\omega}{\partial\beta}&=\frac{\tan\frac{\beta}{2}}{\tan\frac{\omega}{2}},\\
\frac{\partial\Theta}{\partial\alpha}=\frac{\partial\Theta}{\partial\gamma}&=-\frac{\sin\Theta}{2\tan\frac{\omega}{2}},&
\frac{\partial\Theta}{\partial\beta}&=\frac{\cos\Theta\sin\Theta}{\sin\beta},\\
\frac{\partial\Phi}{\partial\alpha}=-\frac{\partial\Phi}{\partial\gamma}&=\frac{1}{2},&
\frac{\partial\Phi}{\partial\beta}&=0. \tag*{(148)}
\end{aligned} ∂ α ∂ ω = ∂ γ ∂ ω ∂ α ∂ Θ = ∂ γ ∂ Θ ∂ α ∂ Φ = − ∂ γ ∂ Φ = cos Θ , = − 2 tan 2 ω sin Θ , = 2 1 , ∂ β ∂ ω ∂ β ∂ Θ ∂ β ∂ Φ = tan 2 ω tan 2 β , = sin β cos Θ sin Θ , = 0. (148)
The Jacobian of the transformation is equal to
∥ ∂ ( α , β , γ ) ∂ ( ω , Θ , Φ ) ∥ = ∥ ∂ ( ω , Θ , Φ ) ∂ ( α , β , γ ) ∥ − 1 = 4 sin Θ sin β sin 2 ω 2 . (149) \left\lVert\frac{\partial(\alpha,\beta,\gamma)}{\partial(\omega,\Theta,\Phi)}\right\rVert
=
\left\lVert\frac{\partial(\omega,\Theta,\Phi)}{\partial(\alpha,\beta,\gamma)}\right\rVert^{-1}
=
\frac{4\sin\Theta}{\sin\beta}\sin^2\frac{\omega}{2}. \tag*{(149)} ∂ ( ω , Θ , Φ ) ∂ ( α , β , γ ) = ∂ ( α , β , γ ) ∂ ( ω , Θ , Φ ) − 1 = sin β 4 sin Θ sin 2 2 ω . (149)
A volume element of the three-dimensional rotation group is given by
d R ≡ sin β d α d β d γ = 4 sin 2 ω 2 d ω sin Θ d Θ d Φ (150) dR\equiv\sin\beta\,d\alpha\,d\beta\,d\gamma
=4\sin^2\frac{\omega}{2}\,d\omega\,\sin\Theta\,d\Theta\,d\Phi \tag*{(150)} d R ≡ sin β d α d β d γ = 4 sin 2 2 ω d ω sin Θ d Θ d Φ (150)
The total volume of the three-dimensional rotation group is equal to
∫ d R = ∫ 0 2 π d α ∫ 0 π sin β d β ∫ 0 2 π d γ = 4 ∫ 0 π sin 2 ω 2 d ω ∫ 0 π sin Θ d Θ ∫ 0 2 π d Φ = 8 π 2 . (151) \int dR
=\int_{0}^{2\pi}d\alpha\int_{0}^{\pi}\sin\beta\,d\beta\int_{0}^{2\pi}d\gamma
=4\int_{0}^{\pi}\sin^2\frac{\omega}{2}\,d\omega\int_{0}^{\pi}\sin\Theta\,d\Theta\int_{0}^{2\pi}d\Phi
=8\pi^2. \tag*{(151)} ∫ d R = ∫ 0 2 π d α ∫ 0 π sin β d β ∫ 0 2 π d γ = 4 ∫ 0 π sin 2 2 ω d ω ∫ 0 π sin Θ d Θ ∫ 0 2 π d Φ = 8 π 2 . (151)
Relations Between Cayley-Klein Parameters a a a , b b b and Euler Angles α \alpha α , β \beta β , γ \gamma γ
The parameters a a a and b b b are expressed in terms of the Euler angles α \alpha α , β \beta β , γ \gamma γ as
a = cos β 2 e i α + γ 2 , b = sin β 2 e i α − γ 2 . (152) a=\cos\frac{\beta}{2}e^{i\frac{\alpha+\gamma}{2}},
\qquad
b=\sin\frac{\beta}{2}e^{i\frac{\alpha-\gamma}{2}}. \tag*{(152)} a = cos 2 β e i 2 α + γ , b = sin 2 β e i 2 α − γ . (152)
while the inverse relations are
cos β = ∣ a ∣ 2 − ∣ b ∣ 2 , cos β 2 = ∣ a ∣ , sin β 2 = ∣ b ∣ \cos\beta=|a|^2-|b|^2,
\qquad
\cos\frac{\beta}{2}=|a|,
\qquad
\sin\frac{\beta}{2}=|b| cos β = ∣ a ∣ 2 − ∣ b ∣ 2 , cos 2 β = ∣ a ∣ , sin 2 β = ∣ b ∣
cot α + γ 2 = − Re a Im a , cot α − γ 2 = Re b Im b . (153) \cot\frac{\alpha+\gamma}{2}=-\frac{\operatorname{Re}a}{\operatorname{Im}a},
\qquad
\cot\frac{\alpha-\gamma}{2}=\frac{\operatorname{Re}b}{\operatorname{Im}b}. \tag*{(153)} cot 2 α + γ = − Im a Re a , cot 2 α − γ = Im b Re b . (153)
The parameters a a a and b b b may be expressed in terms of the Wigner D D D -functions (see Chap. 4)
a = D 1 2 1 2 1 2 ( α , β , γ ) , b = D − 1 2 1 2 1 2 ( α , β , γ ) . (154) a = D^{\frac{1}{2}}_{\frac{1}{2}\frac{1}{2}}(\alpha,\beta,\gamma),\qquad b = D^{\frac{1}{2}}_{-\frac{1}{2}\frac{1}{2}}(\alpha,\beta,\gamma). \tag*{(154)} a = D 2 1 2 1 2 1 ( α , β , γ ) , b = D − 2 1 2 1 2 1 ( α , β , γ ) . (154)
The unitary matrix U U U , according to Eq. (140) , coincides with the transposed rotation matrix D ^ 1 2 ( α , β , γ ) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma) D ^ 2 1 ( α , β , γ ) which transforms spin functions of particles of spin 1 2 \frac{1}{2} 2 1 (see Eq. (326) ).
(c) Relations Between Cayley-Klein Parameters a a a , b b b and Angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ
The parameters a a a , b b b are expressed in terms of the angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ as
a = cos ω 2 − i sin ω 2 cos Θ , a = \cos\frac{\omega}{2}-i\sin\frac{\omega}{2}\cos\Theta, a = cos 2 ω − i sin 2 ω cos Θ ,
b = − i sin ω 2 sin Θ e i Φ . (155) b = -i\sin\frac{\omega}{2}\sin\Theta e^{i\Phi}. \tag*{(155)} b = − i sin 2 ω sin Θ e i Φ . (155)
and the inverse relations have the form
cos ω = 2 ( Re a ) 2 − 1 , cos ω 2 = Re a \cos\omega= 2(\operatorname{Re}a)^2-1,\qquad\cos\frac{\omega}{2}=\operatorname{Re}a cos ω = 2 ( Re a ) 2 − 1 , cos 2 ω = Re a
cos Θ = − Im a 1 − ( Re a ) 2 , \cos\Theta=-\frac{\operatorname{Im}a}{\sqrt{1-(\operatorname{Re}a)^2}}, cos Θ = − 1 − ( Re a ) 2 Im a ,
cot Φ = − Im b Re b . (156) \cot\Phi=-\frac{\operatorname{Im}b}{\operatorname{Re}b}. \tag*{(156)} cot Φ = − Re b Im b . (156)
The unitary matrix U U U , according to Eq. (140) , coincides with the transposed rotation matrix U ^ 1 2 ( ω ; Θ , Φ ) \hat{U}^{\frac{1}{2}}(\omega;\Theta,\Phi) U ^ 2 1 ( ω ; Θ , Φ ) which transforms spin functions of spin 1 2 \frac{1}{2} 2 1 (see Eq. (330) ).
Rotation Operator
Under rotations of coordinate systems quantum-mechanical quantities are transformed by the rotation operators D ^ ( α , β , γ ) \hat{D}(\alpha,\beta,\gamma) D ^ ( α , β , γ ) or U ^ ( ω ; Θ , Φ ) \hat{U}(\omega;\Theta,\Phi) U ^ ( ω ; Θ , Φ ) .
Wave functions (state vectors) Ψ ′ \Psi' Ψ ′ and operators O ^ ′ \hat{O}' O ^ ′ in a rotated coordinate system are related to wave functions Ψ \Psi Ψ and operators O ^ \hat{O} O ^ in an initial coordinate system by
Ψ ′ = D ^ ( α , β , γ ) Ψ , O ^ ′ = D ^ ( α , β , γ ) O ^ [ D ^ ( α , β , γ ) ] − 1 , (157) \Psi'=\hat{D}(\alpha,\beta,\gamma)\Psi,\qquad\hat{O}'=\hat{D}(\alpha,\beta,\gamma)\hat{O}[\hat{D}(\alpha,\beta,\gamma)]^{-1}, \tag*{(157)} Ψ ′ = D ^ ( α , β , γ ) Ψ , O ^ ′ = D ^ ( α , β , γ ) O ^ [ D ^ ( α , β , γ ) ] − 1 , (157)
Ψ ′ = U ^ ( ω ; Θ , Φ ) Ψ , O ^ ′ = U ^ ( ω ; Θ , Φ ) O ^ [ U ^ ( ω ; Θ , Φ ) ] − 1 . (158) \Psi'=\hat{U}(\omega;\Theta,\Phi)\Psi,\qquad\hat{O}'=\hat{U}(\omega;\Theta,\Phi)\hat{O}[\hat{U}(\omega;\Theta,\Phi)]^{-1}. \tag*{(158)} Ψ ′ = U ^ ( ω ; Θ , Φ ) Ψ , O ^ ′ = U ^ ( ω ; Θ , Φ ) O ^ [ U ^ ( ω ; Θ , Φ ) ] − 1 . (158)
If the Euler angles α \alpha α , β \beta β , γ \gamma γ are chosen to describe the rotation, then the rotation operator D ^ ( α , β , γ ) \hat{D}(\alpha,\beta,\gamma) D ^ ( α , β , γ ) may be written as
D ^ ( α , β , γ ) = e − i γ J ^ z ′ e − i β J ^ y ′ e − i α J ^ z (159) \hat{D}(\alpha,\beta,\gamma)=e^{-i\gamma\hat{J}_{z'}}e^{-i\beta\hat{J}_{y'}}e^{-i\alpha\hat{J}_{z}} \tag*{(159)} D ^ ( α , β , γ ) = e − iγ J ^ z ′ e − i β J ^ y ′ e − i α J ^ z (159)
or, equivalently,
D ^ ( α , β , γ ) = e − i α J ^ z e − i β J ^ y e − i γ J ^ z . (160) \hat{D}(\alpha,\beta,\gamma)=e^{-i\alpha\hat{J}_{z}}e^{-i\beta\hat{J}_{y}}e^{-i\gamma\hat{J}_{z}}. \tag*{(160)} D ^ ( α , β , γ ) = e − i α J ^ z e − i β J ^ y e − iγ J ^ z . (160)
Here J ^ i \hat{J}_i J ^ i is the projection of the total angular momentum operator (see Chap. 2) on an i i i -axis. The equivalence of Eqs. (159) and (160) follows from the fact that, according to (1.4.28),
e − i β J ^ y ′ = D ^ ( α , 0 , 0 ) e − i β J ^ y [ D ^ ( α , 0 , 0 ) ] − 1 = e − i α J ^ z e − i β J ^ y e i α J ^ z , (161) e^{-i\beta\hat{J}_{y'}}=\hat{D}(\alpha,0,0)e^{-i\beta\hat{J}_{y}}[\hat{D}(\alpha,0,0)]^{-1}=e^{-i\alpha\hat{J}_{z}}e^{-i\beta\hat{J}_{y}}e^{i\alpha\hat{J}_{z}}, \tag*{(161)} e − i β J ^ y ′ = D ^ ( α , 0 , 0 ) e − i β J ^ y [ D ^ ( α , 0 , 0 ) ] − 1 = e − i α J ^ z e − i β J ^ y e i α J ^ z , (161)
e − i γ J ^ z ′ = D ^ ( α , β , 0 ) e − i γ J ^ z [ D ^ ( α , β , 0 ) ] − 1 = e − i α J ^ z e − i β J ^ y e − i γ J ^ z e i β J ^ y e i α J ^ z . e^{-i\gamma\hat{J}_{z'}}=\hat{D}(\alpha,\beta,0)e^{-i\gamma\hat{J}_{z}}[\hat{D}(\alpha,\beta,0)]^{-1}=e^{-i\alpha\hat{J}_{z}}e^{-i\beta\hat{J}_{y}}e^{-i\gamma\hat{J}_{z}}e^{i\beta\hat{J}_{y}}e^{i\alpha\hat{J}_{z}}. e − iγ J ^ z ′ = D ^ ( α , β , 0 ) e − iγ J ^ z [ D ^ ( α , β , 0 ) ] − 1 = e − i α J ^ z e − i β J ^ y e − iγ J ^ z e i β J ^ y e i α J ^ z .
If the direction of the rotation axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) and the rotation angle ω \omega ω are chosen to describe the coordinate rotation, the rotation operator U ^ \hat{U} U ^ may be written in the form
U ^ ( ω ; Θ , Φ ) = e − i ω n ⋅ J ^ , (162) \hat{U}(\omega;\Theta,\Phi)=e^{-i\omega\mathbf{n}\cdot\hat{\mathbf{J}}}, \tag*{(162)} U ^ ( ω ; Θ , Φ ) = e − iω n ⋅ J ^ , (162)
where J ^ \hat{\mathbf{J}} J ^ is the total angular momentum operator (Chap. 2). Note that D ^ ( α , β , γ ) ≡ U ^ ( ω ; Θ , Φ ) \hat{D}(\alpha,\beta,\gamma)\equiv\hat{U}(\omega;\Theta,\Phi) D ^ ( α , β , γ ) ≡ U ^ ( ω ; Θ , Φ ) . The rotation operator written in the forms (1.4.30), (1.4.31) or (1.4.33) is an unitary operator.
D ^ + ( α , β , γ ) = [ D ^ ( α , β , γ ) ] − 1 = D ^ ( π − γ , β , − π − α ) = D ^ ( − γ , − β , − α ) , \hat{D}^{+}(\alpha,\beta,\gamma)=[\hat{D}(\alpha,\beta,\gamma)]^{-1}=\hat{D}(\pi-\gamma,\beta,-\pi-\alpha)=\hat{D}(-\gamma,-\beta,-\alpha), D ^ + ( α , β , γ ) = [ D ^ ( α , β , γ ) ] − 1 = D ^ ( π − γ , β , − π − α ) = D ^ ( − γ , − β , − α ) ,
U ^ + ( ω ; Θ , Φ ) = [ U ^ ( ω ; Θ , Φ ) ] − 1 = U ^ ( ω ; π − Θ , π + Φ ) = U ^ ( − ω ; Θ , Φ ) . (163) \hat{U}^{+}(\omega;\Theta,\Phi)=[\hat{U}(\omega;\Theta,\Phi)]^{-1}=\hat{U}(\omega;\pi-\Theta,\pi+\Phi)=\hat{U}(-\omega;\Theta,\Phi). \tag*{(163)} U ^ + ( ω ; Θ , Φ ) = [ U ^ ( ω ; Θ , Φ ) ] − 1 = U ^ ( ω ; π − Θ , π + Φ ) = U ^ ( − ω ; Θ , Φ ) . (163)
Matrix elements of D ^ \hat{D} D ^ between eigenstates of the operators J ^ 2 \hat{J}^2 J ^ 2 , J ^ z \hat{J}_z J ^ z are the Wigner D D D -functions (see Chap. 4)
⟨ J ′ M ′ ∣ D ^ ( α , β , γ ) ∣ J M ⟩ = δ J ′ J D M ′ M J ( α , β , γ ) . (164) \langle J'M'|\hat{D}(\alpha,\beta,\gamma)|JM\rangle=\delta_{J'J}D^J_{M'M}(\alpha,\beta,\gamma). \tag*{(164)} ⟨ J ′ M ′ ∣ D ^ ( α , β , γ ) ∣ J M ⟩ = δ J ′ J D M ′ M J ( α , β , γ ) . (164)
Matrix elements of D ^ \hat{D} D ^ between states corresponding to the cartesian basis vectors e i \mathbf{e}_i e i (i = x , y , z i=x,y,z i = x , y , z ) coincide with elements of the rotation matrix a i k a_{ik} a ik (see § 9.10.2 )
⟨ e i ∣ D ^ ( α , β , γ ) ∣ e k ⟩ = a i k , ( i , k = x , y , z ) (165) \langle\mathbf{e}_i|\hat{D}(\alpha,\beta,\gamma)|\mathbf{e}_k\rangle=a_{ik},\qquad(i,k=x,y,z) \tag*{(165)} ⟨ e i ∣ D ^ ( α , β , γ ) ∣ e k ⟩ = a ik , ( i , k = x , y , z ) (165)
Effects of the rotation operator on various wave functions and quantum-mechanical operators are considered in Chaps. 3, 5-7.
Transformation of Cartesian Vectors and Tensors Under Rotations of Coordinate Systems. Rotation Matrix A
An arbitrary vector A \mathbf{A} A may be written as a column
A = ( A x A y A z ) , (166) \mathbf{A}=\begin{pmatrix} A_x \\ A_y \\ A_z \end{pmatrix}, \tag*{(166)} A = A x A y A z , (166)
where A x , A y , A z A_x,A_y,A_z A x , A y , A z are cartesian components of A \mathbf{A} A . In this representation the Cartesian basis vectors e x , e y , e z \mathbf{e}_x,\mathbf{e}_y,\mathbf{e}_z e x , e y , e z have the form
e x = ( 1 0 0 ) , e y = ( 0 1 0 ) , e z = ( 0 0 1 ) . (167) \mathbf{e}_x=\begin{pmatrix}1\\0\\0\end{pmatrix},\qquad\mathbf{e}_y=\begin{pmatrix}0\\1\\0\end{pmatrix},\qquad\mathbf{e}_z=\begin{pmatrix}0\\0\\1\end{pmatrix}. \tag*{(167)} e x = 1 0 0 , e y = 0 1 0 , e z = 0 0 1 . (167)
The effect of the rotation operator on the basis vectors written in such a form is equivalent to an action of some 3 × 3 3\times3 3 × 3 matrix a a a which may be regarded as one of the representations of the rotation operator
a = ( a x x a x y a x z a y x a y y a y z a z x a z y a z z ) . (168) a=\begin{pmatrix}a_{xx}&a_{xy}&a_{xz}\\a_{yx}&a_{yy}&a_{yz}\\a_{zx}&a_{zy}&a_{zz}\end{pmatrix}. \tag*{(168)} a = a xx a y x a z x a x y a y y a z y a x z a y z a z z . (168)
The matrix a a a is real
a ∗ = a , a i k ∗ = a i k , ( i , k = x , y , z ) . (169) a^*=a,\qquad a^*_{ik}=a_{ik},\qquad(i,k=x,y,z). \tag*{(169)} a ∗ = a , a ik ∗ = a ik , ( i , k = x , y , z ) . (169)
and unitary
a + a = a a + = 1. (170) a^+a=aa^+=1. \tag*{(170)} a + a = a a + = 1. (170)
Equations (1.4.40) and (1.4.41) result in the orthogonality condition
a ~ a = a a ~ = 1 , (171) \widetilde{a}a=a\widetilde{a}=1, \tag*{(171)} a a = a a = 1 , (171)
where a ~ \widetilde{a} a is the transpose of a a a . Equation (1.4.42) written in a component form gives six independent relations for the elements a i k a_{ik} a ik
∑ i a i k a i l = δ k l , ( i , k , l = x , y , z ) , (172) \sum_i a_{ik}a_{il}=\delta_{kl},\qquad(i,k,l=x,y,z), \tag*{(172)} i ∑ a ik a i l = δ k l , ( i , k , l = x , y , z ) , (172)
or the equivalent relations
∑ k a i k a l k = δ i l , ( i , k , l = x , y , z ) . (173) \sum_k a_{ik}a_{lk}=\delta_{il},\qquad(i,k,l=x,y,z). \tag*{(173)} k ∑ a ik a l k = δ i l , ( i , k , l = x , y , z ) . (173)
Relations (1.4.43) or (1.4.44) reveal that only three of the nine matrix elements a i k a_{ik} a ik are independent. This result is in agreement with the fact that any rotation of the coordinate system is completely determined by three real parameters.
The matrix a a a is unimodular, i.e.,
det a = ∣ a x x a x y a x z a y x a y y a y z a z x a z y a z z ∣ = 1 (174) \det a=\begin{vmatrix}a_{xx}&a_{xy}&a_{xz}\\a_{yx}&a_{yy}&a_{yz}\\a_{zx}&a_{zy}&a_{zz}\end{vmatrix}=1 \tag*{(174)} det a = a xx a y x a z x a x y a y y a z y a x z a y z a z z = 1 (174)
The relations between cartesian basis vectors e i ′ \mathbf{e}'_i e i ′ in a rotated coordinate system S ′ S' S ′ and basis vectors e i \mathbf{e}_i e i in an initial coordinate system S S S are given by
e i ′ = a e i = ∑ k a k i e k , ( i , k = x , y , z ) . (175) \mathbf{e}'_i=a\mathbf{e}_i=\sum_k a_{ki}\mathbf{e}_k,\qquad(i,k=x,y,z). \tag*{(175)} e i ′ = a e i = k ∑ a k i e k , ( i , k = x , y , z ) . (175)
The transformation properties of cartesian vector components are given by
A i ′ = ∑ k a k i A k , ( i , k = x , y , z ) , (176) A'_i=\sum_k a_{ki}A_k,\qquad(i,k=x,y,z), \tag*{(176)} A i ′ = k ∑ a k i A k , ( i , k = x , y , z ) , (176)
where A k A_k A k are the components of A \mathbf{A} A in the initial coordinate system and A i ′ A'_{i} A i ′ are the components of this vector in the rotated coordinate system. Equations (1.4.43) and (1.4.44) ensure that the absolute value of A \mathbf{A} A is unchanged by the rotation.
The transformation rule for cartesian components of a tensor of rank n n n (n n n is integer) has the form
A i 1 i 2 … i n ′ = ∑ k 1 k 2 … k n a k 1 i 1 a k 2 i 2 … a k n i n A k 1 k 2 … k n . (177) A'_{i_1i_2\ldots i_n}=\sum_{k_1k_2\ldots k_n}a_{k_1i_1}a_{k_2i_2}\ldots a_{k_ni_n}A_{k_1k_2\ldots k_n}. \tag*{(177)} A i 1 i 2 … i n ′ = k 1 k 2 … k n ∑ a k 1 i 1 a k 2 i 2 … a k n i n A k 1 k 2 … k n . (177)
The inverse transformation which corresponds to the rotation S ′ → S S'\to S S ′ → S is performed by the transposed matrix a ~ = a − 1 \tilde{a}=a^{-1} a ~ = a − 1 . The inverse relations are
e k = ∑ i a k i e i ′ , ( i , k = x , y , z ) , (178) \mathbf{e}_k=\sum_i a_{ki}\mathbf{e}'_i,\qquad(i,k=x,y,z), \tag*{(178)} e k = i ∑ a k i e i ′ , ( i , k = x , y , z ) , (178)
A k = ∑ i a k i A i ′ , ( i , k = x , y , z ) , (179) A_k=\sum_i a_{ki}A'_i,\qquad(i,k=x,y,z), \tag*{(179)} A k = i ∑ a k i A i ′ , ( i , k = x , y , z ) , (179)
A k 1 k 2 … k n = ∑ i 1 i 2 … i n a k 1 i 1 a k 2 i 2 … a k n i n A i 1 i 2 … i n ′ , ( i 1 , k 1 , i 2 , k 2 , … , i n , k n = x , y , z ) . (180) A_{k_1k_2\ldots k_n}=\sum_{i_1i_2\ldots i_n}a_{k_1i_1}a_{k_2i_2}\ldots a_{k_ni_n}A'_{i_1i_2\ldots i_n},\qquad(i_1,k_1,i_2,k_2,\ldots,i_n,k_n=x,y,z). \tag*{(180)} A k 1 k 2 … k n = i 1 i 2 … i n ∑ a k 1 i 1 a k 2 i 2 … a k n i n A i 1 i 2 … i n ′ , ( i 1 , k 1 , i 2 , k 2 , … , i n , k n = x , y , z ) . (180)
The elements of the rotation matrix a i k a_{ik} a ik may be evaluated from
a i k = e i e k ′ , ( i , k = x , y , z ) , (181) a_{ik}=\mathbf{e}_i\mathbf{e}'_k,\qquad(i,k=x,y,z), \tag*{(181)} a ik = e i e k ′ , ( i , k = x , y , z ) , (181)
Thus, the elements a i k a_{ik} a ik are cosines of angles between the basis vectors in the initial (S S S ) and rotated (S ′ S' S ′ ) coordinate systems. An equivalent definition of a i k a_{ik} a ik in terms of coordinates in S S S and S ′ S' S ′ has the form
a i k = ∂ x i ′ ∂ x k = ∂ x k ∂ x i ′ , ( x i ′ , x k = x , y , z ) (182) a_{ik}=\frac{\partial x'_i}{\partial x_k}=\frac{\partial x_k}{\partial x'_i},\qquad(x'_i,x_k=x,y,z) \tag*{(182)} a ik = ∂ x k ∂ x i ′ = ∂ x i ′ ∂ x k , ( x i ′ , x k = x , y , z ) (182)
The rotation matrix a a a is given in terms of the Euler angles by
a = ( cos α cos β cos γ − sin α sin γ − cos α cos β sin γ − sin α cos γ cos α sin β sin α cos β cos γ + cos α sin γ − sin α cos β sin γ + cos α cos γ sin α sin β − sin β cos γ sin β sin γ cos β ) . (183) a=\begin{pmatrix}
\cos\alpha\cos\beta\cos\gamma-\sin\alpha\sin\gamma& -\cos\alpha\cos\beta\sin\gamma-\sin\alpha\cos\gamma& \cos\alpha\sin\beta\\
\sin\alpha\cos\beta\cos\gamma+\cos\alpha\sin\gamma& -\sin\alpha\cos\beta\sin\gamma+\cos\alpha\cos\gamma& \sin\alpha\sin\beta\\
-\sin\beta\cos\gamma& \sin\beta\sin\gamma& \cos\beta
\end{pmatrix}. \tag*{(183)} a = cos α cos β cos γ − sin α sin γ sin α cos β cos γ + cos α sin γ − sin β cos γ − cos α cos β sin γ − sin α cos γ − sin α cos β sin γ + cos α cos γ sin β sin γ cos α sin β sin α sin β cos β . (183)
The inverse matrix a − 1 a^{-1} a − 1 may be obtained from Eq. (183) by transposing or, equivalently, by replacing the Euler angles α , β , γ → − γ , − β , − α \alpha,\beta,\gamma\to-\gamma,-\beta,-\alpha α , β , γ → − γ , − β , − α .
The expression for the rotation matrix a a a in terms of the angles Θ , Φ \Theta,\Phi Θ , Φ which describe the direction of the rotation axis, and the rotation angle ω \omega ω has the form
a = ( ( 1 − cos ω ) sin 2 Θ cos 2 Φ + cos ω ( 1 − cos ω ) sin 2 Θ cos Φ sin Φ + sin ω cos Θ ( 1 − cos ω ) sin Θ cos Θ cos Φ − sin ω sin Θ sin Φ ( 1 − cos ω ) sin 2 Θ cos Φ sin Φ − sin ω cos Θ ( 1 − cos ω ) sin 2 Θ sin 2 Φ + cos ω ( 1 − cos ω ) sin Θ cos Θ sin Φ − sin ω sin Θ cos Φ ( 1 − cos ω ) sin Θ cos Θ cos Φ + sin ω sin Θ sin Φ ( 1 − cos ω ) sin Θ cos Θ sin Φ + sin ω sin Θ cos Φ ( 1 − cos ω ) cos 2 Θ + cos ω ) . (184) a=\begin{pmatrix}
(1-\cos\omega)\sin^2\Theta\cos^2\Phi+\cos\omega&
(1-\cos\omega)\sin^2\Theta\cos\Phi\sin\Phi+\sin\omega\cos\Theta&
(1-\cos\omega)\sin\Theta\cos\Theta\cos\Phi-\sin\omega\sin\Theta\sin\Phi\\
(1-\cos\omega)\sin^2\Theta\cos\Phi\sin\Phi-\sin\omega\cos\Theta&
(1-\cos\omega)\sin^2\Theta\sin^2\Phi+\cos\omega&
(1-\cos\omega)\sin\Theta\cos\Theta\sin\Phi-\sin\omega\sin\Theta\cos\Phi\\
(1-\cos\omega)\sin\Theta\cos\Theta\cos\Phi+\sin\omega\sin\Theta\sin\Phi&
(1-\cos\omega)\sin\Theta\cos\Theta\sin\Phi+\sin\omega\sin\Theta\cos\Phi&
(1-\cos\omega)\cos^2\Theta+\cos\omega
\end{pmatrix}. \tag*{(184)} a = ( 1 − cos ω ) sin 2 Θ cos 2 Φ + cos ω ( 1 − cos ω ) sin 2 Θ cos Φ sin Φ − sin ω cos Θ ( 1 − cos ω ) sin Θ cos Θ cos Φ + sin ω sin Θ sin Φ ( 1 − cos ω ) sin 2 Θ cos Φ sin Φ + sin ω cos Θ ( 1 − cos ω ) sin 2 Θ sin 2 Φ + cos ω ( 1 − cos ω ) sin Θ cos Θ sin Φ + sin ω sin Θ cos Φ ( 1 − cos ω ) sin Θ cos Θ cos Φ − sin ω sin Θ sin Φ ( 1 − cos ω ) sin Θ cos Θ sin Φ − sin ω sin Θ cos Φ ( 1 − cos ω ) cos 2 Θ + cos ω . (184)
Equation (1.4.55) may be expanded into
a = cos ω ( 1 0 0 0 1 0 0 0 1 ) + ( 1 − cos ω ) ( n x 2 n x n y n x n z n y n x n y 2 n y n z n z n x n z n y n z 2 ) + sin ω ( 0 − n z n y n z 0 − n x − n y n x 0 ) . (185) a=\cos\omega\begin{pmatrix}1&0&0\\0&1&0\\0&0&1\end{pmatrix}
+(1-\cos\omega)\begin{pmatrix}n_x^2&n_xn_y&n_xn_z\\n_yn_x&n_y^2&n_yn_z\\n_zn_x&n_zn_y&n_z^2\end{pmatrix}
+\sin\omega\begin{pmatrix}0&-n_z&n_y\\n_z&0&-n_x\\-n_y&n_x&0\end{pmatrix}. \tag*{(185)} a = cos ω 1 0 0 0 1 0 0 0 1 + ( 1 − cos ω ) n x 2 n y n x n z n x n x n y n y 2 n z n y n x n z n y n z n z 2 + sin ω 0 n z − n y − n z 0 n x n y − n x 0 . (185)
where n x , n y , n z n_x,n_y,n_z n x , n y , n z are components of the unit vector n \mathbf{n} n which determines direction of the rotation axis. Using (1.4.56) one can easily derive the following expressions for the matrix elements a i k a_{ik} a ik
a i k = cos ω δ i k + ( 1 − cos ω ) n i n k − sin ω ε i k l n l , ( i , k , l = x , y , z ) . (186) a_{ik}=\cos\omega\delta_{ik}+(1-\cos\omega)n_in_k-\sin\omega\varepsilon_{ikl}n_l,\qquad(i,k,l=x,y,z). \tag*{(186)} a ik = cos ω δ ik + ( 1 − cos ω ) n i n k − sin ω ε ik l n l , ( i , k , l = x , y , z ) . (186)
The inverse matrix a − 1 a^{-1} a − 1 may be obtained from (1.4.55) and (1.4.56) by transposing or, equivalently, by replacing ω , Θ , Φ → − ω , Θ , Φ \omega,\Theta,\Phi\to-\omega,\Theta,\Phi ω , Θ , Φ → − ω , Θ , Φ or ω , Θ , Φ → ω , π − Θ , π + Φ \omega,\Theta,\Phi\to\omega,\pi-\Theta,\pi+\Phi ω , Θ , Φ → ω , π − Θ , π + Φ .
The expressions for ω \omega ω , Θ \Theta Θ , Φ \Phi Φ in terms of the matrix elements a i k a_{ik} a ik read
cos ω = 1 2 [ Tr a − 1 ] = 1 2 ( a x x + a y y + a z z − 1 ) , \cos\omega= \frac{1}{2}[\operatorname{Tr} a - 1] = \frac{1}{2}(a_{xx}+a_{yy}+a_{zz}-1), cos ω = 2 1 [ Tr a − 1 ] = 2 1 ( a xx + a y y + a z z − 1 ) ,
n x sin ω ≡ sin ω sin Θ cos Φ = 1 2 ( a z y − a y z ) , n_x\sin\omega\equiv\sin\omega\sin\Theta\cos\Phi= \frac{1}{2}(a_{zy}-a_{yz}), n x sin ω ≡ sin ω sin Θ cos Φ = 2 1 ( a z y − a y z ) ,
n y sin ω ≡ sin ω sin Θ sin Φ = 1 2 ( a x z − a z x ) , n_y\sin\omega\equiv\sin\omega\sin\Theta\sin\Phi= \frac{1}{2}(a_{xz}-a_{zx}), n y sin ω ≡ sin ω sin Θ sin Φ = 2 1 ( a x z − a z x ) ,
n z sin ω ≡ sin ω cos Θ = 1 2 ( a y x − a x y ) , n_z\sin\omega\equiv\sin\omega\cos\Theta= \frac{1}{2}(a_{yx}-a_{xy}), n z sin ω ≡ sin ω cos Θ = 2 1 ( a y x − a x y ) ,
n i sin ω = − 1 2 ∑ k l ε i k l a k l , ( i , k , l = x , y , z ) . (187) n_i\sin\omega= -\frac{1}{2}\sum_{kl}\varepsilon_{ikl}a_{kl}, \qquad(i,k,l=x,y,z). \tag*{(187)} n i sin ω = − 2 1 k l ∑ ε ik l a k l , ( i , k , l = x , y , z ) . (187)
The rotation matrix a a a may be rewritten in terms of the Cayley-Klein parameters as
a = ( 1 2 ( a 2 − b 2 + a ∗ 2 − b ∗ 2 ) i 2 ( − a 2 + b 2 + a ∗ 2 − b ∗ 2 ) a b ∗ + a ∗ b i 2 ( a 2 + b 2 − a ∗ 2 − b ∗ 2 ) 1 2 ( a 2 + b 2 + a ∗ 2 + b ∗ 2 ) i ( a b ∗ − a ∗ b ) − ( a b + a ∗ b ∗ ) i ( a b − a ∗ b ∗ ) a a ∗ − b b ∗ ) . (188) a=\begin{pmatrix}
\frac{1}{2}(a^2-b^2+a^{*2}-b^{*2}) & \frac{i}{2}(-a^2+b^2+a^{*2}-b^{*2}) & ab^*+a^*b \\
\frac{i}{2}(a^2+b^2-a^{*2}-b^{*2}) & \frac{1}{2}(a^2+b^2+a^{*2}+b^{*2}) & i(ab^*-a^*b) \\
-(ab+a^*b^*) & i(ab-a^*b^*) & aa^*-bb^*
\end{pmatrix}. \tag*{(188)} a = 2 1 ( a 2 − b 2 + a ∗ 2 − b ∗ 2 ) 2 i ( a 2 + b 2 − a ∗ 2 − b ∗ 2 ) − ( ab + a ∗ b ∗ ) 2 i ( − a 2 + b 2 + a ∗ 2 − b ∗ 2 ) 2 1 ( a 2 + b 2 + a ∗ 2 + b ∗ 2 ) i ( ab − a ∗ b ∗ ) a b ∗ + a ∗ b i ( a b ∗ − a ∗ b ) a a ∗ − b b ∗ . (188)
One can see that the parameters a , b a,b a , b and − a , − b -a,-b − a , − b correspond to the same rotation matrix.
(a) Rotation through an angle Ψ \Psi Ψ about the x x x -axis:
a x ( Ψ ) = ( 1 0 0 0 cos Ψ − sin Ψ 0 sin Ψ cos Ψ ) . (189) a_x(\Psi)=\begin{pmatrix}
1&0&0\\
0&\cos\Psi&-\sin\Psi\\
0&\sin\Psi&\cos\Psi
\end{pmatrix}. \tag*{(189)} a x ( Ψ ) = 1 0 0 0 cos Ψ sin Ψ 0 − sin Ψ cos Ψ . (189)
(b) Rotation through an angle Ψ \Psi Ψ about the y y y -axis:
a y ( Ψ ) = ( cos Ψ 0 sin Ψ 0 1 0 − sin Ψ 0 cos Ψ ) . (190) a_y(\Psi)=\begin{pmatrix}
\cos\Psi&0&\sin\Psi\\
0&1&0\\
-\sin\Psi&0&\cos\Psi
\end{pmatrix}. \tag*{(190)} a y ( Ψ ) = cos Ψ 0 − sin Ψ 0 1 0 sin Ψ 0 cos Ψ . (190)
(c) Rotation through an angle Ψ \Psi Ψ about the z z z -axis:
a z ( Ψ ) = ( cos Ψ − sin Ψ 0 sin Ψ cos Ψ 0 0 0 1 ) . (191) a_z(\Psi)=\begin{pmatrix}
\cos\Psi&-\sin\Psi&0\\
\sin\Psi&\cos\Psi&0\\
0&0&1
\end{pmatrix}. \tag*{(191)} a z ( Ψ ) = cos Ψ sin Ψ 0 − sin Ψ cos Ψ 0 0 0 1 . (191)
For an arbitrary rotation determined by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ the rotation matrix, in accordance with Eq. (160) , may be written in the form
a = a n ( ω ) = a z ( α ) a y ( β ) a z ( γ ) . (192) a=a_n(\omega)=a_z(\alpha)a_y(\beta)a_z(\gamma). \tag*{(192)} a = a n ( ω ) = a z ( α ) a y ( β ) a z ( γ ) . (192)
Equation (1.4.63) represents a particular case of addition of coordinate rotations (see § 9.10.4 ).
Addition of Rotations
Let us consider two successive rotations of the coordinate system. Let the first rotation transform the coordinate system S { x , y , z } S\{x,y,z\} S { x , y , z } into S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } and the second one transform S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } into S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S''\{x'',y'',z''\} S ′′ { x ′′ , y ′′ , z ′′ } .
Below the parameters describing the resultant rotation S { x , y , z } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S\{x,y,z\}\to S''\{x'',y'',z''\} S { x , y , z } → S ′′ { x ′′ , y ′′ , z ′′ } will be given in terms of the parameters specifying the rotations S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } and S ′ { x ′ , y ′ , z ′ } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S'\{x',y',z'\}\to S''\{x'',y'',z''\} S ′ { x ′ , y ′ , z ′ } → S ′′ { x ′′ , y ′′ , z ′′ } .
(a) (a) Description of rotations in terms of Euler angles
Let both rotations be performed according to the scheme B(p.22). Let the first rotation S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } be described by the Euler angles α 1 , β 1 , γ 1 \alpha_1,\beta_1,\gamma_1 α 1 , β 1 , γ 1 , the second one, S ′ { x ′ , y ′ , z ′ } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S'\{x',y',z'\}\to S''\{x'',y'',z''\} S ′ { x ′ , y ′ , z ′ } → S ′′ { x ′′ , y ′′ , z ′′ } , by the Euler angles α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 and the resultant rotation, S { x , y , z } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S\{x,y,z\}\to S''\{x'',y'',z''\} S { x , y , z } → S ′′ { x ′′ , y ′′ , z ′′ } by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ . The Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ , α 1 , β 1 , γ 1 \alpha_1,\beta_1,\gamma_1 α 1 , β 1 , γ 1 and α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 are supposed to be defined with respect to an initial coordinate system S { x , y , z } S\{x,y,z\} S { x , y , z } .
The operator of resultant rotation has the form
D ^ ( α , β , γ ) = D ^ ( α 2 , β 2 , γ 2 ) D ^ ( α 1 , β 1 , γ 1 ) , (193) \hat{D}(\alpha,\beta,\gamma)=\hat{D}(\alpha_2,\beta_2,\gamma_2)\hat{D}(\alpha_1,\beta_1,\gamma_1), \tag*{(193)} D ^ ( α , β , γ ) = D ^ ( α 2 , β 2 , γ 2 ) D ^ ( α 1 , β 1 , γ 1 ) , (193)
or in more detail
e − i α J ^ z e − i β J ^ y e − i γ J ^ z = e − i α 2 J ^ z e − i β 2 J ^ y e − i γ 2 J ^ z e − i α 1 J ^ z e − i β 1 J ^ y e − i γ 1 J ^ z . (194) e^{-i\alpha\hat{J}_z}e^{-i\beta\hat{J}_y}e^{-i\gamma\hat{J}_z}=e^{-i\alpha_2\hat{J}_z}e^{-i\beta_2\hat{J}_y}e^{-i\gamma_2\hat{J}_z}e^{-i\alpha_1\hat{J}_z}e^{-i\beta_1\hat{J}_y}e^{-i\gamma_1\hat{J}_z}. \tag*{(194)} e − i α J ^ z e − i β J ^ y e − iγ J ^ z = e − i α 2 J ^ z e − i β 2 J ^ y e − i γ 2 J ^ z e − i α 1 J ^ z e − i β 1 J ^ y e − i γ 1 J ^ z . (194)
In Eq. (194) J ^ i \hat{J}_i J ^ i is the projection of the total angular momentum operator on an i i i -axis of the coordinate system S { x , y , z } S\{x,y,z\} S { x , y , z } . The angles of the resultant rotation α , β , γ \alpha,\beta,\gamma α , β , γ are expressed in terms of α 1 , β 1 , γ 1 \alpha_1,\beta_1,\gamma_1 α 1 , β 1 , γ 1 and α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 as
cot ( α − α 2 ) = cos β 2 cot ( α 1 + γ 2 ) + cot β 1 sin β 2 sin ( α 1 + γ 2 ) , \cot(\alpha-\alpha_2)=\cos\beta_2\cot(\alpha_1+\gamma_2)+\cot\beta_1\frac{\sin\beta_2}{\sin(\alpha_1+\gamma_2)}, cot ( α − α 2 ) = cos β 2 cot ( α 1 + γ 2 ) + cot β 1 sin ( α 1 + γ 2 ) sin β 2 ,
cos β = cos β 1 cos β 2 − sin β 1 sin β 2 cos ( α 1 + γ 2 ) , \cos\beta=\cos\beta_1\cos\beta_2-\sin\beta_1\sin\beta_2\cos(\alpha_1+\gamma_2), cos β = cos β 1 cos β 2 − sin β 1 sin β 2 cos ( α 1 + γ 2 ) ,
cot ( γ − γ 1 ) = cos β 1 cot ( α 1 + γ 2 ) + cot β 2 sin β 1 sin ( α 1 + γ 2 ) . (195) \cot(\gamma-\gamma_1)=\cos\beta_1\cot(\alpha_1+\gamma_2)+\cot\beta_2\frac{\sin\beta_1}{\sin(\alpha_1+\gamma_2)}. \tag*{(195)} cot ( γ − γ 1 ) = cos β 1 cot ( α 1 + γ 2 ) + cot β 2 sin ( α 1 + γ 2 ) sin β 1 . (195)
The following relations are useful for evaluation of α , β , γ \alpha,\beta,\gamma α , β , γ .
sin ( α − α 2 ) sin β 1 = sin ( γ − γ 1 ) sin β 2 = sin ( α 1 + γ 2 ) sin β . (196) \frac{\sin(\alpha-\alpha_2)}{\sin\beta_1}=\frac{\sin(\gamma-\gamma_1)}{\sin\beta_2}=\frac{\sin(\alpha_1+\gamma_2)}{\sin\beta}. \tag*{(196)} sin β 1 sin ( α − α 2 ) = sin β 2 sin ( γ − γ 1 ) = sin β sin ( α 1 + γ 2 ) . (196)
cos β 1 = cos β cos β 2 + sin β sin β 2 cos ( α − α 2 ) , \cos\beta_1=\cos\beta\cos\beta_2+\sin\beta\sin\beta_2\cos(\alpha-\alpha_2), cos β 1 = cos β cos β 2 + sin β sin β 2 cos ( α − α 2 ) ,
cos β 2 = cos β cos β 1 + sin β sin β 1 cos ( γ − γ 1 ) , \cos\beta_2=\cos\beta\cos\beta_1+\sin\beta\sin\beta_1\cos(\gamma-\gamma_1), cos β 2 = cos β cos β 1 + sin β sin β 1 cos ( γ − γ 1 ) ,
cos β = cos β 1 cos β 2 − sin β 1 sin β 2 cos ( α 1 + γ 2 ) , (197) \cos\beta=\cos\beta_1\cos\beta_2-\sin\beta_1\sin\beta_2\cos(\alpha_1+\gamma_2), \tag*{(197)} cos β = cos β 1 cos β 2 − sin β 1 sin β 2 cos ( α 1 + γ 2 ) , (197)
cos ( γ − γ 1 ) = cos ( α 1 + γ 2 ) cos ( α − α 2 ) + sin ( α 1 + γ 2 ) sin ( α − α 2 ) cos β 2 , \cos(\gamma-\gamma_1)=\cos(\alpha_1+\gamma_2)\cos(\alpha-\alpha_2)+\sin(\alpha_1+\gamma_2)\sin(\alpha-\alpha_2)\cos\beta_2, cos ( γ − γ 1 ) = cos ( α 1 + γ 2 ) cos ( α − α 2 ) + sin ( α 1 + γ 2 ) sin ( α − α 2 ) cos β 2 ,
cos ( α − α 2 ) = cos ( α 1 + γ 2 ) cos ( γ − γ 1 ) + sin ( α 1 + γ 2 ) sin ( γ − γ 1 ) cos β 1 , \cos(\alpha-\alpha_2)=\cos(\alpha_1+\gamma_2)\cos(\gamma-\gamma_1)+\sin(\alpha_1+\gamma_2)\sin(\gamma-\gamma_1)\cos\beta_1, cos ( α − α 2 ) = cos ( α 1 + γ 2 ) cos ( γ − γ 1 ) + sin ( α 1 + γ 2 ) sin ( γ − γ 1 ) cos β 1 ,
cos ( α 1 + γ 2 ) = cos ( γ − γ 1 ) cos ( α − α 2 ) − sin ( γ − γ 1 ) sin ( α − α 2 ) cos β , (198) \cos(\alpha_1+\gamma_2)=\cos(\gamma-\gamma_1)\cos(\alpha-\alpha_2)-\sin(\gamma-\gamma_1)\sin(\alpha-\alpha_2)\cos\beta, \tag*{(198)} cos ( α 1 + γ 2 ) = cos ( γ − γ 1 ) cos ( α − α 2 ) − sin ( γ − γ 1 ) sin ( α − α 2 ) cos β , (198)
tan β − β 1 2 tan β + β 1 2 = tan α 1 + γ 2 + α − α 2 2 tan α 1 + γ 2 − α + α 2 2 , \frac{\tan\frac{\beta-\beta_1}{2}}{\tan\frac{\beta+\beta_1}{2}}=\frac{\tan\frac{\alpha_1+\gamma_2+\alpha-\alpha_2}{2}}{\tan\frac{\alpha_1+\gamma_2-\alpha+\alpha_2}{2}}, tan 2 β + β 1 tan 2 β − β 1 = tan 2 α 1 + γ 2 − α + α 2 tan 2 α 1 + γ 2 + α − α 2 ,
tan β − β 2 2 tan β + β 2 2 = tan α 1 + γ 2 + γ − γ 1 2 tan α 1 + γ 2 − γ + γ 1 2 , \frac{\tan\frac{\beta-\beta_2}{2}}{\tan\frac{\beta+\beta_2}{2}}=\frac{\tan\frac{\alpha_1+\gamma_2+\gamma-\gamma_1}{2}}{\tan\frac{\alpha_1+\gamma_2-\gamma+\gamma_1}{2}}, tan 2 β + β 2 tan 2 β − β 2 = tan 2 α 1 + γ 2 − γ + γ 1 tan 2 α 1 + γ 2 + γ − γ 1 ,
tan β 1 − β 2 2 tan β 1 + β 2 2 = tan α − α 2 + γ − γ 1 2 tan α − α 2 − γ + γ 1 2 . (199) \frac{\tan\frac{\beta_1-\beta_2}{2}}{\tan\frac{\beta_1+\beta_2}{2}}=\frac{\tan\frac{\alpha-\alpha_2+\gamma-\gamma_1}{2}}{\tan\frac{\alpha-\alpha_2-\gamma+\gamma_1}{2}}. \tag*{(199)} tan 2 β 1 + β 2 tan 2 β 1 − β 2 = tan 2 α − α 2 − γ + γ 1 tan 2 α − α 2 + γ − γ 1 . (199)
Equations (1.4.67)–(1.4.70) may be easily interpreted in terms of the geometry on a sphere. Each rotation may be completely determined by a point of intersection of the z ′ z' z ′ -axis with the spherical surface and by a unit vector in the direction of the x ′ x' x ′ -axis which lies on a plane tangent to the surface at this point. In this case the determination of α , β , γ \alpha,\beta,\gamma α , β , γ is reduced to constructing the corresponding spherical triangle, (Figure 1.8 ). Equations (1.4.67)–(1.4.70) represent the formulas of sines, cosines and tangents for the spherical triangle.
Figure 1.8. Addition of rotations in terms of the spherical geometry.
Another expression for the angles of the resultant rotation will be obtained if successive rotations are performed according to the scheme B (p. 22) but the Euler angles α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 specifying the second rotation S ′ { x ′ , y ′ , z ′ } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S'\{x',y',z'\}\rightarrow S''\{x'',y'',z''\} S ′ { x ′ , y ′ , z ′ } → S ′′ { x ′′ , y ′′ , z ′′ } are defined with respect to the intermediate coordinate system S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } rather than the initial system S { x , y , z } S\{x,y,z\} S { x , y , z } . In this case the operator of the resultant rotation has the form
D ^ ( α , β , γ ) = D ^ ′ ( α 2 , β 2 , γ 2 ) D ^ ( α 1 , β 1 , γ 1 ) , (200) \hat{D}(\alpha,\beta,\gamma)=\hat{D}'(\alpha_2,\beta_2,\gamma_2)\hat{D}(\alpha_1,\beta_1,\gamma_1), \tag*{(200)} D ^ ( α , β , γ ) = D ^ ′ ( α 2 , β 2 , γ 2 ) D ^ ( α 1 , β 1 , γ 1 ) , (200)
where prime indicates that the operator of the second rotation is taken in the coordinate system S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } . According to (1.4.28), the operator D ^ ′ \hat{D}' D ^ ′ is related to the operator in the initial coordinate system by
D ^ ′ ( α 2 , β 2 , γ 2 ) = D ^ ( α 1 , β 1 , γ 1 ) D ^ ( α 2 , β 2 , γ 2 ) [ D ^ ( α 1 , β 1 , γ 1 ) ] − 1 . (201) \hat{D}'(\alpha_2,\beta_2,\gamma_2)=\hat{D}(\alpha_1,\beta_1,\gamma_1)\hat{D}(\alpha_2,\beta_2,\gamma_2)\left[\hat{D}(\alpha_1,\beta_1,\gamma_1)\right]^{-1}. \tag*{(201)} D ^ ′ ( α 2 , β 2 , γ 2 ) = D ^ ( α 1 , β 1 , γ 1 ) D ^ ( α 2 , β 2 , γ 2 ) [ D ^ ( α 1 , β 1 , γ 1 ) ] − 1 . (201)
Substitution of this expression into Eq. (200) yields
D ^ ( α , β , γ ) = D ^ ( α 1 , β 1 , γ 1 ) D ^ ( α 2 , β 2 , γ 2 ) , (202) \hat{D}(\alpha,\beta,\gamma)=\hat{D}(\alpha_1,\beta_1,\gamma_1)\hat{D}(\alpha_2,\beta_2,\gamma_2), \tag*{(202)} D ^ ( α , β , γ ) = D ^ ( α 1 , β 1 , γ 1 ) D ^ ( α 2 , β 2 , γ 2 ) , (202)
i.e., the operator of the resultant rotation differs from (1.4.64) in the order of operators of the first and second rotations. Thus, for such a description of the successive rotations the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ may be obtained from (1.4.66)–(1.4.70) by interchange of indices 1 ↔ 2 1\leftrightarrow2 1 ↔ 2 .
Finally, if successive rotations are performed according to the scheme A (p. 21), i.e., if each rotation is made about the corresponding new axis, the operator of resultant rotation is given by Eq. (202) . In this case the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ of the resultant rotation may also be derived from (1.4.66)–(1.4.70) by the interchange of indices 1 ↔ 2 1\leftrightarrow2 1 ↔ 2 .
(b) (b) Description of Rotations in Terms of Rotation Axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) and Rotation Angle ω \omega ω
Let the first rotation, S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } , be performed about an axis n 1 \mathbf{n}_1 n 1 through an angle ω 1 \omega_1 ω 1 , and the second one, S ′ { x ′ , y ′ , z ′ } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S'\{x',y',z'\}\to S''\{x'',y'',z''\} S ′ { x ′ , y ′ , z ′ } → S ′′ { x ′′ , y ′′ , z ′′ } , be about an axis n 2 \mathbf{n}_2 n 2 through an angle ω 2 \omega_2 ω 2 . The resultant rotation S { x , y , z } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S\{x,y,z\}\to S''\{x'',y'',z''\} S { x , y , z } → S ′′ { x ′′ , y ′′ , z ′′ } may be treated as a rotation about an axis n \mathbf{n} n through an angle ω \omega ω .
The operator of the resultant rotation has the form
e − i ω n ⋅ J ^ = e − i ω 2 n 2 ⋅ J ^ e − i ω 1 n 1 ⋅ J ^ . (203) e^{-i\omega\mathbf{n}\cdot\hat{\mathbf{J}}}=e^{-i\omega_2\mathbf{n}_2\cdot\hat{\mathbf{J}}}e^{-i\omega_1\mathbf{n}_1\cdot\hat{\mathbf{J}}}. \tag*{(203)} e − iω n ⋅ J ^ = e − i ω 2 n 2 ⋅ J ^ e − i ω 1 n 1 ⋅ J ^ . (203)
The angle of the resultant rotation ω \omega ω and the axis of this rotation n \mathbf{n} n are determined by
cos ω 2 = cos ω 1 2 cos ω 2 2 − ( n 1 ⋅ n 2 ) sin ω 1 2 sin ω 2 2 , (204) \cos\frac{\omega}{2}=\cos\frac{\omega_1}{2}\cos\frac{\omega_2}{2}-(\mathbf{n}_1\cdot\mathbf{n}_2)\sin\frac{\omega_1}{2}\sin\frac{\omega_2}{2}, \tag*{(204)} cos 2 ω = cos 2 ω 1 cos 2 ω 2 − ( n 1 ⋅ n 2 ) sin 2 ω 1 sin 2 ω 2 , (204)
n sin ω 2 = n 1 sin ω 1 2 cos ω 2 2 + n 2 sin ω 2 2 cos ω 1 2 − [ n 1 × n 2 ] sin ω 1 2 sin ω 2 2 . \mathbf{n}\sin\frac{\omega}{2}=\mathbf{n}_1\sin\frac{\omega_1}{2}\cos\frac{\omega_2}{2}+\mathbf{n}_2\sin\frac{\omega_2}{2}\cos\frac{\omega_1}{2}-[\mathbf{n}_1\times\mathbf{n}_2]\sin\frac{\omega_1}{2}\sin\frac{\omega_2}{2}. n sin 2 ω = n 1 sin 2 ω 1 cos 2 ω 2 + n 2 sin 2 ω 2 cos 2 ω 1 − [ n 1 × n 2 ] sin 2 ω 1 sin 2 ω 2 .
It follows from Eq. (204) that the resultant rotation is independent of the order of successive rotations (i.e., the rotation operators commute) if and only if n 1 × n 2 = 0 \mathbf{n}_1\times\mathbf{n}_2=0 n 1 × n 2 = 0 , i.e., the axes of both rotations are parallel or antiparallel. In this case
ω = ω 1 ± ω 2 \omega=\omega_1\pm\omega_2 ω = ω 1 ± ω 2
If directions of the rotation axes n 1 , n 2 , n \mathbf{n}_1,\mathbf{n}_2,\mathbf{n} n 1 , n 2 , n are specified by the polar angles Θ 1 , Φ 1 \Theta_1,\Phi_1 Θ 1 , Φ 1 ; Θ 2 , Φ 2 \Theta_2,\Phi_2 Θ 2 , Φ 2 and Θ , Φ \Theta,\Phi Θ , Φ , respectively, and the polar angles are defined with respect to the initial coordinate system S { x , y , z } S\{x,y,z\} S { x , y , z } , then
cos ω 2 = cos ω 1 2 cos ω 2 2 − sin ω 1 2 sin ω 2 2 [ cos Θ 1 cos Θ 2 + sin Θ 1 sin Θ 2 cos ( Φ 1 − Φ 2 ) ] \cos\frac{\omega}{2}=\cos\frac{\omega_1}{2}\cos\frac{\omega_2}{2}-\sin\frac{\omega_1}{2}\sin\frac{\omega_2}{2}[\cos\Theta_1\cos\Theta_2+\sin\Theta_1\sin\Theta_2\cos(\Phi_1-\Phi_2)] cos 2 ω = cos 2 ω 1 cos 2 ω 2 − sin 2 ω 1 sin 2 ω 2 [ cos Θ 1 cos Θ 2 + sin Θ 1 sin Θ 2 cos ( Φ 1 − Φ 2 )]
sin ω 2 cos Θ = sin ω 1 2 cos ω 2 2 cos Θ 1 + sin ω 2 2 cos ω 1 2 cos Θ 2 + sin ω 1 2 sin ω 2 2 sin Θ 1 sin Θ 2 sin ( Φ 1 − Φ 2 ) (205) \sin\frac{\omega}{2}\cos\Theta=\sin\frac{\omega_1}{2}\cos\frac{\omega_2}{2}\cos\Theta_1+\sin\frac{\omega_2}{2}\cos\frac{\omega_1}{2}\cos\Theta_2
+\sin\frac{\omega_1}{2}\sin\frac{\omega_2}{2}\sin\Theta_1\sin\Theta_2\sin(\Phi_1-\Phi_2) \tag*{(205)} sin 2 ω cos Θ = sin 2 ω 1 cos 2 ω 2 cos Θ 1 + sin 2 ω 2 cos 2 ω 1 cos Θ 2 + sin 2 ω 1 sin 2 ω 2 sin Θ 1 sin Θ 2 sin ( Φ 1 − Φ 2 ) (205)
cot Φ = sin Θ 1 ( cot ω 2 2 cos Φ 1 − cos Θ 2 sin Φ 1 ) + sin Θ 2 ( cot ω 1 2 cos Φ 2 + cos Θ 1 sin Φ 2 ) sin Θ 1 ( cot ω 2 2 sin Φ 1 + cos Θ 2 cos Φ 1 ) + sin Θ 2 ( cot ω 1 2 sin Φ 2 − cos Θ 1 cos Φ 2 ) . \cot\Phi=
\frac{\sin\Theta_1\left(\cot\frac{\omega_2}{2}\cos\Phi_1-\cos\Theta_2\sin\Phi_1\right)+\sin\Theta_2\left(\cot\frac{\omega_1}{2}\cos\Phi_2+\cos\Theta_1\sin\Phi_2\right)}
{\sin\Theta_1\left(\cot\frac{\omega_2}{2}\sin\Phi_1+\cos\Theta_2\cos\Phi_1\right)+\sin\Theta_2\left(\cot\frac{\omega_1}{2}\sin\Phi_2-\cos\Theta_1\cos\Phi_2\right)}. cot Φ = sin Θ 1 ( cot 2 ω 2 sin Φ 1 + cos Θ 2 cos Φ 1 ) + sin Θ 2 ( cot 2 ω 1 sin Φ 2 − cos Θ 1 cos Φ 2 ) sin Θ 1 ( cot 2 ω 2 cos Φ 1 − cos Θ 2 sin Φ 1 ) + sin Θ 2 ( cot 2 ω 1 cos Φ 2 + cos Θ 1 sin Φ 2 ) .
If directions of the rotation axes n 1 , n \mathbf{n}_1,\mathbf{n} n 1 , n are defined by the polar angles Θ 1 , Φ 1 \Theta_1,\Phi_1 Θ 1 , Φ 1 and Θ , Φ \Theta,\Phi Θ , Φ with respect to the initial coordinate system S { x , y , z } S\{x,y,z\} S { x , y , z } and the direction of the axis of the second rotation n 2 \mathbf{n}_2 n 2 is defined by the polar angles Θ 2 , Φ 2 \Theta_2,\Phi_2 Θ 2 , Φ 2 with respect to the intermediate coordinate system S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } then the angles ω , Θ , Φ \omega,\Theta,\Phi ω , Θ , Φ of the resultant rotation will be given by Eqs. (205) with interchanged indices 1 ⇄ 2 1\rightleftarrows2 1 ⇄ 2 . This situation is similar to the case when rotations are described by Euler angles.
(c) (c) Description of Rotation in Terms of Cayley-Klein Parameters
Let the first rotation, S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } be determined by the Cayley-Klein parameters a 1 , b 1 a_1,b_1 a 1 , b 1 (1.4.3) and the second one, S ′ { x ′ , y ′ , z ′ } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S'\{x',y',z'\}\to S''\{x'',y'',z''\} S ′ { x ′ , y ′ , z ′ } → S ′′ { x ′′ , y ′′ , z ′′ } , by the parameters a 2 , b 2 a_2,b_2 a 2 , b 2 . Then the resultant rotation S { x , y , z } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S\{x,y,z\}\to S''\{x'',y'',z''\} S { x , y , z } → S ′′ { x ′′ , y ′′ , z ′′ } will be determined by the parameters a , b a,b a , b such as
a = a 1 a 2 − b 1 ∗ b 2 , a=a_1a_2-b_1^*b_2, a = a 1 a 2 − b 1 ∗ b 2 ,
b = a 1 ∗ b 2 + b 1 a 2 . (206) b=a_1^*b_2+b_1a_2. \tag*{(206)} b = a 1 ∗ b 2 + b 1 a 2 . (206)
The matrix U U U (see Eq. (140) ) which describes the resultant rotation is a product of matrices corresponding to the first and second rotations
U ( a , b ) = U ( a 2 , b 2 ) U ( a 1 , b 1 ) . (207) U(a,b)=U(a_2,b_2)U(a_1,b_1). \tag*{(207)} U ( a , b ) = U ( a 2 , b 2 ) U ( a 1 , b 1 ) . (207)
In this case all the matrices are supposed to be given in an initial coordinate system.
(d) (d) Addition Theorem for Rotation Matrices a a a
Let us carry out two successive rotations S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } and S ′ { x ′ , y ′ , z ′ } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S'\{x',y',z'\}\to S''\{x'',y'',z''\} S ′ { x ′ , y ′ , z ′ } → S ′′ { x ′′ , y ′′ , z ′′ } . The matrix which transforms cartesian components of vectors and tensors under the resultant rotation S { x , y , z } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S\{x,y,z\}\to S''\{x'',y'',z''\} S { x , y , z } → S ′′ { x ′′ , y ′′ , z ′′ } represents a product of the matrices a ( 1 ) a(1) a ( 1 ) and a ( 2 ) a(2) a ( 2 ) corresponding to the first and second rotations. The order of these matrices in the product depends on the convention used for the rotation angles. If all angles which describe rotations are referred to the initial coordinate system, i.e., the operator of the resultant rotation is given by Eq. (193) , then
a = a ( 2 ) a ( 1 ) , (208) a=a(2)a(1), \tag*{(208)} a = a ( 2 ) a ( 1 ) , (208)
or, in terms of matrix elements,
a i k = ∑ l a i l ( 2 ) a l k ( 1 ) , ( i , k , l = x , y , z ) . (209) a_{ik}=\sum_l a_{il}(2)a_{lk}(1),\qquad(i,k,l=x,y,z). \tag*{(209)} a ik = l ∑ a i l ( 2 ) a l k ( 1 ) , ( i , k , l = x , y , z ) . (209)
The rotation matrices in terms of the rotation angles are given by Eqs. (183) and (185) . The angles which determine the resultant rotation are related to the angles of the first and second rotations via Eqs. (195) and (205) .
If the angles which determine the first and resultant rotations, S → S ′ S \to S' S → S ′ and S → S ′ ′ S \to S'' S → S ′′ , are defined with respect to the initial system S { x , y , z } S\{x,y,z\} S { x , y , z } but the angles of the second rotation S ′ → S ′ ′ S' \to S'' S ′ → S ′′ are defined with respect to the intermediate system S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } (i.e., the operator of the resultant rotation is given by Eq. (202) ), then
a = a ( 1 ) a ( 2 ) , (210) a = a^{(1)}a^{(2)}, \tag*{(210)} a = a ( 1 ) a ( 2 ) , (210)
or, equivalently,
a i k = ∑ l a i l ( 1 ) a l k ( 2 ) , ( i , k , l = x , y , z ) . (211) a_{ik} = \sum_l a_{il}^{(1)}a_{lk}^{(2)}, \qquad(i,k,l = x,y,z). \tag*{(211)} a ik = l ∑ a i l ( 1 ) a l k ( 2 ) , ( i , k , l = x , y , z ) . (211)
In this case the relationships between the angles of the resultant rotation and the angles of the first and second rotations may be derived from {(xref)}(1.4.66) and (1.4.76) by an interchange of indices 1 ⇌ 2 1 \rightleftharpoons2 1 ⇌ 2 .
Angular momentum operators
Total angular momentum operator
Definition
In quantum mechanics the total angular momentum operator J ^ \hat{J} J ^ is defined as an operator which generates transformations of wave functions (state vectors) and quantum operators under infinitesimal rotations of the coordinate system (see Eqs. (212) and (213) ).
A transformation of an arbitrary wave function Ψ \Psi Ψ under rotation of the coordinate system through an infinitesimal angle δ ω \delta\omega δ ω about an axis n n n may be written as
Ψ → Ψ ′ = ( 1 − i δ ω n ⋅ J ^ ) Ψ . (212) \Psi\rightarrow\Psi' = (1-i\delta\omega n \cdot\hat{J})\Psi. \tag*{(212)} Ψ → Ψ ′ = ( 1 − i δ ω n ⋅ J ^ ) Ψ. (212)
where J ^ \hat{J} J ^ is the total angular momentum operator.
A transformation of an arbitrary quantum operator O ^ \hat{O} O ^ under an infinitesimal rotation has the form
O ^ → O ^ ′ = O ^ − i δ ω n ⋅ [ J ^ , O ^ ] . (213) \hat{O} \rightarrow\hat{O}' = \hat{O}-i\delta\omega n \cdot[\hat{J},\hat{O}]. \tag*{(213)} O ^ → O ^ ′ = O ^ − i δ ω n ⋅ [ J ^ , O ^ ] . (213)
The finite rotation operator can also be written in terms of the total angular momentum operator (see Eqs. (158) , (159) , (161) ). The total angular momentum operator is Hermitian,
J ^ + = J ^ . (214) \hat{J}^{+}=\hat{J}. \tag*{(214)} J ^ + = J ^ . (214)
This property for cartesian and spherical components of the operator J ^ \hat{J} J ^ can be expressed as
( J ^ i ) + = J ^ i , ( i = x , y , z ) ; ( J ^ μ ) + = J ^ μ = ( − 1 ) μ J ^ − μ , ( μ = ± 1 , 0 ) . (215) (\hat{J}_i)^{+}=\hat{J}_i,\quad(i=x,y,z);\qquad(\hat{J}_\mu)^{+}=\hat{J}^{\mu}=(-1)^{\mu}\hat{J}_{-\mu},\quad(\mu=\pm1,0). \tag*{(215)} ( J ^ i ) + = J ^ i , ( i = x , y , z ) ; ( J ^ μ ) + = J ^ μ = ( − 1 ) μ J ^ − μ , ( μ = ± 1 , 0 ) . (215)
The eigenfunctions of the operators J ^ 2 \hat{J}^{2} J ^ 2 and J ^ z \hat{J}_z J ^ z represent well-known tensor spherical harmonics (see Chap. 7).
Commutation Relations
Using the definition of the total angular momentum operator (2.1.1) and equations for the rotation addition (see Sec. a), one may obtain the commutation rules for the operator J ^ \hat{J} J ^ . These rules may be written as
[ ( a ⋅ J ^ ) , ( b ⋅ J ^ ) ] = i ⋅ [ a × b ] ⋅ J ^ , (216) [(a\cdot\hat{J}),(b\cdot\hat{J})]=i\cdot[a\times b]\cdot\hat{J}, \tag*{(216)} [( a ⋅ J ^ ) , ( b ⋅ J ^ )] = i ⋅ [ a × b ] ⋅ J ^ , (216)
a a a and b b b being arbitrary constant vectors. The commutation relations for the total angular momentum operator may also be written symbolically as
[ J ^ × J ^ ] = i J ^ . (217) [\hat{J}\times\hat{J}]=i\hat{J}. \tag*{(217)} [ J ^ × J ^ ] = i J ^ . (217)
Cartesian components of J ^ \hat{J} J ^ satisfy the commutation relations
[ J ^ i , J ^ k ] = i ε i k l J ^ l , [ J ^ 2 , J ^ i ] = 0 ( i , k , l = x , y , z ) . (218) [\hat{J}_i,\hat{J}_k]=i\varepsilon_{ikl}\hat{J}_l,\qquad[\hat{J}^{2},\hat{J}_i]=0\quad(i,k,l=x,y,z). \tag*{(218)} [ J ^ i , J ^ k ] = i ε ik l J ^ l , [ J ^ 2 , J ^ i ] = 0 ( i , k , l = x , y , z ) . (218)
Equations (2.1.7) can be derived from (2.1.5) by substituting a = e i \mathbf{a}=\mathbf{e}_i a = e i , b = e k \mathbf{b}=\mathbf{e}_k b = e k (i , k = x , y , z i,k=x,y,z i , k = x , y , z ) where e i \mathbf{e}_i e i and e k \mathbf{e}_k e k are the cartesian basis vectors. In more detailed form Eq. (2.1.7) reads
[ J ^ x , J ^ x ] = [ J ^ y , J ^ y ] = [ J ^ z , J ^ z ] = 0 , [ J ^ x , J ^ y ] = − [ J ^ y , J ^ x ] = i J ^ z , [ J ^ x , J ^ z ] = − [ J ^ z , J ^ x ] = − i J ^ y , [ J ^ y , J ^ z ] = − [ J ^ z , J ^ y ] = i J ^ x , [ J ^ 2 , J ^ x ] = [ J ^ 2 , J ^ y ] = [ J ^ 2 , J ^ z ] = 0. (219) \begin{aligned}
[\hat{J}_x,\hat{J}_x]&=[\hat{J}_y,\hat{J}_y]=[\hat{J}_z,\hat{J}_z]=0,\\
[\hat{J}_x,\hat{J}_y]&=-[\hat{J}_y,\hat{J}_x]=i\hat{J}_z,\qquad[\hat{J}_x,\hat{J}_z]=-[\hat{J}_z,\hat{J}_x]=-i\hat{J}_y,\\
[\hat{J}_y,\hat{J}_z]&=-[\hat{J}_z,\hat{J}_y]=i\hat{J}_x,\\
[\hat{\mathbf{J}}^2,\hat{J}_x]&=[\hat{\mathbf{J}}^2,\hat{J}_y]=[\hat{\mathbf{J}}^2,\hat{J}_z]=0. \tag*{(219)}
\end{aligned} [ J ^ x , J ^ x ] [ J ^ x , J ^ y ] [ J ^ y , J ^ z ] [ J ^ 2 , J ^ x ] = [ J ^ y , J ^ y ] = [ J ^ z , J ^ z ] = 0 , = − [ J ^ y , J ^ x ] = i J ^ z , [ J ^ x , J ^ z ] = − [ J ^ z , J ^ x ] = − i J ^ y , = − [ J ^ z , J ^ y ] = i J ^ x , = [ J ^ 2 , J ^ y ] = [ J ^ 2 , J ^ z ] = 0. (219)
The square of the total angular momentum operator J ^ 2 \hat{\mathbf{J}}^2 J ^ 2 may be expressed in terms of cartesian components J ^ i ( i = x , y , z ) \hat{J}_i(i=x,y,z) J ^ i ( i = x , y , z ) as
J ^ 2 = ∑ i J ^ i 2 = J ^ x 2 + J ^ y 2 + J ^ z 2 . (220) \hat{\mathbf{J}}^2=\sum_i\hat{J}_i^2=\hat{J}_x^2+\hat{J}_y^2+\hat{J}_z^2. \tag*{(220)} J ^ 2 = i ∑ J ^ i 2 = J ^ x 2 + J ^ y 2 + J ^ z 2 . (220)
Covariant spherical components of J ^ \hat{\mathbf{J}} J ^ satisfy the following commutation relations
[ J ^ μ , J ^ ν ] = − 2 C 1 μ 1 ν 1 λ J ^ λ , [ J ^ 2 , J ^ μ ] = 0 , ( μ , ν , λ = ± 1 , 0 ) . (221) [\hat{J}_\mu,\hat{J}_\nu]=-\sqrt{2}C^{1\lambda}_{1\mu\,1\nu}\hat{J}_\lambda,\qquad[\hat{\mathbf{J}}^2,\hat{J}_\mu]=0,\qquad(\mu,\nu,\lambda=\pm1,0). \tag*{(221)} [ J ^ μ , J ^ ν ] = − 2 C 1 μ 1 ν 1 λ J ^ λ , [ J ^ 2 , J ^ μ ] = 0 , ( μ , ν , λ = ± 1 , 0 ) . (221)
These relations may be obtained from (2.1.5) by putting a = e μ \mathbf{a}=\mathbf{e}_\mu a = e μ , b = e ν \mathbf{b}=\mathbf{e}_\nu b = e ν (μ , ν = ± 1 , 0 \mu,\nu=\pm1,0 μ , ν = ± 1 , 0 ), e μ \mathbf{e}_\mu e μ and e ν \mathbf{e}_\nu e ν being covariant spherical basis vectors. A more detailed form of (2.1.10) is
[ J ^ + 1 , J ^ + 1 ] = [ J ^ 0 , J ^ 0 ] = [ J ^ − 1 , J ^ − 1 ] = 0 , [ J ^ + 1 , J ^ 0 ] = − [ J ^ 0 , J ^ + 1 ] = − J ^ + 1 , [ J ^ + 1 , J ^ − 1 ] = − [ J ^ − 1 , J ^ + 1 ] = − J ^ 0 , [ J ^ 0 , J ^ − 1 ] = − [ J ^ − 1 , J ^ 0 ] = − J ^ − 1 , [ J ^ 2 , J ^ + 1 ] = [ J ^ 2 , J ^ 0 ] = [ J ^ 2 , J ^ − 1 ] = 0. (222) \begin{aligned}
[\hat{J}_{+1},\hat{J}_{+1}]&=[\hat{J}_{0},\hat{J}_{0}]=[\hat{J}_{-1},\hat{J}_{-1}]=0,\\
[\hat{J}_{+1},\hat{J}_{0}]&=-[\hat{J}_{0},\hat{J}_{+1}]=-\hat{J}_{+1},\qquad[\hat{J}_{+1},\hat{J}_{-1}]=-[\hat{J}_{-1},\hat{J}_{+1}]=-\hat{J}_{0},\\
[\hat{J}_{0},\hat{J}_{-1}]&=-[\hat{J}_{-1},\hat{J}_{0}]=-\hat{J}_{-1},\\
[\hat{\mathbf{J}}^2,\hat{J}_{+1}]&=[\hat{\mathbf{J}}^2,\hat{J}_{0}]=[\hat{\mathbf{J}}^2,\hat{J}_{-1}]=0. \tag*{(222)}
\end{aligned} [ J ^ + 1 , J ^ + 1 ] [ J ^ + 1 , J ^ 0 ] [ J ^ 0 , J ^ − 1 ] [ J ^ 2 , J ^ + 1 ] = [ J ^ 0 , J ^ 0 ] = [ J ^ − 1 , J ^ − 1 ] = 0 , = − [ J ^ 0 , J ^ + 1 ] = − J ^ + 1 , [ J ^ + 1 , J ^ − 1 ] = − [ J ^ − 1 , J ^ + 1 ] = − J ^ 0 , = − [ J ^ − 1 , J ^ 0 ] = − J ^ − 1 , = [ J ^ 2 , J ^ 0 ] = [ J ^ 2 , J ^ − 1 ] = 0. (222)
For contravariant spherical components of J ^ \hat{\mathbf{J}} J ^ the commutation relations may be written as
[ J ^ μ , J ^ ν ] = 2 C 1 μ 1 ν 1 λ J ^ λ , [ J ^ 2 , J ^ μ ] = 0 , ( μ , ν , λ = ± 1 , 0 ) (223) [\hat{J}^{\mu},\hat{J}^{\nu}]=\sqrt{2}C^{1\lambda}_{1\mu\,1\nu}\hat{J}^{\lambda},\qquad[\hat{\mathbf{J}}^2,\hat{J}^{\mu}]=0,\qquad(\mu,\nu,\lambda=\pm1,0) \tag*{(223)} [ J ^ μ , J ^ ν ] = 2 C 1 μ 1 ν 1 λ J ^ λ , [ J ^ 2 , J ^ μ ] = 0 , ( μ , ν , λ = ± 1 , 0 ) (223)
or, in a more detailed form
[ J ^ + 1 , J ^ + 1 ] = [ J ^ 0 , J ^ 0 ] = [ J ^ − 1 , J ^ − 1 ] = 0 , [ J ^ + 1 , J ^ 0 ] = − [ J ^ 0 , J ^ + 1 ] = J ^ + 1 , [ J ^ + 1 , J ^ − 1 ] = − [ J ^ − 1 , J ^ + 1 ] = J ^ 0 , [ J ^ 0 , J ^ − 1 ] = − [ J ^ − 1 , J ^ 0 ] = J ^ − 1 , [ J ^ 2 , J ^ + 1 ] = [ J ^ 2 , J ^ 0 ] = [ J ^ 2 , J ^ − 1 ] = 0. (224) \begin{aligned}
[\hat{J}^{+1},\hat{J}^{+1}]&=[\hat{J}^{0},\hat{J}^{0}]=[\hat{J}^{-1},\hat{J}^{-1}]=0,\\
[\hat{J}^{+1},\hat{J}^{0}]&=-[\hat{J}^{0},\hat{J}^{+1}]=\hat{J}^{+1},\qquad[\hat{J}^{+1},\hat{J}^{-1}]=-[\hat{J}^{-1},\hat{J}^{+1}]=\hat{J}^{0},\\
[\hat{J}^{0},\hat{J}^{-1}]&=-[\hat{J}^{-1},\hat{J}^{0}]=\hat{J}^{-1},\\
[\hat{\mathbf{J}}^2,\hat{J}^{+1}]&=[\hat{\mathbf{J}}^2,\hat{J}^{0}]=[\hat{\mathbf{J}}^2,\hat{J}^{-1}]=0. \tag*{(224)}
\end{aligned} [ J ^ + 1 , J ^ + 1 ] [ J ^ + 1 , J ^ 0 ] [ J ^ 0 , J ^ − 1 ] [ J ^ 2 , J ^ + 1 ] = [ J ^ 0 , J ^ 0 ] = [ J ^ − 1 , J ^ − 1 ] = 0 , = − [ J ^ 0 , J ^ + 1 ] = J ^ + 1 , [ J ^ + 1 , J ^ − 1 ] = − [ J ^ − 1 , J ^ + 1 ] = J ^ 0 , = − [ J ^ − 1 , J ^ 0 ] = J ^ − 1 , = [ J ^ 2 , J ^ 0 ] = [ J ^ 2 , J ^ − 1 ] = 0. (224)
The operator J ^ 2 \hat{\mathbf{J}}^2 J ^ 2 is expressed in terms of spherical components J ^ μ \hat{J}_\mu J ^ μ (μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ) as
J ^ 2 = ∑ μ ( − 1 ) μ J ^ − μ J ^ μ = − J ^ + 1 J ^ − 1 + J ^ 0 J ^ 0 − J ^ − 1 J ^ + 1 = J ^ 0 2 − J ^ 0 − 2 J ^ + 1 J ^ − 1 = J ^ 0 2 + J ^ 0 − 2 J ^ − 1 J ^ + 1 . (225) \hat{\mathbf{J}}^2=\sum_\mu(-1)^\mu\hat{J}_{-\mu}\hat{J}_\mu=-\hat{J}_{+1}\hat{J}_{-1}+\hat{J}_0\hat{J}_0-\hat{J}_{-1}\hat{J}_{+1}=\hat{J}_0^2-\hat{J}_0-2\hat{J}_{+1}\hat{J}_{-1}=\hat{J}_0^2+\hat{J}_0-2\hat{J}_{-1}\hat{J}_{+1}. \tag*{(225)} J ^ 2 = μ ∑ ( − 1 ) μ J ^ − μ J ^ μ = − J ^ + 1 J ^ − 1 + J ^ 0 J ^ 0 − J ^ − 1 J ^ + 1 = J ^ 0 2 − J ^ 0 − 2 J ^ + 1 J ^ − 1 = J ^ 0 2 + J ^ 0 − 2 J ^ − 1 J ^ + 1 . (225)
Coordinate Inversion. Time Reversal
The total angular momentum operator J ^ \hat{\mathbf{J}} J ^ is an axial vector, i.e., it is invariant with respect to coordinate inversion (r → − r \mathbf{r}\to-\mathbf{r} r → − r )
P ^ r J ^ i P ^ r − 1 = J ^ i , ( i = x , y , z ) , \hat{P}_r\hat{J}_i\hat{P}_r^{-1}=\hat{J}_i,\qquad(i=x,y,z), P ^ r J ^ i P ^ r − 1 = J ^ i , ( i = x , y , z ) ,
P ^ r J ^ μ P ^ r − 1 = J ^ μ , ( μ = ± 1 , 0 ) . (226) \hat{P}_r\hat{J}_\mu\hat{P}_r^{-1}=\hat{J}_\mu,\qquad(\mu=\pm1,0). \tag*{(226)} P ^ r J ^ μ P ^ r − 1 = J ^ μ , ( μ = ± 1 , 0 ) . (226)
Under time reversal (t → − t t\to-t t → − t ) J ^ \hat{\mathbf{J}} J ^ changes its sign
P ^ t J ^ i P ^ t − 1 = − J ^ i , ( i = x , y , z ) \hat{P}_t\hat{J}_i\hat{P}_t^{-1}=-\hat{J}_i,\qquad(i=x,y,z) P ^ t J ^ i P ^ t − 1 = − J ^ i , ( i = x , y , z )
P ^ t J ^ μ P ^ t − 1 = − J ^ μ , ( μ = ± 1 , 0 ) (227) \hat{P}_t\hat{J}_\mu\hat{P}_t^{-1}=-\hat{J}_\mu,\qquad(\mu=\pm1,0) \tag*{(227)} P ^ t J ^ μ P ^ t − 1 = − J ^ μ , ( μ = ± 1 , 0 ) (227)
In Eqs. (2.1.15) and (2.1.16) P ^ r \hat{P}_r P ^ r and P ^ t \hat{P}_t P ^ t represent the operators of coordinate inversion and time reversal, respectively.
Total Angular Momentum of a System. Orbital and Spin Angular Momenta
Total Angular Momentum of a System. Orbital and Spin Angular Momenta
The total angular momentum of some system consisting of N N N subsystems with angular momenta j ^ ( 1 ) , j ^ ( 2 ) , … , j ^ ( N ) \hat{\mathbf{j}}(1),\hat{\mathbf{j}}(2),\ldots,\hat{\mathbf{j}}(N) j ^ ( 1 ) , j ^ ( 2 ) , … , j ^ ( N ) is the vector sum
J ^ = ∑ n = 1 N j ^ ( n ) . (228) \hat{\mathbf{J}}=\sum_{n=1}^{N}\hat{\mathbf{j}}(n). \tag*{(228)} J ^ = n = 1 ∑ N j ^ ( n ) . (228)
For such a system the following commutation relations hold
[ j ^ i ( n ) , j ^ k ( n ′ ) ] = i δ n n ′ ε i k l j ^ l ( n ) , ( i , k , l = x , y , z ) , (229) [\hat{j}_i(n),\hat{j}_k(n')]=i\delta_{nn'}\varepsilon_{ikl}\hat{j}_l(n),\qquad(i,k,l=x,y,z), \tag*{(229)} [ j ^ i ( n ) , j ^ k ( n ′ )] = i δ n n ′ ε ik l j ^ l ( n ) , ( i , k , l = x , y , z ) , (229)
[ J ^ i , j ^ k ( n ) ] = i ε i k l j ^ l ( n ) , [\hat{J}_i,\hat{j}_k(n)]=i\varepsilon_{ikl}\hat{j}_l(n), [ J ^ i , j ^ k ( n )] = i ε ik l j ^ l ( n ) ,
[ j ^ μ ( n ) , j ^ ν ( n ′ ) ] = − 2 δ n n ′ C 1 μ 1 ν 1 λ j ^ λ ( n ) , ( μ , ν , λ = ± 1 , 0 ) . (230) [\hat{j}_\mu(n),\hat{j}_\nu(n')]=-\sqrt{2}\delta_{nn'}C_{1\mu\,1\nu}^{1\lambda}\hat{j}_\lambda(n),\qquad(\mu,\nu,\lambda=\pm1,0). \tag*{(230)} [ j ^ μ ( n ) , j ^ ν ( n ′ )] = − 2 δ n n ′ C 1 μ 1 ν 1 λ j ^ λ ( n ) , ( μ , ν , λ = ± 1 , 0 ) . (230)
[ J ^ μ , j ^ ν ( n ) ] = − 2 C 1 μ 1 ν 1 λ j ^ λ ( n ) . [\hat{J}_\mu,\hat{j}_\nu(n)]=-\sqrt{2}C_{1\mu\,1\nu}^{1\lambda}\hat{j}_\lambda(n). [ J ^ μ , j ^ ν ( n )] = − 2 C 1 μ 1 ν 1 λ j ^ λ ( n ) .
In particular, Eq. (228) determines the total angular momentum operator as the sum of the orbital angular momentum operator L ^ \hat{\mathbf{L}} L ^ and the spin angular momentum operator S ^ \hat{\mathbf{S}} S ^ .
J ^ = L ^ + S ^ . (231) \hat{\mathbf{J}}=\hat{\mathbf{L}}+\hat{\mathbf{S}}. \tag*{(231)} J ^ = L ^ + S ^ . (231)
The properties of the operators L ^ \hat{\mathbf{L}} L ^ and S ^ \hat{\mathbf{S}} S ^ are considered below (Secs. 2.2 and 2.3). The commutation relations for L ^ \hat{\mathbf{L}} L ^ and J ^ \hat{\mathbf{J}} J ^ are
[ J ^ i , L ^ k ] = i ε i k l L ^ l , ( i , k , l = x , y , z ) , [\hat{J}_i,\hat{L}_k]=i\varepsilon_{ikl}\hat{L}_l,(i,k,l=x,y,z), [ J ^ i , L ^ k ] = i ε ik l L ^ l , ( i , k , l = x , y , z ) ,
[ J ^ μ , L ^ ν ] = − 2 C 1 μ 1 ν 1 λ L ^ λ , ( μ , ν , λ = ± 1 , 0 ) . (232) [\hat{J}_\mu,\hat{L}_\nu]=-\sqrt{2}C_{1\mu\,1\nu}^{1\lambda}\hat{L}_\lambda,(\mu,\nu,\lambda=\pm1,0). \tag*{(232)} [ J ^ μ , L ^ ν ] = − 2 C 1 μ 1 ν 1 λ L ^ λ , ( μ , ν , λ = ± 1 , 0 ) . (232)
or, in more detail
[ J ^ x , L ^ x ] = [ J ^ y , L ^ y ] = [ J ^ z , L ^ z ] = 0 , [\hat{J}_x,\hat{L}_x]=[\hat{J}_y,\hat{L}_y]=[\hat{J}_z,\hat{L}_z]=0, [ J ^ x , L ^ x ] = [ J ^ y , L ^ y ] = [ J ^ z , L ^ z ] = 0 ,
[ J ^ x , L ^ y ] = − [ J ^ y , L ^ x ] = i L ^ z , [ J ^ x , L ^ z ] = − [ J ^ z , L ^ x ] = − i L ^ y , [\hat{J}_x,\hat{L}_y]=-[\hat{J}_y,\hat{L}_x]=i\hat{L}_z,\quad[\hat{J}_x,\hat{L}_z]=-[\hat{J}_z,\hat{L}_x]=-i\hat{L}_y, [ J ^ x , L ^ y ] = − [ J ^ y , L ^ x ] = i L ^ z , [ J ^ x , L ^ z ] = − [ J ^ z , L ^ x ] = − i L ^ y ,
[ J ^ y , L ^ z ] = − [ J ^ z , L ^ y ] = i L ^ x (233) [\hat{J}_y,\hat{L}_z]=-[\hat{J}_z,\hat{L}_y]=i\hat{L}_x \tag*{(233)} [ J ^ y , L ^ z ] = − [ J ^ z , L ^ y ] = i L ^ x (233)
[ J ^ + 1 , L ^ + 1 ] = [ J ^ 0 , L ^ 0 ] = [ J ^ − 1 , L ^ − 1 ] = 0 , [\hat{J}_{+1},\hat{L}_{+1}]=[\hat{J}_0,\hat{L}_0]=[\hat{J}_{-1},\hat{L}_{-1}]=0, [ J ^ + 1 , L ^ + 1 ] = [ J ^ 0 , L ^ 0 ] = [ J ^ − 1 , L ^ − 1 ] = 0 ,
[ J ^ + 1 , L ^ 0 ] = − [ J ^ 0 , L ^ + 1 ] = − L ^ + 1 , [ J ^ + 1 , L ^ − 1 ] = − [ J ^ − 1 , L ^ + 1 ] = − L ^ 0 , [\hat{J}_{+1},\hat{L}_0]=-[\hat{J}_0,\hat{L}_{+1}]=-\hat{L}_{+1},\quad[\hat{J}_{+1},\hat{L}_{-1}]=-[\hat{J}_{-1},\hat{L}_{+1}]=-\hat{L}_0, [ J ^ + 1 , L ^ 0 ] = − [ J ^ 0 , L ^ + 1 ] = − L ^ + 1 , [ J ^ + 1 , L ^ − 1 ] = − [ J ^ − 1 , L ^ + 1 ] = − L ^ 0 ,
[ J ^ 0 , L ^ − 1 ] = − [ J ^ − 1 , L ^ 0 ] = − L ^ − 1 . (234) [\hat{J}_0,\hat{L}_{-1}]=-[\hat{J}_{-1},\hat{L}_0]=-\hat{L}_{-1}. \tag*{(234)} [ J ^ 0 , L ^ − 1 ] = − [ J ^ − 1 , L ^ 0 ] = − L ^ − 1 . (234)
Appropriate relations for the spin operator S ^ \hat{\mathbf{S}} S ^ and the total angular momentum operator J ^ \hat{\mathbf{J}} J ^ may be obtained from Eqs. (2.1.21)–(2.1.23) by replacing L ^ → S ^ \hat{\mathbf{L}}\to\hat{\mathbf{S}} L ^ → S ^ . The orbital angular momentum operator L ^ \hat{\mathbf{L}} L ^ commutes with the spin operator S ^ \hat{\mathbf{S}} S ^
[ L ^ i , S ^ k ] = 0 , ( i , k = x , y , z ) [\hat{L}_i,\hat{S}_k]=0,\qquad(i,k=x,y,z) [ L ^ i , S ^ k ] = 0 , ( i , k = x , y , z )
[ L ^ μ , S ^ ν ] = 0 , ( μ , ν = ± 1 , 0 ) . (235) [\hat{L}_\mu,\hat{S}_\nu]=0,\qquad(\mu,\nu=\pm1,0). \tag*{(235)} [ L ^ μ , S ^ ν ] = 0 , ( μ , ν = ± 1 , 0 ) . (235)
Orbital angular momentum operator
Definition
In classical mechanics the angular momentum L \mathbf{L} L of a particle is defined as
L = [ r × p ] , (236) \mathbf{L}=[\mathbf{r}\times\mathbf{p}], \tag*{(236)} L = [ r × p ] , (236)
where r \mathbf{r} r is the position vector of the particle and p \mathbf{p} p is its linear momentum. In the quantum mechanics the orbital angular momentum operator L ^ \hat{\mathbf{L}} L ^ of a particle is obtained from Eq. (236) by replacing r → r ^ \mathbf{r}\to\hat{\mathbf{r}} r → r ^ and p → p ^ \mathbf{p}\to\hat{\mathbf{p}} p → p ^ , where r ^ \hat{\mathbf{r}} r ^ and p ^ \hat{\mathbf{p}} p ^ are the position and momentum operators, respectively.
In the coordinate representation we have r ^ = r \hat{\mathbf{r}}=\mathbf{r} r ^ = r , p ^ = − i ∇ \hat{\mathbf{p}}=-i\nabla p ^ = − i ∇ ,
L ^ = − i [ r × ∇ ] . (237) \hat{\mathbf{L}}=-i[\mathbf{r}\times\nabla]. \tag*{(237)} L ^ = − i [ r × ∇ ] . (237)
In the momentum representation we have r ^ = i ∇ p \hat{\mathbf{r}}=i\nabla_p r ^ = i ∇ p , p ^ = p \hat{\mathbf{p}}=\mathbf{p} p ^ = p ,
L ^ = − i [ p × ∇ p ] . (238) \hat{\mathbf{L}}=-i[\mathbf{p}\times\nabla_p]. \tag*{(238)} L ^ = − i [ p × ∇ p ] . (238)
The formulas below are valid for both representations. The explicit form of L ^ \hat{\mathbf{L}} L ^ (see § 10.2.3 ) in the momentum representation is obtained by substituting r → p \mathbf{r}\to\mathbf{p} r → p , ∇ ≡ ∂ / ∂ r → ∇ p ≡ ∂ / ∂ p \nabla\equiv\partial/\partial\mathbf{r}\to\nabla_p\equiv\partial/\partial\mathbf{p} ∇ ≡ ∂ / ∂ r → ∇ p ≡ ∂ / ∂ p .
The operator L ^ \hat{\mathbf{L}} L ^ is Hermitian and purely imaginary.
L ^ + = L ^ , L ^ ∗ = − L ^ . (239) \hat{\mathbf{L}}^{+}=\hat{\mathbf{L}},\qquad\hat{\mathbf{L}}^{*}=-\hat{\mathbf{L}}. \tag*{(239)} L ^ + = L ^ , L ^ ∗ = − L ^ . (239)
For cartesian components of L ^ \hat{\mathbf{L}} L ^ , Eqs. (2.2.4) become
L ^ i + = L ^ i , L ^ i ∗ = − L ^ i , ( i = x , y , z ) . (240) \hat{L}_i^{+}=\hat{L}_i,\qquad\hat{L}_i^{*}=-\hat{L}_i,\qquad(i=x,y,z). \tag*{(240)} L ^ i + = L ^ i , L ^ i ∗ = − L ^ i , ( i = x , y , z ) . (240)
and for spherical components Eqs. (2.2.4) yield
( L ^ μ ) + = ( − 1 ) μ ( L ^ + ) − μ = ( − 1 ) μ L ^ − μ , \left(\hat{L}_{\mu}\right)^{+}=(-1)^{\mu}\left(\hat{L}^{+}\right)_{-\mu}=(-1)^{\mu}\hat{L}_{-\mu}, ( L ^ μ ) + = ( − 1 ) μ ( L ^ + ) − μ = ( − 1 ) μ L ^ − μ ,
( L ^ μ ) ∗ = ( − 1 ) μ ( L ^ ∗ ) − μ = ( − 1 ) μ + 1 L ^ − μ , ( μ = ± 1 , 0 ) . (241) \left(\hat{L}_{\mu}\right)^{*}=(-1)^{\mu}\left(\hat{L}^{*}\right)_{-\mu}=(-1)^{\mu+1}\hat{L}_{-\mu},\qquad(\mu=\pm1,0). \tag*{(241)} ( L ^ μ ) ∗ = ( − 1 ) μ ( L ^ ∗ ) − μ = ( − 1 ) μ + 1 L ^ − μ , ( μ = ± 1 , 0 ) . (241)
The orbital angular momentum operator L ^ \hat{\mathbf{L}} L ^ generates transformations of scalar (spinless) wave functions under rotations of the coordinate system. A rotation through an infinitesimal angle δ ω \delta\omega δ ω about the n \mathbf{n} n -axis transforms the position vector r \mathbf{r} r into r + δ r \mathbf{r}+\delta\mathbf{r} r + δ r , with δ r = − δ ω [ n × r ] \delta\mathbf{r}=-\delta\omega[\mathbf{n}\times\mathbf{r}] δ r = − δ ω [ n × r ] . The corresponding transformation of the scalar wave functions reads
Ψ ( r ) → Ψ ( r + δ r ) = ( 1 + δ r ⋅ ∇ ) Ψ ( r ) = ( 1 − i δ ω n ⋅ L ^ ) Ψ ( r ) . (242) \Psi(\mathbf{r})\to\Psi(\mathbf{r}+\delta\mathbf{r})=(1+\delta\mathbf{r}\cdot\nabla)\Psi(\mathbf{r})=(1-i\delta\omega\mathbf{n}\cdot\hat{\mathbf{L}})\Psi(\mathbf{r}). \tag*{(242)} Ψ ( r ) → Ψ ( r + δ r ) = ( 1 + δ r ⋅ ∇ ) Ψ ( r ) = ( 1 − i δ ω n ⋅ L ^ ) Ψ ( r ) . (242)
In the case of a finite rotation through an angle ω \omega ω about the axis n \mathbf{n} n , scalar functions are transformed by the operator e − i ω n ⋅ L ^ e^{-i\omega\mathbf{n}\cdot\hat{\mathbf{L}}} e − iω n ⋅ L ^ which is the particular form of the rotation operator 1.4.32. The eigenfunctions of the operators L ^ 2 \hat{L}^{2} L ^ 2 and L ^ z \hat{L}_{z} L ^ z represent spherical harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) (Chap. 5) which depend on the polar angles ϑ , φ \vartheta,\varphi ϑ , φ .
Commutation Relations
The orbital angular momentum operator L ^ \hat{\mathbf{L}} L ^ satisfies the same commutation relations as the total angular momentum operator J ^ \hat{\mathbf{J}} J ^ . Appropriate expressions for the commutators are readily obtained from Eqs. (2.1.5)–(2.1.14) by substituting J ^ → L ^ \hat{\mathbf{J}}\to\hat{\mathbf{L}} J ^ → L ^ .
The commutation rules for the operator L ^ \hat{\mathbf{L}} L ^ with the position operator r ^ \hat{\mathbf{r}} r ^ and the momentum operator p ^ \hat{\mathbf{p}} p ^ are the following.
For cartesian components
\begin{aligned}
[\hat{L}_i,\hat{x}_k]&=i\varepsilon_{ikl}x_l,\\
[\hat{L}_i,\hat{p}_k]&=i\varepsilon_{ikl}p_l,
\end{aligned}
\qquad(i,k,l=x,y,z). \tag*{(243)}\end{aligned}
Note also that r ^ \hat{\mathbf{r}} r ^ and p ^ \hat{\mathbf{p}} p ^ obey the commutation relations
[ x ^ i , x ^ k ] = 0 , [ p ^ i , p ^ k ] = 0 , [ x ^ i , p ^ k ] = i δ i k , ( i , k = x , y , z ) . (244) [\hat{x}_i,\hat{x}_k]=0,\qquad[\hat{p}_i,\hat{p}_k]=0,\qquad[\hat{x}_i,\hat{p}_k]=i\delta_{ik},\qquad(i,k=x,y,z). \tag*{(244)} [ x ^ i , x ^ k ] = 0 , [ p ^ i , p ^ k ] = 0 , [ x ^ i , p ^ k ] = i δ ik , ( i , k = x , y , z ) . (244)
For spherical components
\begin{aligned}
[\hat{L}_{\mu},\hat{x}_{\nu}]&=-\sqrt{2}C^{1\lambda}_{1\mu1\nu}\hat{x}_{\lambda},\\
[\hat{L}_{\mu},\hat{p}_{\nu}]&=-\sqrt{2}C^{1\lambda}_{1\mu1\nu}\hat{p}_{\lambda},
\end{aligned}
\qquad(\mu,\nu,\lambda=\pm1,0). \tag*{(245)}\end{aligned}
The commutation relations for the spherical components of r ^ \hat{\mathbf{r}} r ^ and p ^ \hat{\mathbf{p}} p ^ are given by
[ x ^ μ , x ^ ν ] = 0 , [ p ^ μ , p ^ ν ] = 0 , [ x ^ μ , p ^ ν ] = i δ μ − ν ( − 1 ) μ , ( μ , ν = ± 1 , 0 ) . (246) [\hat{x}_{\mu},\hat{x}_{\nu}]=0,\qquad[\hat{p}_{\mu},\hat{p}_{\nu}]=0,\qquad[\hat{x}_{\mu},\hat{p}_{\nu}]=i\delta_{\mu-\nu}(-1)^{\mu},\qquad(\mu,\nu=\pm1,0). \tag*{(246)} [ x ^ μ , x ^ ν ] = 0 , [ p ^ μ , p ^ ν ] = 0 , [ x ^ μ , p ^ ν ] = i δ μ − ν ( − 1 ) μ , ( μ , ν = ± 1 , 0 ) . (246)
Equation (2.2.8) and (2.2.10) show that the operators r ^ \hat{\mathbf{r}} r ^ and p ^ \hat{\mathbf{p}} p ^ are vectors.
The commutation relations for square of angular momentum operator L ^ 2 \hat{L}^{2} L ^ 2 are
[ L ^ 2 , r ^ ] = i [ r ^ × L ^ ] − i [ L ^ × r ^ ] , [\hat{L}^{2},\hat{\mathbf{r}}]=i[\hat{\mathbf{r}}\times\hat{\mathbf{L}}]-i[\hat{\mathbf{L}}\times\hat{\mathbf{r}}], [ L ^ 2 , r ^ ] = i [ r ^ × L ^ ] − i [ L ^ × r ^ ] ,
[ L ^ 2 , p ^ ] = i [ p ^ × L ^ ] − i [ L ^ × p ^ ] . (247) [\hat{L}^{2},\hat{\mathbf{p}}]=i[\hat{\mathbf{p}}\times\hat{\mathbf{L}}]-i[\hat{\mathbf{L}}\times\hat{\mathbf{p}}]. \tag*{(247)} [ L ^ 2 , p ^ ] = i [ p ^ × L ^ ] − i [ L ^ × p ^ ] . (247)
The operators r ^ 2 \hat{\mathbf{r}}^2 r ^ 2 and p ^ 2 \hat{\mathbf{p}}^2 p ^ 2 commute with L ^ \hat{\mathbf{L}} L ^ :
[ L ^ , r ^ 2 ] = 0 , [ L ^ , p ^ 2 ] = 0 \left[\hat{\mathbf{L}},\hat{\mathbf{r}}^2\right]=0,\qquad\left[\hat{\mathbf{L}},\hat{\mathbf{p}}^2\right]=0 [ L ^ , r ^ 2 ] = 0 , [ L ^ , p ^ 2 ] = 0
The commutation rules of L ^ \hat{\mathbf{L}} L ^ with the total angular momentum operator J ^ \hat{\mathbf{J}} J ^ and the spin operator S ^ \hat{\mathbf{S}} S ^ have been considered above (Eqs. (232) –(235) ).
Explicit Form
The cartesian components of the operator L ^ \hat{\mathbf{L}} L ^ are given by
L ^ x = − i ( y ∂ ∂ z − z ∂ ∂ y ) , L ^ y = − i ( z ∂ ∂ x − x ∂ ∂ z ) , L ^ z = − i ( x ∂ ∂ y − y ∂ ∂ x ) . (248) \hat{L}_x=-i\left(y\frac{\partial}{\partial z}-z\frac{\partial}{\partial y}\right),\qquad\hat{L}_y=-i\left(z\frac{\partial}{\partial x}-x\frac{\partial}{\partial z}\right),\qquad\hat{L}_z=-i\left(x\frac{\partial}{\partial y}-y\frac{\partial}{\partial x}\right). \tag*{(248)} L ^ x = − i ( y ∂ z ∂ − z ∂ y ∂ ) , L ^ y = − i ( z ∂ x ∂ − x ∂ z ∂ ) , L ^ z = − i ( x ∂ y ∂ − y ∂ x ∂ ) . (248)
or in a more compact form
L ^ i = − i ∑ k , l ε i k l x k ∂ ∂ x l , ( i , k , l = x , y , z ) . (249) \hat{L}_i=-i\sum_{k,l}\varepsilon_{ikl}x_k\frac{\partial}{\partial x_l},\qquad(i,k,l=x,y,z). \tag*{(249)} L ^ i = − i k , l ∑ ε ik l x k ∂ x l ∂ , ( i , k , l = x , y , z ) . (249)
The cartesian components of L ^ \hat{\mathbf{L}} L ^ are expressed in terms of polar angles ϑ , φ \vartheta,\varphi ϑ , φ as
L ^ x = i ( sin φ ∂ ∂ ϑ + cot ϑ cos φ ∂ ∂ φ ) \hat{L}_x=i\left(\sin\varphi\frac{\partial}{\partial\vartheta}+\cot\vartheta\cos\varphi\frac{\partial}{\partial\varphi}\right) L ^ x = i ( sin φ ∂ ϑ ∂ + cot ϑ cos φ ∂ φ ∂ )
L ^ y = i ( − cos φ ∂ ∂ ϑ + cot ϑ sin φ ∂ ∂ φ ) \hat{L}_y=i\left(-\cos\varphi\frac{\partial}{\partial\vartheta}+\cot\vartheta\sin\varphi\frac{\partial}{\partial\varphi}\right) L ^ y = i ( − cos φ ∂ ϑ ∂ + cot ϑ sin φ ∂ φ ∂ )
L ^ z = − i ∂ ∂ φ . (250) \hat{L}_z=-i\frac{\partial}{\partial\varphi}. \tag*{(250)} L ^ z = − i ∂ φ ∂ . (250)
The spherical components of L ^ \hat{\mathbf{L}} L ^ are written as
L ^ + 1 = x 0 ∇ + 1 − x + 1 ∇ 0 , \hat{L}_{+1}=x_0\nabla_{+1}-x_{+1}\nabla_0, L ^ + 1 = x 0 ∇ + 1 − x + 1 ∇ 0 ,
L ^ 0 = x − 1 ∇ + 1 − x + 1 ∇ − 1 , \hat{L}_0=x_{-1}\nabla_{+1}-x_{+1}\nabla_{-1}, L ^ 0 = x − 1 ∇ + 1 − x + 1 ∇ − 1 ,
L ^ − 1 = x − 1 ∇ 0 − x 0 ∇ − 1 . (251) \hat{L}_{-1}=x_{-1}\nabla_0-x_0\nabla_{-1}. \tag*{(251)} L ^ − 1 = x − 1 ∇ 0 − x 0 ∇ − 1 . (251)
or, in a more compact form
L ^ μ = − 2 ∑ ν , λ C 1 ν 1 λ 1 μ x ν ∇ λ , ( μ , ν , λ = ± 1 , 0 ) . (252) \hat{L}_\mu=-\sqrt{2}\sum_{\nu,\lambda}C^{1\mu}_{1\nu1\lambda}x_\nu\nabla_\lambda,\qquad(\mu,\nu,\lambda=\pm1,0). \tag*{(252)} L ^ μ = − 2 ν , λ ∑ C 1 ν 1 λ 1 μ x ν ∇ λ , ( μ , ν , λ = ± 1 , 0 ) . (252)
The spherical components of L ^ \hat{\mathbf{L}} L ^ may also be expressed in terms of polar angles ϑ , φ \vartheta,\varphi ϑ , φ as
L ^ + 1 = − 1 2 e i φ { ∂ ∂ ϑ + i cot ϑ ∂ ∂ φ } , \hat{L}_{+1}=-\frac{1}{\sqrt{2}}e^{i\varphi}\left\{\frac{\partial}{\partial\vartheta}+i\cot\vartheta\frac{\partial}{\partial\varphi}\right\}, L ^ + 1 = − 2 1 e i φ { ∂ ϑ ∂ + i cot ϑ ∂ φ ∂ } ,
L ^ 0 = − i ∂ ∂ φ , \hat{L}_0=-i\frac{\partial}{\partial\varphi}, L ^ 0 = − i ∂ φ ∂ ,
L ^ − 1 = − 1 2 e − i φ { ∂ ∂ ϑ − i cot ϑ ∂ ∂ φ } . (253) \hat{L}_{-1}=-\frac{1}{\sqrt{2}}e^{-i\varphi}\left\{\frac{\partial}{\partial\vartheta}-i\cot\vartheta\frac{\partial}{\partial\varphi}\right\}. \tag*{(253)} L ^ − 1 = − 2 1 e − i φ { ∂ ϑ ∂ − i cot ϑ ∂ φ ∂ } . (253)
For polar components of L ^ \hat{\mathbf{L}} L ^ we have
L ^ r = 0 , L ^ ϑ = i sin ϑ ∂ ∂ φ , L ^ φ = − i ∂ ∂ ϑ . (254) \hat{L}_r=0,\qquad\hat{L}_\vartheta=\frac{i}{\sin\vartheta}\frac{\partial}{\partial\varphi},\qquad\hat{L}_\varphi=-i\frac{\partial}{\partial\vartheta}. \tag*{(254)} L ^ r = 0 , L ^ ϑ = sin ϑ i ∂ φ ∂ , L ^ φ = − i ∂ ϑ ∂ . (254)
The square of the orbital angular momentum operator L ^ 2 \hat{\mathbf{L}}^2 L ^ 2 is expressed in terms of its components as
L ^ 2 = ∑ i L ^ i 2 = L ^ x 2 + L ^ y 2 + L ^ z 2 , \hat{\mathbf{L}}^2=\sum_i\hat{L}_i^2=\hat{L}_x^2+\hat{L}_y^2+\hat{L}_z^2, L ^ 2 = i ∑ L ^ i 2 = L ^ x 2 + L ^ y 2 + L ^ z 2 ,
L ^ 2 = ∑ μ ( − 1 ) μ L ^ μ L ^ − μ = − L ^ + 1 L ^ − 1 + L ^ 0 L ^ 0 − L ^ − 1 L ^ + 1 . (255) \hat{\mathbf{L}}^2=\sum_\mu(-1)^\mu\hat{L}_\mu\hat{L}_{-\mu}=-\hat{L}_{+1}\hat{L}_{-1}+\hat{L}_0\hat{L}_0-\hat{L}_{-1}\hat{L}_{+1}. \tag*{(255)} L ^ 2 = μ ∑ ( − 1 ) μ L ^ μ L ^ − μ = − L ^ + 1 L ^ − 1 + L ^ 0 L ^ 0 − L ^ − 1 L ^ + 1 . (255)
We may also write L ^ 2 \hat{\mathbf{L}}^2 L ^ 2 in the form
L ^ 2 = − [ 1 sin ϑ ∂ ∂ ϑ ( sin ϑ ∂ ∂ ϑ ) + 1 sin 2 ϑ ∂ 2 ∂ φ 2 ] . (256) \hat{\mathbf{L}}^2=-\left[\frac{1}{\sin\vartheta}\frac{\partial}{\partial\vartheta}\left(\sin\vartheta\frac{\partial}{\partial\vartheta}\right)+\frac{1}{\sin^2\vartheta}\frac{\partial^2}{\partial\varphi^2}\right]. \tag*{(256)} L ^ 2 = − [ sin ϑ 1 ∂ ϑ ∂ ( sin ϑ ∂ ϑ ∂ ) + sin 2 ϑ 1 ∂ φ 2 ∂ 2 ] . (256)
This operator differs only in sign from the angular part of the Laplacian operator (Eq. (109) )
L ^ 2 = − Δ Ω . (257) \hat{L}^{2}=-\Delta_{\Omega}. \tag*{(257)} L ^ 2 = − Δ Ω . (257)
The operator L ^ 2 \hat{L}^{2} L ^ 2 is Hermitian and real
( L ^ 2 ) + = ( L ^ 2 ) ∗ = L ^ 2 . (258) \left(\hat{L}^{2}\right)^{+}=\left(\hat{L}^{2}\right)^{*}=\hat{L}^{2}. \tag*{(258)} ( L ^ 2 ) + = ( L ^ 2 ) ∗ = L ^ 2 . (258)
The representation of the orbital angular momentum L ^ \hat{\mathbf{L}} L ^ in the form of a differential operator is only one of possible representations. Alternatively, L ^ \hat{\mathbf{L}} L ^ may be represented by a set of three matrices (because it has three components). These matrix elements of L ^ \hat{\mathbf{L}} L ^ will be given in Chap. 13.
Spin angular momentum operator
Definition
The spin angular momentum operator or briefly the spin operator S ^ \hat{\mathbf{S}} S ^ is usually represented by a set of three (since the vector S ^ \hat{\mathbf{S}} S ^ has three components) square ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrices, S S S being the particle spin. These matrices act on the spin functions (see Chap. 6) and satisfy the same commutation relations as the components of the total angular momentum operator given in Sec. 2.1. The spin operator S ^ \hat{\mathbf{S}} S ^ is Hermitian:
S ^ + = S ^ (259) \hat{\mathbf{S}}^{+}=\hat{\mathbf{S}} \tag*{(259)} S ^ + = S ^ (259)
For the cartesian components of S ^ \hat{\mathbf{S}} S ^ the Hermitian property ((2.3.1)) has the form
( S ^ i ) + = S ^ i , ( i = x , y , z ) (260) \left(\hat{S}_{i}\right)^{+}=\hat{S}_{i},\quad(i=x,y,z) \tag*{(260)} ( S ^ i ) + = S ^ i , ( i = x , y , z ) (260)
and for spherical components
( S ^ μ ) + = ( − 1 ) μ S ^ − μ , ( μ = ± 1 , 0 ) . (261) \left(\hat{S}_{\mu}\right)^{+}=(-1)^{\mu}\hat{S}_{-\mu},\quad(\mu=\pm1,0). \tag*{(261)} ( S ^ μ ) + = ( − 1 ) μ S ^ − μ , ( μ = ± 1 , 0 ) . (261)
The behaviour of S ^ \hat{\mathbf{S}} S ^ under complex conjugation depends on the representation of S ^ \hat{\mathbf{S}} S ^ (see, e.g., Sec. 2.6.2). The eigenfunctions of the operators S ^ 2 \hat{\mathbf{S}}^{2} S ^ 2 and S ^ z \hat{S}_{z} S ^ z are spin functions χ S m \chi_{Sm} χ S m (see Chap. 6) which depend on the spin variable σ \sigma σ . These functions have ( 2 S + 1 ) (2S+1) ( 2 S + 1 ) components and describe polarization states of a particle.
Commutation Relations
The commutation relations for the spin operator S ^ \hat{\mathbf{S}} S ^ are given by Eqs. (2.1.5)–(2.1.14) in which one must replace J ^ \hat{\mathbf{J}} J ^ by S ^ \hat{\mathbf{S}} S ^ . The commutation rules for S ^ \hat{\mathbf{S}} S ^ and the orbital angular momentum operator L ^ \hat{\mathbf{L}} L ^ have already been considered in Sec. 2.1.4. The spin operator S ^ \hat{\mathbf{S}} S ^ commutes with the position operator r ^ \hat{\mathbf{r}} r ^ and the momentum operator p ^ \hat{\mathbf{p}} p ^ ,
[ S ^ i , x ^ k ] = 0 , [ S ^ i , p ^ k ] = 0 , ( i , k = x , y , z ) , (262) [\hat{S}_{i},\hat{x}_{k}]=0,\qquad[\hat{S}_{i},\hat{p}_{k}]=0,\quad(i,k=x,y,z), \tag*{(262)} [ S ^ i , x ^ k ] = 0 , [ S ^ i , p ^ k ] = 0 , ( i , k = x , y , z ) , (262)
[ S ^ μ , x ^ ν ] = 0 , [ S ^ μ , p ^ ν ] = 0 , ( μ , ν = ± 1 , 0 ) . (263) [\hat{S}_{\mu},\hat{x}_{\nu}]=0,\qquad[\hat{S}_{\mu},\hat{p}_{\nu}]=0,\quad(\mu,\nu=\pm1,0). \tag*{(263)} [ S ^ μ , x ^ ν ] = 0 , [ S ^ μ , p ^ ν ] = 0 , ( μ , ν = ± 1 , 0 ) . (263)
Explicit Form
The spherical components of the spin operator S ^ \hat{\mathbf{S}} S ^ may be expressed in terms of the basis spin functions χ S m \chi_{Sm} χ S m (Chap. 6) as
S ^ μ = S ( S + 1 ) ∑ m , m ′ C S m 1 μ S m ′ χ S m ′ χ S m + , ( μ = ± 1 , 0 ) . (264) \hat{S}_{\mu}=\sqrt{S(S+1)}\sum_{m,m'}C^{Sm'}_{Sm\,1\mu}\chi_{Sm'}\chi^{+}_{Sm},\quad(\mu=\pm1,0). \tag*{(264)} S ^ μ = S ( S + 1 ) m , m ′ ∑ C S m 1 μ S m ′ χ S m ′ χ S m + , ( μ = ± 1 , 0 ) . (264)
Cartesian components of S ^ \hat{\mathbf{S}} S ^ may be obtained from (2.3.6), using the relations
S ^ x = 1 2 ( S ^ − 1 − S ^ + 1 ) , S ^ y = i 2 ( S ^ − 1 + S ^ + 1 ) , S ^ z = S ^ 0 . (265) \hat{S}_{x}=\frac{1}{\sqrt{2}}\left(\hat{S}_{-1}-\hat{S}_{+1}\right),\qquad\hat{S}_{y}=\frac{i}{\sqrt{2}}\left(\hat{S}_{-1}+\hat{S}_{+1}\right),\qquad\hat{S}_{z}=\hat{S}_{0}. \tag*{(265)} S ^ x = 2 1 ( S ^ − 1 − S ^ + 1 ) , S ^ y = 2 i ( S ^ − 1 + S ^ + 1 ) , S ^ z = S ^ 0 . (265)
Equations (2.3.6) and (2.3.7) are independent of representation used for spin functions, while an explicit form of the spin matrices depends on this representation. For an arbitrary spin S S S the simplest form of the spin matrices is that in the spherical basis representation. The basis spin functions χ S m \chi_{Sm} χ S m in this representation are given by
χ S m ( σ ) = δ m σ , ( m , σ = − S , − S + 1 , … , S − 1 , S ) , (266) \chi_{Sm}(\sigma)=\delta_{m\sigma},\quad(m,\sigma=-S,-S+1,\ldots,S-1,S), \tag*{(266)} χ S m ( σ ) = δ mσ , ( m , σ = − S , − S + 1 , … , S − 1 , S ) , (266)
and the matrix elements of the spin matrices are
( S ^ μ ) σ ′ σ = S ( S + 1 ) C S σ 1 μ σ ′ , ( σ , σ ′ = − S , − S + 1 , … , S − 1 , S ) (267) \left(\hat{S}_{\mu}\right)_{\sigma'\sigma}=\sqrt{S(S+1)}C^{\sigma'}_{S\sigma\,1\mu},\qquad(\sigma,\sigma'=-S,-S+1,\ldots,S-1,S) \tag*{(267)} ( S ^ μ ) σ ′ σ = S ( S + 1 ) C S σ 1 μ σ ′ , ( σ , σ ′ = − S , − S + 1 , … , S − 1 , S ) (267)
The elements are arranged in the matrices as follows
A = ( A S S ⋯ A S − S ⋮ ⋮ A − S S ⋯ A − S − S ) . (268) A=\begin{pmatrix} A_{SS} & \cdots& A_{S-S} \\ \vdots& & \vdots\\ A_{-SS} & \cdots& A_{-S-S} \end{pmatrix}. \tag*{(268)} A = A S S ⋮ A − S S ⋯ ⋯ A S − S ⋮ A − S − S . (268)
Explicit forms of the spin matrices for spin values 1 2 \frac{1}{2} 2 1 and 1 1 1 are given below in Secs. 2.5 and 2.6.
Traces of Products of Spin Matrices
The traces of products of cartesian components of the spin matrices can be evaluated for an arbitrary spin S S S by use of the following formulas (where i , k , l i,k,l i , k , l , etc. take the values x , y , z x,y,z x , y , z )
Tr { S ^ i } = 0 \operatorname{Tr}\{\hat{S}_i\}=0 Tr { S ^ i } = 0
Tr { S ^ i S ^ k } = S ( S + 1 ) ( 2 S + 1 ) 3 δ i k , \operatorname{Tr}\{\hat{S}_i\hat{S}_k\}=\frac{S(S+1)(2S+1)}{3}\delta_{ik}, Tr { S ^ i S ^ k } = 3 S ( S + 1 ) ( 2 S + 1 ) δ ik ,
Tr { S ^ i S ^ k S ^ l } = i S ( S + 1 ) ( 2 S + 1 ) 6 ε i k l , \operatorname{Tr}\{\hat{S}_i\hat{S}_k\hat{S}_l\}=\mathrm{i}\frac{S(S+1)(2S+1)}{6}\varepsilon_{ikl}, Tr { S ^ i S ^ k S ^ l } = i 6 S ( S + 1 ) ( 2 S + 1 ) ε ik l ,
Tr { S ^ i S ^ k S ^ l S ^ j } = S ( S + 1 ) ( 2 S + 1 ) 15 { [ S ( S + 1 ) + 1 2 ] ( δ i k δ l j + δ i j δ k l ) + [ S ( S + 1 ) − 2 ] δ i l δ k j } . (269) \operatorname{Tr}\{\hat{S}_i\hat{S}_k\hat{S}_l\hat{S}_j\}=\frac{S(S+1)(2S+1)}{15}\left\{\left[S(S+1)+\frac{1}{2}\right](\delta_{ik}\delta_{lj}+\delta_{ij}\delta_{kl})+[S(S+1)-2]\delta_{il}\delta_{kj}\right\}. \tag*{(269)} Tr { S ^ i S ^ k S ^ l S ^ j } = 15 S ( S + 1 ) ( 2 S + 1 ) { [ S ( S + 1 ) + 2 1 ] ( δ ik δ l j + δ ij δ k l ) + [ S ( S + 1 ) − 2 ] δ i l δ k j } . (269)
For spherical components of the spin operator S ^ \hat{S} S ^ the following relations hold (μ , ν , λ \mu,\nu,\lambda μ , ν , λ , etc. take the values ± 1 , 0 \pm1,0 ± 1 , 0 )
Tr { S ^ μ } = 0 , \operatorname{Tr}\{\hat{S}_{\mu}\}=0, Tr { S ^ μ } = 0 ,
Tr { S ^ μ S ^ ν } = S ( S + 1 ) ( 2 S + 1 ) 3 ( − 1 ) μ δ μ − ν , \operatorname{Tr}\{\hat{S}_{\mu}\hat{S}_{\nu}\}=\frac{S(S+1)(2S+1)}{3}(-1)^{\mu}\delta_{\mu-\nu}, Tr { S ^ μ S ^ ν } = 3 S ( S + 1 ) ( 2 S + 1 ) ( − 1 ) μ δ μ − ν ,
Tr { S ^ μ S ^ ν S ^ λ } = − S ( S + 1 ) ( 2 S + 1 ) 6 ( 1 1 1 μ ν λ ) = S ( S + 1 ) ( 2 S + 1 ) 3 2 ( − 1 ) 1 + λ C 1 μ 1 ν 1 − λ , (270) \operatorname{Tr}\{\hat{S}_{\mu}\hat{S}_{\nu}\hat{S}_{\lambda}\}=-\frac{S(S+1)(2S+1)}{\sqrt{6}}\begin{pmatrix}1&1&1\\\mu&\nu&\lambda\end{pmatrix}=\frac{S(S+1)(2S+1)}{3\sqrt{2}}(-1)^{1+\lambda}C^{1-\lambda}_{1\mu\,1\nu}, \tag*{(270)} Tr { S ^ μ S ^ ν S ^ λ } = − 6 S ( S + 1 ) ( 2 S + 1 ) ( 1 μ 1 ν 1 λ ) = 3 2 S ( S + 1 ) ( 2 S + 1 ) ( − 1 ) 1 + λ C 1 μ 1 ν 1 − λ , (270)
Tr { S ^ μ S ^ ν S ^ λ S ^ ρ } = S ( S + 1 ) ( 2 S + 1 ) 15 { [ S ( S + 1 ) + 1 2 ] ( − 1 ) μ + λ ( δ μ − ν δ λ − ρ + δ μ − ρ δ ν − λ ) + [ S ( S + 1 ) − 2 ] ( − 1 ) μ + ν δ μ − λ δ ν − ρ } . \operatorname{Tr}\{\hat{S}_{\mu}\hat{S}_{\nu}\hat{S}_{\lambda}\hat{S}_{\rho}\}=\frac{S(S+1)(2S+1)}{15}\left\{\left[S(S+1)+\frac{1}{2}\right](-1)^{\mu+\lambda}(\delta_{\mu-\nu}\delta_{\lambda-\rho}+\delta_{\mu-\rho}\delta_{\nu-\lambda})+[S(S+1)-2](-1)^{\mu+\nu}\delta_{\mu-\lambda}\delta_{\nu-\rho}\right\}. Tr { S ^ μ S ^ ν S ^ λ S ^ ρ } = 15 S ( S + 1 ) ( 2 S + 1 ) { [ S ( S + 1 ) + 2 1 ] ( − 1 ) μ + λ ( δ μ − ν δ λ − ρ + δ μ − ρ δ ν − λ ) + [ S ( S + 1 ) − 2 ] ( − 1 ) μ + ν δ μ − λ δ ν − ρ } .
Some analytic expressions for traces of products of many spin matrices in the spherical basis are given in Refs. [131, 132, 133].
Polarization operators
Definition
To describe a polarization (i.e., spin) state of a particle the so-called polarization operators are widely used. The polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) (M = − L , − L + 1 , … , L − 1 , L M=-L,-L+1,\ldots,L-1,L M = − L , − L + 1 , … , L − 1 , L and L = 0 , 1 , … , 2 S L=0,1,\ldots,2S L = 0 , 1 , … , 2 S ; L L L and M M M being integers) are ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrices which act on spin functions (see Chap. 6) and transform under rotations of the coordinate system according to the representation D L D^L D L . In other words, T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) are irreducible tensors of rank L L L . Such transformation properties of T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) with respect to rotations of the coordinate system imply the following commutation relations with spherical components of the spin operator S ^ μ \hat{S}_{\mu} S ^ μ (μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 )
[ S ^ μ , T ^ L M ( S ) ] = L ( L + 1 ) C L M 1 μ L M + μ T ^ L M + μ ( S ) . (271) [\hat{S}_{\mu},\hat{T}_{LM}(S)]=\sqrt{L(L+1)}C^{L\,M+\mu}_{L\,M\,1\,\mu}\hat{T}_{L\,M+\mu}(S). \tag*{(271)} [ S ^ μ , T ^ L M ( S )] = L ( L + 1 ) C L M 1 μ L M + μ T ^ L M + μ ( S ) . (271)
Let us normalize the polarization operators by the condition
Tr { T ^ L M + ( S ) T ^ L ′ M ′ ( S ) } = δ L L ′ δ M M ′ . (272) \operatorname{Tr}\{\hat{T}^{+}_{LM}(S)\hat{T}_{L'M'}(S)\}=\delta_{LL'}\delta_{MM'}. \tag*{(272)} Tr { T ^ L M + ( S ) T ^ L ′ M ′ ( S )} = δ L L ′ δ M M ′ . (272)
and choose the phase factors to satisfy the relations
T ^ L M + ( S ) = ( − 1 ) M T ^ L , − M ( S ) . (273) \hat{T}_{LM}^{+}(S)=(-1)^{M}\hat{T}_{L,-M}(S). \tag*{(273)} T ^ L M + ( S ) = ( − 1 ) M T ^ L , − M ( S ) . (273)
The conditions (2.4.1)–(2.4.3) completely determine the polarization operators. The full complement of polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) with − L ≤ M ≤ L -L \le M \le L − L ≤ M ≤ L , 0 ≤ L ≤ 2 S 0 \le L \le2S 0 ≤ L ≤ 2 S constitutes a complete set of ∑ L = 0 2 S ( 2 L + 1 ) = ( 2 S + 1 ) 2 \sum_{L=0}^{2S}(2L+1)=(2S+1)^2 ∑ L = 0 2 S ( 2 L + 1 ) = ( 2 S + 1 ) 2 linearly independent square ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrices.
Explicit Form
The polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) may be constructed from products of the spin matrices as
T ^ L M ( S ) = N L ( S ) ( S ^ ⋅ ∇ ) L { r L Y L M ( ϑ , ϕ ) } . (274) \hat{T}_{LM}(S)=N_L(S)(\hat{\mathbf{S}}\cdot\nabla)^L\{r^LY_{LM}(\vartheta,\phi)\}. \tag*{(274)} T ^ L M ( S ) = N L ( S ) ( S ^ ⋅ ∇ ) L { r L Y L M ( ϑ , ϕ )} . (274)
where S ^ \hat{\mathbf{S}} S ^ is the spin operator, and N L ( S ) N_L(S) N L ( S ) is the normalization factor given by
N L ( S ) = 2 L L ! [ 4 π ( 2 S − L ) ! ( 2 S + L + 1 ) ! ] 1 2 . (275) N_L(S)=\frac{2^L}{L!}\left[\frac{4\pi(2S-L)!}{(2S+L+1)!}\right]^{\frac{1}{2}}. \tag*{(275)} N L ( S ) = L ! 2 L [ ( 2 S + L + 1 )! 4 π ( 2 S − L )! ] 2 1 . (275)
Note that T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) are actually independent of r r r , although the vector r r r enters the right-hand side of Eq. (274) . The operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) may be expressed in terms of the basis spin functions χ S m \chi_{Sm} χ S m (Chap. 6) by
T ^ L M ( S ) = 2 L + 1 2 S + 1 ∑ m , m ′ C S m L M S m ′ χ S m ′ χ S m + . (276) \hat{T}_{LM}(S)=\sqrt{\frac{2L+1}{2S+1}}\sum_{m,m'}C^{Sm'}_{SmLM}\chi_{Sm'}\chi_{Sm}^{+}. \tag*{(276)} T ^ L M ( S ) = 2 S + 1 2 L + 1 m , m ′ ∑ C S m L M S m ′ χ S m ′ χ S m + . (276)
The inverse relation is
χ S m ′ χ S m + = ∑ L 2 L + 1 2 S + 1 C S m L M S m ′ T ^ L M ( S ) . (277) \chi_{Sm'}\chi_{Sm}^{+}=\sum_L\sqrt{\frac{2L+1}{2S+1}}C^{Sm'}_{SmLM}\hat{T}_{LM}(S). \tag*{(277)} χ S m ′ χ S m + = L ∑ 2 S + 1 2 L + 1 C S m L M S m ′ T ^ L M ( S ) . (277)
An explicit form of T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) depends on the representation used for the spin functions. In particular, matrix elements of T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) in the spherical basis representation are given by
[ T ^ L M ( S ) ] σ ′ σ = 2 L + 1 2 S + 1 C S σ L M S σ ′ , ( σ , σ ′ = − S , − S + 1 , … , S ) . (278) \left[\hat{T}_{LM}(S)\right]_{\sigma'\sigma}=\sqrt{\frac{2L+1}{2S+1}}C^{S\sigma'}_{S\sigma LM},\qquad(\sigma,\sigma'=-S,-S+1,\ldots,S). \tag*{(278)} [ T ^ L M ( S ) ] σ ′ σ = 2 S + 1 2 L + 1 C S σ L M S σ ′ , ( σ , σ ′ = − S , − S + 1 , … , S ) . (278)
For L = 0 L=0 L = 0 the operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) are proportional to the unit ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrix
T ^ 00 ( S ) = 1 2 S + 1 I ^ . (279) \hat{T}_{00}(S)=\frac{1}{\sqrt{2S+1}}\hat{I}. \tag*{(279)} T ^ 00 ( S ) = 2 S + 1 1 I ^ . (279)
When L = 1 L=1 L = 1 the operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) are proportional to the spherical components of the spin operator
T ^ 1 M ( S ) = 3 S ( S + 1 ) ( 2 S + 1 ) S ^ M , ( M = ± 1 , 0 ) . (280) \hat{T}_{1M}(S)=\frac{\sqrt{3}}{\sqrt{S(S+1)(2S+1)}}\hat{S}_M,\qquad(M=\pm1,0). \tag*{(280)} T ^ 1 M ( S ) = S ( S + 1 ) ( 2 S + 1 ) 3 S ^ M , ( M = ± 1 , 0 ) . (280)
The polarization operators for a spin value S = 1 S=1 S = 1 are considered in § 11 .
(a) Coordinate inversion (r → − r r\to-r r → − r )
The polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) are invariant under coordinate inversion
P ^ r T ^ L M ( S ) P ^ r − 1 = T ^ L M ( S ) . (281) \hat{P}_r\hat{T}_{LM}(S)\hat{P}_r^{-1}=\hat{T}_{LM}(S). \tag*{(281)} P ^ r T ^ L M ( S ) P ^ r − 1 = T ^ L M ( S ) . (281)
(b) Rotation of coordinate system
Under rotations specified by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ the polarization operators transform as
T ^ L M ′ ( S ) = D ^ ( α , β , γ ) T ^ L M ′ ( S ) [ D ^ ( α , β , γ ) ] − 1 = ∑ M D M M ′ L ( α , β , γ ) T ^ L M ( S ) (282) \hat{T}_{LM'}(S)=\hat{D}(\alpha,\beta,\gamma)\hat{T}_{LM'}(S)[\hat{D}(\alpha,\beta,\gamma)]^{-1}=\sum_M D^L_{MM'}(\alpha,\beta,\gamma)\hat{T}_{LM}(S) \tag*{(282)} T ^ L M ′ ( S ) = D ^ ( α , β , γ ) T ^ L M ′ ( S ) [ D ^ ( α , β , γ ) ] − 1 = M ∑ D M M ′ L ( α , β , γ ) T ^ L M ( S ) (282)
, where D ^ ( α , β , γ ) \hat{D}(\alpha,\beta,\gamma) D ^ ( α , β , γ ) is the rotation operator (see § 9.10.1 ), and D M M ′ L D^L_{MM'} D M M ′ L are the Wigner D D D -functions (Chap. 4).
Expansion Series of Polarization Operators
Expansion Series of Polarization Operators
As has been mentioned above (§ 10.4.1 ) the polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) form a complete set of linearly independent matrices. An arbitrary square ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrix A ^ \hat{A} A ^ (with S S S integer or half-integer) may be expanded in a series of the polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) , i.e., it may be written in the form
A ^ = ∑ L = 0 2 S ∑ M = − L L A L M T ^ L M ( S ) , (283) \hat{A}=\sum_{L=0}^{2S}\sum_{M=-L}^{L}A_{LM}\hat{T}_{LM}(S), \tag*{(283)} A ^ = L = 0 ∑ 2 S M = − L ∑ L A L M T ^ L M ( S ) , (283)
where the expansion coefficients A L M A_{LM} A L M are given by
A L M = Tr { T ^ L M + ( S ) A ^ } . (284) A_{LM}=\operatorname{Tr}\left\{\hat{T}^{+}_{LM}(S)\hat{A}\right\}. \tag*{(284)} A L M = Tr { T ^ L M + ( S ) A ^ } . (284)
If the matrix A ^ \hat{A} A ^ is Hermitian, i.e., A ^ + = A ^ \hat{A}^{+}=\hat{A} A ^ + = A ^ , then
A L M ∗ = ( − 1 ) M A L − M . (285) A^{*}_{LM}=(-1)^{M}A_{L-M}. \tag*{(285)} A L M ∗ = ( − 1 ) M A L − M . (285)
Some examples of such expansions are given below.
(a) Clebsch-Gordan series for polarization operators
Products of two polarization operators T ^ L 1 M 1 ( S ) \hat{T}_{L_{1}M_{1}}(S) T ^ L 1 M 1 ( S ) and T ^ L 2 M 2 ( S ) \hat{T}_{L_{2}M_{2}}(S) T ^ L 2 M 2 ( S ) may be written in the form of a Clebsch-Gordan series
T ^ L 1 M 1 ( S ) T ^ L 2 M 2 ( S ) = ∑ L ( − 1 ) 2 S + L ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) { L 1 L 2 L S S S } C L 1 M 1 L 2 M 2 L M T ^ L M ( S ) . (286) \hat{T}_{L_{1}M_{1}}(S)\hat{T}_{L_{2}M_{2}}(S)=\sum_{L}(-1)^{2S+L}\sqrt{(2L_{1}+1)(2L_{2}+1)}
\left\{\begin{matrix}L_{1}&L_{2}&L\\S&S&S\end{matrix}\right\}
C^{LM}_{L_{1}M_{1}\,L_{2}M_{2}}\hat{T}_{LM}(S). \tag*{(286)} T ^ L 1 M 1 ( S ) T ^ L 2 M 2 ( S ) = L ∑ ( − 1 ) 2 S + L ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) { L 1 S L 2 S L S } C L 1 M 1 L 2 M 2 L M T ^ L M ( S ) . (286)
(b) Expansion of rotation operator
If one is interested only in transformation properties of spin functions and spin operators under rotations of the coordinate system, one may replace the total angular momentum operator J ^ \hat{J} J ^ by the spin operator S ^ \hat{S} S ^ in the general expressions for the rotation operator ((1.4.31) and ((1.4.33). Such a modified operator will be labelled by a superscript S S S . If a rotation is defined by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ , the rotation operator D ^ S ( α , β , γ ) \hat{D}^{S}(\alpha,\beta,\gamma) D ^ S ( α , β , γ ) may be expressed as follows
D ^ S ( α , β , γ ) ≡ e − i α S ^ z e − i β S ^ y e − i γ S ^ z = ∑ L , M , m , m ′ 2 L + 1 2 S + 1 C S m ′ L M S m D m m ′ S ( α , β , γ ) T ^ L M ( S ) . (287) \hat{D}^{S}(\alpha,\beta,\gamma)\equiv e^{-i\alpha\hat{S}_{z}}e^{-i\beta\hat{S}_{y}}e^{-i\gamma\hat{S}_{z}}=\sum_{L,M,m,m'}\frac{2L+1}{2S+1}C^{Sm}_{Sm'\,LM}D^{S}_{mm'}(\alpha,\beta,\gamma)\hat{T}_{LM}(S). \tag*{(287)} D ^ S ( α , β , γ ) ≡ e − i α S ^ z e − i β S ^ y e − iγ S ^ z = L , M , m , m ′ ∑ 2 S + 1 2 L + 1 C S m ′ L M S m D m m ′ S ( α , β , γ ) T ^ L M ( S ) . (287)
If a rotation is defined by the rotation axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) and the rotation angle ω \omega ω , the expansion of the rotation operator U ^ S ( ω ; Θ , Φ ) \hat{U}^{S}(\omega;\Theta,\Phi) U ^ S ( ω ; Θ , Φ ) is given by
U ^ S ( ω ; Θ , Φ ) ≡ e − i ω n ⋅ s ^ = 2 π ∑ L = 0 2 S ( − i ) L 2 S + 1 χ L S ( ω ) ∑ M = − L L Y L M ∗ ( Θ , Φ ) T ^ L M ( S ) , (288) \hat{U}^{S}(\omega;\Theta,\Phi)\equiv e^{-i\omega\mathbf{n}\cdot\hat{\mathbf{s}}}=2\sqrt{\pi}\sum_{L=0}^{2S}\frac{(-i)^{L}}{\sqrt{2S+1}}\chi^{S}_{L}(\omega)\sum_{M=-L}^{L}Y^{*}_{LM}(\Theta,\Phi)\hat{T}_{LM}(S), \tag*{(288)} U ^ S ( ω ; Θ , Φ ) ≡ e − iω n ⋅ s ^ = 2 π L = 0 ∑ 2 S 2 S + 1 ( − i ) L χ L S ( ω ) M = − L ∑ L Y L M ∗ ( Θ , Φ ) T ^ L M ( S ) , (288)
where the functions χ L S ( ω ) \chi^{S}_{L}(\omega) χ L S ( ω ) are the generalized characters (see Sec. 4.15).
Commutators and Anticommutators
The polarization operators satisfy the following commutation relations
[ T ^ L 1 M 1 ( S ) , T ^ L 2 M 2 ( S ) ] = ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) ∑ L 3 ( − 1 ) 2 S + L 3 [ 1 − ( − 1 ) L 1 + L 2 + L 3 ] { L 1 L 2 L 3 S S S } C L 1 M 1 L 2 M 2 L 3 M 3 T ^ L 3 M 3 ( S ) . (289) \left[\hat{T}_{L_{1}M_{1}}(S),\hat{T}_{L_{2}M_{2}}(S)\right]=\sqrt{(2L_{1}+1)(2L_{2}+1)}\sum_{L_{3}}(-1)^{2S+L_{3}}\left[1-(-1)^{L_{1}+L_{2}+L_{3}}\right]\left\{\begin{matrix}L_{1}&L_{2}&L_{3}\\S&S&S\end{matrix}\right\}C^{L_{3}M_{3}}_{L_{1}M_{1}\,L_{2}M_{2}}\hat{T}_{L_{3}M_{3}}(S). \tag*{(289)} [ T ^ L 1 M 1 ( S ) , T ^ L 2 M 2 ( S ) ] = ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) L 3 ∑ ( − 1 ) 2 S + L 3 [ 1 − ( − 1 ) L 1 + L 2 + L 3 ] { L 1 S L 2 S L 3 S } C L 1 M 1 L 2 M 2 L 3 M 3 T ^ L 3 M 3 ( S ) . (289)
{ T ^ L 1 M 1 ( S ) , T ^ L 2 M 2 ( S ) } = ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) ∑ L 3 ( − 1 ) 2 S + L 3 [ 1 + ( − 1 ) L 1 + L 2 + L 3 ] { L 1 L 2 L 3 S S S } C L 1 M 1 L 2 M 2 L 3 M 3 T ^ L 3 M 3 ( S ) . (290) \left\{\hat{T}_{L_{1}M_{1}}(S),\hat{T}_{L_{2}M_{2}}(S)\right\}=\sqrt{(2L_{1}+1)(2L_{2}+1)}\sum_{L_{3}}(-1)^{2S+L_{3}}\left[1+(-1)^{L_{1}+L_{2}+L_{3}}\right]\left\{\begin{matrix}L_{1}&L_{2}&L_{3}\\S&S&S\end{matrix}\right\}C^{L_{3}M_{3}}_{L_{1}M_{1}\,L_{2}M_{2}}\hat{T}_{L_{3}M_{3}}(S). \tag*{(290)} { T ^ L 1 M 1 ( S ) , T ^ L 2 M 2 ( S ) } = ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) L 3 ∑ ( − 1 ) 2 S + L 3 [ 1 + ( − 1 ) L 1 + L 2 + L 3 ] { L 1 S L 2 S L 3 S } C L 1 M 1 L 2 M 2 L 3 M 3 T ^ L 3 M 3 ( S ) . (290)
Equation (2.4.1) is a special case of (2.4.19) for L 1 = 1 L_{1}=1 L 1 = 1 .
Traces of Products of Polarization Operators
Tr { T ^ L M ( S ) } = 2 S + 1 δ L 0 δ M 0 , (291) \operatorname{Tr}\left\{\widehat{T}_{LM}(S)\right\}=\sqrt{2S+1}\delta_{L0}\delta_{M0}, \tag*{(291)} Tr { T L M ( S ) } = 2 S + 1 δ L 0 δ M 0 , (291)
Tr { T ^ L 1 M 1 ( S ) T ^ L 2 M 2 ( S ) } = ( − 1 ) M 1 δ L 1 L 2 δ M 1 − M 2 , (292) \operatorname{Tr}\left\{\widehat{T}_{L_{1}M_{1}}(S)\widehat{T}_{L_{2}M_{2}}(S)\right\}=(-1)^{M_{1}}\delta_{L_{1}L_{2}}\delta_{M_{1}-M_{2}}, \tag*{(292)} Tr { T L 1 M 1 ( S ) T L 2 M 2 ( S ) } = ( − 1 ) M 1 δ L 1 L 2 δ M 1 − M 2 , (292)
Tr { T ^ L 1 M 1 ( S ) T ^ L 2 M 2 ( S ) T ^ L 3 M 3 ( S ) } = ( − 1 ) 2 S + L 3 + M 3 ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) C L 1 M 1 L 2 M 2 L 3 , − M 3 { L 1 L 2 L 3 S S S } . (293) \operatorname{Tr}\left\{\widehat{T}_{L_{1}M_{1}}(S)\widehat{T}_{L_{2}M_{2}}(S)\widehat{T}_{L_{3}M_{3}}(S)\right\}=(-1)^{2S+L_{3}+M_{3}}\sqrt{(2L_{1}+1)(2L_{2}+1)}C^{L_{3},-M_{3}}_{L_{1}M_{1}L_{2}M_{2}}\left\{\begin{matrix}L_{1}&L_{2}&L_{3}\\S&S&S\end{matrix}\right\}. \tag*{(293)} Tr { T L 1 M 1 ( S ) T L 2 M 2 ( S ) T L 3 M 3 ( S ) } = ( − 1 ) 2 S + L 3 + M 3 ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) C L 1 M 1 L 2 M 2 L 3 , − M 3 { L 1 S L 2 S L 3 S } . (293)
In general
Tr { T ^ L 1 M 1 ( S ) T ^ L 2 M 2 ( S ) … T ^ L n M n ( S ) } = [ ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) … ( 2 L n + 1 ) ] 1 / 2 ( 2 S + 1 ) n / 2 , \operatorname{Tr}\left\{\widehat{T}_{L_{1}M_{1}}(S)\widehat{T}_{L_{2}M_{2}}(S)\ldots\widehat{T}_{L_{n}M_{n}}(S)\right\}=\frac{\left[(2L_{1}+1)(2L_{2}+1)\ldots(2L_{n}+1)\right]^{1/2}}{(2S+1)^{n/2}}, Tr { T L 1 M 1 ( S ) T L 2 M 2 ( S ) … T L n M n ( S ) } = ( 2 S + 1 ) n /2 [ ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) … ( 2 L n + 1 ) ] 1/2 ,
× δ M 1 + M 2 + M 3 + … + M n , 0 ∑ m C L 1 M 1 S m S , m + μ 1 C L 2 M 2 S , m + μ 1 S , m + μ 2 … C L n M n S , m + μ n − 1 S , m , (294) {}\times\delta_{M_{1}+M_{2}+M_{3}+\ldots+M_{n},0}\sum_{m}C^{S,m+\mu_{1}}_{L_{1}M_{1}Sm}C^{S,m+\mu_{2}}_{L_{2}M_{2}S,m+\mu_{1}}\ldots C^{S,m}_{L_{n}M_{n}S,m+\mu_{n-1}}, \tag*{(294)} × δ M 1 + M 2 + M 3 + … + M n , 0 m ∑ C L 1 M 1 S m S , m + μ 1 C L 2 M 2 S , m + μ 1 S , m + μ 2 … C L n M n S , m + μ n − 1 S , m , (294)
where
μ k = ∑ i = 1 k M i . \mu_{k}=\sum_{i=1}^{k}M_{i}. μ k = i = 1 ∑ k M i .
Spin matrices for S = 1 / 2 S=1/2 S = 1/2
Explicit Form
For the particular case S = 1 2 S=\frac{1}{2} S = 2 1 , the spin operator S ^ \widehat{S} S is represented by a set of three square 2 × 2 2\times2 2 × 2 matrices. Cartesian components of S ^ \widehat{S} S in the spherical basis representation are given by
S ^ x = 1 2 ( 0 1 1 0 ) , S ^ y = 1 2 ( 0 − i i 0 ) , S ^ z = 1 2 ( 1 0 0 − 1 ) . (295) \widehat{S}_{x}=\frac{1}{2}\begin{pmatrix}0&1\\1&0\end{pmatrix},\qquad\widehat{S}_{y}=\frac{1}{2}\begin{pmatrix}0&-i\\i&0\end{pmatrix},\qquad\widehat{S}_{z}=\frac{1}{2}\begin{pmatrix}1&0\\0&-1\end{pmatrix}. \tag*{(295)} S x = 2 1 ( 0 1 1 0 ) , S y = 2 1 ( 0 i − i 0 ) , S z = 2 1 ( 1 0 0 − 1 ) . (295)
whereas spherical components are written as
S ^ + 1 = − 1 2 ( 0 1 0 0 ) , S ^ 0 = 1 2 ( 1 0 0 − 1 ) , S ^ − 1 = 1 2 ( 0 0 1 0 ) . (296) \widehat{S}_{+1}=-\frac{1}{\sqrt{2}}\begin{pmatrix}0&1\\0&0\end{pmatrix},\qquad\widehat{S}_{0}=\frac{1}{2}\begin{pmatrix}1&0\\0&-1\end{pmatrix},\qquad\widehat{S}_{-1}=\frac{1}{\sqrt{2}}\begin{pmatrix}0&0\\1&0\end{pmatrix}. \tag*{(296)} S + 1 = − 2 1 ( 0 0 1 0 ) , S 0 = 2 1 ( 1 0 0 − 1 ) , S − 1 = 2 1 ( 0 1 0 0 ) . (296)
In addition to the spin matrices, the unit 2 × 2 2\times2 2 × 2 matrix I ^ \widehat{I} I is usually introduced:
I ^ = ( 1 0 0 1 ) . (297) \widehat{I}=\begin{pmatrix}1&0\\0&1\end{pmatrix}. \tag*{(297)} I = ( 1 0 0 1 ) . (297)
Note that the Pauli matrices σ ^ \widehat{\sigma} σ are widely used instead of the spin operator S ^ \widehat{S} S . The Pauli matrices are proportional to the spin operator
S ^ = 1 2 σ ^ . (298) \widehat{S}=\frac{1}{2}\widehat{\sigma}. \tag*{(298)} S = 2 1 σ . (298)
Equations (2.5.1) and (2.5.2) yield the following properties of the spin matrices
S ^ i + = S ^ i , ( i = x , y , z ) , (299) \widehat{S}_{i}^{+}=\widehat{S}_{i},\qquad(i=x,y,z), \tag*{(299)} S i + = S i , ( i = x , y , z ) , (299)
S ^ μ + = ( − 1 ) μ S ^ − μ , ( μ = ± 1 , 0 ) , (300) \widehat{S}_{\mu}^{+}=(-1)^{\mu}\widehat{S}_{-\mu},\qquad(\mu=\pm1,0), \tag*{(300)} S μ + = ( − 1 ) μ S − μ , ( μ = ± 1 , 0 ) , (300)
S ^ x ∗ = S ^ x , S ^ y ∗ = − S ^ y , S ^ z ∗ = S ^ z , (301) \widehat{S}_{x}^{*}=\widehat{S}_{x},\qquad\widehat{S}_{y}^{*}=-\widehat{S}_{y},\qquad\widehat{S}_{z}^{*}=\widehat{S}_{z}, \tag*{(301)} S x ∗ = S x , S y ∗ = − S y , S z ∗ = S z , (301)
S ^ μ ∗ = S ^ μ , ( μ = ± 1 , 0 ) . (302) \widehat{S}_{\mu}^{*}=\widehat{S}_{\mu},\qquad(\mu=\pm1,0). \tag*{(302)} S μ ∗ = S μ , ( μ = ± 1 , 0 ) . (302)
Commutators and Anticommutators
The cartesian components of the spin operator satisfy the following relations
[ S ^ i , S ^ k ] = i ε i k l S ^ l , { S ^ i , S ^ k } = 1 2 δ i k I ^ , ( i , k , l = x , y , z ) . (303) [\widehat{S}_{i},\widehat{S}_{k}]=i\varepsilon_{ikl}\widehat{S}_{l},\qquad\{\widehat{S}_{i},\widehat{S}_{k}\}=\frac{1}{2}\delta_{ik}\widehat{I},\qquad(i,k,l=x,y,z). \tag*{(303)} [ S i , S k ] = i ε ik l S l , { S i , S k } = 2 1 δ ik I , ( i , k , l = x , y , z ) . (303)
For spherical components these relations take the form
[ S ^ μ , S ^ ν ] = − 2 C 1 μ 1 ν 1 λ S ^ λ , { S ^ μ , S ^ ν } = 1 2 ( − 1 ) μ δ μ − ν I ^ , ( μ , ν , λ = ± 1 , 0 ) . (304) [\widehat{S}_{\mu},\widehat{S}_{\nu}]=-\sqrt{2}C^{1\lambda}_{1\mu1\nu}\widehat{S}_{\lambda},\qquad\{\widehat{S}_{\mu},\widehat{S}_{\nu}\}=\frac{1}{2}(-1)^{\mu}\delta_{\mu-\nu}\widehat{I},\qquad(\mu,\nu,\lambda=\pm1,0). \tag*{(304)} [ S μ , S ν ] = − 2 C 1 μ 1 ν 1 λ S λ , { S μ , S ν } = 2 1 ( − 1 ) μ δ μ − ν I , ( μ , ν , λ = ± 1 , 0 ) . (304)
Products of Spin Matrices
Functions of Spin Matrices
The four matrices S ^ x \hat{S}_x S ^ x , S ^ y \hat{S}_y S ^ y , S ^ z \hat{S}_z S ^ z and I ^ \hat{I} I ^ (or S ^ + 1 \hat{S}_{+1} S ^ + 1 , S ^ 0 \hat{S}_0 S ^ 0 , S ^ − 1 \hat{S}_{-1} S ^ − 1 and I ^ \hat{I} I ^ ) constitute a complete set of square 2 × 2 2 \times2 2 × 2 matrices. Any function of the spin operator (S = 1 2 S = \frac{1}{2} S = 2 1 ) may be expanded in a series in terms of these matrices. In particular, products of the spin matrices may also be written in such a form. Products of cartesian components (where the indices i i i , k k k , l l l , etc. take the values x x x , y y y , z z z ) are given by
S ^ i S ^ k = 1 4 δ i k I ^ + i 2 ϵ i k l S ^ l , (305) \hat{S}_i\hat{S}_k = \frac{1}{4}\delta_{ik}\hat{I} + \frac{i}{2}\epsilon_{ikl}\hat{S}_l, \tag*{(305)} S ^ i S ^ k = 4 1 δ ik I ^ + 2 i ϵ ik l S ^ l , (305)
S ^ i S ^ k S ^ l = i 8 ϵ i k l I ^ + 1 4 ( S ^ i δ k l − S ^ k δ i l + S ^ l δ i k ) , (306) \hat{S}_i\hat{S}_k\hat{S}_l = \frac{i}{8}\epsilon_{ikl}\hat{I} + \frac{1}{4}\left(\hat{S}_i\delta_{kl} - \hat{S}_k\delta_{il} + \hat{S}_l\delta_{ik}\right), \tag*{(306)} S ^ i S ^ k S ^ l = 8 i ϵ ik l I ^ + 4 1 ( S ^ i δ k l − S ^ k δ i l + S ^ l δ ik ) , (306)
S ^ i S ^ k S ^ l S ^ j = 1 16 ( δ i k δ l j − δ i l δ k j + δ i j δ k l ) I ^ + i 8 ( δ i k ϵ l j m − δ i l ϵ k j m + δ i j ϵ k l m + δ k l ϵ i j m − δ k j ϵ i l m + δ l j ϵ i k m ) S ^ m . (307) \begin{aligned}
\hat{S}_i\hat{S}_k\hat{S}_l\hat{S}_j
&= \frac{1}{16}\left(\delta_{ik}\delta_{lj} - \delta_{il}\delta_{kj} + \delta_{ij}\delta_{kl}\right)\hat{I} \\
&\quad+ \frac{i}{8}\left(\delta_{ik}\epsilon_{ljm} - \delta_{il}\epsilon_{kjm} + \delta_{ij}\epsilon_{klm} + \delta_{kl}\epsilon_{ijm} - \delta_{kj}\epsilon_{ilm} + \delta_{lj}\epsilon_{ikm}\right)\hat{S}_m. \tag*{(307)}
\end{aligned} S ^ i S ^ k S ^ l S ^ j = 16 1 ( δ ik δ l j − δ i l δ k j + δ ij δ k l ) I ^ + 8 i ( δ ik ϵ l j m − δ i l ϵ k j m + δ ij ϵ k l m + δ k l ϵ ij m − δ k j ϵ i l m + δ l j ϵ ik m ) S ^ m . (307)
In more detailed form, Eq. Eq. (305) is written as
S ^ x 2 = S ^ y 2 = S ^ z 2 = 1 4 I ^ , \hat{S}_x^2 = \hat{S}_y^2 = \hat{S}_z^2 = \frac{1}{4}\hat{I}, S ^ x 2 = S ^ y 2 = S ^ z 2 = 4 1 I ^ ,
S ^ x S ^ y = − S ^ y S ^ x = i 2 S ^ z , S ^ y S ^ z = − S ^ z S ^ y = i 2 S ^ x , \hat{S}_x\hat{S}_y = -\hat{S}_y\hat{S}_x = \frac{i}{2}\hat{S}_z,\qquad\hat{S}_y\hat{S}_z = -\hat{S}_z\hat{S}_y = \frac{i}{2}\hat{S}_x, S ^ x S ^ y = − S ^ y S ^ x = 2 i S ^ z , S ^ y S ^ z = − S ^ z S ^ y = 2 i S ^ x ,
S ^ z S ^ x = − S ^ x S ^ z = i 2 S ^ y . (308) \hat{S}_z\hat{S}_x = -\hat{S}_x\hat{S}_z = \frac{i}{2}\hat{S}_y. \tag*{(308)} S ^ z S ^ x = − S ^ x S ^ z = 2 i S ^ y . (308)
Products of spherical components are as follows:
S ^ + 1 S ^ + 1 = 0 , S ^ 0 S ^ 0 = 1 4 I ^ , S ^ − 1 S ^ − 1 = 0 , \hat{S}_{+1}\hat{S}_{+1}=0,\qquad\hat{S}_0\hat{S}_0=\frac{1}{4}\hat{I},\qquad\hat{S}_{-1}\hat{S}_{-1}=0, S ^ + 1 S ^ + 1 = 0 , S ^ 0 S ^ 0 = 4 1 I ^ , S ^ − 1 S ^ − 1 = 0 ,
S ^ + 1 S ^ − 1 = − 1 2 ( 1 0 0 0 ) = − 1 4 I ^ − 1 2 S ^ 0 , \hat{S}_{+1}\hat{S}_{-1}=-\frac{1}{2}
\begin{pmatrix}
1 & 0\\
0 & 0
\end{pmatrix}
=-\frac{1}{4}\hat{I}-\frac{1}{2}\hat{S}_0, S ^ + 1 S ^ − 1 = − 2 1 ( 1 0 0 0 ) = − 4 1 I ^ − 2 1 S ^ 0 ,
S ^ − 1 S ^ + 1 = − 1 2 ( 0 0 0 1 ) = − 1 4 I ^ + 1 2 S ^ 0 , (309) \hat{S}_{-1}\hat{S}_{+1}=-\frac{1}{2}
\begin{pmatrix}
0 & 0\\
0 & 1
\end{pmatrix}
=-\frac{1}{4}\hat{I}+\frac{1}{2}\hat{S}_0, \tag*{(309)} S ^ − 1 S ^ + 1 = − 2 1 ( 0 0 0 1 ) = − 4 1 I ^ + 2 1 S ^ 0 , (309)
S ^ + 1 S ^ 0 = − 1 2 S ^ + 1 , S ^ − 1 S ^ 0 = 1 2 S ^ − 1 , \hat{S}_{+1}\hat{S}_0=-\frac{1}{2}\hat{S}_{+1},\qquad\hat{S}_{-1}\hat{S}_0=\frac{1}{2}\hat{S}_{-1}, S ^ + 1 S ^ 0 = − 2 1 S ^ + 1 , S ^ − 1 S ^ 0 = 2 1 S ^ − 1 ,
S ^ 0 S ^ + 1 = 1 2 S ^ + 1 , S ^ 0 S ^ − 1 = − 1 2 S ^ − 1 , \hat{S}_0\hat{S}_{+1}=\frac{1}{2}\hat{S}_{+1},\qquad\hat{S}_0\hat{S}_{-1}=-\frac{1}{2}\hat{S}_{-1}, S ^ 0 S ^ + 1 = 2 1 S ^ + 1 , S ^ 0 S ^ − 1 = − 2 1 S ^ − 1 ,
S ^ μ S ^ ν = 1 4 ( − 1 ) μ δ μ , − ν I ^ − 1 2 C 1 μ 1 ν 1 λ S ^ λ , ( μ , ν , λ = ± 1 , 0 ) , \hat{S}_{\mu}\hat{S}_{\nu}=\frac{1}{4}(-1)^{\mu}\delta_{\mu,-\nu}\hat{I}-\frac{1}{\sqrt{2}}C^{1\lambda}_{1\mu\,1\nu}\hat{S}_{\lambda},\qquad(\mu,\nu,\lambda=\pm1,0), S ^ μ S ^ ν = 4 1 ( − 1 ) μ δ μ , − ν I ^ − 2 1 C 1 μ 1 ν 1 λ S ^ λ , ( μ , ν , λ = ± 1 , 0 ) ,
( S ^ + 1 ) n = 0 , ( S ^ − 1 ) n = 0 , ( n = 2 , 3 , … ) , (310) \left(\hat{S}_{+1}\right)^n=0,\qquad\left(\hat{S}_{-1}\right)^n=0,\qquad(n=2,3,\ldots), \tag*{(310)} ( S ^ + 1 ) n = 0 , ( S ^ − 1 ) n = 0 , ( n = 2 , 3 , … ) , (310)
( S ^ 0 ) 2 n = ( 1 4 ) n I ^ , ( S ^ 0 ) 2 n + 1 = ( 1 4 ) n S ^ 0 , ( n = 0 , 1 , 2 , … ) , (311) \left(\hat{S}_0\right)^{2n}=\left(\frac{1}{4}\right)^n\hat{I},\qquad\left(\hat{S}_0\right)^{2n+1}=\left(\frac{1}{4}\right)^n\hat{S}_0,\qquad(n=0,1,2,\ldots), \tag*{(311)} ( S ^ 0 ) 2 n = ( 4 1 ) n I ^ , ( S ^ 0 ) 2 n + 1 = ( 4 1 ) n S ^ 0 , ( n = 0 , 1 , 2 , … ) , (311)
S ^ ⋅ S ^ = S ^ 2 = S ^ x 2 + S ^ y 2 + S ^ z 2 = − S ^ + 1 S ^ − 1 + S ^ 0 S ^ 0 − S ^ − 1 S ^ + 1 = 3 4 I ^ . (312) \hat{\mathbf{S}}\cdot\hat{\mathbf{S}}=\hat{S}^2=\hat{S}_x^2+\hat{S}_y^2+\hat{S}_z^2=-\hat{S}_{+1}\hat{S}_{-1}+\hat{S}_0\hat{S}_0-\hat{S}_{-1}\hat{S}_{+1}=\frac{3}{4}\hat{I}. \tag*{(312)} S ^ ⋅ S ^ = S ^ 2 = S ^ x 2 + S ^ y 2 + S ^ z 2 = − S ^ + 1 S ^ − 1 + S ^ 0 S ^ 0 − S ^ − 1 S ^ + 1 = 4 3 I ^ . (312)
Products of the operators S ^ \hat{\mathbf{S}} S ^ and any vectors a a a and b b b commuting with S ^ \hat{\mathbf{S}} S ^ may be written in the form
S ^ ⋅ ( S ^ ⋅ a ) = 1 4 I ^ a − i 2 [ S ^ × a ] , (313) \hat{\mathbf{S}}\cdot(\hat{\mathbf{S}}\cdot a)=\frac{1}{4}\hat{I}a-\frac{i}{2}[\hat{\mathbf{S}}\times a], \tag*{(313)} S ^ ⋅ ( S ^ ⋅ a ) = 4 1 I ^ a − 2 i [ S ^ × a ] , (313)
( S ^ ⋅ [ S ^ × a ] ) = ( [ S ^ × S ^ ] ⋅ a ) = i ( S ^ ⋅ a ) , (314) (\hat{\mathbf{S}}\cdot[\hat{\mathbf{S}}\times a])=([\hat{\mathbf{S}}\times\hat{\mathbf{S}}]\cdot a)=i(\hat{\mathbf{S}}\cdot a), \tag*{(314)} ( S ^ ⋅ [ S ^ × a ]) = ([ S ^ × S ^ ] ⋅ a ) = i ( S ^ ⋅ a ) , (314)
[ S ^ × [ S ^ × a ] ] = − 1 2 I ^ a + i 2 [ S ^ × a ] , (315) [\hat{\mathbf{S}}\times[\hat{\mathbf{S}}\times a]]=-\frac{1}{2}\hat{I}a+\frac{i}{2}[\hat{\mathbf{S}}\times a], \tag*{(315)} [ S ^ × [ S ^ × a ]] = − 2 1 I ^ a + 2 i [ S ^ × a ] , (315)
( S ^ ⋅ a ) ( S ^ ⋅ b ) = 1 4 ( a ⋅ b ) I ^ + i 2 ( S ^ ⋅ [ a × b ] ) , (316) (\hat{\mathbf{S}}\cdot a)(\hat{\mathbf{S}}\cdot b)=\frac{1}{4}(a\cdot b)\hat{I}+\frac{i}{2}(\hat{\mathbf{S}}\cdot[a\times b]), \tag*{(316)} ( S ^ ⋅ a ) ( S ^ ⋅ b ) = 4 1 ( a ⋅ b ) I ^ + 2 i ( S ^ ⋅ [ a × b ]) , (316)
( [ S ^ × a ] ⋅ [ S ^ × b ] ) = 1 2 ( a ⋅ b ) I ^ + i 2 ( S ^ ⋅ [ a × b ] ) , (317) ([\hat{\mathbf{S}}\times a]\cdot[\hat{\mathbf{S}}\times b])=\frac{1}{2}(a\cdot b)\hat{I}+\frac{i}{2}(\hat{\mathbf{S}}\cdot[a\times b]), \tag*{(317)} ([ S ^ × a ] ⋅ [ S ^ × b ]) = 2 1 ( a ⋅ b ) I ^ + 2 i ( S ^ ⋅ [ a × b ]) , (317)
( S ^ ⋅ a ) [ S ^ × b ] = 1 4 [ a × b ] I ^ + i 2 { ( S ^ ⋅ b ) a − ( a ⋅ b ) S ^ } , (318) (\hat{\mathbf{S}}\cdot a)[\hat{\mathbf{S}}\times b]=\frac{1}{4}[a\times b]\hat{I}+\frac{i}{2}\{(\hat{\mathbf{S}}\cdot b)a-(a\cdot b)\hat{\mathbf{S}}\}, \tag*{(318)} ( S ^ ⋅ a ) [ S ^ × b ] = 4 1 [ a × b ] I ^ + 2 i {( S ^ ⋅ b ) a − ( a ⋅ b ) S ^ } , (318)
[ [ S ^ × a ] × [ S ^ × b ] ] = 1 4 [ a × b ] I ^ + i 2 { ( S ^ ⋅ b ) a + ( S ^ ⋅ a ) b } . (319) [[\hat{\mathbf{S}}\times a]\times[\hat{\mathbf{S}}\times b]]=\frac{1}{4}[a\times b]\hat{I}+\frac{i}{2}\{(\hat{\mathbf{S}}\cdot b)a+(\hat{\mathbf{S}}\cdot a)b\}. \tag*{(319)} [[ S ^ × a ] × [ S ^ × b ]] = 4 1 [ a × b ] I ^ + 2 i {( S ^ ⋅ b ) a + ( S ^ ⋅ a ) b } . (319)
If n n n is a unit vector (n 2 = 1 n^2=1 n 2 = 1 ), then
( S ^ ⋅ n ) 2 k = ( 1 4 ) k I ^ , ( S ^ ⋅ n ) 2 k + 1 = ( 1 4 ) k ( S ^ ⋅ n ) , k = 0 , 1 , 2 , … . (320) (\hat{\mathbf{S}}\cdot n)^{2k}=\left(\frac{1}{4}\right)^k\hat{I},\qquad(\hat{\mathbf{S}}\cdot n)^{2k+1}=\left(\frac{1}{4}\right)^k(\hat{\mathbf{S}}\cdot n),\qquad k=0,1,2,\ldots. \tag*{(320)} ( S ^ ⋅ n ) 2 k = ( 4 1 ) k I ^ , ( S ^ ⋅ n ) 2 k + 1 = ( 4 1 ) k ( S ^ ⋅ n ) , k = 0 , 1 , 2 , … . (320)
Functions of Spin Matrices
e α S ^ i = I ^ cosh α 2 + 2 S ^ i sinh α 2 , (321) e^{\alpha\hat{S}_i}=\hat{I}\cosh\frac{\alpha}{2}+2\hat{S}_i\sinh\frac{\alpha}{2}, \tag*{(321)} e α S ^ i = I ^ cosh 2 α + 2 S ^ i sinh 2 α , (321)
cosh ( α S ^ i ) = I ^ cosh α 2 , sinh ( α S ^ i ) = 2 S ^ i sinh α 2 , (322) \cosh\left(\alpha\hat{S}_i\right)=\hat{I}\cosh\frac{\alpha}{2},\qquad\sinh\left(\alpha\hat{S}_i\right)=2\hat{S}_i\sinh\frac{\alpha}{2}, \tag*{(322)} cosh ( α S ^ i ) = I ^ cosh 2 α , sinh ( α S ^ i ) = 2 S ^ i sinh 2 α , (322)
where i = x , y , z i=x,y,z i = x , y , z .
e α S ^ + 1 = I ^ + α S ^ + 1 , e α S ^ 0 = I ^ cosh α 2 + 2 S ^ 0 sinh α 2 , e α S ^ − 1 = I ^ + α S ^ − 1 , (323) e^{\alpha\hat{S}_{+1}}=\hat{I}+\alpha\hat{S}_{+1},\qquad e^{\alpha\hat{S}_{0}}=\hat{I}\cosh\frac{\alpha}{2}+2\hat{S}_{0}\sinh\frac{\alpha}{2},\qquad e^{\alpha\hat{S}_{-1}}=\hat{I}+\alpha\hat{S}_{-1}, \tag*{(323)} e α S ^ + 1 = I ^ + α S ^ + 1 , e α S ^ 0 = I ^ cosh 2 α + 2 S ^ 0 sinh 2 α , e α S ^ − 1 = I ^ + α S ^ − 1 , (323)
cosh ( α S ^ + 1 ) = I ^ , cosh ( α S ^ 0 ) = I ^ cosh α 2 , cosh ( α S ^ − 1 ) = I ^ , (324) \cosh\left(\alpha\hat{S}_{+1}\right)=\hat{I},\qquad\cosh\left(\alpha\hat{S}_{0}\right)=\hat{I}\cosh\frac{\alpha}{2},\qquad\cosh\left(\alpha\hat{S}_{-1}\right)=\hat{I}, \tag*{(324)} cosh ( α S ^ + 1 ) = I ^ , cosh ( α S ^ 0 ) = I ^ cosh 2 α , cosh ( α S ^ − 1 ) = I ^ , (324)
sinh ( α S ^ + 1 ) = α S ^ + 1 , sinh ( α S ^ 0 ) = 2 S ^ 0 sinh α 2 , sinh ( α S ^ − 1 ) = α S ^ − 1 . (325) \sinh\left(\alpha\hat{S}_{+1}\right)=\alpha\hat{S}_{+1},\qquad\sinh\left(\alpha\hat{S}_{0}\right)=2\hat{S}_{0}\sinh\frac{\alpha}{2},\qquad\sinh\left(\alpha\hat{S}_{-1}\right)=\alpha\hat{S}_{-1}. \tag*{(325)} sinh ( α S ^ + 1 ) = α S ^ + 1 , sinh ( α S ^ 0 ) = 2 S ^ 0 sinh 2 α , sinh ( α S ^ − 1 ) = α S ^ − 1 . (325)
Rotation Operators
Euler angles
Under rotations described by the Euler angles α \alpha α , β \beta β , γ \gamma γ spin functions and spin matrices for S = 1 2 S=\frac{1}{2} S = 2 1 are transformed by a rotation operator D ^ 1 2 ( α , β , γ ) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma) D ^ 2 1 ( α , β , γ ) which is a special case of the operator given by Eq. (160)
D ^ 1 2 ( α , β , γ ) = e − i α S ^ z ⋅ e − i β S ^ y ⋅ e − i γ S ^ z = ( cos β 2 e − i α + γ 2 − sin β 2 e − i α − γ 2 sin β 2 e i α − γ 2 cos β 2 e i α + γ 2 ) . (326) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma)=e^{-i\alpha\hat{S}_z}\cdot e^{-i\beta\hat{S}_y}\cdot e^{-i\gamma\hat{S}_z}=
\begin{pmatrix}
\cos\frac{\beta}{2}e^{-i\frac{\alpha+\gamma}{2}} & -\sin\frac{\beta}{2}e^{-i\frac{\alpha-\gamma}{2}}\\
\sin\frac{\beta}{2}e^{i\frac{\alpha-\gamma}{2}} & \cos\frac{\beta}{2}e^{i\frac{\alpha+\gamma}{2}}
\end{pmatrix}. \tag*{(326)} D ^ 2 1 ( α , β , γ ) = e − i α S ^ z ⋅ e − i β S ^ y ⋅ e − iγ S ^ z = ( cos 2 β e − i 2 α + γ sin 2 β e i 2 α − γ − sin 2 β e − i 2 α − γ cos 2 β e i 2 α + γ ) . (326)
The inverse matrix is
[ D ^ 1 2 ( α , β , γ ) ] − 1 = [ D ^ 1 2 ( α , β , γ ) ] + = D ^ 1 2 ( π − γ , β , − π − α ) = D ^ 1 2 ( − γ , − β , − α ) , = ( cos β 2 e i α + γ 2 sin β 2 e − i α − γ 2 − sin β 2 e i α − γ 2 cos β 2 e − i α + γ 2 ) . (327) \begin{aligned}
\left[\hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma)\right]^{-1}=\left[\hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma)\right]^{+}=\hat{D}^{\frac{1}{2}}(\pi-\gamma,\beta,-\pi-\alpha)=\hat{D}^{\frac{1}{2}}(-\gamma,-\beta,-\alpha), \\
&=\begin{pmatrix}
\cos\frac{\beta}{2}e^{i\frac{\alpha+\gamma}{2}} & \sin\frac{\beta}{2}e^{-i\frac{\alpha-\gamma}{2}}\\
-\sin\frac{\beta}{2}e^{i\frac{\alpha-\gamma}{2}} & \cos\frac{\beta}{2}e^{-i\frac{\alpha+\gamma}{2}}
\end{pmatrix}. \tag*{(327)}
\end{aligned} [ D ^ 2 1 ( α , β , γ ) ] − 1 = [ D ^ 2 1 ( α , β , γ ) ] + = D ^ 2 1 ( π − γ , β , − π − α ) = D ^ 2 1 ( − γ , − β , − α ) , = ( cos 2 β e i 2 α + γ − sin 2 β e i 2 α − γ sin 2 β e − i 2 α − γ cos 2 β e − i 2 α + γ ) . (327)
Matrix elements of the operator D ^ 1 2 ( α , β , γ ) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma) D ^ 2 1 ( α , β , γ ) are the Wigner D D D -functions D M M ′ 1 2 ( α , β , γ ) D^{\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma) D M M ′ 2 1 ( α , β , γ ) (Chap. 4). The expansion of D ^ 1 2 ( α , β , γ ) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma) D ^ 2 1 ( α , β , γ ) in terms of the spin matrices has the form
D ^ 1 2 ( α , β , γ ) = cos β 2 cos α + γ 2 I ^ + 2 i sin β 2 sin α − γ 2 S ^ x − 2 i sin β 2 cos α − γ 2 S ^ y − 2 i cos β 2 sin α + γ 2 S ^ z , (328) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma)=\cos\frac{\beta}{2}\cos\frac{\alpha+\gamma}{2}\hat{I}+2i\sin\frac{\beta}{2}\sin\frac{\alpha-\gamma}{2}\hat{S}_x-2i\sin\frac{\beta}{2}\cos\frac{\alpha-\gamma}{2}\hat{S}_y-2i\cos\frac{\beta}{2}\sin\frac{\alpha+\gamma}{2}\hat{S}_z, \tag*{(328)} D ^ 2 1 ( α , β , γ ) = cos 2 β cos 2 α + γ I ^ + 2 i sin 2 β sin 2 α − γ S ^ x − 2 i sin 2 β cos 2 α − γ S ^ y − 2 i cos 2 β sin 2 α + γ S ^ z , (328)
D ^ 1 2 ( α , β , γ ) = cos β 2 cos α + γ 2 I ^ + 2 sin β 2 e − i α − γ 2 S ^ + 1 − 2 i cos β 2 sin α + γ 2 S ^ 0 + 2 sin β 2 e i α − γ 2 S ^ − 1 . (329) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma)=\cos\frac{\beta}{2}\cos\frac{\alpha+\gamma}{2}\hat{I}+\sqrt{2}\sin\frac{\beta}{2}e^{-i\frac{\alpha-\gamma}{2}}\hat{S}_{+1}-2i\cos\frac{\beta}{2}\sin\frac{\alpha+\gamma}{2}\hat{S}_{0}+\sqrt{2}\sin\frac{\beta}{2}e^{i\frac{\alpha-\gamma}{2}}\hat{S}_{-1}. \tag*{(329)} D ^ 2 1 ( α , β , γ ) = cos 2 β cos 2 α + γ I ^ + 2 sin 2 β e − i 2 α − γ S ^ + 1 − 2 i cos 2 β sin 2 α + γ S ^ 0 + 2 sin 2 β e i 2 α − γ S ^ − 1 . (329)
(a) Axis and angle
Under rotations of the coordinate system through an angle ω \omega ω about an axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) spin functions and spin matrices for S = 1 2 S=\frac{1}{2} S = 2 1 are transformed by the rotation operator U ^ 1 2 ( ω ; Θ , Φ ) \hat{U}^{\frac{1}{2}}(\omega;\Theta,\Phi) U ^ 2 1 ( ω ; Θ , Φ ) which is special case of the operator given by Eq. (162) .
U ^ 1 2 ( ω ; Θ , Φ ) = e − i ω n ⋅ S ^ = ( cos ω 2 − i sin ω 2 cos Θ − i sin ω 2 sin Θ e − i Φ − i sin ω 2 sin Θ e i Φ cos ω 2 + i sin ω 2 cos Θ ) . (330) \hat{U}^{\frac{1}{2}}(\omega;\Theta,\Phi)=e^{-i\omega\mathbf{n}\cdot\hat{\mathbf{S}}}=
\begin{pmatrix}
\cos\frac{\omega}{2}-i\sin\frac{\omega}{2}\cos\Theta& -i\sin\frac{\omega}{2}\sin\Theta e^{-i\Phi}\\
-i\sin\frac{\omega}{2}\sin\Theta e^{i\Phi} & \cos\frac{\omega}{2}+i\sin\frac{\omega}{2}\cos\Theta
\end{pmatrix}. \tag*{(330)} U ^ 2 1 ( ω ; Θ , Φ ) = e − iω n ⋅ S ^ = ( cos 2 ω − i sin 2 ω cos Θ − i sin 2 ω sin Θ e i Φ − i sin 2 ω sin Θ e − i Φ cos 2 ω + i sin 2 ω cos Θ ) . (330)
The inverse matrix is
[ U ^ 1 2 ( ω ; Θ , Φ ) ] − 1 = [ U ^ 1 2 ( ω ; Θ , Φ ) ] + = U ^ 1 2 ( ω ; π − Θ , π + Φ ) = U ^ 1 2 ( − ω ; Θ , Φ ) = ( cos ω 2 + i sin ω 2 cos Θ i sin ω 2 sin Θ e − i Φ i sin ω 2 sin Θ e i Φ cos ω 2 − i sin ω 2 cos Θ ) (331) \begin{aligned}
\left[\hat{U}^{\frac{1}{2}}(\omega;\Theta,\Phi)\right]^{-1}=\left[\hat{U}^{\frac{1}{2}}(\omega;\Theta,\Phi)\right]^{+}=\hat{U}^{\frac{1}{2}}(\omega;\pi-\Theta,\pi+\Phi)=\hat{U}^{\frac{1}{2}}(-\omega;\Theta,\Phi) \\
&=\begin{pmatrix}
\cos\frac{\omega}{2}+i\sin\frac{\omega}{2}\cos\Theta& i\sin\frac{\omega}{2}\sin\Theta e^{-i\Phi}\\
i\sin\frac{\omega}{2}\sin\Theta e^{i\Phi} & \cos\frac{\omega}{2}-i\sin\frac{\omega}{2}\cos\Theta
\end{pmatrix} \tag*{(331)}
\end{aligned} [ U ^ 2 1 ( ω ; Θ , Φ ) ] − 1 = [ U ^ 2 1 ( ω ; Θ , Φ ) ] + = U ^ 2 1 ( ω ; π − Θ , π + Φ ) = U ^ 2 1 ( − ω ; Θ , Φ ) = ( cos 2 ω + i sin 2 ω cos Θ i sin 2 ω sin Θ e i Φ i sin 2 ω sin Θ e − i Φ cos 2 ω − i sin 2 ω cos Θ ) (331)
The expansion of the operator U ^ 1 2 ( ω ; θ , Φ ) \hat{U}^{\frac{1}{2}}(\omega;\theta,\Phi) U ^ 2 1 ( ω ; θ , Φ ) in terms of the spin matrices has the form
U ^ 1 2 ( ω ; θ , Φ ) = e − i ω n ⋅ S ^ = I ^ cos ω 2 − 2 i ( n ⋅ S ^ ) sin ω 2 . (332) \hat{U}^{\frac{1}{2}}(\omega;\theta,\Phi)=e^{-i\omega\mathbf{n}\cdot\hat{\mathbf{S}}}=\hat{I}\cos\frac{\omega}{2}-2i(\mathbf{n}\cdot\hat{\mathbf{S}})\sin\frac{\omega}{2}. \tag*{(332)} U ^ 2 1 ( ω ; θ , Φ ) = e − iω n ⋅ S ^ = I ^ cos 2 ω − 2 i ( n ⋅ S ^ ) sin 2 ω . (332)
The relations between the angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ and the Euler angles α \alpha α , β \beta β , γ \gamma γ are given by Eqs. (145) , (146) . Note that
D ^ 1 2 ( α , β , γ ) = U ^ 1 2 ( ω ; Θ , Φ ) . (333) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma)=\hat{U}^{\frac{1}{2}}(\omega;\Theta,\Phi). \tag*{(333)} D ^ 2 1 ( α , β , γ ) = U ^ 2 1 ( ω ; Θ , Φ ) . (333)
Result of rotation operator
The result of applying the rotation operator D ^ 1 / 2 ( α , β , γ ) \hat{D}^{1/2}(\alpha,\beta,\gamma) D ^ 1/2 ( α , β , γ ) to the spin matrices may be presented as
S ^ i ′ ≡ D ^ 1 / 2 ( α , β , γ ) S ^ i [ D ^ 1 / 2 ( α , β , γ ) ] − 1 = ∑ k a i k S ^ k , ( i , k = x , y , z ) . (334) \hat{S}'_i \equiv\hat{D}^{1/2}(\alpha,\beta,\gamma)\hat{S}_i\left[\hat{D}^{1/2}(\alpha,\beta,\gamma)\right]^{-1} = \sum_k a_{ik}\hat{S}_k,\quad(i,k=x,y,z). \tag*{(334)} S ^ i ′ ≡ D ^ 1/2 ( α , β , γ ) S ^ i [ D ^ 1/2 ( α , β , γ ) ] − 1 = k ∑ a ik S ^ k , ( i , k = x , y , z ) . (334)
where a i k a_{ik} a ik are elements of the rotation matrix (Sec. 1.4.6). For spherical components of the spin matrices we have
S ^ μ ′ ≡ D ^ 1 / 2 ( α , β , γ ) S ^ μ [ D ^ 1 / 2 ( α , β , γ ) ] − 1 = ∑ ν D ν μ 1 ( α , β , γ ) S ^ ν ; (335) \hat{S}'_\mu\equiv\hat{D}^{1/2}(\alpha,\beta,\gamma)\hat{S}_\mu\left[\hat{D}^{1/2}(\alpha,\beta,\gamma)\right]^{-1} = \sum_\nu D^1_{\nu\mu}(\alpha,\beta,\gamma)\hat{S}_\nu; \tag*{(335)} S ^ μ ′ ≡ D ^ 1/2 ( α , β , γ ) S ^ μ [ D ^ 1/2 ( α , β , γ ) ] − 1 = ν ∑ D ν μ 1 ( α , β , γ ) S ^ ν ; (335)
where D ν μ 1 D^1_{\nu\mu} D ν μ 1 are the Wigner D-functions (Chap. 4).
Traces of Products of Spin Matrices ( S = 1 2 S=\frac{1}{2} S = 2 1 )
For cartesian components of the spin operator the following relations hold (where i , k , l i,k,l i , k , l etc. take the values x , y , z x,y,z x , y , z )
Tr { S ^ i } = 0. , \operatorname{Tr}\{\hat{S}_i\}=0., Tr { S ^ i } = 0. ,
Tr { S ^ i S ^ k } = 1 2 δ i k , \operatorname{Tr}\{\hat{S}_i\hat{S}_k\}=\frac{1}{2}\delta_{ik}, Tr { S ^ i S ^ k } = 2 1 δ ik ,
Tr { S ^ i S ^ k S ^ l } = i 4 ϵ i k l , \operatorname{Tr}\{\hat{S}_i\hat{S}_k\hat{S}_l\}=\frac{i}{4}\epsilon_{ikl}, Tr { S ^ i S ^ k S ^ l } = 4 i ϵ ik l ,
Tr { S ^ i S ^ k S ^ l S ^ j } = 1 8 ( δ i k δ l j − δ i l δ k j + δ i j δ k l ) , \operatorname{Tr}\{\hat{S}_i\hat{S}_k\hat{S}_l\hat{S}_j\}=\frac{1}{8}\left(\delta_{ik}\delta_{lj}-\delta_{il}\delta_{kj}+\delta_{ij}\delta_{kl}\right), Tr { S ^ i S ^ k S ^ l S ^ j } = 8 1 ( δ ik δ l j − δ i l δ k j + δ ij δ k l ) ,
Tr { S ^ i S ^ k S ^ l S ^ j S ^ m } = i 16 ( δ i k ϵ l j m + δ j l ϵ i k m + δ m l ϵ i j k + δ j m ϵ i k l ) . (336) \operatorname{Tr}\{\hat{S}_i\hat{S}_k\hat{S}_l\hat{S}_j\hat{S}_m\}=\frac{i}{16}\left(\delta_{ik}\epsilon_{ljm}+\delta_{jl}\epsilon_{ikm}+\delta_{ml}\epsilon_{ijk}+\delta_{jm}\epsilon_{ikl}\right). \tag*{(336)} Tr { S ^ i S ^ k S ^ l S ^ j S ^ m } = 16 i ( δ ik ϵ l j m + δ j l ϵ ik m + δ m l ϵ ij k + δ j m ϵ ik l ) . (336)
The evaluation of traces of products which contain many spin matrices can be facilitated by using the following recurrence relation: if we introduce the definition
T i 1 i 2 … i n ≡ Tr { S ^ i 1 S ^ i 2 … S ^ i n } , ( n ≥ 3 ) , (337) T_{i_1i_2\ldots i_n}\equiv\operatorname{Tr}\{\hat{S}_{i_1}\hat{S}_{i_2}\ldots\hat{S}_{i_n}\},\quad(n\geq3), \tag*{(337)} T i 1 i 2 … i n ≡ Tr { S ^ i 1 S ^ i 2 … S ^ i n } , ( n ≥ 3 ) , (337)
then
T i 1 i 2 … i n = 1 4 δ i 1 i 2 T i 3 i 4 … i n + i 2 ϵ i 1 i 2 i ′ T i ′ i 3 … i n . (338) T_{i_1i_2\ldots i_n}=\frac{1}{4}\delta_{i_1i_2}T_{i_3i_4\ldots i_n}+\frac{i}{2}\epsilon_{i_1i_2i'}T_{i'i_3\ldots i_n}. \tag*{(338)} T i 1 i 2 … i n = 4 1 δ i 1 i 2 T i 3 i 4 … i n + 2 i ϵ i 1 i 2 i ′ T i ′ i 3 … i n . (338)
If n n n is even, one may also use the relation
T i 1 i 2 … i n = 1 4 { δ i 1 i 2 T i 3 i 4 … i n − δ i 1 i 3 T i 2 i 4 … i n + … + δ i 1 i n T i 2 i 3 … i n − 1 } . (339) T_{i_1i_2\ldots i_n}=\frac{1}{4}\left\{\delta_{i_1i_2}T_{i_3i_4\ldots i_n}-\delta_{i_1i_3}T_{i_2i_4\ldots i_n}+\ldots+\delta_{i_1i_n}T_{i_2i_3\ldots i_{n-1}}\right\}. \tag*{(339)} T i 1 i 2 … i n = 4 1 { δ i 1 i 2 T i 3 i 4 … i n − δ i 1 i 3 T i 2 i 4 … i n + … + δ i 1 i n T i 2 i 3 … i n − 1 } . (339)
For spherical components of the spin operator the following relations hold (where μ , ν , λ \mu,\nu,\lambda μ , ν , λ etc. take the values ± 1 , 0 \pm1,0 ± 1 , 0 )
Tr { S ^ μ } = 0 , \operatorname{Tr}\{\hat{S}_\mu\}=0, Tr { S ^ μ } = 0 ,
Tr { S ^ μ S ^ ν } = 1 2 ( − 1 ) μ δ μ − ν , \operatorname{Tr}\{\hat{S}_\mu\hat{S}_\nu\}=\frac{1}{2}(-1)^\mu\delta_{\mu-\nu}, Tr { S ^ μ S ^ ν } = 2 1 ( − 1 ) μ δ μ − ν ,
Tr { S ^ μ S ^ ν S ^ λ } = − 1 2 3 2 ( 1 1 1 μ ν λ ) = ( − 1 ) 1 + λ 1 2 2 C 1 μ 1 ν 1 − λ , \operatorname{Tr}\{\hat{S}_\mu\hat{S}_\nu\hat{S}_\lambda\}=-\frac{1}{2}\sqrt{\frac{3}{2}}\begin{pmatrix}1&1&1\\ \mu&\nu&\lambda\end{pmatrix}=(-1)^{1+\lambda}\frac{1}{2\sqrt{2}}C^{1-\lambda}_{1\mu1\nu}, Tr { S ^ μ S ^ ν S ^ λ } = − 2 1 2 3 ( 1 μ 1 ν 1 λ ) = ( − 1 ) 1 + λ 2 2 1 C 1 μ 1 ν 1 − λ ,
Tr { S ^ μ S ^ ν S ^ λ S ^ ρ } = ( − 1 ) μ 8 { ( − 1 ) λ δ μ − ν δ λ − ρ − ( − 1 ) ν δ μ − λ δ ν − ρ + ( − 1 ) ν δ μ − ρ δ ν − λ } . (340) \operatorname{Tr}\{\hat{S}_\mu\hat{S}_\nu\hat{S}_\lambda\hat{S}_\rho\}=\frac{(-1)^\mu}{8}\left\{(-1)^\lambda\delta_{\mu-\nu}\delta_{\lambda-\rho}-(-1)^\nu\delta_{\mu-\lambda}\delta_{\nu-\rho}+(-1)^\nu\delta_{\mu-\rho}\delta_{\nu-\lambda}\right\}. \tag*{(340)} Tr { S ^ μ S ^ ν S ^ λ S ^ ρ } = 8 ( − 1 ) μ { ( − 1 ) λ δ μ − ν δ λ − ρ − ( − 1 ) ν δ μ − λ δ ν − ρ + ( − 1 ) ν δ μ − ρ δ ν − λ } . (340)
The recurrence relation for spherical components of the spin operator is written as
T μ 1 μ 2 … μ n = ( − 1 ) μ 1 4 δ μ 1 − μ 2 T μ 3 … μ n − 1 2 C 1 μ 2 1 μ 1 T μ μ 3 … μ n (341) T_{\mu_1\mu_2\ldots\mu_n}=\frac{(-1)^{\mu_1}}{4}\delta_{\mu_1-\mu_2}T_{\mu_3\ldots\mu_n}-\frac{1}{\sqrt{2}}C^{1\mu_1}_{1\mu_2}T_{\mu\mu_3\ldots\mu_n} \tag*{(341)} T μ 1 μ 2 … μ n = 4 ( − 1 ) μ 1 δ μ 1 − μ 2 T μ 3 … μ n − 2 1 C 1 μ 2 1 μ 1 T μ μ 3 … μ n (341)
where
T μ 1 μ 2 … μ n ≡ Tr { S ^ μ 1 S ^ μ 2 … S ^ μ n } . (342) T_{\mu_1\mu_2\ldots\mu_n}\equiv\operatorname{Tr}\{\hat{S}_{\mu_1}\hat{S}_{\mu_2}\ldots\hat{S}_{\mu_n}\}. \tag*{(342)} T μ 1 μ 2 … μ n ≡ Tr { S ^ μ 1 S ^ μ 2 … S ^ μ n } . (342)
If n n n is even, one may use the formula
T μ 1 μ 2 … μ n = ( − 1 ) μ 1 4 { δ μ 1 − μ 2 T μ 3 μ 4 … μ n − δ μ 1 − μ 3 T μ 2 μ 4 … μ n + … + δ μ 1 − μ n T μ 2 μ 3 … μ n − 1 } . (343) T_{\mu_1\mu_2\ldots\mu_n}=\frac{(-1)^{\mu_1}}{4}\left\{\delta_{\mu_1-\mu_2}T_{\mu_3\mu_4\ldots\mu_n}-\delta_{\mu_1-\mu_3}T_{\mu_2\mu_4\ldots\mu_n}+\ldots+\delta_{\mu_1-\mu_n}T_{\mu_2\mu_3\ldots\mu_{n-1}}\right\}. \tag*{(343)} T μ 1 μ 2 … μ n = 4 ( − 1 ) μ 1 { δ μ 1 − μ 2 T μ 3 μ 4 … μ n − δ μ 1 − μ 3 T μ 2 μ 4 … μ n + … + δ μ 1 − μ n T μ 2 μ 3 … μ n − 1 } . (343)
Spin matrices and polarization operators for S = 1 S = 1 S = 1
Spin S = 1 S = 1 S = 1
If S = 1 S = 1 S = 1 , the spin operator S ^ \hat{\mathbf{S}} S ^ and the polarization operators T ^ L M \hat{T}_{LM} T ^ L M (L = 0 , 1 , 2 L = 0,1,2 L = 0 , 1 , 2 ; − L ≤ M ≤ L -L \le M \le L − L ≤ M ≤ L ) are square 3 × 3 3 \times3 3 × 3 matrices. The polarization operator T ^ 00 \hat{T}_{00} T ^ 00 is proportional to the unit matrix
T ^ 00 = 1 3 I ^ , (344) \hat{T}_{00} = \frac{1}{\sqrt{3}}\hat{I}, \tag*{(344)} T ^ 00 = 3 1 I ^ , (344)
where
I ^ = ( 1 0 0 0 1 0 0 0 1 ) . (345) \hat{I} =
\begin{pmatrix}
1 & 0 & 0 \\
0 & 1 & 0 \\
0 & 0 & 1
\end{pmatrix}. \tag*{(345)} I ^ = 1 0 0 0 1 0 0 0 1 . (345)
The polarization operators T ^ 1 M \hat{T}_{1M} T ^ 1 M are proportional to spherical components of the spin matrices S ^ M \hat{S}_M S ^ M :
T ^ 1 M = 1 2 S ^ M , ( M = 0 , ± 1 ) . (346) \hat{T}_{1M} = \frac{1}{\sqrt{2}}\hat{S}_M,\qquad(M = 0, \pm1). \tag*{(346)} T ^ 1 M = 2 1 S ^ M , ( M = 0 , ± 1 ) . (346)
The operators T ^ 2 M \hat{T}_{2M} T ^ 2 M (M = 0 , ± 1 , ± 2 M = 0, \pm1, \pm2 M = 0 , ± 1 , ± 2 ) can be expressed in terms of spherical components of the spin matrices as
T ^ 2 M = ∑ μ , ν C 1 μ 1 ν 2 M S ^ μ S ^ ν . (347) \hat{T}_{2M} = \sum_{\mu,\nu} C^{2M}_{1\mu\,1\nu}\hat{S}_{\mu}\hat{S}_{\nu}. \tag*{(347)} T ^ 2 M = μ , ν ∑ C 1 μ 1 ν 2 M S ^ μ S ^ ν . (347)
The polarization operators T ^ 2 M \hat{T}_{2M} T ^ 2 M are equivalent to some symmetric traceless cartesian tensor of second rank Q ^ i k \hat{Q}_{ik} Q ^ ik :
Q ^ i k = 1 2 ( S ^ i S ^ k + S ^ k S ^ i − 4 3 δ i k I ^ ) , ( i , k = x , y , z ) , (348) \hat{Q}_{ik} = \frac{1}{2}\left(\hat{S}_{i}\hat{S}_{k} + \hat{S}_{k}\hat{S}_{i} - \frac{4}{3}\delta_{ik}\hat{I}\right),\qquad(i,k = x,y,z), \tag*{(348)} Q ^ ik = 2 1 ( S ^ i S ^ k + S ^ k S ^ i − 3 4 δ ik I ^ ) , ( i , k = x , y , z ) , (348)
Q ^ i k = Q ^ k i , ∑ i Q ^ i i = 0. (349) \hat{Q}_{ik} = \hat{Q}_{ki},\qquad\sum_i \hat{Q}_{ii} = 0. \tag*{(349)} Q ^ ik = Q ^ k i , i ∑ Q ^ ii = 0. (349)
Q ^ i k \hat{Q}_{ik} Q ^ ik is called the quadrupole tensor. It has 5 linearly independent components. The relations between T ^ 2 M \hat{T}_{2M} T ^ 2 M and Q ^ i k \hat{Q}_{ik} Q ^ ik are given by
T ^ 2 ± 2 = 1 2 ( Q ^ x x − Q ^ y y ± 2 i Q ^ x y ) , \hat{T}_{2\pm2} = \frac{1}{2}\left(\hat{Q}_{xx} - \hat{Q}_{yy} \pm2i\hat{Q}_{xy}\right), T ^ 2 ± 2 = 2 1 ( Q ^ xx − Q ^ y y ± 2 i Q ^ x y ) ,
T ^ 2 ± 1 = ∓ ( Q ^ x z ± i Q ^ y z ) , (350) \hat{T}_{2\pm1} = \mp\left(\hat{Q}_{xz} \pm i\hat{Q}_{yz}\right), \tag*{(350)} T ^ 2 ± 1 = ∓ ( Q ^ x z ± i Q ^ y z ) , (350)
T ^ 20 = 3 2 Q ^ z z . \hat{T}_{20} = \sqrt{\frac{3}{2}}\hat{Q}_{zz}. T ^ 20 = 2 3 Q ^ z z .
The inverse relations are
Q ^ x x = 1 2 ( T ^ 22 + T ^ 2 − 2 ) − 1 6 T ^ 20 , Q ^ x y = Q ^ y x = i 2 ( T ^ 2 − 2 − T ^ 22 ) , \hat{Q}_{xx} = \frac{1}{2}\left(\hat{T}_{22} + \hat{T}_{2-2}\right) - \frac{1}{\sqrt{6}}\hat{T}_{20},\qquad
\hat{Q}_{xy} = \hat{Q}_{yx} = \frac{i}{2}\left(\hat{T}_{2-2} - \hat{T}_{22}\right), Q ^ xx = 2 1 ( T ^ 22 + T ^ 2 − 2 ) − 6 1 T ^ 20 , Q ^ x y = Q ^ y x = 2 i ( T ^ 2 − 2 − T ^ 22 ) ,
Q ^ y y = − 1 2 ( T ^ 22 + T ^ 2 − 2 ) − 1 6 T ^ 20 , Q ^ x z = Q ^ z x = 1 2 ( T ^ 2 − 1 − T ^ 21 ) , \hat{Q}_{yy} = -\frac{1}{2}\left(\hat{T}_{22} + \hat{T}_{2-2}\right) - \frac{1}{\sqrt{6}}\hat{T}_{20},\qquad
\hat{Q}_{xz} = \hat{Q}_{zx} = \frac{1}{2}\left(\hat{T}_{2-1} - \hat{T}_{21}\right), Q ^ y y = − 2 1 ( T ^ 22 + T ^ 2 − 2 ) − 6 1 T ^ 20 , Q ^ x z = Q ^ z x = 2 1 ( T ^ 2 − 1 − T ^ 21 ) ,
Q ^ z z = 2 3 T ^ 20 , Q ^ y z = Q ^ z y = i 2 ( T ^ 2 − 1 + T ^ 21 ) . (351) \hat{Q}_{zz} = \sqrt{\frac{2}{3}}\hat{T}_{20},\qquad
\hat{Q}_{yz} = \hat{Q}_{zy} = \frac{i}{2}\left(\hat{T}_{2-1} + \hat{T}_{21}\right). \tag*{(351)} Q ^ z z = 3 2 T ^ 20 , Q ^ y z = Q ^ z y = 2 i ( T ^ 2 − 1 + T ^ 21 ) . (351)
Explicit Form
Explicit forms of the spin matrices and the polarization operators depend on the representation used for basis spin functions (Chap. 6). We shall consider the spherical basis representation and the cartesian one. These representations are used very frequently.
(a) Spherical basis representation
Cartesian components of the spin operator S ^ \hat{\mathbf{S}} S ^ are given by
S ^ x = 1 2 ( 0 1 0 1 0 1 0 1 0 ) , S ^ y = i 2 ( 0 − 1 0 1 0 − 1 0 1 0 ) , S ^ z = ( 1 0 0 0 0 0 0 0 − 1 ) (352) \hat{S}_{x} = \frac{1}{\sqrt{2}}
\begin{pmatrix}
0 & 1 & 0 \\
1 & 0 & 1 \\
0 & 1 & 0
\end{pmatrix},
\qquad
\hat{S}_{y} = \frac{i}{\sqrt{2}}
\begin{pmatrix}
0 & -1 & 0 \\
1 & 0 & -1 \\
0 & 1 & 0
\end{pmatrix},
\qquad
\hat{S}_{z} =
\begin{pmatrix}
1 & 0 & 0 \\
0 & 0 & 0 \\
0 & 0 & -1
\end{pmatrix} \tag*{(352)} S ^ x = 2 1 0 1 0 1 0 1 0 1 0 , S ^ y = 2 i 0 1 0 − 1 0 1 0 − 1 0 , S ^ z = 1 0 0 0 0 0 0 0 − 1 (352)
and the spherical components of S ^ \hat{S} S ^ are
S ^ + 1 = − ( 0 1 0 0 0 1 0 0 0 ) , S ^ 0 = ( 1 0 0 0 0 0 0 0 − 1 ) , S ^ − 1 = ( 0 0 0 1 0 0 0 1 0 ) . (353) \hat{S}_{+1}=-\begin{pmatrix} 0 & 1 & 0 \\ 0 & 0 & 1 \\ 0 & 0 & 0 \end{pmatrix},\qquad\hat{S}_{0}=\begin{pmatrix} 1 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & -1 \end{pmatrix},\qquad\hat{S}_{-1}=\begin{pmatrix} 0 & 0 & 0 \\ 1 & 0 & 0 \\ 0 & 1 & 0 \end{pmatrix}. \tag*{(353)} S ^ + 1 = − 0 0 0 1 0 0 0 1 0 , S ^ 0 = 1 0 0 0 0 0 0 0 − 1 , S ^ − 1 = 0 1 0 0 0 1 0 0 0 . (353)
The order of rows and columns in these matrices is shown in Eq. (268) . Matrix elements of the spin operators S ^ μ \hat{S}_{\mu} S ^ μ in the spherical basis representation are given by
( S ^ μ ) σ ′ σ = 2 C 1 σ 1 μ 1 σ ′ ( μ , σ , σ ′ = ± 1 , 0 ) . (354) \left(\hat{S}_{\mu}\right)_{\sigma'\sigma}=\sqrt{2}C^{1\sigma'}_{1\sigma\,1\mu}\qquad(\mu,\sigma,\sigma'=\pm1,0). \tag*{(354)} ( S ^ μ ) σ ′ σ = 2 C 1 σ 1 μ 1 σ ′ ( μ , σ , σ ′ = ± 1 , 0 ) . (354)
The spin matrices in the spherical basis representation satisfy the relations
S ^ i + = S ^ i , ( i = x , y , z ) , (355) \hat{S}_{i}^{+}=\hat{S}_{i},\qquad(i=x,y,z), \tag*{(355)} S ^ i + = S ^ i , ( i = x , y , z ) , (355)
S ^ x ∗ = S ^ x , S ^ y ∗ = − S ^ y , S ^ z ∗ = S ^ z , (356) \hat{S}_{x}^{*}=\hat{S}_{x},\qquad\hat{S}_{y}^{*}=-\hat{S}_{y},\qquad\hat{S}_{z}^{*}=\hat{S}_{z}, \tag*{(356)} S ^ x ∗ = S ^ x , S ^ y ∗ = − S ^ y , S ^ z ∗ = S ^ z , (356)
S ^ μ ∗ = S ^ μ , S ^ μ + = ( − 1 ) μ S ^ − μ ( μ = ± 1 , 0 ) . (357) \hat{S}_{\mu}^{*}=\hat{S}_{\mu},\qquad\hat{S}_{\mu}^{+}=(-1)^{\mu}\hat{S}_{-\mu}\qquad(\mu=\pm1,0). \tag*{(357)} S ^ μ ∗ = S ^ μ , S ^ μ + = ( − 1 ) μ S ^ − μ ( μ = ± 1 , 0 ) . (357)
The quadrupole tensor Q ^ i k ( i , k = x , y , z ) \hat{Q}_{ik}(i,k=x,y,z) Q ^ ik ( i , k = x , y , z ) in the spherical basis representation has the form
Q ^ x x = 1 6 ( − 1 0 3 0 2 0 3 0 − 1 ) , Q ^ y y = 1 6 ( − 1 0 − 3 0 2 0 − 3 0 − 1 ) , Q ^ z z = 1 3 ( 1 0 0 0 − 2 0 0 0 1 ) , (358) \hat{Q}_{xx}=\frac{1}{6}\begin{pmatrix} -1 & 0 & 3 \\ 0 & 2 & 0 \\ 3 & 0 & -1 \end{pmatrix},\qquad\hat{Q}_{yy}=\frac{1}{6}\begin{pmatrix} -1 & 0 & -3 \\ 0 & 2 & 0 \\ -3 & 0 & -1 \end{pmatrix},\qquad\hat{Q}_{zz}=\frac{1}{3}\begin{pmatrix} 1 & 0 & 0 \\ 0 & -2 & 0 \\ 0 & 0 & 1 \end{pmatrix}, \tag*{(358)} Q ^ xx = 6 1 − 1 0 3 0 2 0 3 0 − 1 , Q ^ y y = 6 1 − 1 0 − 3 0 2 0 − 3 0 − 1 , Q ^ z z = 3 1 1 0 0 0 − 2 0 0 0 1 , (358)
Q ^ x y = Q ^ y x = i 2 ( 0 0 − 1 0 0 0 1 0 0 ) , Q ^ x z = Q ^ z x = 1 2 2 ( 0 1 0 1 0 − 1 0 − 1 0 ) , Q ^ y z = Q ^ z y = i 2 2 ( 0 − 1 0 1 0 1 0 − 1 0 ) . \hat{Q}_{xy}=\hat{Q}_{yx}=\frac{i}{2}\begin{pmatrix} 0 & 0 & -1 \\ 0 & 0 & 0 \\ 1 & 0 & 0 \end{pmatrix},\qquad\hat{Q}_{xz}=\hat{Q}_{zx}=\frac{1}{2\sqrt{2}}\begin{pmatrix} 0 & 1 & 0 \\ 1 & 0 & -1 \\ 0 & -1 & 0 \end{pmatrix},\qquad\hat{Q}_{yz}=\hat{Q}_{zy}=\frac{i}{2\sqrt{2}}\begin{pmatrix} 0 & -1 & 0 \\ 1 & 0 & 1 \\ 0 & -1 & 0 \end{pmatrix}. Q ^ x y = Q ^ y x = 2 i 0 0 1 0 0 0 − 1 0 0 , Q ^ x z = Q ^ z x = 2 2 1 0 1 0 1 0 − 1 0 − 1 0 , Q ^ y z = Q ^ z y = 2 2 i 0 1 0 − 1 0 − 1 0 1 0 .
The components of the quadrupole tensor Q ^ i k \hat{Q}_{ik} Q ^ ik in the spherical basis representation have the following properties
Q ^ i k + = Q ^ i k , ( i , k = x , y , z ) , (359) \hat{Q}_{ik}^{+}=\hat{Q}_{ik},\qquad(i,k=x,y,z), \tag*{(359)} Q ^ ik + = Q ^ ik , ( i , k = x , y , z ) , (359)
The matrices Q ^ x x , Q ^ y y , Q ^ z z , Q ^ x z , Q ^ z x \hat{Q}_{xx},\hat{Q}_{yy},\hat{Q}_{zz},\hat{Q}_{xz},\hat{Q}_{zx} Q ^ xx , Q ^ y y , Q ^ z z , Q ^ x z , Q ^ z x are real, whereas Q ^ x y , Q ^ y x , Q ^ y z , Q ^ z y \hat{Q}_{xy},\hat{Q}_{yx},\hat{Q}_{yz},\hat{Q}_{zy} Q ^ x y , Q ^ y x , Q ^ y z , Q ^ z y are purely imaginary. The polarization operators T ^ L M \hat{T}_{LM} T ^ L M in the spherical basis representation are given by
T ^ 00 = 1 3 ( 1 0 0 0 1 0 0 0 1 ) , (360) \hat{T}_{00}=\frac{1}{\sqrt{3}}\begin{pmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{pmatrix}, \tag*{(360)} T ^ 00 = 3 1 1 0 0 0 1 0 0 0 1 , (360)
T ^ + 1 = − 1 2 ( 0 1 0 0 0 1 0 0 0 ) , T ^ 10 = 1 2 ( 1 0 0 0 0 0 0 0 − 1 ) , T ^ 1 − 1 = 1 2 ( 0 0 0 1 0 0 0 1 0 ) , (361) \hat{T}_{+1}=-\frac{1}{\sqrt{2}}\begin{pmatrix} 0 & 1 & 0 \\ 0 & 0 & 1 \\ 0 & 0 & 0 \end{pmatrix},\qquad\hat{T}_{10}=\frac{1}{\sqrt{2}}\begin{pmatrix} 1 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & -1 \end{pmatrix},\qquad\hat{T}_{1-1}=\frac{1}{\sqrt{2}}\begin{pmatrix} 0 & 0 & 0 \\ 1 & 0 & 0 \\ 0 & 1 & 0 \end{pmatrix}, \tag*{(361)} T ^ + 1 = − 2 1 0 0 0 1 0 0 0 1 0 , T ^ 10 = 2 1 1 0 0 0 0 0 0 0 − 1 , T ^ 1 − 1 = 2 1 0 1 0 0 0 1 0 0 0 , (361)
T ^ + 2 = ( 0 0 1 0 0 0 0 0 0 ) , T ^ 21 = 1 2 ( 0 − 1 0 0 0 1 0 0 0 ) , T ^ 20 = 1 6 ( 1 0 0 0 − 2 0 0 0 1 ) , \hat{T}_{+2}=\begin{pmatrix} 0 & 0 & 1 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \end{pmatrix},\qquad\hat{T}_{21}=\frac{1}{\sqrt{2}}\begin{pmatrix} 0 & -1 & 0 \\ 0 & 0 & 1 \\ 0 & 0 & 0 \end{pmatrix},\qquad\hat{T}_{20}=\frac{1}{\sqrt{6}}\begin{pmatrix} 1 & 0 & 0 \\ 0 & -2 & 0 \\ 0 & 0 & 1 \end{pmatrix}, T ^ + 2 = 0 0 0 0 0 0 1 0 0 , T ^ 21 = 2 1 0 0 0 − 1 0 0 0 1 0 , T ^ 20 = 6 1 1 0 0 0 − 2 0 0 0 1 ,
T ^ 2 − 1 = 1 2 ( 0 0 0 1 0 0 0 − 1 0 ) , T ^ 2 − 2 = ( 0 0 0 0 0 0 1 0 0 ) . (362) \hat{T}_{2-1}=\frac{1}{\sqrt{2}}\begin{pmatrix} 0 & 0 & 0 \\ 1 & 0 & 0 \\ 0 & -1 & 0 \end{pmatrix},\qquad\hat{T}_{2-2}=\begin{pmatrix} 0 & 0 & 0 \\ 0 & 0 & 0 \\ 1 & 0 & 0 \end{pmatrix}. \tag*{(362)} T ^ 2 − 1 = 2 1 0 1 0 0 0 − 1 0 0 0 , T ^ 2 − 2 = 0 0 1 0 0 0 0 0 0 . (362)
Matrix elements of the polarization operators in the spherical basis representation may be written as
( T ^ L M ) σ ′ σ = 2 L + 1 3 C 1 σ L M 1 σ ′ , ( σ , σ ′ = ± 1 , 0 ; − L ≤ M ≤ L ) . (363) \left(\hat{T}_{LM}\right)_{\sigma'\sigma}=\sqrt{\frac{2L+1}{3}}C^{1\sigma'}_{1\sigma\,LM},\qquad(\sigma,\sigma'=\pm1,0;-L\leq M\leq L). \tag*{(363)} ( T ^ L M ) σ ′ σ = 3 2 L + 1 C 1 σ L M 1 σ ′ , ( σ , σ ′ = ± 1 , 0 ; − L ≤ M ≤ L ) . (363)
The polarization operators T ^ L M \hat{T}_{LM} T ^ L M in the spherical basis representation are real, i.e.,
T ^ L M ∗ = T ^ L M . (364) \hat{T}_{LM}^{*}=\hat{T}_{LM}. \tag*{(364)} T ^ L M ∗ = T ^ L M . (364)
and satisfy the relations
T ^ L M + = ( − 1 ) M T ^ L − M . (365) \hat{T}_{LM}^{+}=(-1)^{M}\hat{T}_{L-M}. \tag*{(365)} T ^ L M + = ( − 1 ) M T ^ L − M . (365)
Cartesian basis representation
Cartesian components of the spin operator S ^ \hat{S} S ^ in the cartesian basis representation are given by
S ^ x = ( 0 0 0 0 0 − i 0 i 0 ) , S ^ y = ( 0 0 i 0 0 0 − i 0 0 ) , S ^ z = ( 0 − i 0 i 0 0 0 0 0 ) . (366) \hat{S}_x=\begin{pmatrix}0&0&0\\0&0&-\mathrm{i}\\0&\mathrm{i}&0\end{pmatrix},\qquad
\hat{S}_y=\begin{pmatrix}0&0&\mathrm{i}\\0&0&0\\-\mathrm{i}&0&0\end{pmatrix},\qquad
\hat{S}_z=\begin{pmatrix}0&-\mathrm{i}&0\\\mathrm{i}&0&0\\0&0&0\end{pmatrix}. \tag*{(366)} S ^ x = 0 0 0 0 0 i 0 − i 0 , S ^ y = 0 0 − i 0 0 0 i 0 0 , S ^ z = 0 i 0 − i 0 0 0 0 0 . (366)
The matrix elements of S ^ i \hat{S}_i S ^ i in this representation may be written as
( S ^ i ) k l = − i ε i k l , ( i , k , l = x , y , z ) . (367) \left(\hat{S}_i\right)_{kl}=-\mathrm{i}\varepsilon_{ikl},\qquad(i,k,l=x,y,z). \tag*{(367)} ( S ^ i ) k l = − i ε ik l , ( i , k , l = x , y , z ) . (367)
The spherical components of S ^ \hat{S} S ^ are as follows
S ^ + 1 = 1 2 ( 0 0 1 0 0 i − 1 − i 0 ) , S ^ 0 = ( 0 − i 0 i 0 0 0 0 0 ) , S ^ − 1 = 1 2 ( 0 0 1 0 0 − i − 1 i 0 ) . (368) \hat{S}_{+1}=\frac{1}{\sqrt{2}}\begin{pmatrix}0&0&1\\0&0&\mathrm{i}\\-1&-\mathrm{i}&0\end{pmatrix},\qquad
\hat{S}_0=\begin{pmatrix}0&-\mathrm{i}&0\\\mathrm{i}&0&0\\0&0&0\end{pmatrix},\qquad
\hat{S}_{-1}=\frac{1}{\sqrt{2}}\begin{pmatrix}0&0&1\\0&0&-\mathrm{i}\\-1&\mathrm{i}&0\end{pmatrix}. \tag*{(368)} S ^ + 1 = 2 1 0 0 − 1 0 0 − i 1 i 0 , S ^ 0 = 0 i 0 − i 0 0 0 0 0 , S ^ − 1 = 2 1 0 0 − 1 0 0 i 1 − i 0 . (368)
The spin matrices in the cartesian basis representation satisfy the following relations
( S ^ μ ) + = ( − 1 ) μ S ^ − μ , S ^ μ ∗ = ( − 1 ) 1 + μ S ^ − μ , ( μ = ± 1 , 0 ) , (369) \left(\hat{S}_{\mu}\right)^{+}=(-1)^{\mu}\hat{S}_{-\mu},\qquad
\hat{S}_{\mu}^{*}=(-1)^{1+\mu}\hat{S}_{-\mu},\qquad(\mu=\pm1,0), \tag*{(369)} ( S ^ μ ) + = ( − 1 ) μ S ^ − μ , S ^ μ ∗ = ( − 1 ) 1 + μ S ^ − μ , ( μ = ± 1 , 0 ) , (369)
( S ^ i ) + = S ^ i , S ^ i ∗ = − S ^ i , ( i = x , y , z ) . (370) \left(\hat{S}_{i}\right)^{+}=\hat{S}_{i},\qquad\hat{S}_{i}^{*}=-\hat{S}_{i},\qquad(i=x,y,z). \tag*{(370)} ( S ^ i ) + = S ^ i , S ^ i ∗ = − S ^ i , ( i = x , y , z ) . (370)
The quadrupole tensor Q ^ i k ( i , k = x , y , z ) \hat{Q}_{ik}(i,k=x,y,z) Q ^ ik ( i , k = x , y , z ) in the cartesian basis representation is given by
Q ^ x x = 1 3 ( − 2 0 0 0 1 0 0 0 1 ) , Q ^ y y = 1 3 ( 1 0 0 0 − 2 0 0 0 1 ) , Q ^ z z = 1 3 ( 1 0 0 0 1 0 0 0 − 2 ) , \hat{Q}_{xx}=\frac{1}{3}\begin{pmatrix}-2&0&0\\0&1&0\\0&0&1\end{pmatrix},\qquad
\hat{Q}_{yy}=\frac{1}{3}\begin{pmatrix}1&0&0\\0&-2&0\\0&0&1\end{pmatrix},\qquad
\hat{Q}_{zz}=\frac{1}{3}\begin{pmatrix}1&0&0\\0&1&0\\0&0&-2\end{pmatrix}, Q ^ xx = 3 1 − 2 0 0 0 1 0 0 0 1 , Q ^ y y = 3 1 1 0 0 0 − 2 0 0 0 1 , Q ^ z z = 3 1 1 0 0 0 1 0 0 0 − 2 ,
Q ^ x y = Q ^ y x = 1 2 ( 0 − 1 0 − 1 0 0 0 0 0 ) , Q ^ x z = Q ^ z x = 1 2 ( 0 0 − 1 0 0 0 − 1 0 0 ) , Q ^ y z = Q ^ z y = 1 2 ( 0 0 0 0 0 − 1 0 − 1 0 ) . (371) \hat{Q}_{xy}=\hat{Q}_{yx}=\frac{1}{2}\begin{pmatrix}0&-1&0\\-1&0&0\\0&0&0\end{pmatrix},\qquad
\hat{Q}_{xz}=\hat{Q}_{zx}=\frac{1}{2}\begin{pmatrix}0&0&-1\\0&0&0\\-1&0&0\end{pmatrix},\qquad
\hat{Q}_{yz}=\hat{Q}_{zy}=\frac{1}{2}\begin{pmatrix}0&0&0\\0&0&-1\\0&-1&0\end{pmatrix}. \tag*{(371)} Q ^ x y = Q ^ y x = 2 1 0 − 1 0 − 1 0 0 0 0 0 , Q ^ x z = Q ^ z x = 2 1 0 0 − 1 0 0 0 − 1 0 0 , Q ^ y z = Q ^ z y = 2 1 0 0 0 0 0 − 1 0 − 1 0 . (371)
The matrix elements of Q ^ i k \hat{Q}_{ik} Q ^ ik in the cartesian basis representation may be written in the form
( Q ^ i k ) l m = − 1 2 ( δ i l δ k m + δ i m δ k l − 2 3 δ i k δ l m ) , ( i , k , l , m = x , y , z ) . (372) \left(\hat{Q}_{ik}\right)_{lm}=-\frac{1}{2}\left(\delta_{il}\delta_{km}+\delta_{im}\delta_{kl}-\frac{2}{3}\delta_{ik}\delta_{lm}\right),\qquad(i,k,l,m=x,y,z). \tag*{(372)} ( Q ^ ik ) l m = − 2 1 ( δ i l δ k m + δ im δ k l − 3 2 δ ik δ l m ) , ( i , k , l , m = x , y , z ) . (372)
In the cartesian basis representation the matrices Q ^ i k \hat{Q}_{ik} Q ^ ik are Hermitian and real, i.e.,
Q ^ i k + = Q ^ i k , Q ^ i k ∗ = Q ^ i k , ( i , k = x , y , z ) . (373) \hat{Q}_{ik}^{+}=\hat{Q}_{ik},\qquad\hat{Q}_{ik}^{*}=\hat{Q}_{ik},\qquad(i,k=x,y,z). \tag*{(373)} Q ^ ik + = Q ^ ik , Q ^ ik ∗ = Q ^ ik , ( i , k = x , y , z ) . (373)
The polarization operators T ^ L M \hat{T}_{LM} T ^ L M in the cartesian basis representation have the form
T ^ 00 = 1 3 ( 1 0 0 0 1 0 0 0 1 ) . (374) \hat{T}_{00}=\frac{1}{\sqrt{3}}\begin{pmatrix}1&0&0\\0&1&0\\0&0&1\end{pmatrix}. \tag*{(374)} T ^ 00 = 3 1 1 0 0 0 1 0 0 0 1 . (374)
T ^ 1 + 1 = 1 2 ( 0 0 1 0 0 i − 1 − i 0 ) , T ^ 10 = 1 2 ( 0 − i 0 i 0 0 0 0 0 ) , T ^ 1 − 1 = 1 2 ( 0 0 1 0 0 − i − 1 i 0 ) , \hat{T}_{1+1}=\frac{1}{2}\begin{pmatrix}0&0&1\\0&0&\mathrm{i}\\-1&-\mathrm{i}&0\end{pmatrix},\qquad
\hat{T}_{10}=\frac{1}{\sqrt{2}}\begin{pmatrix}0&-\mathrm{i}&0\\\mathrm{i}&0&0\\0&0&0\end{pmatrix},\qquad
\hat{T}_{1-1}=\frac{1}{2}\begin{pmatrix}0&0&1\\0&0&-\mathrm{i}\\-1&\mathrm{i}&0\end{pmatrix}, T ^ 1 + 1 = 2 1 0 0 − 1 0 0 − i 1 i 0 , T ^ 10 = 2 1 0 i 0 − i 0 0 0 0 0 , T ^ 1 − 1 = 2 1 0 0 − 1 0 0 i 1 − i 0 ,
T ^ 2 + 2 = 1 2 ( − 1 − i 0 − i 1 0 0 0 0 ) , T ^ 2 + 1 = 1 2 ( 0 0 1 0 0 i 1 i 0 ) , T ^ 20 = 1 6 ( 1 0 0 0 1 0 0 0 − 2 ) . (375) \hat{T}_{2+2}=\frac{1}{2}\begin{pmatrix}-1&-\mathrm{i}&0\\-\mathrm{i}&1&0\\0&0&0\end{pmatrix},\qquad
\hat{T}_{2+1}=\frac{1}{2}\begin{pmatrix}0&0&1\\0&0&\mathrm{i}\\1&\mathrm{i}&0\end{pmatrix},\qquad
\hat{T}_{20}=\frac{1}{\sqrt{6}}\begin{pmatrix}1&0&0\\0&1&0\\0&0&-2\end{pmatrix}. \tag*{(375)} T ^ 2 + 2 = 2 1 − 1 − i 0 − i 1 0 0 0 0 , T ^ 2 + 1 = 2 1 0 0 1 0 0 i 1 i 0 , T ^ 20 = 6 1 1 0 0 0 1 0 0 0 − 2 . (375)
T ^ 2 − 1 = 1 2 ( 0 0 − 1 0 0 i − 1 i 0 ) , T ^ 2 − 2 = 1 2 ( − 1 i 0 i 1 0 0 0 0 ) . (376) \hat{T}_{2-1}=\frac{1}{2}\begin{pmatrix}0&0&-1\\0&0&\mathrm{i}\\-1&\mathrm{i}&0\end{pmatrix},\qquad
\hat{T}_{2-2}=\frac{1}{2}\begin{pmatrix}-1&\mathrm{i}&0\\\mathrm{i}&1&0\\0&0&0\end{pmatrix}. \tag*{(376)} T ^ 2 − 1 = 2 1 0 0 − 1 0 0 i − 1 i 0 , T ^ 2 − 2 = 2 1 − 1 i 0 i 1 0 0 0 0 . (376)
The polarization operators T ^ L M \hat{T}_{LM} T ^ L M in the cartesian basis representation satisfy the relations
T ^ L M + = ( − 1 ) M T ^ L − M , (377) \hat{T}_{LM}^{+}=(-1)^{M}\hat{T}_{L-M}, \tag*{(377)} T ^ L M + = ( − 1 ) M T ^ L − M , (377)
T ^ L M ∗ = ( − 1 ) L + M T ^ L − M . (378) \hat{T}_{LM}^{*}=(-1)^{L+M}\hat{T}_{L-M}. \tag*{(378)} T ^ L M ∗ = ( − 1 ) L + M T ^ L − M . (378)
One can easily transform any matrix in the cartesian basis representation into a matrix in the spherical basis representation and vice versa with the aid of the unitary matrix U U U ,
A ^ ( spherical basis ) = U A ^ ( cartesian basis ) U − 1 , A ^ ( cartesian basis ) = U − 1 A ^ ( spherical basis ) U . (379) \begin{aligned}
\hat{A}(\text{spherical basis}) &= U\hat{A}(\text{cartesian basis})U^{-1},\\
\hat{A}(\text{cartesian basis}) &= U^{-1}\hat{A}(\text{spherical basis})U. \tag*{(379)}
\end{aligned} A ^ ( spherical basis ) A ^ ( cartesian basis ) = U A ^ ( cartesian basis ) U − 1 , = U − 1 A ^ ( spherical basis ) U . (379)
In this case A ^ \hat{A} A ^ is any spin or polarization operator (i.e., S ^ i \hat{S}_i S ^ i , S ^ μ \hat{S}_\mu S ^ μ , Q ^ i k \hat{Q}_{ik} Q ^ ik , T ^ L M \hat{T}_{LM} T ^ L M ),
U = ( − 1 2 i 2 0 0 0 1 1 2 i 2 0 ) , U − 1 = U + = ( − 1 2 0 1 2 i 2 0 i 2 0 1 0 ) . (380) U=\begin{pmatrix}
-\frac{1}{\sqrt{2}} & \frac{i}{\sqrt{2}} & 0\\
0 & 0 & 1\\
\frac{1}{\sqrt{2}} & \frac{i}{\sqrt{2}} & 0
\end{pmatrix},\qquad
U^{-1}=U^{+}=\begin{pmatrix}
-\frac{1}{\sqrt{2}} & 0 & \frac{1}{\sqrt{2}}\\
\frac{i}{\sqrt{2}} & 0 & \frac{i}{\sqrt{2}}\\
0 & 1 & 0
\end{pmatrix}. \tag*{(380)} U = − 2 1 0 2 1 2 i 0 2 i 0 1 0 , U − 1 = U + = − 2 1 2 i 0 0 0 1 2 1 2 i 0 . (380)
The matrix U U U coincides with M ( + 1 , 0 , − 1 ← x , y , z ) M(+1,0,-1\leftarrow x,y,z) M ( + 1 , 0 , − 1 ← x , y , z ) and U − 1 U^{-1} U − 1 coincides with M ( x , y , z ← + 1 , 0 , − 1 ) M(x,y,z\leftarrow+1,0,-1) M ( x , y , z ← + 1 , 0 , − 1 ) (Table 1.2 ).
Table 1.2. Matrices of transformations between cartesian, spherical contravariant and polar components of vectors.
The formulas given in § 11 (except those of § 11.0.2 ) are independent of the representation unless the contrary is indicated.
Products of Spin and Polarization Matrices
For S = 1 S=1 S = 1 , the nine matrices T ^ L M ( L = 0 , 1 , 2 ; − L ≤ M ≤ L ) \hat{T}_{LM}(L=0,1,2;\,-L\le M\le L) T ^ L M ( L = 0 , 1 , 2 ; − L ≤ M ≤ L ) or, equivalently, the matrices I ^ , S ^ , Q ^ i k \hat{I},\hat{S},\hat{Q}_{ik} I ^ , S ^ , Q ^ ik constitute a complete set of square 3 × 3 3\times3 3 × 3 matrices for expanding any function of the spin and polarization operators. In particular, products of the spin and polarization operators may be expanded in a series in terms of these matrices.
Products of cartesian components of the matrices S ^ \hat{S} S ^ and Q ^ i k \hat{Q}_{ik} Q ^ ik may be expressed as follows (i , k , l i,k,l i , k , l , etc. take the values x , y , z x,y,z x , y , z )
(a) Products of two cartesian components of spin matrices
S ^ i S ^ k = 2 3 δ i k I ^ + i 2 ε i k l S ^ l + Q ^ i k . (381) \hat{S}_i\hat{S}_k=\frac{2}{3}\delta_{ik}\hat{I}+\frac{i}{2}\varepsilon_{ikl}\hat{S}_l+\hat{Q}_{ik}. \tag*{(381)} S ^ i S ^ k = 3 2 δ ik I ^ + 2 i ε ik l S ^ l + Q ^ ik . (381)
In particular,
S ^ 2 = S ^ x 2 + S ^ y 2 + S ^ z 2 = 2 I ^ , (382) \hat{S}^{2}=\hat{S}_{x}^{2}+\hat{S}_{y}^{2}+\hat{S}_{z}^{2}=2\hat{I}, \tag*{(382)} S ^ 2 = S ^ x 2 + S ^ y 2 + S ^ z 2 = 2 I ^ , (382)
[ S ^ i , S ^ k ] ≡ S ^ i S ^ k − S ^ k S ^ i = i ε i k l S ^ l , (383) [\hat{S}_i,\hat{S}_k]\equiv\hat{S}_i\hat{S}_k-\hat{S}_k\hat{S}_i=i\varepsilon_{ikl}\hat{S}_l, \tag*{(383)} [ S ^ i , S ^ k ] ≡ S ^ i S ^ k − S ^ k S ^ i = i ε ik l S ^ l , (383)
{ S ^ i , S ^ k } ≡ S ^ i S ^ k + S ^ k S ^ i = 4 3 δ i k I ^ + 2 Q ^ i k . (384) \{\hat{S}_i,\hat{S}_k\}\equiv\hat{S}_i\hat{S}_k+\hat{S}_k\hat{S}_i=\frac{4}{3}\delta_{ik}\hat{I}+2\hat{Q}_{ik}. \tag*{(384)} { S ^ i , S ^ k } ≡ S ^ i S ^ k + S ^ k S ^ i = 3 4 δ ik I ^ + 2 Q ^ ik . (384)
(b) Products of cartesian components of three spin matrices
S ^ i S ^ k S ^ l = i 3 ε i k l I ^ + 1 2 ( δ i k S ^ l + δ k l S ^ i ) + i ε i l m Q ^ k m = i 3 ε i k l I ^ + 1 2 ( δ i k S ^ l + δ k l S ^ i ) + i 2 ( ε i k m Q ^ l m + ε i l m Q ^ k m + ε k l m Q ^ i m ) . (385) \begin{aligned}
\hat{S}_i\hat{S}_k\hat{S}_l
&=\frac{i}{3}\varepsilon_{ikl}\hat{I}+\frac{1}{2}\left(\delta_{ik}\hat{S}_l+\delta_{kl}\hat{S}_i\right)+i\varepsilon_{ilm}\hat{Q}_{km}\\
&=\frac{i}{3}\varepsilon_{ikl}\hat{I}+\frac{1}{2}\left(\delta_{ik}\hat{S}_l+\delta_{kl}\hat{S}_i\right)+\frac{i}{2}\left(\varepsilon_{ikm}\hat{Q}_{lm}+\varepsilon_{ilm}\hat{Q}_{km}+\varepsilon_{klm}\hat{Q}_{im}\right). \tag*{(385)}
\end{aligned} S ^ i S ^ k S ^ l = 3 i ε ik l I ^ + 2 1 ( δ ik S ^ l + δ k l S ^ i ) + i ε i l m Q ^ k m = 3 i ε ik l I ^ + 2 1 ( δ ik S ^ l + δ k l S ^ i ) + 2 i ( ε ik m Q ^ l m + ε i l m Q ^ k m + ε k l m Q ^ im ) . (385)
From Eq. (385) one may obtain the Duffin-Kemmer relation
S ^ i S ^ k S ^ l + S ^ l S ^ k S ^ i = δ i k S ^ l + δ k l S ^ i . (386) \hat{S}_i\hat{S}_k\hat{S}_l+\hat{S}_l\hat{S}_k\hat{S}_i=\delta_{ik}\hat{S}_l+\delta_{kl}\hat{S}_i. \tag*{(386)} S ^ i S ^ k S ^ l + S ^ l S ^ k S ^ i = δ ik S ^ l + δ k l S ^ i . (386)
Other particular cases of (2.6.42) are
S ^ i S ^ k S ^ i = δ i k S ^ i (no sum over i ) ; ∑ i S ^ i S ^ k S ^ i = S ^ k (387) \hat{S}_i\hat{S}_k\hat{S}_i=\delta_{ik}\hat{S}_i\quad\text{(no sum over }i\text{)};\qquad\sum_i\hat{S}_i\hat{S}_k\hat{S}_i=\hat{S}_k \tag*{(387)} S ^ i S ^ k S ^ i = δ ik S ^ i (no sum over i ) ; i ∑ S ^ i S ^ k S ^ i = S ^ k (387)
S ^ i S ^ k S ^ k + S ^ k S ^ k S ^ i = S ^ i + δ i k S ^ k (no sum over k ) , (388) \hat{S}_i\hat{S}_k\hat{S}_k+\hat{S}_k\hat{S}_k\hat{S}_i=\hat{S}_i+\delta_{ik}\hat{S}_k\quad\text{(no sum over }k\text{)}, \tag*{(388)} S ^ i S ^ k S ^ k + S ^ k S ^ k S ^ i = S ^ i + δ ik S ^ k (no sum over k ) , (388)
S ^ x S ^ y S ^ z + S ^ y S ^ z S ^ x + S ^ z S ^ x S ^ y = i I ^ , S ^ x S ^ z S ^ y + S ^ z S ^ y S ^ x + S ^ y S ^ x S ^ z = − i I ^ , (389) \begin{aligned}
\hat{S}_x\hat{S}_y\hat{S}_z+\hat{S}_y\hat{S}_z\hat{S}_x+\hat{S}_z\hat{S}_x\hat{S}_y&=i\hat{I},\\
\hat{S}_x\hat{S}_z\hat{S}_y+\hat{S}_z\hat{S}_y\hat{S}_x+\hat{S}_y\hat{S}_x\hat{S}_z&=-i\hat{I}, \tag*{(389)}
\end{aligned} S ^ x S ^ y S ^ z + S ^ y S ^ z S ^ x + S ^ z S ^ x S ^ y S ^ x S ^ z S ^ y + S ^ z S ^ y S ^ x + S ^ y S ^ x S ^ z = i I ^ , = − i I ^ , (389)
S ^ x S ^ z S ^ z = S ^ y S ^ y S ^ x = 1 2 S ^ x − i Q ^ y z , (390) \hat{S}_x\hat{S}_z\hat{S}_z=\hat{S}_y\hat{S}_y\hat{S}_x=\frac{1}{2}\hat{S}_x-i\hat{Q}_{yz}, \tag*{(390)} S ^ x S ^ z S ^ z = S ^ y S ^ y S ^ x = 2 1 S ^ x − i Q ^ y z , (390)
S ^ x S ^ y S ^ y = S ^ z S ^ z S ^ x = 1 2 S ^ x + i Q ^ y z , S ^ y S ^ x S ^ x = S ^ z S ^ z S ^ y = 1 2 S ^ y − i Q ^ x z , S ^ y S ^ z S ^ z = S ^ x S ^ x S ^ y = 1 2 S ^ y + i Q ^ x z , (391) \begin{aligned}
\hat{S}_x\hat{S}_y\hat{S}_y&=\hat{S}_z\hat{S}_z\hat{S}_x=\frac{1}{2}\hat{S}_x+i\hat{Q}_{yz},\\
\hat{S}_y\hat{S}_x\hat{S}_x&=\hat{S}_z\hat{S}_z\hat{S}_y=\frac{1}{2}\hat{S}_y-i\hat{Q}_{xz},\\
\hat{S}_y\hat{S}_z\hat{S}_z&=\hat{S}_x\hat{S}_x\hat{S}_y=\frac{1}{2}\hat{S}_y+i\hat{Q}_{xz}, \tag*{(391)}
\end{aligned} S ^ x S ^ y S ^ y S ^ y S ^ x S ^ x S ^ y S ^ z S ^ z = S ^ z S ^ z S ^ x = 2 1 S ^ x + i Q ^ y z , = S ^ z S ^ z S ^ y = 2 1 S ^ y − i Q ^ x z , = S ^ x S ^ x S ^ y = 2 1 S ^ y + i Q ^ x z , (391)
S ^ z S ^ y S ^ y = S ^ x S ^ x S ^ z = 1 2 S ^ z − i Q ^ x y , S ^ z S ^ x S ^ x = S ^ y S ^ y S ^ z = 1 2 S ^ z + i Q ^ x y . (392) \begin{aligned}
\hat{S}_z\hat{S}_y\hat{S}_y&=\hat{S}_x\hat{S}_x\hat{S}_z=\frac{1}{2}\hat{S}_z-i\hat{Q}_{xy},\\
\hat{S}_z\hat{S}_x\hat{S}_x&=\hat{S}_y\hat{S}_y\hat{S}_z=\frac{1}{2}\hat{S}_z+i\hat{Q}_{xy}. \tag*{(392)}
\end{aligned} S ^ z S ^ y S ^ y S ^ z S ^ x S ^ x = S ^ x S ^ x S ^ z = 2 1 S ^ z − i Q ^ x y , = S ^ y S ^ y S ^ z = 2 1 S ^ z + i Q ^ x y . (392)
General powers of cartesian components of spin matrices
S ^ i 2 n = 2 3 I ^ + Q ^ i i , ( n = 1 , 2 , … ) , S ^ i 2 n + 1 = S ^ i , ( n = 0 , 1 , 2 , … ) . (393) \hat{S}_i^{2n}=\frac{2}{3}\hat{I}+\hat{Q}_{ii},\quad(n=1,2,\ldots),\qquad\hat{S}_i^{2n+1}=\hat{S}_i,\quad(n=0,1,2,\ldots). \tag*{(393)} S ^ i 2 n = 3 2 I ^ + Q ^ ii , ( n = 1 , 2 , … ) , S ^ i 2 n + 1 = S ^ i , ( n = 0 , 1 , 2 , … ) . (393)
In particular, if n n n is the unit vector, then
( S ^ ⋅ n ) 2 k = 2 3 I ^ + ∑ i , l n i n l Q ^ i l , ( k = 1 , 2 , 3 , … ) , (394) (\hat{\boldsymbol{S}}\cdot\boldsymbol{n})^{2k}=\frac{2}{3}\hat{I}+\sum_{i,l}n_i n_l\hat{Q}_{il},\quad(k=1,2,3,\ldots), \tag*{(394)} ( S ^ ⋅ n ) 2 k = 3 2 I ^ + i , l ∑ n i n l Q ^ i l , ( k = 1 , 2 , 3 , … ) , (394)
( S ^ ⋅ n ) 2 k + 1 = ( S ^ ⋅ n ) , ( k = 0 , 1 , 2 , … ) . (\hat{\boldsymbol{S}}\cdot\boldsymbol{n})^{2k+1}=(\hat{\boldsymbol{S}}\cdot\boldsymbol{n}),\quad(k=0,1,2,\ldots). ( S ^ ⋅ n ) 2 k + 1 = ( S ^ ⋅ n ) , ( k = 0 , 1 , 2 , … ) .
Products of cartesian components of quadrupole tensors and spin matrices
Q ^ i k S ^ l = 1 4 ( δ i l S ^ k + δ k l S ^ i − 2 3 δ i k S ^ l ) + i 2 ( ε i l m Q ^ k m + ε k l m Q ^ i m ) , (395) \hat{Q}_{ik}\hat{S}_l=\frac{1}{4}\left(\delta_{il}\hat{S}_k+\delta_{kl}\hat{S}_i-\frac{2}{3}\delta_{ik}\hat{S}_l\right)+\frac{i}{2}\left(\varepsilon_{ilm}\hat{Q}_{km}+\varepsilon_{klm}\hat{Q}_{im}\right), \tag*{(395)} Q ^ ik S ^ l = 4 1 ( δ i l S ^ k + δ k l S ^ i − 3 2 δ ik S ^ l ) + 2 i ( ε i l m Q ^ k m + ε k l m Q ^ im ) , (395)
S ^ i Q ^ k l = 1 4 ( δ i k S ^ l + δ i l S ^ k − 2 3 δ k l S ^ i ) + i 2 ( ε i k m Q ^ l m + ε i l m Q ^ k m ) . (396) \hat{S}_i\hat{Q}_{kl}=\frac{1}{4}\left(\delta_{ik}\hat{S}_l+\delta_{il}\hat{S}_k-\frac{2}{3}\delta_{kl}\hat{S}_i\right)+\frac{i}{2}\left(\varepsilon_{ikm}\hat{Q}_{lm}+\varepsilon_{ilm}\hat{Q}_{km}\right). \tag*{(396)} S ^ i Q ^ k l = 4 1 ( δ ik S ^ l + δ i l S ^ k − 3 2 δ k l S ^ i ) + 2 i ( ε ik m Q ^ l m + ε i l m Q ^ k m ) . (396)
Products of cartesian components of quadrupole tensors
Q ^ i k Q ^ l m = 1 6 ( δ i l δ k m + δ i m δ k l − 2 3 δ i k δ l m ) I ^ − 1 4 ( δ i l Q ^ k m + δ i m Q ^ k l + δ k m Q ^ i l + δ k l Q ^ i m − 4 3 δ i k Q ^ l m − 4 3 δ l m Q ^ i k ) + i 8 ( δ i l ε k m p S ^ p + δ i m ε k l p S ^ p + δ k l ε i m p S ^ p + δ k m ε i l p S ^ p ) (397) \hat{Q}_{ik}\hat{Q}_{lm}=\frac{1}{6}\left(\delta_{il}\delta_{km}+\delta_{im}\delta_{kl}-\frac{2}{3}\delta_{ik}\delta_{lm}\right)\hat{I}-\frac{1}{4}\left(\delta_{il}\hat{Q}_{km}+\delta_{im}\hat{Q}_{kl}+\delta_{km}\hat{Q}_{il}+\delta_{kl}\hat{Q}_{im}-\frac{4}{3}\delta_{ik}\hat{Q}_{lm}-\frac{4}{3}\delta_{lm}\hat{Q}_{ik}\right)+\frac{i}{8}\left(\delta_{il}\varepsilon_{kmp}\hat{S}_p+\delta_{im}\varepsilon_{klp}\hat{S}_p+\delta_{kl}\varepsilon_{imp}\hat{S}_p+\delta_{km}\varepsilon_{ilp}\hat{S}_p\right) \tag*{(397)} Q ^ ik Q ^ l m = 6 1 ( δ i l δ k m + δ im δ k l − 3 2 δ ik δ l m ) I ^ − 4 1 ( δ i l Q ^ k m + δ im Q ^ k l + δ k m Q ^ i l + δ k l Q ^ im − 3 4 δ ik Q ^ l m − 3 4 δ l m Q ^ ik ) + 8 i ( δ i l ε k m p S ^ p + δ im ε k l p S ^ p + δ k l ε im p S ^ p + δ k m ε i l p S ^ p ) (397)
Products involving spherical components of the spin operator S ^ \hat{\boldsymbol{S}} S ^ and the polarization operators T ^ L M \hat{T}_{LM} T ^ L M may be expanded in a series as given below (μ , ν = ± 1 , 0 \mu,\nu=\pm1,0 μ , ν = ± 1 , 0 )
Products of spherical components spin matrices
S ^ μ S ^ ν = 2 3 ( − 1 ) μ δ μ − ν I ^ − 1 2 C 1 μ 1 ν 1 λ S ^ λ + C 1 μ 1 ν 2 M T ^ 2 M . (398) \hat{S}_\mu\hat{S}_\nu=\frac{2}{3}(-1)^\mu\delta_{\mu-\nu}\hat{I}-\frac{1}{\sqrt{2}}C^{1\lambda}_{1\mu\,1\nu}\hat{S}_\lambda+C^{2M}_{1\mu\,1\nu}\hat{T}_{2M}. \tag*{(398)} S ^ μ S ^ ν = 3 2 ( − 1 ) μ δ μ − ν I ^ − 2 1 C 1 μ 1 ν 1 λ S ^ λ + C 1 μ 1 ν 2 M T ^ 2 M . (398)
In particular
S ^ 2 = − S ^ + 1 S ^ − 1 + S ^ 0 S ^ 0 − S ^ − 1 S ^ + 1 = 2 I ^ (399) \hat{S}^2=-\hat{S}_{+1}\hat{S}_{-1}+\hat{S}_0\hat{S}_0-\hat{S}_{-1}\hat{S}_{+1}=2\hat{I} \tag*{(399)} S ^ 2 = − S ^ + 1 S ^ − 1 + S ^ 0 S ^ 0 − S ^ − 1 S ^ + 1 = 2 I ^ (399)
[ S ^ μ , S ^ ν ] ≡ S ^ μ S ^ ν − S ^ ν S ^ μ = − 2 C 1 μ 1 ν 1 λ S ^ λ (400) [\hat{S}_\mu,\hat{S}_\nu]\equiv\hat{S}_\mu\hat{S}_\nu-\hat{S}_\nu\hat{S}_\mu=-\sqrt{2}C^{1\lambda}_{1\mu\,1\nu}\hat{S}_\lambda \tag*{(400)} [ S ^ μ , S ^ ν ] ≡ S ^ μ S ^ ν − S ^ ν S ^ μ = − 2 C 1 μ 1 ν 1 λ S ^ λ (400)
{ S ^ μ , S ^ ν } ≡ S ^ μ S ^ ν + S ^ ν S ^ μ = 4 3 ( − 1 ) μ δ μ − ν + 2 C 1 μ 1 ν 2 M T ^ 2 M (401) \{\hat{S}_\mu,\hat{S}_\nu\}\equiv\hat{S}_\mu\hat{S}_\nu+\hat{S}_\nu\hat{S}_\mu=\frac{4}{3}(-1)^\mu\delta_{\mu-\nu}+2C^{2M}_{1\mu\,1\nu}\hat{T}_{2M} \tag*{(401)} { S ^ μ , S ^ ν } ≡ S ^ μ S ^ ν + S ^ ν S ^ μ = 3 4 ( − 1 ) μ δ μ − ν + 2 C 1 μ 1 ν 2 M T ^ 2 M (401)
(g) Products of spherical components spin matrices and polarisation operators
S ^ ± 1 2 = T ^ ± 2 2 , S ^ ± 1 3 = S ^ ± 1 4 = S ^ ± 1 5 = ⋯ = 0. (402) \hat{S}_{\pm1}^{2}=\hat{T}_{\pm2}^{2},\ \hat{S}_{\pm1}^{3}=\hat{S}_{\pm1}^{4}=\hat{S}_{\pm1}^{5}=\cdots=0. \tag*{(402)} S ^ ± 1 2 = T ^ ± 2 2 , S ^ ± 1 3 = S ^ ± 1 4 = S ^ ± 1 5 = ⋯ = 0. (402)
S ^ 0 2 = 2 3 I ^ + 2 3 T ^ 20 (403) \hat{S}_{0}^{2}=\frac{2}{3}\hat{I}+\sqrt{\frac{2}{3}}\hat{T}_{20} \tag*{(403)} S ^ 0 2 = 3 2 I ^ + 3 2 T ^ 20 (403)
S ^ 0 2 n = S ^ 0 2 , ( n = 1 , 2 , … ) , S ^ 0 2 n + 1 = S ^ 0 , ( n = 0 , 1 , 2 , … ) . (404) \hat{S}_{0}^{2n}=\hat{S}_{0}^{2},\ (n=1,2,\ldots),\qquad\hat{S}_{0}^{2n+1}=\hat{S}_{0},\ (n=0,1,2,\ldots). \tag*{(404)} S ^ 0 2 n = S ^ 0 2 , ( n = 1 , 2 , … ) , S ^ 0 2 n + 1 = S ^ 0 , ( n = 0 , 1 , 2 , … ) . (404)
(h) Products of polarisation operators
T ^ L 1 M 1 T ^ L 2 M 2 = ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) ∑ L ( − 1 ) L { L 1 L 2 L 1 1 1 } C L 1 M 1 L 2 M 2 L M T ^ L M . (405) \hat{T}_{L_{1}M_{1}}\hat{T}_{L_{2}M_{2}}=\sqrt{(2L_{1}+1)(2L_{2}+1)}\sum_{L}(-1)^{L}\left\{\begin{matrix}L_{1}&L_{2}&L\\1&1&1\end{matrix}\right\}C_{L_{1}M_{1}L_{2}M_{2}}^{LM}\hat{T}_{LM}. \tag*{(405)} T ^ L 1 M 1 T ^ L 2 M 2 = ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) L ∑ ( − 1 ) L { L 1 1 L 2 1 L 1 } C L 1 M 1 L 2 M 2 L M T ^ L M . (405)
Particularly,
T ^ L M T ^ 00 = T ^ 00 T ^ L M = 1 3 T ^ L M , ( L = 0 , 1 , 2 ; − L ≤ M ≤ L ) , (406) \hat{T}_{LM}\hat{T}_{00}=\hat{T}_{00}\hat{T}_{LM}=\frac{1}{\sqrt{3}}\hat{T}_{LM},\qquad(L=0,1,2;\ -L\leq M\leq L), \tag*{(406)} T ^ L M T ^ 00 = T ^ 00 T ^ L M = 3 1 T ^ L M , ( L = 0 , 1 , 2 ; − L ≤ M ≤ L ) , (406)
S ^ μ T ^ 2 M = − 1 2 5 3 C 1 μ 2 M 1 ν S ^ ν − 3 2 C 1 μ 2 M 2 N T ^ 2 N , (407) \hat{S}_{\mu}\hat{T}_{2M}=-\frac{1}{2}\sqrt{\frac{5}{3}}C_{1\mu2M}^{1\nu}\hat{S}_{\nu}-\sqrt{\frac{3}{2}}C_{1\mu2M}^{2N}\hat{T}_{2N}, \tag*{(407)} S ^ μ T ^ 2 M = − 2 1 3 5 C 1 μ 2 M 1 ν S ^ ν − 2 3 C 1 μ 2 M 2 N T ^ 2 N , (407)
T ^ 2 M S ^ μ = − 1 2 5 3 C 1 μ 2 M 1 ν S ^ ν + 3 2 C 1 μ 2 M 2 N T ^ 2 N , (408) \hat{T}_{2M}\hat{S}_{\mu}=-\frac{1}{2}\sqrt{\frac{5}{3}}C_{1\mu2M}^{1\nu}\hat{S}_{\nu}+\sqrt{\frac{3}{2}}C_{1\mu2M}^{2N}\hat{T}_{2N}, \tag*{(408)} T ^ 2 M S ^ μ = − 2 1 3 5 C 1 μ 2 M 1 ν S ^ ν + 2 3 C 1 μ 2 M 2 N T ^ 2 N , (408)
T ^ 2 M T ^ 2 N = ( − 1 ) M ⋅ 1 3 δ M , − N I ^ + 1 2 5 2 C 2 M 2 N 1 μ S ^ μ + 1 2 7 3 C 2 M 2 N 2 A T ^ 2 A . (409) \hat{T}_{2M}\hat{T}_{2N}=(-1)^{M}\cdot\frac{1}{3}\delta_{M,-N}\hat{I}+\frac{1}{2}\sqrt{\frac{5}{2}}C_{2M2N}^{1\mu}\hat{S}_{\mu}+\frac{1}{2}\sqrt{\frac{7}{3}}C_{2M2N}^{2A}\hat{T}_{2A}. \tag*{(409)} T ^ 2 M T ^ 2 N = ( − 1 ) M ⋅ 3 1 δ M , − N I ^ + 2 1 2 5 C 2 M 2 N 1 μ S ^ μ + 2 1 3 7 C 2 M 2 N 2 A T ^ 2 A . (409)
Functions of Spin Matrices
e α S ^ i = 1 3 ( 1 + 2 cosh α ) I ^ + sinh α S ^ i + ( cosh α − 1 ) Q ^ i i (no sum over i ) , (410) e^{\alpha\hat{S}_{i}}=\frac{1}{3}(1+2\cosh\alpha)\hat{I}+\sinh\alpha\hat{S}_{i}+(\cosh\alpha-1)\hat{Q}_{ii}\qquad\text{(no sum over }i\text{)}, \tag*{(410)} e α S ^ i = 3 1 ( 1 + 2 cosh α ) I ^ + sinh α S ^ i + ( cosh α − 1 ) Q ^ ii (no sum over i ) , (410)
cosh ( α S ^ i ) = 1 3 ( 1 + 2 cosh α ) I ^ + ( cosh α − 1 ) Q ^ i i , (411) \cosh\left(\alpha\hat{S}_{i}\right)=\frac{1}{3}(1+2\cosh\alpha)\hat{I}+(\cosh\alpha-1)\hat{Q}_{ii}, \tag*{(411)} cosh ( α S ^ i ) = 3 1 ( 1 + 2 cosh α ) I ^ + ( cosh α − 1 ) Q ^ ii , (411)
sinh ( α S ^ i ) = sinh α S ^ i , (412) \sinh\left(\alpha\hat{S}_{i}\right)=\sinh\alpha\hat{S}_{i}, \tag*{(412)} sinh ( α S ^ i ) = sinh α S ^ i , (412)
where i = x , y , z i=x,y,z i = x , y , z .
e α S ^ + 1 = I ^ + α S ^ + 1 + α 2 2 T ^ 22 , (413) e^{\alpha\hat{S}_{+1}}=\hat{I}+\alpha\hat{S}_{+1}+\frac{\alpha^{2}}{2}\hat{T}_{22}, \tag*{(413)} e α S ^ + 1 = I ^ + α S ^ + 1 + 2 α 2 T ^ 22 , (413)
e α S ^ 0 = 1 3 ( 1 + 2 cosh α ) I ^ + sinh α S ^ 0 + 2 3 ( cosh α − 1 ) T ^ 20 , (414) e^{\alpha\hat{S}_{0}}=\frac{1}{3}(1+2\cosh\alpha)\hat{I}+\sinh\alpha\hat{S}_{0}+\sqrt{\frac{2}{3}}(\cosh\alpha-1)\hat{T}_{20}, \tag*{(414)} e α S ^ 0 = 3 1 ( 1 + 2 cosh α ) I ^ + sinh α S ^ 0 + 3 2 ( cosh α − 1 ) T ^ 20 , (414)
e α S ^ − 1 = I ^ + α S ^ − 1 + α 2 2 T ^ 2 − 2 , (415) e^{\alpha\hat{S}_{-1}}=\hat{I}+\alpha\hat{S}_{-1}+\frac{\alpha^{2}}{2}\hat{T}_{2-2}, \tag*{(415)} e α S ^ − 1 = I ^ + α S ^ − 1 + 2 α 2 T ^ 2 − 2 , (415)
sinh ( α S ^ + 1 ) = α S ^ + 1 , sinh ( α S ^ 0 ) = sinh α S ^ 0 , sinh ( α S ^ − 1 ) = α S ^ − 1 . (416) \sinh\left(\alpha\hat{S}_{+1}\right)=\alpha\hat{S}_{+1},\qquad\sinh\left(\alpha\hat{S}_{0}\right)=\sinh\alpha\hat{S}_{0},\qquad\sinh\left(\alpha\hat{S}_{-1}\right)=\alpha\hat{S}_{-1}. \tag*{(416)} sinh ( α S ^ + 1 ) = α S ^ + 1 , sinh ( α S ^ 0 ) = sinh α S ^ 0 , sinh ( α S ^ − 1 ) = α S ^ − 1 . (416)
cosh ( α S ^ + 1 ) = I ^ + α 2 2 T ^ 22 , \cosh\left(\alpha\hat{S}_{+1}\right)=\hat{I}+\frac{\alpha^{2}}{2}\hat{T}_{22}, cosh ( α S ^ + 1 ) = I ^ + 2 α 2 T ^ 22 ,
cosh ( α S ^ 0 ) = 1 3 ( 1 + 2 cosh α ) I ^ + 2 3 ( cosh α − 1 ) T ^ 20 , (417) \cosh\left(\alpha\hat{S}_{0}\right)=\frac{1}{3}(1+2\cosh\alpha)\hat{I}+\sqrt{\frac{2}{3}}(\cosh\alpha-1)\hat{T}_{20}, \tag*{(417)} cosh ( α S ^ 0 ) = 3 1 ( 1 + 2 cosh α ) I ^ + 3 2 ( cosh α − 1 ) T ^ 20 , (417)
cosh ( α S ^ − 1 ) = I ^ + α 2 2 T ^ 2 − 2 . \cosh\left(\alpha\hat{S}_{-1}\right)=\hat{I}+\frac{\alpha^{2}}{2}\hat{T}_{2-2}. cosh ( α S ^ − 1 ) = I ^ + 2 α 2 T ^ 2 − 2 .
Operators of Coordinate Rotations
(a) Coordinate rotations defined be Euler angles
Under rotations of coordinate systems defined by the Euler angles α \alpha α , β \beta β , γ \gamma γ spin functions and spin matrices for S = 1 S=1 S = 1 are transformed by the rotation operator D ^ 1 ( α , β , γ ) \hat{D}^{1}(\alpha,\beta,\gamma) D ^ 1 ( α , β , γ ) which is a special case of the operator given by Eq. (160)
D ^ 1 ( α , β , γ ) = e − i α S ^ x e − i β S ^ y e − i γ S ^ z . (418) \hat{D}^{1}(\alpha,\beta,\gamma)=e^{-i\alpha\hat{S}_{x}}e^{-i\beta\hat{S}_{y}}e^{-i\gamma\hat{S}_{z}}. \tag*{(418)} D ^ 1 ( α , β , γ ) = e − i α S ^ x e − i β S ^ y e − iγ S ^ z . (418)
In the spherical basis representation D ^ 1 ( α , β , γ ) \hat{D}^{1}(\alpha,\beta,\gamma) D ^ 1 ( α , β , γ ) has the form
D ^ 1 ( α , β , γ ) = ( 1 + cos β 2 e − i ( α + γ ) − sin β 2 e − i α 1 − cos β 2 e i ( γ − α ) sin β 2 e − i γ cos β − sin β 2 e i γ 1 − cos β 2 e i ( α − γ ) sin β 2 e i α 1 + cos β 2 e i ( γ + α ) ) . (419) \hat{D}^{1}(\alpha,\beta,\gamma)=
\begin{pmatrix}
\frac{1+\cos\beta}{2}e^{-i(\alpha+\gamma)} & -\frac{\sin\beta}{\sqrt{2}}e^{-i\alpha} & \frac{1-\cos\beta}{2}e^{i(\gamma-\alpha)}\\
\frac{\sin\beta}{\sqrt{2}}e^{-i\gamma} & \cos\beta& -\frac{\sin\beta}{\sqrt{2}}e^{i\gamma}\\
\frac{1-\cos\beta}{2}e^{i(\alpha-\gamma)} & \frac{\sin\beta}{\sqrt{2}}e^{i\alpha} & \frac{1+\cos\beta}{2}e^{i(\gamma+\alpha)}
\end{pmatrix}. \tag*{(419)} D ^ 1 ( α , β , γ ) = 2 1 + c o s β e − i ( α + γ ) 2 s i n β e − iγ 2 1 − c o s β e i ( α − γ ) − 2 s i n β e − i α cos β 2 s i n β e i α 2 1 − c o s β e i ( γ − α ) − 2 s i n β e iγ 2 1 + c o s β e i ( γ + α ) . (419)
The matrix elements of D ^ 1 ( α , β , γ ) \hat{D}^{1}(\alpha,\beta,\gamma) D ^ 1 ( α , β , γ ) in this representation are the Wigner D D D -functions D M M ′ 1 ( α , β , γ ) D^{1}_{MM'}(\alpha,\beta,\gamma) D M M ′ 1 ( α , β , γ ) (see Chap. 4).
In the cartesian basis representation we have
D ^ 1 ( α , β , γ ) = ( cos β cos α cos γ − sin α sin γ − cos β cos α sin γ − sin α cos γ sin β cos α cos β sin α cos γ + cos α sin γ − cos β sin α sin γ + cos α cos γ sin β sin α − sin β cos γ sin β sin γ cos β ) . (420) \hat{D}^{1}(\alpha,\beta,\gamma)=
\begin{pmatrix}
\cos\beta\cos\alpha\cos\gamma-\sin\alpha\sin\gamma& -\cos\beta\cos\alpha\sin\gamma-\sin\alpha\cos\gamma& \sin\beta\cos\alpha\\
\cos\beta\sin\alpha\cos\gamma+\cos\alpha\sin\gamma& -\cos\beta\sin\alpha\sin\gamma+\cos\alpha\cos\gamma& \sin\beta\sin\alpha\\
-\sin\beta\cos\gamma& \sin\beta\sin\gamma& \cos\beta
\end{pmatrix}. \tag*{(420)} D ^ 1 ( α , β , γ ) = cos β cos α cos γ − sin α sin γ cos β sin α cos γ + cos α sin γ − sin β cos γ − cos β cos α sin γ − sin α cos γ − cos β sin α sin γ + cos α cos γ sin β sin γ sin β cos α sin β sin α cos β . (420)
The matrix elements of D ^ 1 ( α , β , γ ) \hat{D}^{1}(\alpha,\beta,\gamma) D ^ 1 ( α , β , γ ) in the cartesian basis representation coincide with the elements of the rotation matrix a i k a_{ik} a ik (see Sec. 1.4.6).
The inverse rotation of the coordinate system is performed by the matrix
[ D ^ 1 ( α , β , γ ) ] − 1 = [ D ^ 1 ( α , β , γ ) ] + = D ^ 1 ( − γ , − β , − α ) = D ^ 1 ( π − γ ; β , − π − α ) . (421) \left[\hat{D}^{1}(\alpha,\beta,\gamma)\right]^{-1}
=
\left[\hat{D}^{1}(\alpha,\beta,\gamma)\right]^{+}
=
\hat{D}^{1}(-\gamma,-\beta,-\alpha)
=
\hat{D}^{1}(\pi-\gamma;\beta,-\pi-\alpha). \tag*{(421)} [ D ^ 1 ( α , β , γ ) ] − 1 = [ D ^ 1 ( α , β , γ ) ] + = D ^ 1 ( − γ , − β , − α ) = D ^ 1 ( π − γ ; β , − π − α ) . (421)
Coordinate rotations defined be an axis and angle of rotation
Under rotation of the coordinate system through an angle ω \omega ω about an axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) the spin functions and spin matrices for S = 1 S=1 S = 1 are transformed by the rotation operator U ^ 1 ( ω ; Θ , Φ ) \hat{U}^{1}(\omega;\Theta,\Phi) U ^ 1 ( ω ; Θ , Φ ) which is a special case of the operator given by Eq. (1.4.33)
U ^ 1 ( ω ; Θ , Φ ) = e − i ω n ⋅ s ^ . (422) \hat{U}^{1}(\omega;\Theta,\Phi)=e^{-i\omega\mathbf{n}\cdot\hat{\mathbf{s}}}. \tag*{(422)} U ^ 1 ( ω ; Θ , Φ ) = e − iω n ⋅ s ^ . (422)
In the spherical basis representation this operator may be written as
U ^ 1 ( ω ; Θ , Φ ) = ( 1 2 cos ω ( 1 + cos 2 Θ ) − i sin ω cos Θ + 1 2 sin 2 Θ 1 2 sin Θ e i Φ [ ( cos ω − 1 ) cos Θ − i sin ω ] 1 2 ( cos ω − 1 ) sin 2 Θ e i 2 Φ 1 2 sin Θ e − i Φ [ ( cos ω − 1 ) cos Θ − i sin ω ] cos ω sin 2 Θ + cos 2 Θ − 1 2 sin Θ e i Φ [ ( cos ω − 1 ) cos Θ + i sin ω ] − 1 2 sin Θ e − i Φ [ ( cos ω − 1 ) cos Θ + i sin ω ] 1 2 cos ω ( 1 + cos 2 Θ ) + i sin ω cos Θ + 1 2 sin 2 Θ 1 2 ( cos ω − 1 ) sin 2 Θ e − i 2 Φ ) . (423) \hat{U}^{1}(\omega;\Theta,\Phi)=
\begin{pmatrix}
\frac{1}{2}\cos\omega(1+\cos^{2}\Theta)-i\sin\omega\cos\Theta+\frac{1}{2}\sin^{2}\Theta&
\frac{1}{\sqrt{2}}\sin\Theta e^{i\Phi}[(\cos\omega-1)\cos\Theta-i\sin\omega] &
\frac{1}{2}(\cos\omega-1)\sin^{2}\Theta e^{i2\Phi}\\
\frac{1}{\sqrt{2}}\sin\Theta e^{-i\Phi}[(\cos\omega-1)\cos\Theta-i\sin\omega] &
\cos\omega\sin^{2}\Theta+\cos^{2}\Theta&
-\frac{1}{\sqrt{2}}\sin\Theta e^{i\Phi}[(\cos\omega-1)\cos\Theta+i\sin\omega]\\
-\frac{1}{\sqrt{2}}\sin\Theta e^{-i\Phi}[(\cos\omega-1)\cos\Theta+i\sin\omega] &
\frac{1}{2}\cos\omega(1+\cos^{2}\Theta)+i\sin\omega\cos\Theta+\frac{1}{2}\sin^{2}\Theta&
\frac{1}{2}(\cos\omega-1)\sin^{2}\Theta e^{-i2\Phi}
\end{pmatrix}. \tag*{(423)} U ^ 1 ( ω ; Θ , Φ ) = 2 1 cos ω ( 1 + cos 2 Θ ) − i sin ω cos Θ + 2 1 sin 2 Θ 2 1 sin Θ e − i Φ [( cos ω − 1 ) cos Θ − i sin ω ] − 2 1 sin Θ e − i Φ [( cos ω − 1 ) cos Θ + i sin ω ] 2 1 sin Θ e i Φ [( cos ω − 1 ) cos Θ − i sin ω ] cos ω sin 2 Θ + cos 2 Θ 2 1 cos ω ( 1 + cos 2 Θ ) + i sin ω cos Θ + 2 1 sin 2 Θ 2 1 ( cos ω − 1 ) sin 2 Θ e i 2Φ − 2 1 sin Θ e i Φ [( cos ω − 1 ) cos Θ + i sin ω ] 2 1 ( cos ω − 1 ) sin 2 Θ e − i 2Φ . (423)
In the cartesian basis representation U ^ 1 ( ω ; Θ , Φ ) \hat{U}^{1}(\omega;\Theta,\Phi) U ^ 1 ( ω ; Θ , Φ ) has the form
U ^ 1 ( ω ; Θ , Φ ) = ( ( 1 − cos ω ) sin 2 Θ cos 2 Φ + cos ω ( 1 − cos ω ) sin 2 Θ cos Φ sin Φ + sin ω cos Θ ( 1 − cos ω ) sin Θ cos Θ cos Φ − sin ω sin Θ sin Φ ( 1 − cos ω ) sin 2 Θ cos Φ sin Φ − sin ω cos Θ ( 1 − cos ω ) sin 2 Θ sin 2 Φ + cos ω ( 1 − cos ω ) sin Θ cos Θ sin Φ − sin ω sin Θ cos Φ ( 1 − cos ω ) sin Θ cos Θ cos Φ + sin ω sin Θ sin Φ ( 1 − cos ω ) sin Θ cos Θ sin Φ + sin ω sin Θ cos Φ ( 1 − cos ω ) cos 2 Θ + cos ω ) . (424) \hat{U}^{1}(\omega;\Theta,\Phi)=
\begin{pmatrix}
(1-\cos\omega)\sin^{2}\Theta\cos^{2}\Phi+\cos\omega&
(1-\cos\omega)\sin^{2}\Theta\cos\Phi\sin\Phi+\sin\omega\cos\Theta&
(1-\cos\omega)\sin\Theta\cos\Theta\cos\Phi-\sin\omega\sin\Theta\sin\Phi\\
(1-\cos\omega)\sin^{2}\Theta\cos\Phi\sin\Phi-\sin\omega\cos\Theta&
(1-\cos\omega)\sin^{2}\Theta\sin^{2}\Phi+\cos\omega&
(1-\cos\omega)\sin\Theta\cos\Theta\sin\Phi-\sin\omega\sin\Theta\cos\Phi\\
(1-\cos\omega)\sin\Theta\cos\Theta\cos\Phi+\sin\omega\sin\Theta\sin\Phi&
(1-\cos\omega)\sin\Theta\cos\Theta\sin\Phi+\sin\omega\sin\Theta\cos\Phi&
(1-\cos\omega)\cos^{2}\Theta+\cos\omega
\end{pmatrix}. \tag*{(424)} U ^ 1 ( ω ; Θ , Φ ) = ( 1 − cos ω ) sin 2 Θ cos 2 Φ + cos ω ( 1 − cos ω ) sin 2 Θ cos Φ sin Φ − sin ω cos Θ ( 1 − cos ω ) sin Θ cos Θ cos Φ + sin ω sin Θ sin Φ ( 1 − cos ω ) sin 2 Θ cos Φ sin Φ + sin ω cos Θ ( 1 − cos ω ) sin 2 Θ sin 2 Φ + cos ω ( 1 − cos ω ) sin Θ cos Θ sin Φ + sin ω sin Θ cos Φ ( 1 − cos ω ) sin Θ cos Θ cos Φ − sin ω sin Θ sin Φ ( 1 − cos ω ) sin Θ cos Θ sin Φ − sin ω sin Θ cos Φ ( 1 − cos ω ) cos 2 Θ + cos ω . (424)
The inverse rotation of the coordinate system is performed by the matrix
[ U ^ 1 ( ω ; Θ , Φ ) ] − 1 = [ U ^ 1 ( ω ; Θ , Φ ) ] + = U ^ 1 ( − ω ; Θ , Φ ) = U ^ 1 ( ω ; π − Θ , π + Φ ) (425) \left[\hat{U}^{1}(\omega;\Theta,\Phi)\right]^{-1}
=
\left[\hat{U}^{1}(\omega;\Theta,\Phi)\right]^{+}
=
\hat{U}^{1}(-\omega;\Theta,\Phi)
=
\hat{U}^{1}(\omega;\pi-\Theta,\pi+\Phi) \tag*{(425)} [ U ^ 1 ( ω ; Θ , Φ ) ] − 1 = [ U ^ 1 ( ω ; Θ , Φ ) ] + = U ^ 1 ( − ω ; Θ , Φ ) = U ^ 1 ( ω ; π − Θ , π + Φ ) (425)
The expansion of the rotation operator U ^ 1 ( ω ; Θ , Φ ) \hat{U}^{1}(\omega;\Theta,\Phi) U ^ 1 ( ω ; Θ , Φ ) in terms of the spin matrices and the polarization operators has the form
U ^ 1 ( ω ; Θ , Φ ) = 1 3 ( 2 cos ω + 1 ) I ^ − i sin ω ( n ⋅ S ^ ) + ( cos ω − 1 ) ∑ i , k n i n k Q ^ i k . (426) \hat{U}^{1}(\omega;\Theta,\Phi)
=
\frac{1}{3}(2\cos\omega+1)\hat{I}
-i\sin\omega(\mathbf{n}\cdot\hat{\mathbf{S}})
+(\cos\omega-1)\sum_{i,k}n_i n_k\hat{Q}_{ik}. \tag*{(426)} U ^ 1 ( ω ; Θ , Φ ) = 3 1 ( 2 cos ω + 1 ) I ^ − i sin ω ( n ⋅ S ^ ) + ( cos ω − 1 ) i , k ∑ n i n k Q ^ ik . (426)
(See also Eqs. (2.4.17) and (2.4.18).)
The relations between the angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ and the Euler angles α \alpha α , β \beta β , γ \gamma γ are given by Eqs. (1.4.16), (1.4.17). Note that
D ^ 1 ( α , β , γ ) = U ^ 1 ( ω ; Θ , Φ ) . (427) \hat{D}^{1}(\alpha,\beta,\gamma)=\hat{U}^{1}(\omega;\Theta,\Phi). \tag*{(427)} D ^ 1 ( α , β , γ ) = U ^ 1 ( ω ; Θ , Φ ) . (427)
Effect of a rotation operator on a spin matrix
Applying the rotation operator D ^ 1 ( α , β , γ ) \hat{D}^{1}(\alpha,\beta,\gamma) D ^ 1 ( α , β , γ ) to the spin matrices yields (i , k , j , l = x , y , z i,k,j,l=x,y,z i , k , j , l = x , y , z )
S ^ i = D ^ 1 ( α , β , γ ) S ^ i [ D ^ 1 ( α , β , γ ) ] − 1 = ∑ k a i k S ^ k , (428) \hat{S}_{i}=\hat{D}^{1}(\alpha,\beta,\gamma)\hat{S}_{i}\left[\hat{D}^{1}(\alpha,\beta,\gamma)\right]^{-1}=\sum_{k}a_{ik}\hat{S}_{k}, \tag*{(428)} S ^ i = D ^ 1 ( α , β , γ ) S ^ i [ D ^ 1 ( α , β , γ ) ] − 1 = k ∑ a ik S ^ k , (428)
Q ^ i k = D ^ 1 ( α , β , γ ) Q ^ i k [ D ^ 1 ( α , β , γ ) ] − 1 = ∑ j , l a i j a k l Q ^ j l , (429) \hat{Q}_{ik}=\hat{D}^{1}(\alpha,\beta,\gamma)\hat{Q}_{ik}\left[\hat{D}^{1}(\alpha,\beta,\gamma)\right]^{-1}=\sum_{j,l}a_{ij}a_{kl}\hat{Q}_{jl}, \tag*{(429)} Q ^ ik = D ^ 1 ( α , β , γ ) Q ^ ik [ D ^ 1 ( α , β , γ ) ] − 1 = j , l ∑ a ij a k l Q ^ j l , (429)
where a i k a_{ik} a ik is the rotation matrix (§ 9.10.2 )
S ^ μ ≡ D ^ 1 ( α , β , γ ) S ^ μ [ D ^ 1 ( α , β , γ ) ] − 1 = ∑ ν D ν μ 1 ( α , β , γ ) S ^ ν , ( μ , ν = ± 1 , 0 ) , (430) \hat{S}_{\mu}\equiv\hat{D}^{1}(\alpha,\beta,\gamma)\hat{S}_{\mu}\left[\hat{D}^{1}(\alpha,\beta,\gamma)\right]^{-1}=\sum_{\nu}D^{1}_{\nu\mu}(\alpha,\beta,\gamma)\hat{S}_{\nu},\qquad(\mu,\nu=\pm1,0), \tag*{(430)} S ^ μ ≡ D ^ 1 ( α , β , γ ) S ^ μ [ D ^ 1 ( α , β , γ ) ] − 1 = ν ∑ D ν μ 1 ( α , β , γ ) S ^ ν , ( μ , ν = ± 1 , 0 ) , (430)
T ^ L M = D ^ 1 ( α , β , γ ) T ^ L M [ D ^ 1 ( α , β , γ ) ] − 1 = ∑ M ′ D M ′ M L ( α , β , γ ) T ^ L M ′ , ( − L ≤ M , M ′ ≤ L ) . (431) \hat{T}_{LM}=\hat{D}^{1}(\alpha,\beta,\gamma)\hat{T}_{LM}\left[\hat{D}^{1}(\alpha,\beta,\gamma)\right]^{-1}=\sum_{M'}D^{L}_{M'M}(\alpha,\beta,\gamma)\hat{T}_{LM'},\qquad(-L\leq M,M'\leq L). \tag*{(431)} T ^ L M = D ^ 1 ( α , β , γ ) T ^ L M [ D ^ 1 ( α , β , γ ) ] − 1 = M ′ ∑ D M ′ M L ( α , β , γ ) T ^ L M ′ , ( − L ≤ M , M ′ ≤ L ) . (431)
D M ′ M L ( α , β , γ ) D^{L}_{M'M}(\alpha,\beta,\gamma) D M ′ M L ( α , β , γ ) being the Wigner D D D -functions (see Chap. 4).
Traces of Products of Spin Matrices
For cartesian components of the spin matrices S ^ i \hat{S}_{i} S ^ i and Q ^ i k \hat{Q}_{ik} Q ^ ik the following relations hold (i , k , l i,k,l i , k , l , etc. take the values x , y , z x,y,z x , y , z ).
Tr { S ^ i } = 0 , Tr { Q ^ i k } = 0 , Tr { S ^ i S ^ k } = 2 δ i k , Tr { S ^ i Q ^ k l } = 0 , Tr { Q ^ i k S ^ l } = 0 , Tr { S ^ i S ^ k S ^ l } = i ε i k l , Tr { Q ^ i k Q ^ l m } = 1 2 ( δ i l δ k m + δ i m δ k l − 2 3 δ i k δ l m ) , Tr { S ^ i S ^ k S ^ l S ^ m } = δ i k δ l m + δ i m δ k l . (432) \begin{aligned}
\operatorname{Tr}\{\hat{S}_{i}\}&=0,\qquad\operatorname{Tr}\{\hat{Q}_{ik}\}=0,\\
\operatorname{Tr}\{\hat{S}_{i}\hat{S}_{k}\}&=2\delta_{ik},\qquad\operatorname{Tr}\{\hat{S}_{i}\hat{Q}_{kl}\}=0,\qquad\operatorname{Tr}\{\hat{Q}_{ik}\hat{S}_{l}\}=0,\\
\operatorname{Tr}\{\hat{S}_{i}\hat{S}_{k}\hat{S}_{l}\}&=i\varepsilon_{ikl},\\
\operatorname{Tr}\{\hat{Q}_{ik}\hat{Q}_{lm}\}&=\frac{1}{2}\left(\delta_{il}\delta_{km}+\delta_{im}\delta_{kl}-\frac{2}{3}\delta_{ik}\delta_{lm}\right),\qquad\operatorname{Tr}\{\hat{S}_{i}\hat{S}_{k}\hat{S}_{l}\hat{S}_{m}\}=\delta_{ik}\delta_{lm}+\delta_{im}\delta_{kl}. \tag*{(432)}
\end{aligned} Tr { S ^ i } Tr { S ^ i S ^ k } Tr { S ^ i S ^ k S ^ l } Tr { Q ^ ik Q ^ l m } = 0 , Tr { Q ^ ik } = 0 , = 2 δ ik , Tr { S ^ i Q ^ k l } = 0 , Tr { Q ^ ik S ^ l } = 0 , = i ε ik l , = 2 1 ( δ i l δ k m + δ im δ k l − 3 2 δ ik δ l m ) , Tr { S ^ i S ^ k S ^ l S ^ m } = δ ik δ l m + δ im δ k l . (432)
For spherical components of the spin operator S ^ μ \hat{S}_{\mu} S ^ μ we have (μ , ν , λ \mu,\nu,\lambda μ , ν , λ , etc. take the values ± 1 , 0 \pm1,0 ± 1 , 0 )
Tr { S ^ μ } = 0 , \operatorname{Tr}\{\hat{S}_{\mu}\}=0, Tr { S ^ μ } = 0 ,
Tr { S ^ μ S ^ ν } = ( − 1 ) μ 2 δ μ , − ν , \operatorname{Tr}\{\hat{S}_{\mu}\hat{S}_{\nu}\}=(-1)^{\mu}2\delta_{\mu,-\nu}, Tr { S ^ μ S ^ ν } = ( − 1 ) μ 2 δ μ , − ν ,
Tr { S ^ μ S ^ ν S ^ λ } = − 6 ( 1 1 1 μ ν λ ) = ( − 1 ) 1 + λ 2 C 1 μ 1 ν 1 , − λ , (433) \operatorname{Tr}\{\hat{S}_{\mu}\hat{S}_{\nu}\hat{S}_{\lambda}\}=-\sqrt{6}\begin{pmatrix}1&1&1\\ \mu&\nu&\lambda\end{pmatrix}=(-1)^{1+\lambda}\sqrt{2}C^{1,-\lambda}_{1\mu1\nu}, \tag*{(433)} Tr { S ^ μ S ^ ν S ^ λ } = − 6 ( 1 μ 1 ν 1 λ ) = ( − 1 ) 1 + λ 2 C 1 μ 1 ν 1 , − λ , (433)
Tr { S ^ μ S ^ ν S ^ λ S ^ ρ } = ( − 1 ) μ + λ ( δ μ , − ν δ λ , − ρ + δ μ , − ρ δ ν , − λ ) , \operatorname{Tr}\{\hat{S}_{\mu}\hat{S}_{\nu}\hat{S}_{\lambda}\hat{S}_{\rho}\}=(-1)^{\mu+\lambda}(\delta_{\mu,-\nu}\delta_{\lambda,-\rho}+\delta_{\mu,-\rho}\delta_{\nu,-\lambda}), Tr { S ^ μ S ^ ν S ^ λ S ^ ρ } = ( − 1 ) μ + λ ( δ μ , − ν δ λ , − ρ + δ μ , − ρ δ ν , − λ ) ,
Tr { S ^ μ 1 S ^ μ 2 … S ^ μ n } = 0 if μ 1 + μ 2 + … + μ n ≠ 0. (434) \operatorname{Tr}\{\hat{S}_{\mu_{1}}\hat{S}_{\mu_{2}}\ldots\hat{S}_{\mu_{n}}\}=0\quad\text{if }\mu_{1}+\mu_{2}+\ldots+\mu_{n}\ne0. \tag*{(434)} Tr { S ^ μ 1 S ^ μ 2 … S ^ μ n } = 0 if μ 1 + μ 2 + … + μ n = 0. (434)
Traces of products of the polarization operators T ^ L M \hat{T}_{LM} T ^ L M (L = 0 , 1 , 2 L=0,1,2 L = 0 , 1 , 2 ; − L ≤ M ≤ L -L\leq M\leq L − L ≤ M ≤ L ) can be evaluated by using Eqs. (291) –(294) at S = 1 S=1 S = 1 .
Irreducible tensors
Definition and properties of irreducible tensors
Definition
Irreducible tensors occupy a central position in angular momentum theory. Under rotations of coordinate systems these tensors transform in the same manner as eigenfunctions of the angular momentum operator. The use of this property permits us to develop very effective methods for calculating matrix elements of different quantum-mechanical operators.
An irreducible tensor M J \mathcal{M}_{J} M J of rank J J J (with J J J integer or half-integer) is defined as a set of 2 J + 1 2J+1 2 J + 1 functions (components) M J M \mathcal{M}_{JM} M J M (where M = − J , − J + 1 , … , J − 1 , J M=-J,-J+1,\ldots,J-1,J M = − J , − J + 1 , … , J − 1 , J ) which satisfy the following commutation rules with spherical components of the angular momentum operator
[ J ^ ± 1 , M J M ] = ± 1 2 e ± i δ J ( J + 1 ) − M ( M ± 1 ) M J , M ± 1 , \left[\hat{J}_{\pm1},\mathcal{M}_{JM}\right]=\pm\frac{1}{\sqrt{2}}e^{\pm i\delta}\sqrt{J(J+1)-M(M\pm1)}\mathcal{M}_{J,M\pm1}, [ J ^ ± 1 , M J M ] = ± 2 1 e ± i δ J ( J + 1 ) − M ( M ± 1 ) M J , M ± 1 ,
[ J ^ 0 , M J M ] = M M J M . (435) \left[\hat{J}_{0},\mathcal{M}_{JM}\right]=M\mathcal{M}_{JM}. \tag*{(435)} [ J ^ 0 , M J M ] = M M J M . (435)
In compact form
[ J ^ μ , M J M ] = e i μ δ J ( J + 1 ) C J M , 1 μ J , M + μ M J , M + μ . (436) \left[\hat{J}_{\mu},\mathcal{M}_{JM}\right]=e^{i\mu\delta}\sqrt{J(J+1)}C^{J,M+\mu}_{JM,1\mu}\mathcal{M}_{J,M+\mu}. \tag*{(436)} [ J ^ μ , M J M ] = e i μ δ J ( J + 1 ) C J M , 1 μ J , M + μ M J , M + μ . (436)
From these relations it follows that
[ J ^ 2 , M J M ] = J ( J + 1 ) M J M . (437) \left[\hat{J}^{2},\mathcal{M}_{JM}\right]=J(J+1)\mathcal{M}_{JM}. \tag*{(437)} [ J ^ 2 , M J M ] = J ( J + 1 ) M J M . (437)
The quantity δ \delta δ in Eq. (435) which determines relative phases of different M J M \mathcal{M}_{JM} M J M components is arbitrary. Let us adopt δ = 0 \delta=0 δ = 0 , i.e., e ± i δ = 1 e^{\pm i\delta}=1 e ± i δ = 1 , and choose the positive sign of the square root. The linear equations (3.1.1) define the components of the irreducible tensor M J M \mathcal{M}_{JM} M J M within an arbitrary scalar factor, which is the same for all the components. This factor can be a real or complex number, function or operator. In the case of integer rank J J J the overall phase of the M J M \mathcal{M}_{JM} M J M components is usually defined in such manner that
( M J M ) ∗ = ( − 1 ) − M M J , − M . (438) (\mathcal{M}_{JM})^{*}=(-1)^{-M}\mathcal{M}_{J,-M}. \tag*{(438)} ( M J M ) ∗ = ( − 1 ) − M M J , − M . (438)
This choice of the phase coincides with that for spherical harmonics (Chap. 5). However, sometimes in quantum-mechanical applications it is convenient to redefine irreducible tensors as
M ~ J = i J M J . (439) \widetilde{\mathcal{M}}_{J}=i^{J}\mathcal{M}_{J}. \tag*{(439)} M J = i J M J . (439)
Then one has
( M ~ J M ) ∗ = ( − 1 ) J − M M ~ J , − M . (440) (\widetilde{\mathcal{M}}_{JM})^{*}=(-1)^{J-M}\widetilde{\mathcal{M}}_{J,-M}. \tag*{(440)} ( M J M ) ∗ = ( − 1 ) J − M M J , − M . (440)
The choice of the phase (3.1.6) can be used for tensors of integer as well as half-integer rank J J J . Making use of this phase convention for tensor operators as well as for wavefunctions describing initial ∣ a ⟩ |a\rangle ∣ a ⟩ and final states ∣ b ⟩ |b\rangle ∣ b ⟩ we get the following relations for matrix elements of Hermitian M ~ J † = M ~ J \widetilde{\mathcal{M}}_{J}^{\dagger}=\widetilde{\mathcal{M}}_{J} M J † = M J operators
⟨ b ∣ M ~ J M ∣ a ⟩ = ( ⟨ a ∣ ( M ~ J M ) ∗ ∣ b ⟩ ) ∗ . (441) \langle b|\widetilde{\mathcal{M}}_{JM}|a\rangle=\left(\langle a|(\widetilde{\mathcal{M}}_{JM})^{*}|b\rangle\right)^{*}. \tag*{(441)} ⟨ b ∣ M J M ∣ a ⟩ = ( ⟨ a ∣ ( M J M ) ∗ ∣ b ⟩ ) ∗ . (441)
Covariant and Contravariant Components
Any irreducible tensor M J \mathcal{M}_{J} M J of rank J J J can be expanded in a series based on a complete set of the unit orthonormalized irreducible tensors e J M \mathbf{e}_{JM} e J M of rank J J J
e J M ⋅ e J ′ M ′ = δ J J ′ δ M M ′ . (442) \mathbf{e}_{J}^{M}\cdot\mathbf{e}_{J'M'}=\delta_{JJ'}\delta_{MM'}. \tag*{(442)} e J M ⋅ e J ′ M ′ = δ J J ′ δ M M ′ . (442)
The tensors e J M \mathbf{e}_{JM} e J M can be composed, for example, of the basis spin functions. The expansion of M J \mathcal{M}_{J} M J is written as
M J = ∑ M e J M ⋅ M J M = ∑ M e J M ⋅ M J M . (443) \mathcal{M}_{J}=\sum_{M}\mathbf{e}_{J}^{M}\cdot\mathcal{M}_{JM}=\sum_{M}\mathbf{e}_{JM}\cdot\mathcal{M}_{J}^{M}. \tag*{(443)} M J = M ∑ e J M ⋅ M J M = M ∑ e J M ⋅ M J M . (443)
M J M \mathcal{M}_{JM} M J M represents the covariant component of the tensor M J \mathcal{M}_{J} M J and M J M \mathcal{M}_{J}^{M} M J M denotes the contravariant component. These components are related by
M J M = ( M J M ) ∗ = ( − 1 ) − M M J , − M , \mathcal{M}_{J}^{M}=\left(\mathcal{M}_{JM}\right)^{*}=(-1)^{-M}\mathcal{M}_{J,-M}, M J M = ( M J M ) ∗ = ( − 1 ) − M M J , − M ,
M ~ J M = ( M ~ J M ) ∗ = ( − 1 ) J − M M ~ J , − M . (444) \widetilde{\mathcal{M}}_{J}^{M}=\left(\widetilde{\mathcal{M}}_{JM}\right)^{*}=(-1)^{J-M}\widetilde{\mathcal{M}}_{J,-M}. \tag*{(444)} M J M = ( M J M ) ∗ = ( − 1 ) J − M M J , − M . (444)
Transformation of Irreducible Tensors Under a Rotation of the Coordinate System
Under rotations of the coordinate system described by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ , the components of irreducible tensors M J M \mathcal{M}_{JM} M J M and M ~ J M \widetilde{\mathcal{M}}_{JM} M J M undergo linear transformation. The coefficients of such transformation are the Wigner D D D -functions (Chap. 4)
M J M ′ ( X ′ ) = D ^ ( α , β , γ ) M J M ′ ( X ) [ D ^ ( α , β , γ ) ] − 1 = ∑ M M J M ( X ) D M M ′ J ( α , β , γ ) , \mathcal{M}_{JM'}(X')=\hat{D}(\alpha,\beta,\gamma)\mathcal{M}_{JM'}(X)\left[\hat{D}(\alpha,\beta,\gamma)\right]^{-1}=\sum_{M}\mathcal{M}_{JM}(X)D_{MM'}^{J}(\alpha,\beta,\gamma), M J M ′ ( X ′ ) = D ^ ( α , β , γ ) M J M ′ ( X ) [ D ^ ( α , β , γ ) ] − 1 = M ∑ M J M ( X ) D M M ′ J ( α , β , γ ) ,
M ~ J M ′ ( X ′ ) = D ^ ( α , β , γ ) M ~ J M ′ ( X ) [ D ^ ( α , β , γ ) ] − 1 = ∑ M M ~ J M ( X ) D M M ′ J ( α , β , γ ) . (445) \widetilde{\mathcal{M}}_{JM'}(X')=\hat{D}(\alpha,\beta,\gamma)\widetilde{\mathcal{M}}_{JM'}(X)\left[\hat{D}(\alpha,\beta,\gamma)\right]^{-1}=\sum_{M}\widetilde{\mathcal{M}}_{JM}(X)D_{MM'}^{J}(\alpha,\beta,\gamma). \tag*{(445)} M J M ′ ( X ′ ) = D ^ ( α , β , γ ) M J M ′ ( X ) [ D ^ ( α , β , γ ) ] − 1 = M ∑ M J M ( X ) D M M ′ J ( α , β , γ ) . (445)
Here X X X and X ′ X' X ′ denote sets of all arguments of the tensor in the initial and final coordinate systems, respectively.
Transformation of Irreducible Tensors Under Inversion of the Coordinate System
The transformation properties of irreducible tensor components under inversion of the coordinate system, r → − r \mathbf{r}\to-\mathbf{r} r → − r , permit us generally to represent an irreducible tensor M J \mathcal{M}_{J} M J of integer rank J J J as the sum of two tensors M J ( + 1 ) \mathcal{M}_{J}^{(+1)} M J ( + 1 ) and M J ( − 1 ) \mathcal{M}_{J}^{(-1)} M J ( − 1 ) , i.e.,
M J = M J ( + 1 ) + M J ( − 1 ) . (446) \mathcal{M}_{J}=\mathcal{M}_{J}^{(+1)}+\mathcal{M}_{J}^{(-1)}. \tag*{(446)} M J = M J ( + 1 ) + M J ( − 1 ) . (446)
Each of these tensors has definite parity.
Under inversion of the coordinate system the tensors M J ( ± 1 ) \mathcal{M}_{J}^{(\pm1)} M J ( ± 1 ) transform in the following way
P ^ r M J ( π J ) P ^ r − 1 = π J M J ( π J ) , ( π J = ± 1 ) . (447) \hat{P}_{\mathbf{r}}\mathcal{M}_{J}^{(\pi_{J})}\hat{P}_{\mathbf{r}}^{-1}=\pi_{J}\mathcal{M}_{J}^{(\pi_{J})},\qquad(\pi_{J}=\pm1). \tag*{(447)} P ^ r M J ( π J ) P ^ r − 1 = π J M J ( π J ) , ( π J = ± 1 ) . (447)
The tensor M J ( π J ) \mathcal{M}_{J}^{(\pi_{J})} M J ( π J ) with parity π J = ( − 1 ) J \pi_{J}=(-1)^{J} π J = ( − 1 ) J is called the true (or polar) irreducible tensor of rank J J J . Tensor M J ( π J ) \mathcal{M}_{J}^{(\pi_{J})} M J ( π J ) with parity π J = ( − 1 ) J + 1 \pi_{J}=(-1)^{J+1} π J = ( − 1 ) J + 1 is the pseudotensor (or axial tensor) of rank J J J . The relations (3.1.12) and (3.1.13) are valid for the tensors M ~ J \widetilde{\mathcal{M}}_{J} M J as well.
Double Tensors
A double tensor W J 1 J 2 ( 1 , 2 ) W_{J_{1}J_{2}}(1,2) W J 1 J 2 ( 1 , 2 ) of ranks J 1 J_{1} J 1 and J 2 J_{2} J 2 has ( 2 J 1 + 1 ) ( 2 J 2 + 1 ) (2J_{1}+1)(2J_{2}+1) ( 2 J 1 + 1 ) ( 2 J 2 + 1 ) components and depends on variables of two different subsystems, 1 and 2. The components of the double tensor satisfy the following commutation relations
[ J ^ ± 1 ( 1 ) , W J 1 M 1 J 2 M 2 ( 1 , 2 ) ] = ∓ 1 2 J 1 ( J 1 + 1 ) − M 1 ( M 1 ± 1 ) W J 1 M 1 ± 1 J 2 M 2 ( 1 , 2 ) , (448) \left[\hat{J}_{\pm1}(1),W_{J_{1}M_{1}J_{2}M_{2}}(1,2)\right]=\mp\frac{1}{\sqrt{2}}\sqrt{J_{1}(J_{1}+1)-M_{1}(M_{1}\pm1)}W_{J_{1}M_{1}\pm1J_{2}M_{2}}(1,2), \tag*{(448)} [ J ^ ± 1 ( 1 ) , W J 1 M 1 J 2 M 2 ( 1 , 2 ) ] = ∓ 2 1 J 1 ( J 1 + 1 ) − M 1 ( M 1 ± 1 ) W J 1 M 1 ± 1 J 2 M 2 ( 1 , 2 ) , (448)
[ J ^ 0 ( 1 ) , W J 1 M 1 J 2 M 2 ( 1 , 2 ) ] = M 1 W J 1 M 1 J 2 M 2 ( 1 , 2 ) , (449) \left[\hat{J}_{0}(1),W_{J_{1}M_{1}J_{2}M_{2}}(1,2)\right]=M_{1}W_{J_{1}M_{1}J_{2}M_{2}}(1,2), \tag*{(449)} [ J ^ 0 ( 1 ) , W J 1 M 1 J 2 M 2 ( 1 , 2 ) ] = M 1 W J 1 M 1 J 2 M 2 ( 1 , 2 ) , (449)
[ J ^ ± 1 ( 2 ) , W J 1 M 1 J 2 M 2 ( 1 , 2 ) ] = ∓ 1 2 J 2 ( J 2 + 1 ) − M 2 ( M 2 ± 1 ) W J 1 M 1 J 2 M 2 ± 1 ( 1 , 2 ) , (450) \left[\hat{J}_{\pm1}(2),W_{J_{1}M_{1}J_{2}M_{2}}(1,2)\right]=\mp\frac{1}{\sqrt{2}}\sqrt{J_{2}(J_{2}+1)-M_{2}(M_{2}\pm1)}W_{J_{1}M_{1}J_{2}M_{2}\pm1}(1,2), \tag*{(450)} [ J ^ ± 1 ( 2 ) , W J 1 M 1 J 2 M 2 ( 1 , 2 ) ] = ∓ 2 1 J 2 ( J 2 + 1 ) − M 2 ( M 2 ± 1 ) W J 1 M 1 J 2 M 2 ± 1 ( 1 , 2 ) , (450)
[ J ^ 0 ( 2 ) , W J 1 M 1 J 2 M 2 ( 1 , 2 ) ] = M 2 W J 1 M 1 J 2 M 2 ( 1 , 2 ) . (451) \left[\hat{J}_{0}(2),W_{J_{1}M_{1}J_{2}M_{2}}(1,2)\right]=M_{2}W_{J_{1}M_{1}J_{2}M_{2}}(1,2). \tag*{(451)} [ J ^ 0 ( 2 ) , W J 1 M 1 J 2 M 2 ( 1 , 2 ) ] = M 2 W J 1 M 1 J 2 M 2 ( 1 , 2 ) . (451)
Here J ^ ( 1 ) \hat{J}(1) J ^ ( 1 ) and J ^ ( 2 ) \hat{J}(2) J ^ ( 2 ) are the operators of the total angular momenta of subsystems 1 and 2, respectively. Under rotation of subsystem 1 and 2 the double tensor W J 1 J 2 ( 1 , 2 ) W_{J_1J_2}(1,2) W J 1 J 2 ( 1 , 2 ) transforms according to the representation D J 1 D^{J_1} D J 1 or D J 2 D^{J_2} D J 2 , respectively.
W J 1 M 1 ′ J 2 M 2 ( 1 ′ , 2 ) = ∑ M 1 W J 1 M 1 J 2 M 2 ( 1 , 2 ) D M 1 M 1 ′ J 1 ( α 1 , β 1 , γ 1 ) . (452) W_{J_1M'_1J_2M_2}(1',2)=\sum_{M_1}W_{J_1M_1J_2M_2}(1,2)D^{J_1}_{M_1M'_1}(\alpha_1,\beta_1,\gamma_1). \tag*{(452)} W J 1 M 1 ′ J 2 M 2 ( 1 ′ , 2 ) = M 1 ∑ W J 1 M 1 J 2 M 2 ( 1 , 2 ) D M 1 M 1 ′ J 1 ( α 1 , β 1 , γ 1 ) . (452)
W J 1 M 1 J 2 M 2 ′ ( 1 , 2 ′ ) = ∑ M 2 W J 1 M 1 J 2 M 2 ( 1 , 2 ) D M 2 M 2 ′ J 2 ( α 2 , β 2 , γ 2 ) . (453) W_{J_1M_1J_2M'_2}(1,2')=\sum_{M_2}W_{J_1M_1J_2M_2}(1,2)D^{J_2}_{M_2M'_2}(\alpha_2,\beta_2,\gamma_2). \tag*{(453)} W J 1 M 1 J 2 M 2 ′ ( 1 , 2 ′ ) = M 2 ∑ W J 1 M 1 J 2 M 2 ( 1 , 2 ) D M 2 M 2 ′ J 2 ( α 2 , β 2 , γ 2 ) . (453)
Examples of Irreducible Tensors
In this section we present some examples of irreducible tensors considered in this book.
(a) The operators of the angular momenta J ^ , L ^ , S ^ \hat{\mathbf{J}},\hat{\mathbf{L}},\hat{\mathbf{S}} J ^ , L ^ , S ^ (Chap. 2) are irreducible tensor operators of rank 1.
(b) The polarization operator T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) (Sec. 2.4) is an irreducible tensor operator of rank L L L .
(c) Spherical harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) (Chap. 5) are irreducible tensors of rank l l l .
(d) The spin wave functions of a particle of spin S S S (Chap. 6) are irreducible tensors of rank S S S .
(e) Tensor spherical harmonics Y J M L S ( ϑ , φ ) Y^{{L S}}_{JM}(\vartheta,\varphi) Y J M L S ( ϑ , φ ) (Chap. 7) are irreducible tensors of rank J J J .
We may apply the results given in this chapter for all the tensors mentioned above.
Direct and Irreducible Tensor Products. Commutators of Tensor Products
An irreducible tensor product L J \mathfrak{L}_J L J of two irreducible tensors M J 1 \mathfrak{M}_{J_1} M J 1 and N J 2 \mathfrak{N}_{J_2} N J 2 of ranks J 1 J_1 J 1 and J 2 J_2 J 2 is defined as the tensor of rank J J J whose components L J M \mathfrak{L}_{JM} L J M can be expressed in terms of M J 1 M 1 \mathfrak{M}_{J_1M_1} M J 1 M 1 and N J 2 M 2 \mathfrak{N}_{J_2M_2} N J 2 M 2 according to
L J M = ∑ M 1 M 2 C J 1 M 1 J 2 M 2 J M M J 1 M 1 N J 2 M 2 (454) \mathfrak{L}_{JM}=\sum_{M_1M_2}C^{JM}_{J_1M_1J_2M_2}\mathfrak{M}_{J_1M_1}\mathfrak{N}_{J_2M_2} \tag*{(454)} L J M = M 1 M 2 ∑ C J 1 M 1 J 2 M 2 J M M J 1 M 1 N J 2 M 2 (454)
Irreducible tensor product is denoted as
L J ≡ { M J 1 ⊗ N J 2 } J . (455) \mathfrak{L}_J\equiv\{\mathfrak{M}_{J_1}\otimes\mathfrak{N}_{J_2}\}_J. \tag*{(455)} L J ≡ { M J 1 ⊗ N J 2 } J . (455)
The direct product of two irreducible tensors M J 1 \mathfrak{M}_{J_1} M J 1 and N J 2 \mathfrak{N}_{J_2} N J 2 is defined as a set of ( 2 J 1 + 1 ) ( 2 J 2 + 1 ) (2J_1+1)(2J_2+1) ( 2 J 1 + 1 ) ( 2 J 2 + 1 ) components M J 1 M 1 N J 2 M 2 \mathfrak{M}_{J_1M_1}\mathfrak{N}_{J_2M_2} M J 1 M 1 N J 2 M 2 . This tensor is generally reducible and can be decomposed into parts which themselves transform independently under a rotation of the coordinate system. In other words, the direct product can be represented as a sum of irreducible tensors L J M \mathfrak{L}_{JM} L J M
M J 1 M 1 N J 2 M 2 = ∑ J = ∣ J 1 − J 2 ∣ J 1 + J 2 C J 1 M 1 J 2 M 2 J M L J M . (456) \mathfrak{M}_{J_1M_1}\mathfrak{N}_{J_2M_2}=\sum_{J=\lvert J_1-J_2\rvert}^{J_1+J_2}C^{JM}_{J_1M_1J_2M_2}\mathfrak{L}_{JM}. \tag*{(456)} M J 1 M 1 N J 2 M 2 = J = ∣ J 1 − J 2 ∣ ∑ J 1 + J 2 C J 1 M 1 J 2 M 2 J M L J M . (456)
It is essential that the irreducible tensor product L ~ J M \widetilde{\mathfrak{L}}_{JM} L J M satisfies the same equality (3.1.6) as M ~ J 1 M 1 \widetilde{\mathfrak{M}}_{J_1M_1} M J 1 M 1 and N ~ J 2 M 2 \widetilde{\mathfrak{N}}_{J_2M_2} N J 2 M 2
( L ~ J M ) ∗ = ( − 1 ) J − M L ~ J , − M ( L ~ J M = { M ~ J 1 ⊗ N ~ J 2 } J M ) (457) \left(\widetilde{\mathfrak{L}}_{JM}\right)^*=(-1)^{J-M}\widetilde{\mathfrak{L}}_{J,-M}\qquad\left(\widetilde{\mathfrak{L}}_{JM}=\{\widetilde{\mathfrak{M}}_{J_1}\otimes\widetilde{\mathfrak{N}}_{J_2}\}_{JM}\right) \tag*{(457)} ( L J M ) ∗ = ( − 1 ) J − M L J , − M ( L J M = { M J 1 ⊗ N J 2 } J M ) (457)
This property is specific for tensors of the M ~ J M \widetilde{\mathfrak{M}}_{JM} M J M -type. The tensor product L J M \mathfrak{L}_{JM} L J M does not satisfy the relation (3.1.4) although M J 1 M 1 \mathfrak{M}_{J_1M_1} M J 1 M 1 and N J 2 M 2 \mathfrak{N}_{J_2M_2} N J 2 M 2 satisfy this relation.
Let us introduce some definitions and notations which will be widely used in the consideration of products of non-commuting tensor operators.
The commutator of components of two irreducible tensors is defined by
R J 1 M 1 J 2 M 2 ≡ [ M J 1 M 1 , N J 2 M 2 ] ≡ M J 1 M 1 N J 2 M 2 − N J 2 M 2 M J 1 M 1 . (458) \mathfrak{R}_{J_1M_1J_2M_2}\equiv[\mathfrak{M}_{J_1M_1},\mathfrak{N}_{J_2M_2}]\equiv\mathfrak{M}_{J_1M_1}\mathfrak{N}_{J_2M_2}-\mathfrak{N}_{J_2M_2}\mathfrak{M}_{J_1M_1}. \tag*{(458)} R J 1 M 1 J 2 M 2 ≡ [ M J 1 M 1 , N J 2 M 2 ] ≡ M J 1 M 1 N J 2 M 2 − N J 2 M 2 M J 1 M 1 . (458)
The commutator of an irreducible tensor product is written as
R J M J 1 J 2 ≡ { M J 1 ⊗ N J 2 } J M − ( − 1 ) J 1 + J 2 − J { N J 2 ⊗ M J 1 } J M . (459) \mathfrak{R}^{J_1J_2}_{JM}\equiv\{\mathfrak{M}_{J_1}\otimes\mathfrak{N}_{J_2}\}_{JM}-(-1)^{J_1+J_2-J}\{\mathfrak{N}_{J_2}\otimes\mathfrak{M}_{J_1}\}_{JM}. \tag*{(459)} R J M J 1 J 2 ≡ { M J 1 ⊗ N J 2 } J M − ( − 1 ) J 1 + J 2 − J { N J 2 ⊗ M J 1 } J M . (459)
The functions R J M J 1 J 2 \mathfrak{R}^{J_1J_2}_{JM} R J M J 1 J 2 are the components of some irreducible tensor. They may be expressed in the form
R J M J 1 J 2 = ∑ M 1 M 2 C J 1 M 1 J 2 M 2 J M R J 1 M 1 J 2 M 2 (460) \mathfrak{R}^{J_1J_2}_{JM}=\sum_{M_1M_2}C^{JM}_{J_1M_1J_2M_2}\mathfrak{R}_{J_1M_1J_2M_2} \tag*{(460)} R J M J 1 J 2 = M 1 M 2 ∑ C J 1 M 1 J 2 M 2 J M R J 1 M 1 J 2 M 2 (460)
For commuting tensors we get
{ M J 1 ⊗ N J 2 } J M = ( − 1 ) J 1 + J 2 − J { N J 2 ⊗ M J 1 } J M . (461) \{\mathcal{M}_{J_1} \otimes\mathcal{N}_{J_2}\}_{JM} = (-1)^{J_1+J_2-J} \{\mathcal{N}_{J_2} \otimes\mathcal{M}_{J_1}\}_{JM}. \tag*{(461)} { M J 1 ⊗ N J 2 } J M = ( − 1 ) J 1 + J 2 − J { N J 2 ⊗ M J 1 } J M . (461)
On the other hand, for non-commuting tensors we have
{ M J 1 ⊗ N J 2 } J M = ( − 1 ) J 1 + J 2 − J { N J 2 ⊗ M J 1 } J M + R J M J 1 J 2 . (462) \{\mathcal{M}_{J_1} \otimes\mathcal{N}_{J_2}\}_{JM} = (-1)^{J_1+J_2-J} \{\mathcal{N}_{J_2} \otimes\mathcal{M}_{J_1}\}_{JM} + \mathcal{R}^{J_1J_2}_{JM}. \tag*{(462)} { M J 1 ⊗ N J 2 } J M = ( − 1 ) J 1 + J 2 − J { N J 2 ⊗ M J 1 } J M + R J M J 1 J 2 . (462)
In particular, from these equations one can see that an irreducible (rank-1) tensor product of two identical tensors M J \mathcal{M}_J M J is equal to zero, if I = 2 J − 1 , 2 J − 3 , … I = 2J - 1, 2J - 3, \ldots I = 2 J − 1 , 2 J − 3 , … and the tensor components commute
{ M J ⊗ M J } I = 0 (463) \{\mathcal{M}_J \otimes\mathcal{M}_J\}_I = 0 \tag*{(463)} { M J ⊗ M J } I = 0 (463)
Scalar Products of Irreducible Tensors
The scalar product of two irreducible tensors M J \mathcal{M}_J M J and N J \mathcal{N}_J N J of the same rank is defined as
( M J ⋅ N J ) = ∑ M ( − 1 ) − M M J M N J , − M = ∑ M M J M N J M ∗ = ∑ M M J M N J M . (464) (\mathcal{M}_J \cdot\mathcal{N}_J) = \sum_M (-1)^{-M}\mathcal{M}_{JM}\mathcal{N}_{J,-M} = \sum_M \mathcal{M}_{JM}\mathcal{N}_{JM}^{*} = \sum_M \mathcal{M}_{JM}\mathcal{N}_{J}^{M}. \tag*{(464)} ( M J ⋅ N J ) = M ∑ ( − 1 ) − M M J M N J , − M = M ∑ M J M N J M ∗ = M ∑ M J M N J M . (464)
Similarly, the scalar product of two irreducible tensors M ~ J \widetilde{\mathcal{M}}_J M J and N ~ J \widetilde{\mathcal{N}}_J N J is given by
( M ~ J ⋅ N ~ J ) = ∑ M ( − 1 ) J − M M ~ J M N ~ J , − M = ∑ M M ~ J M N ~ J M ∗ = ∑ M M ~ J M N ~ J M . (\widetilde{\mathcal{M}}_J \cdot\widetilde{\mathcal{N}}_J) = \sum_M (-1)^{J-M}\widetilde{\mathcal{M}}_{JM}\widetilde{\mathcal{N}}_{J,-M} = \sum_M \widetilde{\mathcal{M}}_{JM}\widetilde{\mathcal{N}}_{JM}^{*} = \sum_M \widetilde{\mathcal{M}}_{JM}\widetilde{\mathcal{N}}_{J}^{M}. ( M J ⋅ N J ) = M ∑ ( − 1 ) J − M M J M N J , − M = M ∑ M J M N J M ∗ = M ∑ M J M N J M .
Note that
( M J ⋅ N J ) = ( M ~ J ⋅ N ~ J ) . (465) (\mathcal{M}_J \cdot\mathcal{N}_J) = (\widetilde{\mathcal{M}}_J \cdot\widetilde{\mathcal{N}}_J). \tag*{(465)} ( M J ⋅ N J ) = ( M J ⋅ N J ) . (465)
Scalar products differ from irreducible tensor products of zero rank only by some numerical factor. The latter tensor products read
{ M J ⊗ N J } 00 = ∑ M 1 M 2 C J M 1 J M 2 00 M J M 1 N J M 2 = 1 2 J + 1 ∑ M ( − 1 ) J − M M J M N J , − M , (466) \{\mathcal{M}_J \otimes\mathcal{N}_J\}_{00} = \sum_{M_1M_2} C^{00}_{J M_1 J M_2}\mathcal{M}_{JM_1}\mathcal{N}_{JM_2} = \frac{1}{\sqrt{2J+1}}\sum_M (-1)^{J-M}\mathcal{M}_{JM}\mathcal{N}_{J,-M}, \tag*{(466)} { M J ⊗ N J } 00 = M 1 M 2 ∑ C J M 1 J M 2 00 M J M 1 N J M 2 = 2 J + 1 1 M ∑ ( − 1 ) J − M M J M N J , − M , (466)
{ M ~ J ⊗ N ~ J } 00 = ∑ M 1 M 2 C J M 1 J M 2 00 M ~ J M 1 N ~ J M 2 = 1 2 J + 1 ∑ M ( − 1 ) J − M M ~ J M N ~ J , − M . (467) \{\widetilde{\mathcal{M}}_J \otimes\widetilde{\mathcal{N}}_J\}_{00} = \sum_{M_1M_2} C^{00}_{J M_1 J M_2}\widetilde{\mathcal{M}}_{JM_1}\widetilde{\mathcal{N}}_{JM_2} = \frac{1}{\sqrt{2J+1}}\sum_M (-1)^{J-M}\widetilde{\mathcal{M}}_{JM}\widetilde{\mathcal{N}}_{J,-M}. \tag*{(467)} { M J ⊗ N J } 00 = M 1 M 2 ∑ C J M 1 J M 2 00 M J M 1 N J M 2 = 2 J + 1 1 M ∑ ( − 1 ) J − M M J M N J , − M . (467)
Hence, the relations between scalar products and tensor products of zero rank are determined by
( M J ⋅ N J ) = ( − 1 ) − J 2 J + 1 { M J ⊗ N J } 00 , (\mathcal{M}_J \cdot\mathcal{N}_J) = (-1)^{-J}\sqrt{2J+1}\{\mathcal{M}_J \otimes\mathcal{N}_J\}_{00}, ( M J ⋅ N J ) = ( − 1 ) − J 2 J + 1 { M J ⊗ N J } 00 ,
( M ~ J ⋅ N ~ J ) = 2 J + 1 { M ~ J ⊗ N ~ J } 00 . (468) (\widetilde{\mathcal{M}}_J \cdot\widetilde{\mathcal{N}}_J) = \sqrt{2J+1}\{\widetilde{\mathcal{M}}_J \otimes\widetilde{\mathcal{N}}_J\}_{00}. \tag*{(468)} ( M J ⋅ N J ) = 2 J + 1 { M J ⊗ N J } 00 . (468)
For the scalar product of double tensors we have
( W ~ J 1 J 2 ( 1 , 2 ) ⋅ U ~ J 1 J 2 ( 1 , 2 ) ) = ∑ M 1 M 2 ( − 1 ) J 1 − M 1 + J 2 − M 2 W ~ J 1 M 1 J 2 M 2 ( 1 , 2 ) U ~ J 1 , − M 1 J 2 , − M 2 ( 1 , 2 ) . (469) \left(\widetilde{W}_{J_1J_2}(1,2) \cdot\widetilde{U}_{J_1J_2}(1,2)\right) = \sum_{M_1M_2}(-1)^{J_1-M_1+J_2-M_2}\widetilde{W}_{J_1M_1J_2M_2}(1,2)\widetilde{U}_{J_1,-M_1J_2,-M_2}(1,2). \tag*{(469)} ( W J 1 J 2 ( 1 , 2 ) ⋅ U J 1 J 2 ( 1 , 2 ) ) = M 1 M 2 ∑ ( − 1 ) J 1 − M 1 + J 2 − M 2 W J 1 M 1 J 2 M 2 ( 1 , 2 ) U J 1 , − M 1 J 2 , − M 2 ( 1 , 2 ) . (469)
Relation between the irreducible tensor algebra and vector and tensor theory
Vectors and Irreducible Tensors
An arbitrary vector A \mathbf{A} A is an irreducible tensor of rank one. Its spherical components may be treated as components of an irreducible tensor A 1 A_1 A 1 , of rank one, for which A 1 μ = ( − 1 ) μ A 1 , − μ A_1^\mu= (-1)^\mu A_{1,-\mu} A 1 μ = ( − 1 ) μ A 1 , − μ .
A 1 μ ≡ A μ , A 1 μ ≡ A μ . (470) A_{1\mu} \equiv A_\mu,\qquad A^{1\mu} \equiv A^\mu. \tag*{(470)} A 1 μ ≡ A μ , A 1 μ ≡ A μ . (470)
A polar vector is a true tensor of rank one. Under inversion of the coordinate system its components change their sign. An axial vector is a pseudotensor of rank one. Under coordinate inversion its components remain unchanged.
Using two vectors A = A 1 \mathbf{A} = \mathbf{A}_1 A = A 1 and B = B 1 \mathbf{B} = \mathbf{B}_1 B = B 1 one can construct three irreducible tensor products of ranks 0, 1, 2,
{ A 1 ⊗ B 1 } 00 , { A 1 ⊗ B 1 } 1 μ , { A 1 ⊗ B 1 } 2 μ . \{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{00},\qquad\{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{1\mu},\qquad\{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{2\mu}. { A 1 ⊗ B 1 } 00 , { A 1 ⊗ B 1 } 1 μ , { A 1 ⊗ B 1 } 2 μ .
If one of the vectors A \mathbf{A} A or B \mathbf{B} B is polar and the other axial, then { A 1 ⊗ B 1 } 00 \{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{00} { A 1 ⊗ B 1 } 00 is a pseudoscalar, { A 1 ⊗ B 1 } 1 μ \{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{1\mu} { A 1 ⊗ B 1 } 1 μ is a polar vector, { A 1 ⊗ B 1 } 2 μ \{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{2\mu} { A 1 ⊗ B 1 } 2 μ is a pseudotensor of rank two. If both vectors A \mathbf{A} A and B \mathbf{B} B are either polar or axial, then { A 1 ⊗ B 1 } 00 \{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{00} { A 1 ⊗ B 1 } 00 is a scalar, { A 1 ⊗ B 1 } 1 μ \{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{1\mu} { A 1 ⊗ B 1 } 1 μ is an axial vector and { A 1 ⊗ B 1 } 2 μ \{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{2\mu} { A 1 ⊗ B 1 } 2 μ is a true tensor of rank two.
Rank zero tensor and scalar product
The irreducible tensor product of rank zero differs from the scalar product of vectors A \mathbf{A} A and B \mathbf{B} B by a numerical factor:
{ A 1 ⊗ B 1 } 00 = − 1 3 ( A ⋅ B ) (471) \{\mathbf{A}_1 \otimes\mathbf{B}_1\}_{00}=-\frac{1}{\sqrt{3}}(\mathbf{A}\cdot\mathbf{B}) \tag*{(471)} { A 1 ⊗ B 1 } 00 = − 3 1 ( A ⋅ B ) (471)
The scalar product of irreducible tensors A 1 \mathbf{A}_1 A 1 and B 1 \mathbf{B}_1 B 1 introduced in § 12.1.8 coincides with the scalar product of vectors,
( A 1 ⋅ B 1 ) = ( A ⋅ B ) . (472) (\mathbf{A}_1\cdot\mathbf{B}_1)=(\mathbf{A}\cdot\mathbf{B}). \tag*{(472)} ( A 1 ⋅ B 1 ) = ( A ⋅ B ) . (472)
Rank one tensor and vector product
The irreducible tensor product of rank one is related to the vector product of vectors A \mathbf{A} A and B \mathbf{B} B by
{ A 1 ⊗ B 1 } 1 = i 2 [ A × B ] . (473) \{\mathbf{A}_1\otimes\mathbf{B}_1\}_1=\frac{i}{\sqrt{2}}[\mathbf{A}\times\mathbf{B}]. \tag*{(473)} { A 1 ⊗ B 1 } 1 = 2 i [ A × B ] . (473)
The components of the tensor product of rank one can be expressed in terms of products of spherical components of vectors A \mathbf{A} A and B \mathbf{B} B as
{ A 1 ⊗ B 1 } 1 M = i 2 [ A × B ] M = ∑ μ , ν C 1 μ 1 ν 1 M A μ B ν . (474) \{\mathbf{A}_1\otimes\mathbf{B}_1\}_{1M}=\frac{i}{\sqrt{2}}[\mathbf{A}\times\mathbf{B}]_M=\sum_{\mu,\nu}C^{1M}_{1\mu1\nu}A_\mu B_\nu. \tag*{(474)} { A 1 ⊗ B 1 } 1 M = 2 i [ A × B ] M = μ , ν ∑ C 1 μ 1 ν 1 M A μ B ν . (474)
Rank two tensor in terms of vector components
The components of a tensor product of rank two are also related to products of spherical components of A \mathbf{A} A and B \mathbf{B} B :
{ A 1 ⊗ B 1 } 2 M = ∑ μ , ν C 1 μ 1 ν 2 M A μ B ν = 3 ∣ M ∣ − 2 14 ∣ M ∣ − 12 ∑ μ + ν = M μ ≥ ν ( A μ B ν + A ν B μ ) . (475) \{\mathbf{A}_1\otimes\mathbf{B}_1\}_{2M}=\sum_{\mu,\nu}C^{2M}_{1\mu1\nu}A_\mu B_\nu=\sqrt{\frac{3|M|-2}{14|M|-12}}\sum_{\substack{\mu+\nu=M\\\mu\geq\nu}}(A_\mu B_\nu+A_\nu B_\mu). \tag*{(475)} { A 1 ⊗ B 1 } 2 M = μ , ν ∑ C 1 μ 1 ν 2 M A μ B ν = 14∣ M ∣ − 12 3∣ M ∣ − 2 μ + ν = M μ ≥ ν ∑ ( A μ B ν + A ν B μ ) . (475)
In a more detailed form Eq. (475) may be written as
{ A 1 ⊗ B 1 } 2 + 2 = A + 1 B + 1 , { A 1 ⊗ B 1 } 2 + 1 = 1 2 ( A + 1 B 0 + A 0 B + 1 ) , { A 1 ⊗ B 1 } 20 = 1 6 ( A + 1 B − 1 + 2 A 0 B 0 + A − 1 B + 1 ) , { A 1 ⊗ B 1 } 2 − 1 = 1 2 ( A − 1 B 0 + A 0 B − 1 ) , { A 1 ⊗ B 1 } 2 − 2 = A − 1 B − 1 . (476) \begin{aligned}
\{\mathbf{A}_1\otimes\mathbf{B}_1\}_{2+2}&=A_{+1}B_{+1},\\
\{\mathbf{A}_1\otimes\mathbf{B}_1\}_{2+1}&=\frac{1}{\sqrt{2}}(A_{+1}B_0+A_0B_{+1}),\\
\{\mathbf{A}_1\otimes\mathbf{B}_1\}_{20}&=\frac{1}{\sqrt{6}}(A_{+1}B_{-1}+2A_0B_0+A_{-1}B_{+1}),\\
\{\mathbf{A}_1\otimes\mathbf{B}_1\}_{2-1}&=\frac{1}{\sqrt{2}}(A_{-1}B_0+A_0B_{-1}),\\
\{\mathbf{A}_1\otimes\mathbf{B}_1\}_{2-2}&=A_{-1}B_{-1}. \tag*{(476)}
\end{aligned} { A 1 ⊗ B 1 } 2 + 2 { A 1 ⊗ B 1 } 2 + 1 { A 1 ⊗ B 1 } 20 { A 1 ⊗ B 1 } 2 − 1 { A 1 ⊗ B 1 } 2 − 2 = A + 1 B + 1 , = 2 1 ( A + 1 B 0 + A 0 B + 1 ) , = 6 1 ( A + 1 B − 1 + 2 A 0 B 0 + A − 1 B + 1 ) , = 2 1 ( A − 1 B 0 + A 0 B − 1 ) , = A − 1 B − 1 . (476)
Given three commuting vectors A 1 , B 1 , C 1 \mathbf{A}_1,\mathbf{B}_1,\mathbf{C}_1 A 1 , B 1 , C 1 we can compose the following irreducible tensor products of ranks 0 and 1:
{ { A 1 ⊗ B 1 } 0 ⊗ C 1 } 1 = − 1 3 ( A ⋅ B ) ⋅ C , (477) \{\{\mathbf{A}_1\otimes\mathbf{B}_1\}_0\otimes\mathbf{C}_1\}_1=-\frac{1}{\sqrt{3}}(\mathbf{A}\cdot\mathbf{B})\cdot\mathbf{C}, \tag*{(477)} {{ A 1 ⊗ B 1 } 0 ⊗ C 1 } 1 = − 3 1 ( A ⋅ B ) ⋅ C , (477)
{ { A 1 ⊗ B 1 } 1 ⊗ C 1 } 0 = − i 6 [ A × B ] ⋅ C , (478) \{\{\mathbf{A}_1\otimes\mathbf{B}_1\}_1\otimes\mathbf{C}_1\}_0=-\frac{i}{\sqrt{6}}[\mathbf{A}\times\mathbf{B}]\cdot\mathbf{C}, \tag*{(478)} {{ A 1 ⊗ B 1 } 1 ⊗ C 1 } 0 = − 6 i [ A × B ] ⋅ C , (478)
{ { A 1 ⊗ B 1 } 1 ⊗ C 1 } 1 = − 1 2 [ [ A × B ] × C ] = 1 2 A ( B ⋅ C ) − 1 2 B ( A ⋅ C ) , (479) \{\{\mathbf{A}_1\otimes\mathbf{B}_1\}_1\otimes\mathbf{C}_1\}_1=-\frac{1}{2}[[\mathbf{A}\times\mathbf{B}]\times\mathbf{C}]=\frac{1}{2}\mathbf{A}(\mathbf{B}\cdot\mathbf{C})-\frac{1}{2}\mathbf{B}(\mathbf{A}\cdot\mathbf{C}), \tag*{(479)} {{ A 1 ⊗ B 1 } 1 ⊗ C 1 } 1 = − 2 1 [[ A × B ] × C ] = 2 1 A ( B ⋅ C ) − 2 1 B ( A ⋅ C ) , (479)
{ { A 1 ⊗ B 1 } 2 ⊗ C 1 } 1 = 3 5 { 1 3 C ( A ⋅ B ) − 1 2 B ( A ⋅ C ) − 1 2 A ( B ⋅ C ) } . (480) \{\{\mathbf{A}_1\otimes\mathbf{B}_1\}_2\otimes\mathbf{C}_1\}_1=\sqrt{\frac{3}{5}}\left\{\frac{1}{3}\mathbf{C}(\mathbf{A}\cdot\mathbf{B})-\frac{1}{2}\mathbf{B}(\mathbf{A}\cdot\mathbf{C})-\frac{1}{2}\mathbf{A}(\mathbf{B}\cdot\mathbf{C})\right\}. \tag*{(480)} {{ A 1 ⊗ B 1 } 2 ⊗ C 1 } 1 = 5 3 { 3 1 C ( A ⋅ B ) − 2 1 B ( A ⋅ C ) − 2 1 A ( B ⋅ C ) } . (480)
In addition to the foregoing formulas, there exist products which differ from Eqs. (477) –(480) by vector coupling schemes. The problems concerned with recoupling in tensor products are considered below (§ 13 ).
Given four commuting vectors A 1 , B 1 , C 1 \mathbf{A}_1,\mathbf{B}_1,\mathbf{C}_1 A 1 , B 1 , C 1 and D 1 \mathbf{D}_1 D 1 we can compose the following irreducible tensor products of ranks 0 and 1:
{ { A 1 ⊗ B 1 } 0 ⊗ { C 1 ⊗ D 1 } 0 } 0 = 1 3 ( A ⋅ B ) ( D ⋅ C ) , (481) \left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_0\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_0\right\}_0=\frac{1}{3}(\mathbf{A}\cdot\mathbf{B})(\mathbf{D}\cdot\mathbf{C}), \tag*{(481)} { { A 1 ⊗ B 1 } 0 ⊗ { C 1 ⊗ D 1 } 0 } 0 = 3 1 ( A ⋅ B ) ( D ⋅ C ) , (481)
{ { A 1 ⊗ B 1 } 1 ⊗ { C 1 ⊗ D 1 } 0 } 1 = − i 6 [ A × B ] ( C ⋅ D ) , (482) \left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_1\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_0\right\}_1=-\frac{i}{\sqrt{6}}[\mathbf{A}\times\mathbf{B}](\mathbf{C}\cdot\mathbf{D}), \tag*{(482)} { { A 1 ⊗ B 1 } 1 ⊗ { C 1 ⊗ D 1 } 0 } 1 = − 6 i [ A × B ] ( C ⋅ D ) , (482)
{ { A 1 ⊗ B 1 } 0 ⊗ { C 1 ⊗ D 1 } 1 } 1 = − i 6 ( A ⋅ B ) [ C × D ] , (483) \left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_0\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_1\right\}_1=-\frac{i}{\sqrt{6}}(\mathbf{A}\cdot\mathbf{B})[\mathbf{C}\times\mathbf{D}], \tag*{(483)} { { A 1 ⊗ B 1 } 0 ⊗ { C 1 ⊗ D 1 } 1 } 1 = − 6 i ( A ⋅ B ) [ C × D ] , (483)
{ { A 1 ⊗ B 1 } 1 ⊗ { C 1 ⊗ D 1 } 1 } 0 = 1 2 3 { ( A ⋅ C ) ( B ⋅ D ) − ( A ⋅ D ) ( B ⋅ C ) } , (484) \left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_1\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_1\right\}_0=\frac{1}{2\sqrt{3}}\{(\mathbf{A}\cdot\mathbf{C})(\mathbf{B}\cdot\mathbf{D})-(\mathbf{A}\cdot\mathbf{D})(\mathbf{B}\cdot\mathbf{C})\}, \tag*{(484)} { { A 1 ⊗ B 1 } 1 ⊗ { C 1 ⊗ D 1 } 1 } 0 = 2 3 1 {( A ⋅ C ) ( B ⋅ D ) − ( A ⋅ D ) ( B ⋅ C )} , (484)
{ { A 1 ⊗ B 1 } 1 ⊗ { C 1 ⊗ D 1 } 1 } 1 = − i 2 2 { C ( D ⋅ [ A × B ] ) − D ( C ⋅ [ A × B ] ) } = − i 2 2 { B ( A ⋅ [ C × D ] ) − A ( B ⋅ [ C × D ] ) } , (485) \begin{aligned}
\left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_1\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_1\right\}_1
&=-\frac{i}{2\sqrt{2}}\{\mathbf{C}(\mathbf{D}\cdot[\mathbf{A}\times\mathbf{B}])-\mathbf{D}(\mathbf{C}\cdot[\mathbf{A}\times\mathbf{B}])\}\\
&=-\frac{i}{2\sqrt{2}}\{\mathbf{B}(\mathbf{A}\cdot[\mathbf{C}\times\mathbf{D}])-\mathbf{A}(\mathbf{B}\cdot[\mathbf{C}\times\mathbf{D}])\}, \tag*{(485)}
\end{aligned} { { A 1 ⊗ B 1 } 1 ⊗ { C 1 ⊗ D 1 } 1 } 1 = − 2 2 i { C ( D ⋅ [ A × B ]) − D ( C ⋅ [ A × B ])} = − 2 2 i { B ( A ⋅ [ C × D ]) − A ( B ⋅ [ C × D ])} , (485)
{ { A 1 ⊗ B 1 } 2 ⊗ { C 1 ⊗ D 1 } 1 } 1 = i 3 2 ⋅ 5 { 1 3 ( A ⋅ B ) [ C × D ] − 1 2 B ( D ⋅ [ A × C ] ) − 1 2 A ( D ⋅ [ B × C ] ) } , (486) \left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_2\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_1\right\}_1=\frac{i\sqrt{3}}{\sqrt{2}\cdot5}\left\{\frac{1}{3}(\mathbf{A}\cdot\mathbf{B})[\mathbf{C}\times\mathbf{D}]-\frac{1}{2}\mathbf{B}(\mathbf{D}\cdot[\mathbf{A}\times\mathbf{C}])-\frac{1}{2}\mathbf{A}(\mathbf{D}\cdot[\mathbf{B}\times\mathbf{C}])\right\}, \tag*{(486)} { { A 1 ⊗ B 1 } 2 ⊗ { C 1 ⊗ D 1 } 1 } 1 = 2 ⋅ 5 i 3 { 3 1 ( A ⋅ B ) [ C × D ] − 2 1 B ( D ⋅ [ A × C ]) − 2 1 A ( D ⋅ [ B × C ]) } , (486)
{ { A 1 ⊗ B 1 } 1 ⊗ { C 1 ⊗ D 1 } 2 } 1 = i 3 2 ⋅ 5 { 1 3 ( C ⋅ D ) [ A × B ] − 1 2 C ( B ⋅ [ D × A ] ) − 1 2 D ( B ⋅ [ C × A ] ) } , (487) \left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_1\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_2\right\}_1=\frac{i\sqrt{3}}{\sqrt{2}\cdot5}\left\{\frac{1}{3}(\mathbf{C}\cdot\mathbf{D})[\mathbf{A}\times\mathbf{B}]-\frac{1}{2}\mathbf{C}(\mathbf{B}\cdot[\mathbf{D}\times\mathbf{A}])-\frac{1}{2}\mathbf{D}(\mathbf{B}\cdot[\mathbf{C}\times\mathbf{A}])\right\}, \tag*{(487)} { { A 1 ⊗ B 1 } 1 ⊗ { C 1 ⊗ D 1 } 2 } 1 = 2 ⋅ 5 i 3 { 3 1 ( C ⋅ D ) [ A × B ] − 2 1 C ( B ⋅ [ D × A ]) − 2 1 D ( B ⋅ [ C × A ]) } , (487)
{ { A 1 ⊗ B 1 } 2 ⊗ { C 1 ⊗ D 1 } 2 } 0 = 1 5 { 1 2 ( A ⋅ C ) ( B ⋅ D ) − 1 3 ( A ⋅ B ) ( C ⋅ D ) + 1 2 ( A ⋅ D ) ( B ⋅ C ) } , (488) \left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_2\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_2\right\}_0=\frac{1}{\sqrt{5}}\left\{\frac{1}{2}(\mathbf{A}\cdot\mathbf{C})(\mathbf{B}\cdot\mathbf{D})-\frac{1}{3}(\mathbf{A}\cdot\mathbf{B})(\mathbf{C}\cdot\mathbf{D})+\frac{1}{2}(\mathbf{A}\cdot\mathbf{D})(\mathbf{B}\cdot\mathbf{C})\right\}, \tag*{(488)} { { A 1 ⊗ B 1 } 2 ⊗ { C 1 ⊗ D 1 } 2 } 0 = 5 1 { 2 1 ( A ⋅ C ) ( B ⋅ D ) − 3 1 ( A ⋅ B ) ( C ⋅ D ) + 2 1 ( A ⋅ D ) ( B ⋅ C ) } , (488)
{ { A 1 ⊗ B 1 } 2 ⊗ { C 1 ⊗ D 1 } 2 } 1 = − i 2 2 ⋅ 5 { ( A ⋅ C ) [ B × D ] + ( A ⋅ D ) [ B × C ] + ( B ⋅ C ) [ A × D ] + ( B ⋅ D ) [ A × C ] } . (489) \left\{\left\{\mathbf{A}_1\otimes\mathbf{B}_1\right\}_2\otimes\left\{\mathbf{C}_1\otimes\mathbf{D}_1\right\}_2\right\}_1=-\frac{i}{2\sqrt{2}\cdot5}\{(\mathbf{A}\cdot\mathbf{C})[\mathbf{B}\times\mathbf{D}]+(\mathbf{A}\cdot\mathbf{D})[\mathbf{B}\times\mathbf{C}]+(\mathbf{B}\cdot\mathbf{C})[\mathbf{A}\times\mathbf{D}]+(\mathbf{B}\cdot\mathbf{D})[\mathbf{A}\times\mathbf{C}]\}. \tag*{(489)} { { A 1 ⊗ B 1 } 2 ⊗ { C 1 ⊗ D 1 } 2 } 1 = − 2 2 ⋅ 5 i {( A ⋅ C ) [ B × D ] + ( A ⋅ D ) [ B × C ] + ( B ⋅ C ) [ A × D ] + ( B ⋅ D ) [ A × C ]} . (489)
The change of coupling schemes in products of four operators is discussed in § 13 .
Products constructed from the components of identical vectors have the following property
{ ⋯ { { A 1 ⊗ A 1 } l 2 ⊗ A 1 } l 3 ⋯ ⊗ A 1 } l n = { A 1 ⊗ ⋯ { A 1 ⊗ { A 1 ⊗ A 1 } l 2 } l 3 ⋯ } l n . (490) \left\{\cdots\left\{\left\{\mathbf{A}_1\otimes\mathbf{A}_1\right\}_{l_2}\otimes\mathbf{A}_1\right\}_{l_3}\cdots\otimes\mathbf{A}_1\right\}_{l_n}
=
\left\{\mathbf{A}_1\otimes\cdots\left\{\mathbf{A}_1\otimes\left\{\mathbf{A}_1\otimes\mathbf{A}_1\right\}_{l_2}\right\}_{l_3}\cdots\right\}_{l_n}. \tag*{(490)} { ⋯ { { A 1 ⊗ A 1 } l 2 ⊗ A 1 } l 3 ⋯ ⊗ A 1 } l n = { A 1 ⊗ ⋯ { A 1 ⊗ { A 1 ⊗ A 1 } l 2 } l 3 ⋯ } l n . (490)
If a vector A \mathbf{A} A does not contain any spin variable and differential operator one can express such products in terms of spherical harmonics (Chap. 5), i.e.,
{ ⋯ { { A 1 ⊗ A 1 } l 2 ⊗ A 1 } l 3 ⋯ ⊗ A 1 } l n m n = 4 π 2 l n + 1 ∣ A ∣ n Y l n m n ( ϑ , φ ) ∏ i = 2 n ( l i − 1 1 l i 0 0 0 ) , (491) \left\{\cdots\left\{\left\{\mathbf{A}_1\otimes\mathbf{A}_1\right\}_{l_2}\otimes\mathbf{A}_1\right\}_{l_3}\cdots\otimes\mathbf{A}_1\right\}_{l_n m_n}=\sqrt{\frac{4\pi}{2l_n+1}}|\mathbf{A}|^nY_{l_n m_n}(\vartheta,\varphi)\prod_{i=2}^{n}\left(\begin{array}{ccc}l_{i-1}&1&l_i\\0&0&0\end{array}\right), \tag*{(491)} { ⋯ { { A 1 ⊗ A 1 } l 2 ⊗ A 1 } l 3 ⋯ ⊗ A 1 } l n m n = 2 l n + 1 4 π ∣ A ∣ n Y l n m n ( ϑ , φ ) i = 2 ∏ n ( l i − 1 0 1 0 l i 0 ) , (491)
where ϑ , φ \vartheta,\varphi ϑ , φ are the polar angles of the vector A \mathbf{A} A and l 1 = 1 l_1=1 l 1 = 1 . In particular, if l 2 = 2 l_2=2 l 2 = 2 , l 3 = 3 , … , l n = n l_3=3,\ldots,l_n=n l 3 = 3 , … , l n = n , we have
{ ⋯ { { A 1 ⊗ A 1 } 2 ⊗ A 1 } 3 ⋯ ⊗ A 1 } n m = 4 π n ! ( 2 n + 1 ) ! ! ∣ A ∣ n Y n m ( ϑ , φ ) . (492) \left\{\cdots\left\{\left\{\mathbf{A}_1\otimes\mathbf{A}_1\right\}_2\otimes\mathbf{A}_1\right\}_3\cdots\otimes\mathbf{A}_1\right\}_{nm}=\sqrt{\frac{4\pi n!}{(2n+1)!!}}|\mathbf{A}|^nY_{nm}(\vartheta,\varphi). \tag*{(492)} { ⋯ { { A 1 ⊗ A 1 } 2 ⊗ A 1 } 3 ⋯ ⊗ A 1 } nm = ( 2 n + 1 )!! 4 π n ! ∣ A ∣ n Y nm ( ϑ , φ ) . (492)
Cartesian Tensors of Second and Third Ranks
An arbitrary cartesian tensor of second rank T i k T_{ik} T ik (i , k = x , y , z i,k=x,y,z i , k = x , y , z ) is, generally, reducible and may be decomposed into three irreducible parts:
(a) a tensor which is proportional to the unit tensor, E i k = E δ i k E_{ik}=E\delta_{ik} E ik = E δ ik ;
(b) an antisymmetric tensor A i k = − A k i A_{ik}=-A_{ki} A ik = − A k i .
(c) a symmetric traceless tensor S i k = S k i S_{ik}=S_{ki} S ik = S k i , ∑ i S i i = 0 \sum_i S_{ii}=0 ∑ i S ii = 0 .
Thus,
T i k = E δ i k + A i k + S i k , (493) T_{ik}=E\delta_{ik}+A_{ik}+S_{ik}, \tag*{(493)} T ik = E δ ik + A ik + S ik , (493)
E = 1 3 Tr ( T i k ) = 1 3 ∑ i T i i , (494) E=\frac{1}{3}\operatorname{Tr}(T_{ik})=\frac{1}{3}\sum_i T_{ii}, \tag*{(494)} E = 3 1 Tr ( T ik ) = 3 1 i ∑ T ii , (494)
A i k = 1 2 ( T i k − T k i ) , (495) A_{ik}=\frac{1}{2}(T_{ik}-T_{ki}), \tag*{(495)} A ik = 2 1 ( T ik − T k i ) , (495)
S i k = 1 2 ( T i k + T k i − 2 3 δ i k ∑ l T l l ) . (496) S_{ik}=\frac{1}{2}\left(T_{ik}+T_{ki}-\frac{2}{3}\delta_{ik}\sum_l T_{ll}\right). \tag*{(496)} S ik = 2 1 ( T ik + T k i − 3 2 δ ik l ∑ T l l ) . (496)
The value E E E is invariant under rotations of coordinate systems. It is an irreducible tensor of zero rank
I 00 = E . (497) \mathcal{I}_{00}=E. \tag*{(497)} I 00 = E . (497)
The antisymmetric tensor A i k A_{ik} A ik is equivalent to the axial vector
A i k = ε i k l U l , U i = 1 2 ∑ k l ε i k l A k l . (498) A_{ik}=\varepsilon_{ikl}\mathcal{U}_l,\qquad\mathcal{U}_i=\frac{1}{2}\sum_{kl}\varepsilon_{ikl}A_{kl}. \tag*{(498)} A ik = ε ik l U l , U i = 2 1 k l ∑ ε ik l A k l . (498)
Using the components A i k A_{ik} A ik , one can form an irreducible pseudotensor of rank one:
I 10 = U z = A x y , \mathcal{I}_{10}=\mathcal{U}_z=A_{xy}, I 10 = U z = A x y ,
I 1 ± 1 = ∓ 1 2 ( U x ± i U y ) = ∓ 1 2 ( A y z ± i A z x ) . (499) \mathcal{I}_{1\pm1}=\mp\frac{1}{\sqrt{2}}(\mathcal{U}_x\pm i\mathcal{U}_y)=\mp\frac{1}{\sqrt{2}}(A_{yz}\pm iA_{zx}). \tag*{(499)} I 1 ± 1 = ∓ 2 1 ( U x ± i U y ) = ∓ 2 1 ( A y z ± i A z x ) . (499)
Using the components of the symmetric traceless tensor S i j S_{ij} S ij , we may construct an irreducible tensor of second rank.
I 20 = S z z , \mathcal{I}_{20}=S_{zz}, I 20 = S z z ,
I 2 ± 1 = ∓ 2 3 ( S x z ± i S y z ) , \mathcal{I}_{2\pm1}=\mp\sqrt{\frac{2}{3}}(S_{xz}\pm iS_{yz}), I 2 ± 1 = ∓ 3 2 ( S x z ± i S y z ) ,
I 2 ± 2 = 1 6 ( S x x − S y y ± 2 i S x y ) . (500) \mathcal{I}_{2\pm2}=\sqrt{\frac{1}{6}}(S_{xx}-S_{yy}\pm2iS_{xy}). \tag*{(500)} I 2 ± 2 = 6 1 ( S xx − S y y ± 2 i S x y ) . (500)
Thus, from nine cartesian components T i k T_{ik} T ik of a tensor of second rank we can compose one tensor of zero rank (scalar which has one component), one pseudotensor of the first rank (three components) and one irreducible tensor of second rank (five components).
From the 27 cartesian components T i k l T_{ikl} T ik l of a tensor of third rank we can compose one pseudotensor of zero rank T 00 \mathcal{T}_{00} T 00 (scalar, which has one component), three tensors of first rank T 1 μ \mathcal{T}_{1\mu} T 1 μ (3 × 3 = 9 3 \times3 = 9 3 × 3 = 9 components), two irreducible tensors of second rank T 2 μ \mathcal{T}_{2\mu} T 2 μ (2 × 5 = 10 2 \times5 = 10 2 × 5 = 10 components) and one irreducible tensor of third rank T 3 μ \mathcal{T}_{3\mu} T 3 μ (seven components). In this case construction of the tensors of first and second ranks is not unique.
Differential Operations as Irreducible Tensor Products
Differential operations on scalars and vectors (see Sec. 1.3) can be written in the form of irreducible tensor products of the operator ∇ \nabla ∇ and corresponding scalars and vectors,
grad Φ = { ∇ 1 ⊗ Φ } 1 , (501) \operatorname{grad}\Phi=\{\nabla_1\otimes\Phi\}_1, \tag*{(501)} grad Φ = { ∇ 1 ⊗ Φ } 1 , (501)
div A = − 3 { ∇ 1 ⊗ A 1 } 0 , (502) \operatorname{div}\mathbf{A}=-\sqrt{3}\{\nabla_1\otimes A_1\}_0, \tag*{(502)} div A = − 3 { ∇ 1 ⊗ A 1 } 0 , (502)
curl A = − i 2 { ∇ 1 ⊗ A 1 } 1 , (503) \operatorname{curl}\mathbf{A}=-i\sqrt{2}\{\nabla_1\otimes A_1\}_1, \tag*{(503)} curl A = − i 2 { ∇ 1 ⊗ A 1 } 1 , (503)
Δ = ∇ 2 = − 3 { ∇ 1 ⊗ ∇ 1 } 0 , (504) \Delta=\nabla^2=-\sqrt{3}\{\nabla_1\otimes\nabla_1\}_0, \tag*{(504)} Δ = ∇ 2 = − 3 { ∇ 1 ⊗ ∇ 1 } 0 , (504)
grad div A = − 3 { ∇ 1 ⊗ { ∇ 1 ⊗ A 1 } 0 } 1 , (505) \operatorname{grad}\operatorname{div}\mathbf{A}=-\sqrt{3}\{\nabla_1\otimes\{\nabla_1\otimes A_1\}_0\}_1, \tag*{(505)} grad div A = − 3 { ∇ 1 ⊗ { ∇ 1 ⊗ A 1 } 0 } 1 , (505)
curl curl A = − 2 { ∇ 1 ⊗ { ∇ 1 ⊗ A 1 } 1 } 1 , (506) \operatorname{curl}\operatorname{curl}\mathbf{A}=-2\{\nabla_1\otimes\{\nabla_1\otimes A_1\}_1\}_1, \tag*{(506)} curl curl A = − 2 { ∇ 1 ⊗ { ∇ 1 ⊗ A 1 } 1 } 1 , (506)
div grad Φ = − 3 { ∇ 1 ⊗ { ∇ 1 ⊗ Φ } 1 } 0 = − 3 { ∇ 1 ⊗ ∇ 1 } 0 Φ , (507) \operatorname{div}\operatorname{grad}\Phi=-\sqrt{3}\{\nabla_1\otimes\{\nabla_1\otimes\Phi\}_1\}_0=-\sqrt{3}\{\nabla_1\otimes\nabla_1\}_0\Phi, \tag*{(507)} div grad Φ = − 3 { ∇ 1 ⊗ { ∇ 1 ⊗ Φ } 1 } 0 = − 3 { ∇ 1 ⊗ ∇ 1 } 0 Φ , (507)
curl grad Φ = − i 2 { ∇ 1 ⊗ { ∇ 1 ⊗ Φ } 1 } 1 = 0 , (508) \operatorname{curl}\operatorname{grad}\Phi=-i\sqrt{2}\{\nabla_1\otimes\{\nabla_1\otimes\Phi\}_1\}_1=0, \tag*{(508)} curl grad Φ = − i 2 { ∇ 1 ⊗ { ∇ 1 ⊗ Φ } 1 } 1 = 0 , (508)
div curl A = i 6 { ∇ 1 ⊗ { ∇ 1 ⊗ A 1 } 1 } 0 = 0. (509) \operatorname{div}\operatorname{curl}\mathbf{A}=i\sqrt{6}\{\nabla_1\otimes\{\nabla_1\otimes A_1\}_1\}_0=0. \tag*{(509)} div curl A = i 6 { ∇ 1 ⊗ { ∇ 1 ⊗ A 1 } 1 } 0 = 0. (509)
As follows from Sec. 1.3.1 the operator ∇ \nabla ∇ is given by
∇ = n ∂ ∂ r − i r [ n × L ^ ] (510) \nabla=\mathbf{n}\frac{\partial}{\partial r}-\frac{i}{r}[\mathbf{n}\times\hat{\mathbf{L}}] \tag*{(510)} ∇ = n ∂ r ∂ − r i [ n × L ^ ] (510)
In spherical component form it is written as
∇ μ = 4 π 3 ( Y 1 μ ∂ ∂ r − 2 r { Y 1 ⊗ L ^ 1 } 1 μ ) . (511) \nabla_\mu=\sqrt{\frac{4\pi}{3}}\left(Y_{1\mu}\frac{\partial}{\partial r}-\frac{\sqrt{2}}{r}\{Y_1\otimes\hat{L}_1\}_{1\mu}\right). \tag*{(511)} ∇ μ = 3 4 π ( Y 1 μ ∂ r ∂ − r 2 { Y 1 ⊗ L ^ 1 } 1 μ ) . (511)
Here Y 1 Y_1 Y 1 is an irreducible tensor, whose components are spherical harmonics Y 1 μ ( n ) Y_{1\mu}(\mathbf{n}) Y 1 μ ( n ) (Chap. 5); L ^ 1 = L ^ \hat{L}_1=\hat{L} L ^ 1 = L ^ is the orbital angular momentum operator (see § 10.2 ).
When expanding any scalar function Φ ( r ) \Phi(\mathbf{r}) Φ ( r ) in a Taylor series, one will deal with arbitrarily large powers of operator ∇ \nabla ∇ :
Φ ( r + δ r ) = e ( δ r ⋅ ∇ ) Φ ( r ) = Φ ( r ) + ( δ r ⋅ ∇ ) Φ ( r ) + 1 2 ! ( δ r ⋅ ∇ ) 2 Φ ( r ) + … = Φ ( r ) + δ r ( u ⋅ ∇ ) Φ ( r ) + 1 2 ! ( δ r ) 2 ( u ⋅ ∇ ) 2 Φ ( r ) + … (512) \begin{aligned}
\Phi(\mathbf{r}+\delta\mathbf{r})=e^{(\delta\mathbf{r}\cdot\nabla)}\Phi(\mathbf{r})=\Phi(\mathbf{r})+(\delta\mathbf{r}\cdot\nabla)\Phi(\mathbf{r})+\frac{1}{2!}(\delta\mathbf{r}\cdot\nabla)^2\Phi(\mathbf{r})+\ldots\\
&=\Phi(\mathbf{r})+\delta r(\mathbf{u}\cdot\nabla)\Phi(\mathbf{r})+\frac{1}{2!}(\delta r)^2(\mathbf{u}\cdot\nabla)^2\Phi(\mathbf{r})+\ldots \tag*{(512)}
\end{aligned} Φ ( r + δ r ) = e ( δ r ⋅ ∇ ) Φ ( r ) = Φ ( r ) + ( δ r ⋅ ∇ ) Φ ( r ) + 2 ! 1 ( δ r ⋅ ∇ ) 2 Φ ( r ) + … = Φ ( r ) + δ r ( u ⋅ ∇ ) Φ ( r ) + 2 ! 1 ( δ r ) 2 ( u ⋅ ∇ ) 2 Φ ( r ) + … (512)
Here u = δ r / ∣ δ r ∣ \mathbf{u}=\delta\mathbf{r}/|\delta\mathbf{r}| u = δ r /∣ δ r ∣ and ( u ⋅ ∇ ) = d / d s (\mathbf{u}\cdot\nabla)=d/ds ( u ⋅ ∇ ) = d / d s is the operator of directional differentiation in the direction u \mathbf{u} u (see Sec. 1.3). The operator ( u ⋅ ∇ ) n ≡ d n / d s n (\mathbf{u}\cdot\nabla)^n\equiv d^n/ds^n ( u ⋅ ∇ ) n ≡ d n / d s n can be written in the form of tensor product
( u ⋅ ∇ ) n = ( − 1 ) n ∑ l 2 , l 3 , … l n ( − 1 ) l n ( { … { { u 1 ⊗ u 1 } l 2 ⊗ u 1 } l 3 … ⊗ u 1 } l n ⋅ { … { { ∇ 1 ⊗ ∇ 1 } l 2 ⊗ ∇ 1 } l 3 … ⊗ ∇ 1 } l n ) . (513) (\mathbf{u}\cdot\nabla)^n=(-1)^n\sum_{l_2,l_3,\ldots l_n}(-1)^{l_n}\left(\{\ldots\{\{u_1\otimes u_1\}_{l_2}\otimes u_1\}_{l_3}\ldots\otimes u_1\}_{l_n}\cdot\{\ldots\{\{\nabla_1\otimes\nabla_1\}_{l_2}\otimes\nabla_1\}_{l_3}\ldots\otimes\nabla_1\}_{l_n}\right). \tag*{(513)} ( u ⋅ ∇ ) n = ( − 1 ) n l 2 , l 3 , … l n ∑ ( − 1 ) l n ( { … {{ u 1 ⊗ u 1 } l 2 ⊗ u 1 } l 3 … ⊗ u 1 } l n ⋅ { … {{ ∇ 1 ⊗ ∇ 1 } l 2 ⊗ ∇ 1 } l 3 … ⊗ ∇ 1 } l n ) . (513)
Using Eqs. (491) and (492) one can express multiple tensor products of the unit vector u \mathbf{u} u which enters Eq. (513) in terms of spherical harmonics.
Recoupling in irreducible tensor products
The irreducible tensor product of two irreducible tensors is defined in § 12.1.7 . By making use of the same relations one can form irreducible tensor products of three and more irreducible tensors. However, in these cases different orders and different coupling schemes of tensors in products are possible.
The recoupling tensors without change of their order may be carried out by some real (for a given definition of the vector addition coefficients) and orthogonal matrix which performs the direct and inverse transformation. When the tensor order has to be changed one should take account of the commutation rules (Eqs. (458) –(462) ). We will use the notation
Π a b c … d = [ ( 2 a + 1 ) ( 2 b + 1 ) ( 2 c + 1 ) … ( 2 d + 1 ) ] 1 2 \Pi_{abc\ldots d}=[(2a+1)(2b+1)(2c+1)\ldots(2d+1)]^{\frac{1}{2}} Π ab c … d = [( 2 a + 1 ) ( 2 b + 1 ) ( 2 c + 1 ) … ( 2 d + 1 ) ] 2 1
Relations Valid for Commuting as well as Non-Commuting Tensors
For recoupling tensors without changing their order in irreducible tensor products of three and four tensors, one has the following relations:
{ { P a ⊗ Q b } c ⊗ R d } f = ( − 1 ) a + b + f + d ∑ h Π h c { a b c d f h } { P a ⊗ { Q b ⊗ R d } h } f (514) \{\{\mathbf{P}_a\otimes\mathbf{Q}_b\}_c\otimes\mathbf{R}_d\}_f=(-1)^{a+b+f+d}\sum_h\Pi_{hc}\left\{\begin{array}{ccc}a&b&c\\d&f&h\end{array}\right\}\{\mathbf{P}_a\otimes\{\mathbf{Q}_b\otimes\mathbf{R}_d\}_h\}_f \tag*{(514)} {{ P a ⊗ Q b } c ⊗ R d } f = ( − 1 ) a + b + f + d h ∑ Π h c { a d b f c h } { P a ⊗ { Q b ⊗ R d } h } f (514)
( { P a ⊗ Q b } c ⋅ R c ) = ( − 1 ) − c + a Π c Π a ( P a ⋅ { Q b ⊗ R c } a ) (515) (\{\mathbf{P}_a\otimes\mathbf{Q}_b\}_c\cdot\mathbf{R}_c)=(-1)^{-c+a}\frac{\Pi_c}{\Pi_a}(\mathbf{P}_a\cdot\{\mathbf{Q}_b\otimes\mathbf{R}_c\}_a) \tag*{(515)} ({ P a ⊗ Q b } c ⋅ R c ) = ( − 1 ) − c + a Π a Π c ( P a ⋅ { Q b ⊗ R c } a ) (515)
{ { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = ( − 1 ) d + c + e + k ∑ h Π h f { d c h k e f } { { { P a ⊗ Q b } c ⊗ R d } h ⊗ S e } k = ( − 1 ) a + f + b + k ∑ h Π h c { a b c f k h } { P a ⊗ { Q b ⊗ { R d ⊗ S e } f } h } k . (516) \begin{aligned}
\{\{\mathbf{P}_a\otimes\mathbf{Q}_b\}_c\otimes\{\mathbf{R}_d\otimes\mathbf{S}_e\}_f\}_k=(-1)^{d+c+e+k}\sum_h\Pi_{hf}\left\{\begin{array}{ccc}d&c&h\\k&e&f\end{array}\right\}\{\{\{\mathbf{P}_a\otimes\mathbf{Q}_b\}_c\otimes\mathbf{R}_d\}_h\otimes\mathbf{S}_e\}_k \\
&=(-1)^{a+f+b+k}\sum_h\Pi_{hc}\left\{\begin{array}{ccc}a&b&c\\f&k&h\end{array}\right\}\{\mathbf{P}_a\otimes\{\mathbf{Q}_b\otimes\{\mathbf{R}_d\otimes\mathbf{S}_e\}_f\}_h\}_k. \tag*{(516)}
\end{aligned} {{ P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = ( − 1 ) d + c + e + k h ∑ Π h f { d k c e h f } {{{ P a ⊗ Q b } c ⊗ R d } h ⊗ S e } k = ( − 1 ) a + f + b + k h ∑ Π h c { a f b k c h } { P a ⊗ { Q b ⊗ { R d ⊗ S e } f } h } k . (516)
( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = ( − 1 ) − c + e Π c Π e ( { { P a ⊗ Q b } c ⊗ R d } e ⋅ S e ) = ( − 1 ) − c + a Π c Π a ( P a ⋅ { Q b ⊗ { R d ⊗ S e } f } a ) . (517) \begin{aligned}
(\{\mathbf{P}_a\otimes\mathbf{Q}_b\}_c\cdot\{\mathbf{R}_d\otimes\mathbf{S}_e\}_c)=\frac{(-1)^{-c+e}\Pi_c}{\Pi_e}(\{\{\mathbf{P}_a\otimes\mathbf{Q}_b\}_c\otimes\mathbf{R}_d\}_e\cdot\mathbf{S}_e) \\
&=(-1)^{-c+a}\frac{\Pi_c}{\Pi_a}(\mathbf{P}_a\cdot\{\mathbf{Q}_b\otimes\{\mathbf{R}_d\otimes\mathbf{S}_e\}_f\}_a). \tag*{(517)}
\end{aligned} ({ P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = Π e ( − 1 ) − c + e Π c ({{ P a ⊗ Q b } c ⊗ R d } e ⋅ S e ) = ( − 1 ) − c + a Π a Π c ( P a ⋅ { Q b ⊗ { R d ⊗ S e } f } a ) . (517)
{ { { P a ⊗ Q b } c ⊗ R d } h ⊗ S e } k = ∑ f , q ( − 1 ) a + f + b − d − c − e Π c q h f { a b c f k q } { d e f k c h } { P a ⊗ { Q b ⊗ { R d ⊗ S e } f } q } k (518) \{\{\{\mathbf{P}_a\otimes\mathbf{Q}_b\}_c\otimes\mathbf{R}_d\}_h\otimes\mathbf{S}_e\}_k=\sum_{f,q}(-1)^{a+f+b-d-c-e}\Pi_{cqhf}\left\{\begin{array}{ccc}a&b&c\\f&k&q\end{array}\right\}\left\{\begin{array}{ccc}d&e&f\\k&c&h\end{array}\right\}\{\mathbf{P}_a\otimes\{\mathbf{Q}_b\otimes\{\mathbf{R}_d\otimes\mathbf{S}_e\}_f\}_q\}_k \tag*{(518)} {{{ P a ⊗ Q b } c ⊗ R d } h ⊗ S e } k = f , q ∑ ( − 1 ) a + f + b − d − c − e Π c q h f { a f b k c q } { d k e c f h } { P a ⊗ { Q b ⊗ { R d ⊗ S e } f } q } k (518)
( { { P a ⊗ Q b } c ⊗ R d } e ⋅ S e ) = ( − 1 ) − c + a Π e Π a ( P a ⋅ { Q b ⊗ { R d ⊗ S e } f } a ) . (519) (\{\{\mathbf{P}_a\otimes\mathbf{Q}_b\}_c\otimes\mathbf{R}_d\}_e\cdot\mathbf{S}_e)=(-1)^{-c+a}\frac{\Pi_e}{\Pi_a}(\mathbf{P}_a\cdot\{\mathbf{Q}_b\otimes\{\mathbf{R}_d\otimes\mathbf{S}_e\}_f\}_a). \tag*{(519)} ({{ P a ⊗ Q b } c ⊗ R d } e ⋅ S e ) = ( − 1 ) − c + a Π a Π e ( P a ⋅ { Q b ⊗ { R d ⊗ S e } f } a ) . (519)
Relations for Commuting Tensors
Relations for Non-Commuting Tensors
To change the coupling scheme for irreducible products of three and four commuting tensors one can use the following relations
{ { P a ⊗ { Q b ⊗ R d } f } e } = ( − 1 ) b + d − f { { P a ⊗ { R d ⊗ Q b } f } e } = ( − 1 ) a + f − e { { Q b ⊗ R d } f ⊗ P a } e = ( − 1 ) a + b + d − e { { R d ⊗ Q b } f ⊗ P a } e (520) \left\{\left\{P_a \otimes\left\{Q_b \otimes R_d\right\}_f\right\}_e\right\}
= (-1)^{b+d-f}\left\{\left\{P_a \otimes\left\{R_d \otimes Q_b\right\}_f\right\}_e\right\}
= (-1)^{a+f-e}\left\{\left\{Q_b \otimes R_d\right\}_f \otimes P_a\right\}_e
= (-1)^{a+b+d-e}\left\{\left\{R_d \otimes Q_b\right\}_f \otimes P_a\right\}_e \tag*{(520)} { { P a ⊗ { Q b ⊗ R d } f } e } = ( − 1 ) b + d − f { { P a ⊗ { R d ⊗ Q b } f } e } = ( − 1 ) a + f − e { { Q b ⊗ R d } f ⊗ P a } e = ( − 1 ) a + b + d − e { { R d ⊗ Q b } f ⊗ P a } e (520)
{ { P a ⊗ Q b } c ⊗ R d } f = ( − 1 ) c + d + f ∑ h Π c h { a b c f d h } { Q b ⊗ { P a ⊗ R d } h } f (521) \left\{\left\{P_a \otimes Q_b\right\}_c \otimes R_d\right\}_f
= (-1)^{c+d+f}\sum_h \Pi_{ch}
\left\{\begin{array}{ccc}
a & b & c\\
f & d & h
\end{array}\right\}
\left\{Q_b \otimes\left\{P_a \otimes R_d\right\}_h\right\}_f \tag*{(521)} { { P a ⊗ Q b } c ⊗ R d } f = ( − 1 ) c + d + f h ∑ Π c h { a f b d c h } { Q b ⊗ { P a ⊗ R d } h } f (521)
( { P a ⊗ Q b } c ⋅ R c ) = ( − 1 ) − a Π c Π b ( Q b ⋅ { P a ⊗ R c } b ) (522) \left(\left\{P_a \otimes Q_b\right\}_c \mathbin{\cdot} R_c\right)
= (-1)^{-a}\frac{\Pi_c}{\Pi_b}
\left(Q_b \mathbin{\cdot} \left\{P_a \otimes R_c\right\}_b\right) \tag*{(522)} ( { P a ⊗ Q b } c ⋅ R c ) = ( − 1 ) − a Π b Π c ( Q b ⋅ { P a ⊗ R c } b ) (522)
{ P a ⊗ { Q b ⊗ { R d ⊗ S e } f } h } k = ( − 1 ) d + e − f { P a ⊗ { Q b ⊗ { S e ⊗ R d } f } h } k = ( − 1 ) d + b + e − h { P a ⊗ { { S e ⊗ R d } f ⊗ Q b } h } k = ( − 1 ) b + f − h { P a ⊗ { { R d ⊗ S e } f ⊗ Q b } h } k = ( − 1 ) a + h − k { { Q b ⊗ { R d ⊗ S e } f } h ⊗ P a } k = ( − 1 ) d + e + a + h − f − k { { Q b ⊗ { S e ⊗ R d } f } h ⊗ P a } k = ( − 1 ) a + b + d + e − h { { { S e ⊗ R d } f ⊗ Q b } h ⊗ P a } k = ( − 1 ) a + b + f − k { { { R d ⊗ S e } f ⊗ Q b } h ⊗ P a } k (523) \begin{aligned}
\left\{P_a \otimes\left\{Q_b \otimes\left\{R_d \otimes S_e\right\}_f\right\}_h\right\}_k
&= (-1)^{d+e-f}
\left\{P_a \otimes\left\{Q_b \otimes\left\{S_e \otimes R_d\right\}_f\right\}_h\right\}_k\\
&= (-1)^{d+b+e-h}
\left\{P_a \otimes\left\{\left\{S_e \otimes R_d\right\}_f \otimes Q_b\right\}_h\right\}_k\\
&= (-1)^{b+f-h}
\left\{P_a \otimes\left\{\left\{R_d \otimes S_e\right\}_f \otimes Q_b\right\}_h\right\}_k\\
&= (-1)^{a+h-k}
\left\{\left\{Q_b \otimes\left\{R_d \otimes S_e\right\}_f\right\}_h \otimes P_a\right\}_k\\
&= (-1)^{d+e+a+h-f-k}
\left\{\left\{Q_b \otimes\left\{S_e \otimes R_d\right\}_f\right\}_h \otimes P_a\right\}_k\\
&= (-1)^{a+b+d+e-h}
\left\{\left\{\left\{S_e \otimes R_d\right\}_f \otimes Q_b\right\}_h \otimes P_a\right\}_k\\
&= (-1)^{a+b+f-k}
\left\{\left\{\left\{R_d \otimes S_e\right\}_f \otimes Q_b\right\}_h \otimes P_a\right\}_k \tag*{(523)}
\end{aligned} { P a ⊗ { Q b ⊗ { R d ⊗ S e } f } h } k = ( − 1 ) d + e − f { P a ⊗ { Q b ⊗ { S e ⊗ R d } f } h } k = ( − 1 ) d + b + e − h { P a ⊗ { { S e ⊗ R d } f ⊗ Q b } h } k = ( − 1 ) b + f − h { P a ⊗ { { R d ⊗ S e } f ⊗ Q b } h } k = ( − 1 ) a + h − k { { Q b ⊗ { R d ⊗ S e } f } h ⊗ P a } k = ( − 1 ) d + e + a + h − f − k { { Q b ⊗ { S e ⊗ R d } f } h ⊗ P a } k = ( − 1 ) a + b + d + e − h { { { S e ⊗ R d } f ⊗ Q b } h ⊗ P a } k = ( − 1 ) a + b + f − k { { { R d ⊗ S e } f ⊗ Q b } h ⊗ P a } k (523)
{ { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = ∑ g h Π c f g h { a b c d e f g h k } { { P a ⊗ R d } g ⊗ { Q b ⊗ S e } h } k (524) \left\{\left\{P_a \otimes Q_b\right\}_c \otimes\left\{R_d \otimes S_e\right\}_f\right\}_k
= \sum_{gh}\Pi_{cfgh}
\left\{\begin{array}{ccc}
a & b & c\\
d & e & f\\
g & h & k
\end{array}\right\}
\left\{\left\{P_a \otimes R_d\right\}_g \otimes\left\{Q_b \otimes S_e\right\}_h\right\}_k \tag*{(524)} { { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = g h ∑ Π c f g h ⎩ ⎨ ⎧ a d g b e h c f k ⎭ ⎬ ⎫ { { P a ⊗ R d } g ⊗ { Q b ⊗ S e } h } k (524)
{ { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = ∑ g h ( − 1 ) h + b − k − e Π c f g h { a b c g d h k c g } { { { P a ⊗ R d } h ⊗ Q b } g ⊗ S e } k (525) \left\{\left\{P_a \otimes Q_b\right\}_c \otimes\left\{R_d \otimes S_e\right\}_f\right\}_k
= \sum_{gh}(-1)^{h+b-k-e}\Pi_{cfgh}
\left\{\begin{array}{ccc}
a & b & c\\
g & d & h\\
k & c & g
\end{array}\right\}
\left\{\left\{\left\{P_a \otimes R_d\right\}_h \otimes Q_b\right\}_g \otimes S_e\right\}_k \tag*{(525)} { { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = g h ∑ ( − 1 ) h + b − k − e Π c f g h ⎩ ⎨ ⎧ a g k b d c c h g ⎭ ⎬ ⎫ { { { P a ⊗ R d } h ⊗ Q b } g ⊗ S e } k (525)
( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = ( − 1 ) 2 a + b − d ∑ g Π c { a b c e d g } ( { P a ⊗ R d } g ⋅ { Q b ⊗ S e } g ) (526) \left(\left\{P_a \otimes Q_b\right\}_c \mathbin{\cdot} \left\{R_d \otimes S_e\right\}_c\right)
= (-1)^{2a+b-d}\sum_g \Pi_c
\left\{\begin{array}{ccc}
a & b & c\\
e & d & g
\end{array}\right\}
\left(\left\{P_a \otimes R_d\right\}_g \mathbin{\cdot} \left\{Q_b \otimes S_e\right\}_g\right) \tag*{(526)} ( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = ( − 1 ) 2 a + b − d g ∑ Π c { a e b d c g } ( { P a ⊗ R d } g ⋅ { Q b ⊗ S e } g ) (526)
( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = ∑ h ( − 1 ) e + b + d + h Π c c h Π e { a b c e d h } ( { { P a ⊗ R d } h ⊗ Q b } e ⋅ S e ) (527) \left(\left\{P_a \otimes Q_b\right\}_c \mathbin{\cdot} \left\{R_d \otimes S_e\right\}_c\right)
= \sum_h (-1)^{e+b+d+h}\frac{\Pi_{cch}}{\Pi_e}
\left\{\begin{array}{ccc}
a & b & c\\
e & d & h
\end{array}\right\}
\left(\left\{\left\{P_a \otimes R_d\right\}_h \otimes Q_b\right\}_e \mathbin{\cdot} S_e\right) \tag*{(527)} ( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = h ∑ ( − 1 ) e + b + d + h Π e Π cc h { a e b d c h } ( { { P a ⊗ R d } h ⊗ Q b } e ⋅ S e ) (527)
Equation (3.1.27) and Eqs. (3.3.11)–(3.3.14) permit us to obtain all other permutations. The product { P a ⊗ Q b } b \left\{P_a \otimes Q_b\right\}_b { P a ⊗ Q b } b is orthogonal to tensor Q b Q_b Q b , if a = 2 b − 1 , 2 b − 2 , … a=2b-1,2b-2,\ldots a = 2 b − 1 , 2 b − 2 , …
( { P a ⊗ Q b } b ⋅ Q b ) = 0. (528) \left(\left\{P_a \otimes Q_b\right\}_b \mathbin{\cdot} Q_b\right)=0. \tag*{(528)} ( { P a ⊗ Q b } b ⋅ Q b ) = 0. (528)
Relations for Non-Commuting Tensors
A change of the coupling scheme for irreducible tensor products of three and four non-commuting tensors gives
{ { P a ⊗ Q b } c ⊗ R d } f = ( − 1 ) f + d + c ∑ h Π c h { a b c f d h } { Q b ⊗ { P a ⊗ R d } h } f + { R c a b ⊗ R d } f (529) \left\{\left\{\mathbf{P}_a\otimes\mathbf{Q}_b\right\}_c\otimes\mathbf{R}_d\right\}_f=(-1)^{f+d+c}\sum_h\Pi_{ch}\left\{\begin{matrix}a&b&c\\f&d&h\end{matrix}\right\}\left\{\mathbf{Q}_b\otimes\left\{\mathbf{P}_a\otimes\mathbf{R}_d\right\}_h\right\}_f+\left\{\mathcal{R}^{ab}_c\otimes\mathbf{R}_d\right\}_f \tag*{(529)} { { P a ⊗ Q b } c ⊗ R d } f = ( − 1 ) f + d + c h ∑ Π c h { a f b d c h } { Q b ⊗ { P a ⊗ R d } h } f + { R c ab ⊗ R d } f (529)
( { P a ⊗ Q b } c ⋅ R c ) = ( − 1 ) − a Π c Π b ( Q b ⋅ { P a ⊗ R c } b ) + ( R c a b ⋅ R c ) (530) \left(\left\{\mathbf{P}_a\otimes\mathbf{Q}_b\right\}_c\cdot\mathbf{R}_c\right)=(-1)^{-a}\frac{\Pi_c}{\Pi_b}\left(\mathbf{Q}_b\cdot\left\{\mathbf{P}_a\otimes\mathbf{R}_c\right\}_b\right)+\left(\mathcal{R}^{ab}_c\cdot\mathbf{R}_c\right) \tag*{(530)} ( { P a ⊗ Q b } c ⋅ R c ) = ( − 1 ) − a Π b Π c ( Q b ⋅ { P a ⊗ R c } b ) + ( R c ab ⋅ R c ) (530)
{ P a ⊗ { Q b ⊗ R d } h } f = ∑ c g ( − 1 ) a + b − c Π c c g h { a b c f d g } { a b c d f h } { Q b ⊗ { P a ⊗ R d } g } f + ∑ c ( − 1 ) a + b + f + d Π c h { a b c d f h } { R c a b ⊗ R d } f (531) \begin{aligned}
\left\{\mathbf{P}_a\otimes\left\{\mathbf{Q}_b\otimes\mathbf{R}_d\right\}_h\right\}_f
={}&\sum_{cg}(-1)^{a+b-c}\Pi_{ccgh}\left\{\begin{matrix}a&b&c\\f&d&g\end{matrix}\right\}\left\{\begin{matrix}a&b&c\\d&f&h\end{matrix}\right\}\left\{\mathbf{Q}_b\otimes\left\{\mathbf{P}_a\otimes\mathbf{R}_d\right\}_g\right\}_f\\
&+\sum_c(-1)^{a+b+f+d}\Pi_{ch}\left\{\begin{matrix}a&b&c\\d&f&h\end{matrix}\right\}\left\{\mathcal{R}^{ab}_c\otimes\mathbf{R}_d\right\}_f \tag*{(531)}
\end{aligned} { P a ⊗ { Q b ⊗ R d } h } f = c g ∑ ( − 1 ) a + b − c Π cc g h { a f b d c g } { a d b f c h } { Q b ⊗ { P a ⊗ R d } g } f + c ∑ ( − 1 ) a + b + f + d Π c h { a d b f c h } { R c ab ⊗ R d } f (531)
( P a ⋅ { Q b ⊗ R d } a ) = ( − 1 ) 2 b − d Π a Π b ( Q b ⋅ { P a ⊗ R d } b ) + ( − 1 ) − a + d Π a Π d ( R d a b ⋅ R d ) (532) \left(\mathbf{P}_a\cdot\left\{\mathbf{Q}_b\otimes\mathbf{R}_d\right\}_a\right)=(-1)^{2b-d}\frac{\Pi_a}{\Pi_b}\left(\mathbf{Q}_b\cdot\left\{\mathbf{P}_a\otimes\mathbf{R}_d\right\}_b\right)+(-1)^{-a+d}\frac{\Pi_a}{\Pi_d}\left(\mathcal{R}^{ab}_d\cdot\mathbf{R}_d\right) \tag*{(532)} ( P a ⋅ { Q b ⊗ R d } a ) = ( − 1 ) 2 b − d Π b Π a ( Q b ⋅ { P a ⊗ R d } b ) + ( − 1 ) − a + d Π d Π a ( R d ab ⋅ R d ) (532)
{ { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = ∑ g r Π c f g r { a b c d e f r g k } { { P a ⊗ R d } r ⊗ { Q b ⊗ S e } g } k + ∑ h q ( − 1 ) a + k + d + e − f − h Π c h q f { a b c f k h } { d e f h b q } { P a ⊗ { R q b d ⊗ S e } h } k (533) \begin{aligned}
\left\{\left\{\mathbf{P}_a\otimes\mathbf{Q}_b\right\}_c\otimes\left\{\mathbf{R}_d\otimes\mathbf{S}_e\right\}_f\right\}_k
={}&\sum_{gr}\Pi_{cfgr}\left\{\begin{matrix}a&b&c\\d&e&f\\r&g&k\end{matrix}\right\}\left\{\left\{\mathbf{P}_a\otimes\mathbf{R}_d\right\}_r\otimes\left\{\mathbf{Q}_b\otimes\mathbf{S}_e\right\}_g\right\}_k\\
&+\sum_{hq}(-1)^{a+k+d+e-f-h}\Pi_{chqf}\left\{\begin{matrix}a&b&c\\f&k&h\end{matrix}\right\}\left\{\begin{matrix}d&e&f\\h&b&q\end{matrix}\right\}\left\{\mathbf{P}_a\otimes\left\{\mathcal{R}^{bd}_q\otimes\mathbf{S}_e\right\}_h\right\}_k \tag*{(533)}
\end{aligned} { { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = g r ∑ Π c f g r ⎩ ⎨ ⎧ a d r b e g c f k ⎭ ⎬ ⎫ { { P a ⊗ R d } r ⊗ { Q b ⊗ S e } g } k + h q ∑ ( − 1 ) a + k + d + e − f − h Π c h q f { a f b k c h } { d h e b f q } { P a ⊗ { R q b d ⊗ S e } h } k (533)
( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = ∑ g ( − 1 ) d + b + 2 g Π c 2 Π c Π g { a b c e d g } ( { P a ⊗ R d } g ⋅ { Q b ⊗ S e } g ) + ∑ q ( − 1 ) d + e + c − b Π c c q Π a { a b c d e q } ( P a ⋅ { R q b d ⊗ S e } a ) (534) \begin{aligned}
\left(\left\{\mathbf{P}_a\otimes\mathbf{Q}_b\right\}_c\cdot\left\{\mathbf{R}_d\otimes\mathbf{S}_e\right\}_c\right)
={}&\sum_g(-1)^{d+b+2g}\Pi_c^2\frac{\Pi_c}{\Pi_g}\left\{\begin{matrix}a&b&c\\e&d&g\end{matrix}\right\}\left(\left\{\mathbf{P}_a\otimes\mathbf{R}_d\right\}_g\cdot\left\{\mathbf{Q}_b\otimes\mathbf{S}_e\right\}_g\right)\\
&+\sum_q(-1)^{d+e+c-b}\frac{\Pi_{ccq}}{\Pi_a}\left\{\begin{matrix}a&b&c\\d&e&q\end{matrix}\right\}\left(\mathbf{P}_a\cdot\left\{\mathcal{R}^{bd}_q\otimes\mathbf{S}_e\right\}_a\right) \tag*{(534)}
\end{aligned} ( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = g ∑ ( − 1 ) d + b + 2 g Π c 2 Π g Π c { a e b d c g } ( { P a ⊗ R d } g ⋅ { Q b ⊗ S e } g ) + q ∑ ( − 1 ) d + e + c − b Π a Π cc q { a d b e c q } ( P a ⋅ { R q b d ⊗ S e } a ) (534)
{ { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = ∑ h g ( − 1 ) a − b + k − d − g Π h c f g { a b c f k h } { b e g d h f } { P a ⊗ { R d ⊗ { Q b ⊗ S e } g } h } k + ∑ h q ( − 1 ) a + k + d + e − f − h Π h c q f { a b c f k h } { d e f h b q } { P a ⊗ { R q b d ⊗ S e } h } k (535) \begin{aligned}
\left\{\left\{\mathbf{P}_a\otimes\mathbf{Q}_b\right\}_c\otimes\left\{\mathbf{R}_d\otimes\mathbf{S}_e\right\}_f\right\}_k
={}&\sum_{hg}(-1)^{a-b+k-d-g}\Pi_{hcf g}\left\{\begin{matrix}a&b&c\\f&k&h\end{matrix}\right\}\left\{\begin{matrix}b&e&g\\d&h&f\end{matrix}\right\}\left\{\mathbf{P}_a\otimes\left\{\mathbf{R}_d\otimes\left\{\mathbf{Q}_b\otimes\mathbf{S}_e\right\}_g\right\}_h\right\}_k\\
&+\sum_{hq}(-1)^{a+k+d+e-f-h}\Pi_{hcqf}\left\{\begin{matrix}a&b&c\\f&k&h\end{matrix}\right\}\left\{\begin{matrix}d&e&f\\h&b&q\end{matrix}\right\}\left\{\mathbf{P}_a\otimes\left\{\mathcal{R}^{bd}_q\otimes\mathbf{S}_e\right\}_h\right\}_k \tag*{(535)}
\end{aligned} { { P a ⊗ Q b } c ⊗ { R d ⊗ S e } f } k = h g ∑ ( − 1 ) a − b + k − d − g Π h c f g { a f b k c h } { b d e h g f } { P a ⊗ { R d ⊗ { Q b ⊗ S e } g } h } k + h q ∑ ( − 1 ) a + k + d + e − f − h Π h c q f { a f b k c h } { d h e b f q } { P a ⊗ { R q b d ⊗ S e } h } k (535)
( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = ∑ g ( − 1 ) a + d + g + b Π c c g Π a { b e g d a c } ( P a ⋅ { R d ⊗ { Q b ⊗ S e } g } a ) + ∑ q ( − 1 ) d + e + c − b Π c c q Π a { a b c d e q } ( P a ⋅ { R q b d ⊗ S e } a ) (536) \begin{aligned}
\left(\left\{\mathbf{P}_a\otimes\mathbf{Q}_b\right\}_c\cdot\left\{\mathbf{R}_d\otimes\mathbf{S}_e\right\}_c\right)
={}&\sum_g(-1)^{a+d+g+b}\frac{\Pi_{ccg}}{\Pi_a}\left\{\begin{matrix}b&e&g\\d&a&c\end{matrix}\right\}\left(\mathbf{P}_a\cdot\left\{\mathbf{R}_d\otimes\left\{\mathbf{Q}_b\otimes\mathbf{S}_e\right\}_g\right\}_a\right)\\
&+\sum_q(-1)^{d+e+c-b}\frac{\Pi_{ccq}}{\Pi_a}\left\{\begin{matrix}a&b&c\\d&e&q\end{matrix}\right\}\left(\mathbf{P}_a\cdot\left\{\mathcal{R}^{bd}_q\otimes\mathbf{S}_e\right\}_a\right) \tag*{(536)}
\end{aligned} ( { P a ⊗ Q b } c ⋅ { R d ⊗ S e } c ) = g ∑ ( − 1 ) a + d + g + b Π a Π cc g { b d e a g c } ( P a ⋅ { R d ⊗ { Q b ⊗ S e } g } a ) + q ∑ ( − 1 ) d + e + c − b Π a Π cc q { a d b e c q } ( P a ⋅ { R q b d ⊗ S e } a ) (536)
The definition of the commutator R q b d \mathcal{R}^{bd}_q R q b d is given in Eq. (460) . By making use of the foregoing formulas, one can obtain all other relations for irreducible products of non-commuting tensors.
Wigner D-functions
Definition of D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
(a) The Wigner D D D -functions D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) may be defined as the matrix elements of the rotation operator D ^ ( α , β , γ ) \hat{D}(\alpha,\beta,\gamma) D ^ ( α , β , γ ) in the JM-representation. The arguments α , β , γ \alpha,\beta,\gamma α , β , γ are the Euler angles which specify the rotation
⟨ J M ∣ D ^ ( α , β , γ ) ∣ J ′ M ′ ⟩ = δ J J ′ D M M ′ J ( α , β , γ ) . (537) \langle JM|\hat{D}(\alpha,\beta,\gamma)|J'M'\rangle=\delta_{JJ'}D^J_{MM'}(\alpha,\beta,\gamma). \tag*{(537)} ⟨ J M ∣ D ^ ( α , β , γ ) ∣ J ′ M ′ ⟩ = δ J J ′ D M M ′ J ( α , β , γ ) . (537)
The D D D -functions realize transformations of covariant components of any irreducible tensor of rank J J J (e.g., the wave function Ψ J M \Psi_{JM} Ψ J M of a quantum mechanical system with angular momentum J J J and its projection M M M ) under coordinate rotations.
Ψ J M ′ ( ϑ ′ , φ ′ , σ ′ ) = ∑ M = − J J Ψ J M ( ϑ , φ , σ ) D M M ′ J ( α , β , γ ) , Ψ J M ′ ∗ ( ϑ ′ , φ ′ , σ ′ ) = ∑ M = − J J Ψ J M ∗ ( ϑ , φ , σ ) D M M ′ J ∗ ( α , β , γ ) . (538) \begin{aligned}
\Psi_{JM'}(\vartheta',\varphi',\sigma')&=\sum_{M=-J}^{J}\Psi_{JM}(\vartheta,\varphi,\sigma)D^J_{MM'}(\alpha,\beta,\gamma),\\
\Psi^*_{JM'}(\vartheta',\varphi',\sigma')&=\sum_{M=-J}^{J}\Psi^*_{JM}(\vartheta,\varphi,\sigma)D^{J*}_{MM'}(\alpha,\beta,\gamma). \tag*{(538)}
\end{aligned} Ψ J M ′ ( ϑ ′ , φ ′ , σ ′ ) Ψ J M ′ ∗ ( ϑ ′ , φ ′ , σ ′ ) = M = − J ∑ J Ψ J M ( ϑ , φ , σ ) D M M ′ J ( α , β , γ ) , = M = − J ∑ J Ψ J M ∗ ( ϑ , φ , σ ) D M M ′ J ∗ ( α , β , γ ) . (538)
Here ϑ , φ \vartheta,\varphi ϑ , φ and ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ are polar angles in the initial and rotated coordinate systems, S S S and S ′ S' S ′ , respectively. The angles ϑ , φ \vartheta,\varphi ϑ , φ and ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ are related by Eqs. (131) and (1.4.3). Similarly, σ \sigma σ and σ ′ \sigma' σ ′ are spin variables in the initial and new systems.
The inverse transformation S ′ → S S'\to S S ′ → S is performed by the inverse matrix [ D ^ − 1 ( α , β , γ ) ] M M ′ J \left[\hat{D}^{-1}(\alpha,\beta,\gamma)\right]^J_{MM'} [ D ^ − 1 ( α , β , γ ) ] M M ′ J . Owing to the unitarity of the rotation operator,
D ^ − 1 ( α , β , γ ) = D ^ + ( α , β , γ ) . (539) \hat{D}^{-1}(\alpha,\beta,\gamma)=\hat{D}^{+}(\alpha,\beta,\gamma). \tag*{(539)} D ^ − 1 ( α , β , γ ) = D ^ + ( α , β , γ ) . (539)
the elements of inverse matrix are given by
[ D ^ − 1 ( α , β , γ ) ] M M ′ J = D M ′ M J ∗ ( α , β , γ ) . (540) \left[\hat{D}^{-1}(\alpha,\beta,\gamma)\right]^J_{MM'}=D^{J*}_{M'M}(\alpha,\beta,\gamma). \tag*{(540)} [ D ^ − 1 ( α , β , γ ) ] M M ′ J = D M ′ M J ∗ ( α , β , γ ) . (540)
Hence, under the inverse rotation, S ′ → S S'\to S S ′ → S , wave functions transform as
Ψ J M ( ϑ , φ , σ ) = ∑ M ′ = − J J D M M ′ J ∗ ( α , β , γ ) Ψ J M ′ ( ϑ ′ , φ ′ , σ ′ ) , Ψ J M ∗ ( ϑ , φ , σ ) = ∑ M ′ = − J J D M M ′ J ( α , β , γ ) Ψ J M ′ ∗ ( ϑ ′ , φ ′ , σ ′ ) . (541) \begin{aligned}
\Psi_{JM}(\vartheta,\varphi,\sigma)&=\sum_{M'=-J}^{J}D^{J*}_{MM'}(\alpha,\beta,\gamma)\Psi_{JM'}(\vartheta',\varphi',\sigma'),\\
\Psi^*_{JM}(\vartheta,\varphi,\sigma)&=\sum_{M'=-J}^{J}D^J_{MM'}(\alpha,\beta,\gamma)\Psi^*_{JM'}(\vartheta',\varphi',\sigma'). \tag*{(541)}
\end{aligned} Ψ J M ( ϑ , φ , σ ) Ψ J M ∗ ( ϑ , φ , σ ) = M ′ = − J ∑ J D M M ′ J ∗ ( α , β , γ ) Ψ J M ′ ( ϑ ′ , φ ′ , σ ′ ) , = M ′ = − J ∑ J D M M ′ J ( α , β , γ ) Ψ J M ′ ∗ ( ϑ ′ , φ ′ , σ ′ ) . (541)
The unitarity condition for the Wigner D D D -functions may be written as
∑ M = − J J D M M ′ J ( α , β , γ ) D M M ~ ′ J ∗ ( α , β , γ ) = δ M ′ M ~ ′ , ∑ M ′ = − J J D M M ′ J ∗ ( α , β , γ ) D M ~ ′ M ′ J ( α , β , γ ) = δ M M ~ . (542) \begin{aligned}
\sum_{M=-J}^{J}D^J_{MM'}(\alpha,\beta,\gamma)D^{J*}_{M\widetilde{M}'}(\alpha,\beta,\gamma)&=\delta_{M'\widetilde{M}'},\\
\sum_{M'=-J}^{J}D^{J*}_{MM'}(\alpha,\beta,\gamma)D^J_{\widetilde{M}'M'}(\alpha,\beta,\gamma)&=\delta_{M\widetilde{M}}. \tag*{(542)}
\end{aligned} M = − J ∑ J D M M ′ J ( α , β , γ ) D M M ′ J ∗ ( α , β , γ ) M ′ = − J ∑ J D M M ′ J ∗ ( α , β , γ ) D M ′ M ′ J ( α , β , γ ) = δ M ′ M ′ , = δ M M . (542)
The matrix D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) is unimodular, i.e.,
det ∥ D M M ′ J ( α , β , γ ) ∥ = + 1. (543) \det\left\|D^J_{MM'}(\alpha,\beta,\gamma)\right\|=+1. \tag*{(543)} det D M M ′ J ( α , β , γ ) = + 1. (543)
(b) A set of ( 2 J + 1 ) (2J+1) ( 2 J + 1 ) functions Ψ J M \Psi_{JM} Ψ J M with different M M M ’s constitute a basis for expansion of an arbitrary function Ψ J \Psi_J Ψ J with the same J J J :
Ψ J ( ϑ , φ ) = ∑ M = − J J C J M Ψ J M ( ϑ , φ ) = ( C J ⋅ Ψ J ) . (544) \Psi_J(\vartheta,\varphi)=\sum_{M=-J}^{J}C_J^M\Psi_{JM}(\vartheta,\varphi)=({\boldsymbol C}_J\cdot{\boldsymbol\Psi}_J). \tag*{(544)} Ψ J ( ϑ , φ ) = M = − J ∑ J C J M Ψ J M ( ϑ , φ ) = ( C J ⋅ Ψ J ) . (544)
The expansion coefficients C J M C_J^M C J M are contravariant components of some irreducible tensor of rank J J J . Under rotations the quantities C J M C_J^M C J M transform by means of functions D M M ′ J ∗ ( α , β , γ ) D_{MM'}^{J*}(\alpha,\beta,\gamma) D M M ′ J ∗ ( α , β , γ ) .
The effect of the operator D ^ ( α , β , γ ) \hat{D}(\alpha,\beta,\gamma) D ^ ( α , β , γ ) on Ψ J \Psi_J Ψ J may be interpreted in two different ways:
as a rotation of the coordinate system without rotation of the physical body (this is the passive interpretation; D ^ \hat{D} D ^ acts on the basis functions Ψ J M \Psi_{JM} Ψ J M while C J M C_J^M C J M remain unchanged);
as a rotation of the physical body without rotation of coordinate system (active interpretation; D ^ \hat{D} D ^ acts on C J M C_J^M C J M but does not affect Ψ J M \Psi_{JM} Ψ J M ).
Any rotation of a physical body in combination with the same rotation of coordinate system leaves the wave function Ψ J \Psi_J Ψ J unchanged:
{ D ^ ( α , β , γ ) } p h y s . b o d y ⋅ { D ^ ( α , β , γ ) } c o o r d . s y s t e m = 1 , (545) \{\hat{D}(\alpha,\beta,\gamma)\}_{\mathrm{phys.\ body}}\cdot\{\hat{D}(\alpha,\beta,\gamma)\}_{\mathrm{coord.\ system}}=1, \tag*{(545)} { D ^ ( α , β , γ ) } phys. body ⋅ { D ^ ( α , β , γ ) } coord. system = 1 , (545)
i.e.,
{ D ^ M M ′ J ( α , β , γ ) } p h y s . b o d y = { [ D ^ − 1 ( α , β , γ ) ] M M ′ J } c o o r d . s y s t e m . (546) \{\hat{D}_{MM'}^J(\alpha,\beta,\gamma)\}_{\mathrm{phys.\ body}}=\{[\hat{D}^{-1}(\alpha,\beta,\gamma)]_{MM'}^J\}_{\mathrm{coord.\ system}}. \tag*{(546)} { D ^ M M ′ J ( α , β , γ ) } phys. body = {[ D ^ − 1 ( α , β , γ ) ] M M ′ J } coord. system . (546)
Moreover, a rotation of the coordinate system (or physical body) described by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ may also be realized in two ways:
by rotating about the initial axes (case B in § 9.9.1 ), or
by rotating about the new (turned) axes (case A in § 9.9.1 ).
Thus, any transformation of wave functions described by Eq. (538) can be treated in four different ways (Table 4.1 ).
Table 4.1. Effect of the Operator D ^ ( α , β , γ ) \hat{D}(\alpha,\beta,\gamma) D ^ ( α , β , γ )
Differential equations for DM MJ(, , )
The Wigner D D D -functions are complex. They depend on three real arguments α , β , γ \alpha,\beta,\gamma α , β , γ and are defined in the domain
0 ≤ α < 2 π , 0 ≤ β ≤ π , 0 ≤ γ < 2 π . (547) 0\leq\alpha<2\pi,\qquad0\leq\beta\leq\pi,\qquad0\leq\gamma<2\pi. \tag*{(547)} 0 ≤ α < 2 π , 0 ≤ β ≤ π , 0 ≤ γ < 2 π . (547)
These functions, as well as their derivatives, are single-valued, finite and continuous. Sometimes it is convenient to change the domain (4.1.11). This can be done using the symmetries of D M M ′ J ( α , β , γ ) D_{MM'}^J(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) (Sec. 4.4). For example, the matrix of the inverse rotation satisfies the equation
[ D − 1 ( α , β , γ ) ] M M ′ J = D M M ′ J ( π − γ , β , − π − α ) = D M M ′ J ( − γ , − β , − α ) . (548) \left[D^{-1}(\alpha,\beta,\gamma)\right]_{MM'}^J=D_{MM'}^J(\pi-\gamma,\beta,-\pi-\alpha)=D_{MM'}^J(-\gamma,-\beta,-\alpha). \tag*{(548)} [ D − 1 ( α , β , γ ) ] M M ′ J = D M M ′ J ( π − γ , β , − π − α ) = D M M ′ J ( − γ , − β , − α ) . (548)
This means that the inverse transformation S ′ → S S'\to S S ′ → S may be realized by the Euler angles
α ′ = π − γ , β ′ = β , γ ′ = − π − α , (549) \alpha'=\pi-\gamma,\qquad\beta'=\beta,\qquad\gamma'=-\pi-\alpha, \tag*{(549)} α ′ = π − γ , β ′ = β , γ ′ = − π − α , (549)
as well as by
α ′ = − γ , β ′ = − β , γ ′ = − α . (550) \alpha'=-\gamma,\qquad\beta'=-\beta,\qquad\gamma'=-\alpha. \tag*{(550)} α ′ = − γ , β ′ = − β , γ ′ = − α . (550)
Differential equations for D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
(a) The Wigner D D D -functions represent wave functions of a rigid symmetric top. They are eigenfunctions of three operators
J ^ z = − i ∂ ∂ α , J ^ z ′ = − i ∂ ∂ γ , J ^ 2 = − [ ∂ 2 ∂ β 2 + cot β ∂ ∂ β + 1 sin 2 β ( ∂ 2 ∂ α 2 − 2 cos β ∂ 2 ∂ α ∂ γ + ∂ 2 ∂ γ 2 ) ] (551) \hat{J}_z=-i\frac{\partial}{\partial\alpha},\qquad\hat{J}'_z=-i\frac{\partial}{\partial\gamma},\qquad\hat{J}^2=-\left[\frac{\partial^2}{\partial\beta^2}+\cot\beta\frac{\partial}{\partial\beta}+\frac{1}{\sin^2\beta}\left(\frac{\partial^2}{\partial\alpha^2}-2\cos\beta\frac{\partial^2}{\partial\alpha\partial\gamma}+\frac{\partial^2}{\partial\gamma^2}\right)\right] \tag*{(551)} J ^ z = − i ∂ α ∂ , J ^ z ′ = − i ∂ γ ∂ , J ^ 2 = − [ ∂ β 2 ∂ 2 + cot β ∂ β ∂ + sin 2 β 1 ( ∂ α 2 ∂ 2 − 2 cos β ∂ α ∂ γ ∂ 2 + ∂ γ 2 ∂ 2 ) ] (551)
where J ^ \hat{J} J ^ is the operator of angular momentum of the top; J ^ z ′ \hat{J}'_z J ^ z ′ and J ^ z \hat{J}_z J ^ z are projections of J ^ \hat{J} J ^ onto the z z z -axis of the rotating (body-fixed) and non-rotating (lab-fixed) coordinate systems, respectively. The eigenvalues of the operators (4.2.1) are defined by the equations
J ^ x D M M ′ J ( α , β , γ ) = − M D M M ′ J ( α , β , γ ) , \hat{J}_xD^J_{MM'}(\alpha,\beta,\gamma)=-MD^J_{MM'}(\alpha,\beta,\gamma), J ^ x D M M ′ J ( α , β , γ ) = − M D M M ′ J ( α , β , γ ) ,
J ^ z D M M ′ J ( α , β , γ ) = − M ′ D M M ′ J ( α , β , γ ) , \hat{J}_zD^J_{MM'}(\alpha,\beta,\gamma)=-M'D^J_{MM'}(\alpha,\beta,\gamma), J ^ z D M M ′ J ( α , β , γ ) = − M ′ D M M ′ J ( α , β , γ ) ,
J ^ 2 D M M ′ J ( α , β , γ ) = { − 1 sin β ∂ ∂ β ( sin β ∂ ∂ β ) + M 2 − 2 M M ′ cos β + M ′ 2 sin 2 β } D M M ′ J ( α , β , γ ) = J ( J + 1 ) D M M ′ J ( α , β , γ ) . (552) \begin{aligned}
\hat{J}^2D^J_{MM'}(\alpha,\beta,\gamma)=\left\{-\frac{1}{\sin\beta}\frac{\partial}{\partial\beta}\left(\sin\beta\frac{\partial}{\partial\beta}\right)+\frac{M^2-2MM'\cos\beta+M'^2}{\sin^2\beta}\right\}D^J_{MM'}(\alpha,\beta,\gamma) \\
&=J(J+1)D^J_{MM'}(\alpha,\beta,\gamma). \tag*{(552)}
\end{aligned} J ^ 2 D M M ′ J ( α , β , γ ) = { − sin β 1 ∂ β ∂ ( sin β ∂ β ∂ ) + sin 2 β M 2 − 2 M M ′ cos β + M ′2 } D M M ′ J ( α , β , γ ) = J ( J + 1 ) D M M ′ J ( α , β , γ ) . (552)
Periodicity conditions for D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) are as follows
D M M ′ J ( α ± 2 k π , β , γ ) = D M M ′ J ( α , β , γ ) , D^J_{MM'}(\alpha\pm2k\pi,\beta,\gamma)=D^J_{MM'}(\alpha,\beta,\gamma), D M M ′ J ( α ± 2 k π , β , γ ) = D M M ′ J ( α , β , γ ) ,
D M M ′ J ( α , β , γ ± 2 k π ) = D M M ′ J ( α , β , γ ) , D^J_{MM'}(\alpha,\beta,\gamma\pm2k\pi)=D^J_{MM'}(\alpha,\beta,\gamma), D M M ′ J ( α , β , γ ± 2 k π ) = D M M ′ J ( α , β , γ ) ,
D M M ′ J ( α , β ± 2 k π , γ ) = D M M ′ J ( α , β , γ ) . (553) D^J_{MM'}(\alpha,\beta\pm2k\pi,\gamma)=D^J_{MM'}(\alpha,\beta,\gamma). \tag*{(553)} D M M ′ J ( α , β ± 2 k π , γ ) = D M M ′ J ( α , β , γ ) . (553)
where
k = 0 , 1 , 2 , … if J is integer , k=0,1,2,\ldots\qquad\text{if }J\text{ is integer}, k = 0 , 1 , 2 , … if J is integer ,
k = 0 , 2 , 4 , … if J is half-integer . k=0,2,4,\ldots\qquad\text{if }J\text{ is half-integer}. k = 0 , 2 , 4 , … if J is half-integer .
(b) The functions D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) can also be defined as solutions of the differential equations
[ J ^ ν , D M M ′ J ( α , β , γ ) ] = J ^ ν D M M ′ J ( α , β , γ ) = ( − 1 ) 1 + ν J ( J + 1 ) C J M − ν J M − ν D M − ν M ′ J ( α , β , γ ) = { − M D M M ′ J ( α , β , γ ) , ν = 0 , ± J ( J + 1 ) − M ( M ∓ 1 ) 2 D M ∓ 1 M ′ J ( α , β , γ ) , ν = ± 1 , (554) \begin{aligned}
\left[\hat{J}_\nu,D^J_{MM'}(\alpha,\beta,\gamma)\right]=\hat{J}_\nu D^J_{MM'}(\alpha,\beta,\gamma)=(-1)^{1+\nu}\sqrt{J(J+1)}C^{JM-\nu}_{JM-\nu}D^J_{M-\nu M'}(\alpha,\beta,\gamma) \\
&=\begin{cases}
-MD^J_{MM'}(\alpha,\beta,\gamma), & \nu=0,\\
\pm\sqrt{\dfrac{J(J+1)-M(M\mp1)}{2}}D^J_{M\mp1M'}(\alpha,\beta,\gamma), & \nu=\pm1,
\end{cases} \tag*{(554)}
\end{aligned} [ J ^ ν , D M M ′ J ( α , β , γ ) ] = J ^ ν D M M ′ J ( α , β , γ ) = ( − 1 ) 1 + ν J ( J + 1 ) C J M − ν J M − ν D M − ν M ′ J ( α , β , γ ) = ⎩ ⎨ ⎧ − M D M M ′ J ( α , β , γ ) , ± 2 J ( J + 1 ) − M ( M ∓ 1 ) D M ∓ 1 M ′ J ( α , β , γ ) , ν = 0 , ν = ± 1 , (554)
[ J ^ ν ′ , D M M ′ J ( α , β , γ ) ] = J ^ ν ′ D M M ′ J ( α , β , γ ) = − J ( J + 1 ) C J M ′ + ν J M ′ + ν D M M ′ + ν J ( α , β , γ ) = { − M ′ D M M ′ J ( α , β , γ ) , ν = 0 , ± J ( J + 1 ) − M ′ ( M ′ ± 1 ) 2 D M M ′ ± 1 J ( α , β , γ ) , ν = ± 1. (555) \begin{aligned}
\left[\hat{J}'_\nu,D^J_{MM'}(\alpha,\beta,\gamma)\right]=\hat{J}'_\nu D^J_{MM'}(\alpha,\beta,\gamma)=-\sqrt{J(J+1)}C^{JM'+\nu}_{JM'+\nu}D^J_{MM'+\nu}(\alpha,\beta,\gamma) \\
&=\begin{cases}
-M'D^J_{MM'}(\alpha,\beta,\gamma), & \nu=0,\\
\pm\sqrt{\dfrac{J(J+1)-M'(M'\pm1)}{2}}D^J_{MM'\pm1}(\alpha,\beta,\gamma), & \nu=\pm1.
\end{cases} \tag*{(555)}
\end{aligned} [ J ^ ν ′ , D M M ′ J ( α , β , γ ) ] = J ^ ν ′ D M M ′ J ( α , β , γ ) = − J ( J + 1 ) C J M ′ + ν J M ′ + ν D M M ′ + ν J ( α , β , γ ) = ⎩ ⎨ ⎧ − M ′ D M M ′ J ( α , β , γ ) , ± 2 J ( J + 1 ) − M ′ ( M ′ ± 1 ) D M M ′ ± 1 J ( α , β , γ ) , ν = 0 , ν = ± 1. (555)
Here J ^ ν \hat{J}_\nu J ^ ν is a covariant spherical component of J ^ \hat{J} J ^ in the non-rotating (lab-fixed) system
J ^ ± 1 = i 2 e ± i α [ ∓ cot β ∂ ∂ α + i ∂ ∂ β ± 1 sin β ∂ ∂ γ ] \hat{J}_{\pm1}=\frac{i}{\sqrt{2}}e^{\pm i\alpha}\left[\mp\cot\beta\frac{\partial}{\partial\alpha}+i\frac{\partial}{\partial\beta}\pm\frac{1}{\sin\beta}\frac{\partial}{\partial\gamma}\right] J ^ ± 1 = 2 i e ± i α [ ∓ cot β ∂ α ∂ + i ∂ β ∂ ± sin β 1 ∂ γ ∂ ]
J ^ 0 = − i ∂ ∂ α . (556) \hat{J}_0=-i\frac{\partial}{\partial\alpha}. \tag*{(556)} J ^ 0 = − i ∂ α ∂ . (556)
and J ^ ′ ν \hat{J}'^\nu J ^ ′ ν is a contravariant component of J ^ \hat{J} J ^ in the rotating (body-fixed) system
J ^ ′ ± 1 = i 2 e ∓ i γ [ ± cot β ∂ ∂ γ + i ∂ ∂ β ± 1 sin β ∂ ∂ α ] \hat{J}'^{\pm1}=\frac{i}{\sqrt{2}}e^{\mp i\gamma}\left[\pm\cot\beta\frac{\partial}{\partial\gamma}+i\frac{\partial}{\partial\beta}\pm\frac{1}{\sin\beta}\frac{\partial}{\partial\alpha}\right] J ^ ′ ± 1 = 2 i e ∓ iγ [ ± cot β ∂ γ ∂ + i ∂ β ∂ ± sin β 1 ∂ α ∂ ]
J ^ ′ 0 = − i ∂ ∂ γ . (557) \hat{J}'^0=-i\frac{\partial}{\partial\gamma}. \tag*{(557)} J ^ ′0 = − i ∂ γ ∂ . (557)
The operator J ^ 2 \hat{J}^2 J ^ 2 can be expressed in terms of J ^ ν \hat{J}_\nu J ^ ν or J ^ ′ ν \hat{J}'^\nu J ^ ′ ν as
J ^ 2 = − J ^ − 1 J ^ + 1 + J ^ 0 J ^ 0 − J ^ + 1 J ^ − 1 = − J ^ − 1 J ^ + 1 + J ^ 0 J ^ 0 − J ^ + 1 J ^ − 1 = J ^ 0 ( J ^ 0 + 1 ) − 2 J ^ − 1 J ^ + 1 = J ^ 0 ( J ^ 0 − 1 ) − 2 J ^ − 1 J ^ + 1 = J ^ 0 ( J ^ 0 − 1 ) − 2 J ^ + 1 J ^ − 1 = J ^ ′ 0 ( J ^ ′ 0 + 1 ) − 2 J ^ ′ + 1 J ^ ′ − 1 . (558) \begin{aligned}
\hat{J}^2=-\hat{J}_{-1}\hat{J}_{+1}+\hat{J}_0\hat{J}_0-\hat{J}_{+1}\hat{J}_{-1}=-\hat{J}^{-1}\hat{J}^{+1}+\hat{J}^0\hat{J}^0-\hat{J}^{+1}\hat{J}^{-1}=\hat{J}_0(\hat{J}_0+1)-2\hat{J}_{-1}\hat{J}_{+1} \\
&=\hat{J}^0(\hat{J}^0-1)-2\hat{J}^{-1}\hat{J}^{+1}=\hat{J}_0(\hat{J}_0-1)-2\hat{J}_{+1}\hat{J}_{-1}=\hat{J}'^0(\hat{J}'^0+1)-2\hat{J}'^{+1}\hat{J}'^{-1}. \tag*{(558)}
\end{aligned} J ^ 2 = − J ^ − 1 J ^ + 1 + J ^ 0 J ^ 0 − J ^ + 1 J ^ − 1 = − J ^ − 1 J ^ + 1 + J ^ 0 J ^ 0 − J ^ + 1 J ^ − 1 = J ^ 0 ( J ^ 0 + 1 ) − 2 J ^ − 1 J ^ + 1 = J ^ 0 ( J ^ 0 − 1 ) − 2 J ^ − 1 J ^ + 1 = J ^ 0 ( J ^ 0 − 1 ) − 2 J ^ + 1 J ^ − 1 = J ^ ′0 ( J ^ ′0 + 1 ) − 2 J ^ ′ + 1 J ^ ′ − 1 . (558)
The relationships between J ^ ν \hat{J}_{\nu} J ^ ν and J ^ ′ μ \hat{J}^{\prime\mu} J ^ ′ μ read
J ^ ′ μ ( α , β , γ ) = ∑ ν D ν μ 1 ( α , β , γ ) J ^ ν ( α , β , γ ) J ^ ν ( α , β , γ ) = ∑ μ D ν μ 1 ∗ ( α , β , γ ) J ^ ′ μ ( α , β , γ ) (559) \begin{aligned}
\hat{J}^{\prime\mu}(\alpha,\beta,\gamma)&=\sum_{\nu}D^{1}_{\nu\mu}(\alpha,\beta,\gamma)\hat{J}_{\nu}(\alpha,\beta,\gamma)\\
\hat{J}_{\nu}(\alpha,\beta,\gamma)&=\sum_{\mu}D^{1*}_{\nu\mu}(\alpha,\beta,\gamma)\hat{J}^{\prime\mu}(\alpha,\beta,\gamma) \tag*{(559)}
\end{aligned} J ^ ′ μ ( α , β , γ ) J ^ ν ( α , β , γ ) = ν ∑ D ν μ 1 ( α , β , γ ) J ^ ν ( α , β , γ ) = μ ∑ D ν μ 1 ∗ ( α , β , γ ) J ^ ′ μ ( α , β , γ ) (559)
Hence
J ^ ′ ν ( α , β , γ ) = − J ^ ν ( − γ , − β , − α ) , J ^ ′ ν ( α , β , γ ) = J ^ − ν ( γ , β , α ) . (560) \hat{J}^{\prime\nu}(\alpha,\beta,\gamma)=-\hat{J}_{\nu}(-\gamma,-\beta,-\alpha),\qquad\hat{J}^{\prime\nu}(\alpha,\beta,\gamma)=\hat{J}_{-\nu}(\gamma,\beta,\alpha). \tag*{(560)} J ^ ′ ν ( α , β , γ ) = − J ^ ν ( − γ , − β , − α ) , J ^ ′ ν ( α , β , γ ) = J ^ − ν ( γ , β , α ) . (560)
Commutators of spherical components J ^ μ \hat{J}_{\mu} J ^ μ and J ^ ′ ν \hat{J}^{\prime\nu} J ^ ′ ν (μ , ν = − 1 , 0 , 1 \mu,\nu=-1,0,1 μ , ν = − 1 , 0 , 1 ) are given by
[ J ^ μ , J ^ ν ] = − 2 C 1 μ 1 ν 1 μ + ν J ^ μ + ν [ J ^ ′ μ , J ^ ′ ν ] = 2 C 1 μ 1 ν 1 μ + ν J ^ ′ μ + ν [ J ^ μ , J ^ ′ ν ] = 0 (561) \begin{aligned}
[\hat{J}_{\mu},\hat{J}_{\nu}]&=-\sqrt{2}C^{1\mu+\nu}_{1\mu1\nu}\hat{J}_{\mu+\nu}\\
[\hat{J}^{\prime\mu},\hat{J}^{\prime\nu}]&=\sqrt{2}C^{1\mu+\nu}_{1\mu1\nu}\hat{J}^{\prime\mu+\nu}\\
[\hat{J}_{\mu},\hat{J}^{\prime\nu}]&=0 \tag*{(561)}
\end{aligned} [ J ^ μ , J ^ ν ] [ J ^ ′ μ , J ^ ′ ν ] [ J ^ μ , J ^ ′ ν ] = − 2 C 1 μ 1 ν 1 μ + ν J ^ μ + ν = 2 C 1 μ 1 ν 1 μ + ν J ^ ′ μ + ν = 0 (561)
The commutators of cartesian components J ^ k \hat{J}_{k} J ^ k and J ^ l \hat{J}_{l} J ^ l (i , k , l = x , y , z i,k,l=x,y,z i , k , l = x , y , z ) are
[ J ^ i , J ^ k ] = i ϵ i k l J ^ l , [ J ^ i ′ , J ^ k ′ ] = i ϵ i k l J ^ l ′ , [ J ^ i , J ^ k ′ ] = 0. (562) \begin{aligned}
[\hat{J}_{i},\hat{J}_{k}]&=i\epsilon_{ikl}\hat{J}_{l},\\
[\hat{J}^{\prime}_{i},\hat{J}^{\prime}_{k}]&=i\epsilon_{ikl}\hat{J}^{\prime}_{l},\\
[\hat{J}_{i},\hat{J}^{\prime}_{k}]&=0. \tag*{(562)}
\end{aligned} [ J ^ i , J ^ k ] [ J ^ i ′ , J ^ k ′ ] [ J ^ i , J ^ k ′ ] = i ϵ ik l J ^ l , = i ϵ ik l J ^ l ′ , = 0. (562)
The operator of orbital angular momentum L ^ ( ϑ , φ ) \hat{L}(\vartheta,\varphi) L ^ ( ϑ , φ ) (§ 10.2 ) may be regarded as a special case of J ^ ( α , β , γ ) \hat{J}(\alpha,\beta,\gamma) J ^ ( α , β , γ ) , viz. at α = φ \alpha=\varphi α = φ , β = ϑ \beta=\vartheta β = ϑ , γ = 0 \gamma=0 γ = 0
J ^ ν ( φ , ϑ , 0 ) = L ^ ν ( ϑ , φ ) , ν = − 1 , 0 , 1 , J ^ i ( φ , ϑ , 0 ) = L ^ i ( ϑ , φ ) , i = x , y , z . (563) \begin{aligned}
\hat{J}_{\nu}(\varphi,\vartheta,0)&=\hat{L}_{\nu}(\vartheta,\varphi),\qquad\nu=-1,0,1,\\
\hat{J}_{i}(\varphi,\vartheta,0)&=\hat{L}_{i}(\vartheta,\varphi),\qquad i=x,y,z. \tag*{(563)}
\end{aligned} J ^ ν ( φ , ϑ , 0 ) J ^ i ( φ , ϑ , 0 ) = L ^ ν ( ϑ , φ ) , ν = − 1 , 0 , 1 , = L ^ i ( ϑ , φ ) , i = x , y , z . (563)
at α = 0 \alpha=0 α = 0 , β = ϑ \beta=\vartheta β = ϑ , γ = φ \gamma=\varphi γ = φ
J ^ ′ ν ( 0 , ϑ , φ ) = L ^ ′ ν ( ϑ , φ ) , ν = − 1 , 0 , 1 , J ^ i ′ ( 0 , ϑ , φ ) = L ^ i ′ ( ϑ , φ ) , i = x , y , z . (564) \begin{aligned}
\hat{J}^{\prime\nu}(0,\vartheta,\varphi)&=\hat{L}^{\prime\nu}(\vartheta,\varphi),\qquad\nu=-1,0,1,\\
\hat{J}^{\prime}_{i}(0,\vartheta,\varphi)&=\hat{L}^{\prime}_{i}(\vartheta,\varphi),\qquad i=x,y,z. \tag*{(564)}
\end{aligned} J ^ ′ ν ( 0 , ϑ , φ ) J ^ i ′ ( 0 , ϑ , φ ) = L ^ ′ ν ( ϑ , φ ) , ν = − 1 , 0 , 1 , = L ^ i ′ ( ϑ , φ ) , i = x , y , z . (564)
(c) Equations (4.2.2) define D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) only within normalization and phase factors. To fix these factors some additional conditions are required. In the case of diagonal elements of the rotation matrix these factors are completely determined by the boundary condition
D M M ′ J ( 0 , 0 , 0 ) = δ M M ′ . (565) D^{J}_{MM'}(0,0,0)=\delta_{MM'}. \tag*{(565)} D M M ′ J ( 0 , 0 , 0 ) = δ M M ′ . (565)
As for non-diagonal elements, they are determined by Eqs. (4.2.4) and (4.2.5) which relate the D M M ′ J D^{J}_{MM'} D M M ′ J of different M M M and M ′ M' M ′ . This phase convention corresponds to the condition
D M M ′ J ( 0 , π , 0 ) = ( − 1 ) J + M δ M , − M ′ = ( − 1 ) J − M ′ δ − M , M ′ . (566) D^{J}_{MM'}(0,\pi,0)=(-1)^{J+M}\delta_{M,-M'}=(-1)^{J-M'}\delta_{-M,M'}. \tag*{(566)} D M M ′ J ( 0 , π , 0 ) = ( − 1 ) J + M δ M , − M ′ = ( − 1 ) J − M ′ δ − M , M ′ . (566)
Explicit forms of the Wigner D-functions
D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) may be represented as a product of three functions, each of which depends only on one argument α \alpha α , β \beta β or γ \gamma γ ,
D M M ′ J ( α , β , γ ) = e − i M α d M M ′ J ( β ) e − i M ′ γ . (567) D^{J}_{MM'}(\alpha,\beta,\gamma)=e^{-iM\alpha}d^{J}_{MM'}(\beta)e^{-iM'\gamma}. \tag*{(567)} D M M ′ J ( α , β , γ ) = e − i M α d M M ′ J ( β ) e − i M ′ γ . (567)
where d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) is a real function whose explicit forms are given below.
Expressions for d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) Involving Trigonometric Functions
d M M ′ J ( β ) = ( − 1 ) J − M ′ [ ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ] 1 2 ∑ k ( − 1 ) k ( cos β 2 ) M + M ′ + 2 k ( sin β 2 ) 2 J − M − M ′ − 2 k k ! ( J − M − k ) ! ( J − M ′ − k ) ! ( M + M ′ + k ) ! (568) d^{J}_{MM'}(\beta)=(-1)^{J-M'}\left[(J+M)!(J-M)!(J+M')!(J-M')!\right]^{\frac{1}{2}}
\sum_{k}(-1)^{k}\frac{\left(\cos\frac{\beta}{2}\right)^{M+M'+2k}\left(\sin\frac{\beta}{2}\right)^{2J-M-M'-2k}}{k!(J-M-k)!(J-M'-k)!(M+M'+k)!} \tag*{(568)} d M M ′ J ( β ) = ( − 1 ) J − M ′ [ ( J + M )! ( J − M )! ( J + M ′ )! ( J − M ′ )! ] 2 1 k ∑ ( − 1 ) k k ! ( J − M − k )! ( J − M ′ − k )! ( M + M ′ + k )! ( cos 2 β ) M + M ′ + 2 k ( sin 2 β ) 2 J − M − M ′ − 2 k (568)
d M M ′ J ( β ) = ( − 1 ) J + M [ ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ] 1 2 × ∑ k ( − 1 ) k ( cos β 2 ) 2 k − M − M ′ ( sin β 2 ) 2 J + M + M ′ − 2 k k ! ( J + M − k ) ! ( J + M ′ − k ) ! ( k − M − M ′ ) ! (569) d^{J}_{MM'}(\beta)=(-1)^{J+M}\left[(J+M)!(J-M)!(J+M')!(J-M')!\right]^{\frac{1}{2}}
\times\sum_{k}(-1)^k\frac{\left(\cos\frac{\beta}{2}\right)^{2k-M-M'}\left(\sin\frac{\beta}{2}\right)^{2J+M+M'-2k}}{k!(J+M-k)!(J+M'-k)!(k-M-M')!} \tag*{(569)} d M M ′ J ( β ) = ( − 1 ) J + M [ ( J + M )! ( J − M )! ( J + M ′ )! ( J − M ′ )! ] 2 1 × k ∑ ( − 1 ) k k ! ( J + M − k )! ( J + M ′ − k )! ( k − M − M ′ )! ( cos 2 β ) 2 k − M − M ′ ( sin 2 β ) 2 J + M + M ′ − 2 k (569)
d M M ′ J ( β ) = [ ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ] 1 2 × ∑ k ( − 1 ) k ( cos β 2 ) 2 J − 2 k + M − M ′ ( sin β 2 ) 2 k − M + M ′ k ! ( J + M − k ) ! ( J − M ′ − k ) ! ( M ′ − M + k ) ! (570) d^{J}_{MM'}(\beta)=\left[(J+M)!(J-M)!(J+M')!(J-M')!\right]^{\frac{1}{2}}
\times\sum_{k}(-1)^k\frac{\left(\cos\frac{\beta}{2}\right)^{2J-2k+M-M'}\left(\sin\frac{\beta}{2}\right)^{2k-M+M'}}{k!(J+M-k)!(J-M'-k)!(M'-M+k)!} \tag*{(570)} d M M ′ J ( β ) = [ ( J + M )! ( J − M )! ( J + M ′ )! ( J − M ′ )! ] 2 1 × k ∑ ( − 1 ) k k ! ( J + M − k )! ( J − M ′ − k )! ( M ′ − M + k )! ( cos 2 β ) 2 J − 2 k + M − M ′ ( sin 2 β ) 2 k − M + M ′ (570)
d M M ′ J ( β ) = ( − 1 ) M − M ′ [ ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ] 1 2 × ∑ k ( − 1 ) k ( cos β 2 ) 2 J − 2 k − M + M ′ ( sin β 2 ) 2 k + M − M ′ k ! ( J − M − k ) ! ( J + M ′ − k ) ! ( M − M ′ + k ) ! (571) d^{J}_{MM'}(\beta)=(-1)^{M-M'}\left[(J+M)!(J-M)!(J+M')!(J-M')!\right]^{\frac{1}{2}}
\times\sum_{k}(-1)^k\frac{\left(\cos\frac{\beta}{2}\right)^{2J-2k-M+M'}\left(\sin\frac{\beta}{2}\right)^{2k+M-M'}}{k!(J-M-k)!(J+M'-k)!(M-M'+k)!} \tag*{(571)} d M M ′ J ( β ) = ( − 1 ) M − M ′ [ ( J + M )! ( J − M )! ( J + M ′ )! ( J − M ′ )! ] 2 1 × k ∑ ( − 1 ) k k ! ( J − M − k )! ( J + M ′ − k )! ( M − M ′ + k )! ( cos 2 β ) 2 J − 2 k − M + M ′ ( sin 2 β ) 2 k + M − M ′ (571)
In Eqs. (568) –(571) k k k runs over all integer values for which the factorial arguments are non-negative. Each of these sums contains ( N + 1 ) (N+1) ( N + 1 ) terms, where N N N is the minimum of J + M J+M J + M , J − M J-M J − M , J + M ′ J+M' J + M ′ and J − M ′ J-M' J − M ′ . Equations (568) –(571) are not independent, but may be transformed into one another by changing summation variables. Equations (568) –(571) may be regarded as special cases of the more general expression
d M M ′ J ( β ) = [ ( J 1 + J 2 + J + 1 ) ! ( J 1 + J 2 − J ) ! 2 J + 1 ] 1 2 ∑ m 1 m 2 m 1 + m 2 = M ( − 1 ) J 2 + m 2 C J 1 m 1 J 2 m 2 J M ( cos β 2 ) J 1 + J 2 + m 1 − m 2 ( sin β 2 ) J 1 + J 2 − m 1 + m 2 [ ( J 1 + m 1 ) ! ( J 1 − m 1 ) ! ( J 2 + m 2 ) ! ( J 2 − m 2 ) ! ] 1 2 (572) d^{J}_{MM'}(\beta)=\left[\frac{(J_{1}+J_{2}+J+1)!(J_{1}+J_{2}-J)!}{2J+1}\right]^{\frac{1}{2}}
\sum_{\substack{m_{1}m_{2}\\m_{1}+m_{2}=M}}
(-1)^{J_{2}+m_{2}}C^{JM}_{J_{1}m_{1}J_{2}m_{2}}
\frac{\left(\cos\frac{\beta}{2}\right)^{J_{1}+J_{2}+m_{1}-m_{2}}\left(\sin\frac{\beta}{2}\right)^{J_{1}+J_{2}-m_{1}+m_{2}}}{\left[(J_{1}+m_{1})!(J_{1}-m_{1})!(J_{2}+m_{2})!(J_{2}-m_{2})!\right]^{\frac{1}{2}}} \tag*{(572)} d M M ′ J ( β ) = [ 2 J + 1 ( J 1 + J 2 + J + 1 )! ( J 1 + J 2 − J )! ] 2 1 m 1 m 2 m 1 + m 2 = M ∑ ( − 1 ) J 2 + m 2 C J 1 m 1 J 2 m 2 J M [ ( J 1 + m 1 )! ( J 1 − m 1 )! ( J 2 + m 2 )! ( J 2 − m 2 )! ] 2 1 ( cos 2 β ) J 1 + J 2 + m 1 − m 2 ( sin 2 β ) J 1 + J 2 − m 1 + m 2 (572)
where J 1 J_{1} J 1 and J 2 J_{2} J 2 are arbitrary integer or half-integer numbers which satisfy the conditions J 1 − J 2 = M ′ J_{1}-J_{2}=M' J 1 − J 2 = M ′ and ∣ J 1 − J 2 ∣ ≤ J ≤ J 1 + J 2 \lvert J_{1}-J_{2}\rvert\leq J\leq J_{1}+J_{2} ∣ J 1 − J 2 ∣ ≤ J ≤ J 1 + J 2 . The sum in Eq. (572) is over all possible (positive and negative) values of m 1 m_{1} m 1 and m 2 m_{2} m 2 which correspond to nonzero Clebsch-Gordan coefficients. In particular, Eq. (572) reduces to Eq. (568) , when J 1 = ( J + M ′ ) / 2 J_{1}=(J+M')/2 J 1 = ( J + M ′ ) /2 , J 2 = ( J − M ′ ) / 2 J_{2}=(J-M')/2 J 2 = ( J − M ′ ) /2 , m 1 = M + k − ( J − M ′ ) / 2 m_{1}=M+k-(J-M')/2 m 1 = M + k − ( J − M ′ ) /2 and m 2 = − k + ( J − M ′ ) / 2 m_{2}=-k+(J-M')/2 m 2 = − k + ( J − M ′ ) /2 .
Differential Representations of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β )
d M M ′ J ( β ) = ( − 1 ) J − M ′ 1 2 J [ ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ] 1 2 ( 1 − cos β ) M ′ − M 2 ( 1 + cos β ) − M + M ′ 2 × d J − M ( d cos β ) J − M [ ( 1 − cos β ) J − M ′ ( 1 + cos β ) J + M ′ ] (573) d^{J}_{MM'}(\beta)=(-1)^{J-M'}\frac{1}{2^{J}}\left[\frac{(J+M)!}{(J-M)!(J+M')!(J-M')!}\right]^{\frac{1}{2}}(1-\cos\beta)^{\frac{M'-M}{2}}(1+\cos\beta)^{-\frac{M+M'}{2}}
\times\frac{d^{J-M}}{(d\cos\beta)^{J-M}}\left[(1-\cos\beta)^{J-M'}(1+\cos\beta)^{J+M'}\right] \tag*{(573)} d M M ′ J ( β ) = ( − 1 ) J − M ′ 2 J 1 [ ( J − M )! ( J + M ′ )! ( J − M ′ )! ( J + M )! ] 2 1 ( 1 − cos β ) 2 M ′ − M ( 1 + cos β ) − 2 M + M ′ × ( d cos β ) J − M d J − M [ ( 1 − cos β ) J − M ′ ( 1 + cos β ) J + M ′ ] (573)
d M M ′ J ( β ) = ( − 1 ) J + M 1 2 J [ ( J − M ) ! ( J + M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ] 1 2 ( 1 − cos β ) M − M ′ 2 ( 1 + cos β ) M + M ′ 2 × d J + M ( d cos β ) J + M [ ( 1 − cos β ) J + M ′ ( 1 + cos β ) J − M ′ ] (574) d^{J}_{MM'}(\beta)=(-1)^{J+M}\frac{1}{2^{J}}\left[\frac{(J-M)!}{(J+M)!(J+M')!(J-M')!}\right]^{\frac{1}{2}}(1-\cos\beta)^{\frac{M-M'}{2}}(1+\cos\beta)^{\frac{M+M'}{2}}
\times\frac{d^{J+M}}{(d\cos\beta)^{J+M}}\left[(1-\cos\beta)^{J+M'}(1+\cos\beta)^{J-M'}\right] \tag*{(574)} d M M ′ J ( β ) = ( − 1 ) J + M 2 J 1 [ ( J + M )! ( J + M ′ )! ( J − M ′ )! ( J − M )! ] 2 1 ( 1 − cos β ) 2 M − M ′ ( 1 + cos β ) 2 M + M ′ × ( d cos β ) J + M d J + M [ ( 1 − cos β ) J + M ′ ( 1 + cos β ) J − M ′ ] (574)
d M M ′ J ( β ) = ( − 1 ) J − M ′ 1 2 J [ ( J + M ′ ) ! ( J + M ) ! ( J − M ) ! ( J − M ′ ) ! ] 1 2 ( 1 − cos β ) M − M ′ 2 ( 1 + cos β ) − M + M ′ 2 × d J − M ′ ( d cos β ) J − M ′ [ ( 1 − cos β ) J − M ( 1 + cos β ) J + M ] (575) d^{J}_{MM'}(\beta)=(-1)^{J-M'}\frac{1}{2^{J}}\left[\frac{(J+M')!}{(J+M)!(J-M)!(J-M')!}\right]^{\frac{1}{2}}(1-\cos\beta)^{\frac{M-M'}{2}}(1+\cos\beta)^{-\frac{M+M'}{2}}
\times\frac{d^{J-M'}}{(d\cos\beta)^{J-M'}}\left[(1-\cos\beta)^{J-M}(1+\cos\beta)^{J+M}\right] \tag*{(575)} d M M ′ J ( β ) = ( − 1 ) J − M ′ 2 J 1 [ ( J + M )! ( J − M )! ( J − M ′ )! ( J + M ′ )! ] 2 1 ( 1 − cos β ) 2 M − M ′ ( 1 + cos β ) − 2 M + M ′ × ( d cos β ) J − M ′ d J − M ′ [ ( 1 − cos β ) J − M ( 1 + cos β ) J + M ] (575)
d M M ′ J ( β ) = ( − 1 ) J + M 1 2 J [ ( J − M ′ ) ! ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ] 1 2 ( 1 − cos β ) M ′ − M 2 ( 1 + cos β ) M + M ′ 2 × d J + M ′ ( d cos β ) J + M ′ [ ( 1 − cos β ) J + M ( 1 + cos β ) J − M ] (576) d^{J}_{MM'}(\beta)=(-1)^{J+M}\frac{1}{2^{J}}\left[\frac{(J-M')!}{(J+M)!(J-M)!(J+M')!}\right]^{\frac{1}{2}}(1-\cos\beta)^{\frac{M'-M}{2}}(1+\cos\beta)^{\frac{M+M'}{2}}
\times\frac{d^{J+M'}}{(d\cos\beta)^{J+M'}}\left[(1-\cos\beta)^{J+M}(1+\cos\beta)^{J-M}\right] \tag*{(576)} d M M ′ J ( β ) = ( − 1 ) J + M 2 J 1 [ ( J + M )! ( J − M )! ( J + M ′ )! ( J − M ′ )! ] 2 1 ( 1 − cos β ) 2 M ′ − M ( 1 + cos β ) 2 M + M ′ × ( d cos β ) J + M ′ d J + M ′ [ ( 1 − cos β ) J + M ( 1 + cos β ) J − M ] (576)
In practice it is convenient to use such equation from Eqs. (573) –(576) in which the order of derivative is the lowest.
Integral Representations of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β )
d M M ′ J ( β ) = i M − M ′ 1 2 π [ ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ] 1 2 × ∫ 0 2 π ( e i ϕ 2 cos β 2 + i e − i ϕ 2 sin β 2 ) J − M ′ ( e − i ϕ 2 cos β 2 + i e i ϕ 2 sin β 2 ) J + M ′ e i M ′ ϕ d ϕ . (577) d^{J}_{MM'}(\beta)=i^{M-M'}\frac{1}{2\pi}\left[\frac{(J+M)!(J-M)!}{(J+M')!(J-M')!}\right]^{\frac{1}{2}}
\times\int_{0}^{2\pi}\left(e^{i\frac{\phi}{2}}\cos\frac{\beta}{2}+ie^{-i\frac{\phi}{2}}\sin\frac{\beta}{2}\right)^{J-M'}
\left(e^{-i\frac{\phi}{2}}\cos\frac{\beta}{2}+ie^{i\frac{\phi}{2}}\sin\frac{\beta}{2}\right)^{J+M'}e^{iM'\phi}d\phi. \tag*{(577)} d M M ′ J ( β ) = i M − M ′ 2 π 1 [ ( J + M ′ )! ( J − M ′ )! ( J + M )! ( J − M )! ] 2 1 × ∫ 0 2 π ( e i 2 ϕ cos 2 β + i e − i 2 ϕ sin 2 β ) J − M ′ ( e − i 2 ϕ cos 2 β + i e i 2 ϕ sin 2 β ) J + M ′ e i M ′ ϕ d ϕ . (577)
Equation (4.3.11) can be rewritten as a contour integral
d M M ′ J ( β ) = i M − M ′ − 1 2 π [ ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ] ∣ z ∣ = 1 1 2 × ∮ ( z cos β 2 + i sin β 2 ) J − M ′ ( i z sin β 2 + cos β 2 ) J + M ′ z M − M ′ − 1 d z . (578) d^{J}_{MM'}(\beta)=\frac{i^{M-M'-1}}{2\pi}\left[\frac{(J+M)!(J-M)!}{(J+M')!(J-M')!}\right]^{\frac{1}{2}}_{|z|=1}
\times\oint\left(z\cos\frac{\beta}{2}+i\sin\frac{\beta}{2}\right)^{J-M'}
\left(iz\sin\frac{\beta}{2}+\cos\frac{\beta}{2}\right)^{J+M'}z^{M-M'-1}dz. \tag*{(578)} d M M ′ J ( β ) = 2 π i M − M ′ − 1 [ ( J + M ′ )! ( J − M ′ )! ( J + M )! ( J − M )! ] ∣ z ∣ = 1 2 1 × ∮ ( z cos 2 β + i sin 2 β ) J − M ′ ( i z sin 2 β + cos 2 β ) J + M ′ z M − M ′ − 1 d z . (578)
The integration contour in Eq. (4.3.12) is a circle of unit radius about the origin of the z z z -plane.
Relation Between d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) and the Jacobi Polynomials
The functions d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) can be expressed in terms of the Jacobi polynomials
d M M ′ J ( β ) = ξ M M ′ [ s ! ( s + μ + ν ) ! ( s + μ ) ! ( s + ν ) ! ] 1 2 ( sin β 2 ) μ ( cos β 2 ) ν P s ( μ , ν ) ( cos β ) , (579) d^{J}_{MM'}(\beta)=\xi_{MM'}\left[\frac{s!(s+\mu+\nu)!}{(s+\mu)!(s+\nu)!}\right]^{\frac{1}{2}}
\left(\sin\frac{\beta}{2}\right)^{\mu}\left(\cos\frac{\beta}{2}\right)^{\nu}P_{s}^{(\mu,\nu)}(\cos\beta), \tag*{(579)} d M M ′ J ( β ) = ξ M M ′ [ ( s + μ )! ( s + ν )! s ! ( s + μ + ν )! ] 2 1 ( sin 2 β ) μ ( cos 2 β ) ν P s ( μ , ν ) ( cos β ) , (579)
where μ , ν \mu,\nu μ , ν and s s s are related to M , M ′ M,M' M , M ′ and J J J by
μ = ∣ M − M ′ ∣ , ν = ∣ M + M ′ ∣ , s = J − 1 2 ( μ + ν ) . (580) \mu=|M-M'|,\quad\nu=|M+M'|,\quad s=J-\frac{1}{2}(\mu+\nu). \tag*{(580)} μ = ∣ M − M ′ ∣ , ν = ∣ M + M ′ ∣ , s = J − 2 1 ( μ + ν ) . (580)
and
ξ M M ′ = { 1 if M ′ ≥ M , ( − 1 ) M ′ − M if M ′ < M . (581) \xi_{MM'}=\begin{cases}
1 & \text{if }M'\ge M,\\
(-1)^{M'-M} & \text{if }M'<M.
\end{cases} \tag*{(581)} ξ M M ′ = { 1 ( − 1 ) M ′ − M if M ′ ≥ M , if M ′ < M . (581)
Relations Between d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) and Hypergeometric Functions
d M M ′ J ( β ) = ξ M M ′ μ ! [ ( s + μ + ν ) ! ( s + μ ) ! s ! ( s + ν ) ! ] 1 2 ( sin β 2 ) μ ( cos β 2 ) ν F ( − s , s + μ + ν + 1 ; μ + 1 ; sin 2 β 2 ) , (582) d^{J}_{MM'}(\beta)=\frac{\xi_{MM'}}{\mu!}\left[\frac{(s+\mu+\nu)!(s+\mu)!}{s!(s+\nu)!}\right]^{\frac{1}{2}}
\left(\sin\frac{\beta}{2}\right)^{\mu}\left(\cos\frac{\beta}{2}\right)^{\nu}
F\left(-s,s+\mu+\nu+1;\mu+1;\sin^{2}\frac{\beta}{2}\right), \tag*{(582)} d M M ′ J ( β ) = μ ! ξ M M ′ [ s ! ( s + ν )! ( s + μ + ν )! ( s + μ )! ] 2 1 ( sin 2 β ) μ ( cos 2 β ) ν F ( − s , s + μ + ν + 1 ; μ + 1 ; sin 2 2 β ) , (582)
d M M ′ J ( β ) = ξ M M ′ μ ! [ ( s + μ + ν ) ! ( s + μ ) ! s ! ( s + ν ) ! ] 1 2 ( sin β 2 ) μ ( cos β 2 ) − ν F ( s + μ + 1 , − s − ν ; μ + 1 ; sin 2 β 2 ) , (583) d^{J}_{MM'}(\beta)=\frac{\xi_{MM'}}{\mu!}\left[\frac{(s+\mu+\nu)!(s+\mu)!}{s!(s+\nu)!}\right]^{\frac{1}{2}}
\left(\sin\frac{\beta}{2}\right)^{\mu}\left(\cos\frac{\beta}{2}\right)^{-\nu}
F\left(s+\mu+1,-s-\nu;\mu+1;\sin^{2}\frac{\beta}{2}\right), \tag*{(583)} d M M ′ J ( β ) = μ ! ξ M M ′ [ s ! ( s + ν )! ( s + μ + ν )! ( s + μ )! ] 2 1 ( sin 2 β ) μ ( cos 2 β ) − ν F ( s + μ + 1 , − s − ν ; μ + 1 ; sin 2 2 β ) , (583)
d M M ′ J ( β ) = ( − 1 ) s ξ M M ′ ν ! [ ( s + μ + ν ) ! ( s + ν ) ! s ! ( s + μ ) ! ] 1 2 ( sin β 2 ) μ ( cos β 2 ) ν F ( − s , s + μ + ν + 1 ; ν + 1 ; cos 2 β 2 ) , (584) d^{J}_{MM'}(\beta)=\frac{(-1)^{s}\xi_{MM'}}{\nu!}\left[\frac{(s+\mu+\nu)!(s+\nu)!}{s!(s+\mu)!}\right]^{\frac{1}{2}}
\left(\sin\frac{\beta}{2}\right)^{\mu}\left(\cos\frac{\beta}{2}\right)^{\nu}
F\left(-s,s+\mu+\nu+1;\nu+1;\cos^{2}\frac{\beta}{2}\right), \tag*{(584)} d M M ′ J ( β ) = ν ! ( − 1 ) s ξ M M ′ [ s ! ( s + μ )! ( s + μ + ν )! ( s + ν )! ] 2 1 ( sin 2 β ) μ ( cos 2 β ) ν F ( − s , s + μ + ν + 1 ; ν + 1 ; cos 2 2 β ) , (584)
d M ′ M J ( β ) = ( − 1 ) s ξ M M ′ ν ! [ ( s + μ + ν ) ! ( s + ν ) ! s ! ( s + μ ) ! ] 1 2 ( sin β 2 ) − μ ( cos β 2 ) ν F ( s + ν + 1 ; − s − μ ; ν + 1 ; cos 2 β 2 ) , (585) d^{J}_{M'M}(\beta)=\frac{(-1)^{s}\xi_{MM'}}{\nu!}\left[\frac{(s+\mu+\nu)!(s+\nu)!}{s!(s+\mu)!}\right]^{\frac{1}{2}}
\left(\sin\frac{\beta}{2}\right)^{-\mu}\left(\cos\frac{\beta}{2}\right)^{\nu}
F\left(s+\nu+1;-s-\mu;\nu+1;\cos^{2}\frac{\beta}{2}\right), \tag*{(585)} d M ′ M J ( β ) = ν ! ( − 1 ) s ξ M M ′ [ s ! ( s + μ )! ( s + μ + ν )! ( s + ν )! ] 2 1 ( sin 2 β ) − μ ( cos 2 β ) ν F ( s + ν + 1 ; − s − μ ; ν + 1 ; cos 2 2 β ) , (585)
d M M ′ J ( β ) = ξ M M ′ μ ! [ ( s + μ + ν ) ! ( s + μ ) ! s ! ( s + ν ) ! ] 1 2 ( sin β 2 ) μ ( cos β 2 ) 2 s + ν F ( − s , − s − ν ; μ + 1 ; − tan 2 β 2 ) , (586) d^{J}_{MM'}(\beta)=\frac{\xi_{MM'}}{\mu!}\left[\frac{(s+\mu+\nu)!(s+\mu)!}{s!(s+\nu)!}\right]^{\frac{1}{2}}
\left(\sin\frac{\beta}{2}\right)^{\mu}\left(\cos\frac{\beta}{2}\right)^{2s+\nu}
F\left(-s,-s-\nu;\mu+1;-\tan^{2}\frac{\beta}{2}\right), \tag*{(586)} d M M ′ J ( β ) = μ ! ξ M M ′ [ s ! ( s + ν )! ( s + μ + ν )! ( s + μ )! ] 2 1 ( sin 2 β ) μ ( cos 2 β ) 2 s + ν F ( − s , − s − ν ; μ + 1 ; − tan 2 2 β ) , (586)
d M M ′ J ( β ) = ( − 1 ) s ξ M M ′ ν ! [ ( s + μ + ν ) ! ( s + ν ) ! s ! ( s + μ ) ! ] 1 2 ( sin β 2 ) 2 s + μ ( cos β 2 ) ν F ( − s , − s − μ ; ν + 1 ; − cot 2 β 2 ) , (587) d^{J}_{MM'}(\beta)=\frac{(-1)^{s}\xi_{MM'}}{\nu!}\left[\frac{(s+\mu+\nu)!(s+\nu)!}{s!(s+\mu)!}\right]^{\frac{1}{2}}
\left(\sin\frac{\beta}{2}\right)^{2s+\mu}\left(\cos\frac{\beta}{2}\right)^{\nu}
F\left(-s,-s-\mu;\nu+1;-\cot^{2}\frac{\beta}{2}\right), \tag*{(587)} d M M ′ J ( β ) = ν ! ( − 1 ) s ξ M M ′ [ s ! ( s + μ )! ( s + μ + ν )! ( s + ν )! ] 2 1 ( sin 2 β ) 2 s + μ ( cos 2 β ) ν F ( − s , − s − μ ; ν + 1 ; − cot 2 2 β ) , (587)
d M M ′ J ( β ) = ( − 1 ) s ξ M M ′ ( 2 s + μ + ν ) ! [ s ! ( s + μ + ν ) ! ( s + μ ) ! ( s + ν ) ! ] 1 2 ( sin β 2 ) 2 s + μ ( cos β 2 ) ν F ( − s , − s − μ ; − 2 s − μ − ν ; 1 sin 2 β 2 ) , (588) d^{J}_{MM'}(\beta)=\frac{(-1)^{s}\xi_{MM'}(2s+\mu+\nu)!}{[s!(s+\mu+\nu)!(s+\mu)!(s+\nu)!]^{\frac{1}{2}}}
\left(\sin\frac{\beta}{2}\right)^{2s+\mu}\left(\cos\frac{\beta}{2}\right)^{\nu}
F\left(-s,-s-\mu;-2s-\mu-\nu;\frac{1}{\sin^{2}\frac{\beta}{2}}\right), \tag*{(588)} d M M ′ J ( β ) = [ s ! ( s + μ + ν )! ( s + μ )! ( s + ν )! ] 2 1 ( − 1 ) s ξ M M ′ ( 2 s + μ + ν )! ( sin 2 β ) 2 s + μ ( cos 2 β ) ν F ( − s , − s − μ ; − 2 s − μ − ν ; sin 2 2 β 1 ) , (588)
d M M ′ J ( β ) = ξ M M ′ ( 2 s + μ + ν ) ! [ s ! ( s + μ + ν ) ! ( s + μ ) ! ( s + ν ) ! ] 1 2 ( sin β 2 ) μ ( cos β 2 ) 2 s + ν F ( − s , − s − ν ; − 2 s − μ − ν ; 1 cos 2 β 2 ) . (589) d^{J}_{MM'}(\beta)=\frac{\xi_{MM'}(2s+\mu+\nu)!}{[s!(s+\mu+\nu)!(s+\mu)!(s+\nu)!]^{\frac{1}{2}}}
\left(\sin\frac{\beta}{2}\right)^{\mu}\left(\cos\frac{\beta}{2}\right)^{2s+\nu}
F\left(-s,-s-\nu;-2s-\mu-\nu;\frac{1}{\cos^{2}\frac{\beta}{2}}\right). \tag*{(589)} d M M ′ J ( β ) = [ s ! ( s + μ + ν )! ( s + μ )! ( s + ν )! ] 2 1 ξ M M ′ ( 2 s + μ + ν )! ( sin 2 β ) μ ( cos 2 β ) 2 s + ν F ( − s , − s − ν ; − 2 s − μ − ν ; cos 2 2 β 1 ) . (589)
Parameters μ , ν , s \mu,\nu,s μ , ν , s and a phase factor ξ M M ′ \xi_{MM'} ξ M M ′ in Eqs. (4.3.16)–(4.3.23) are defined by Eqs. (4.3.14) and (4.3.15).
Symmetries of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) and D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
Symmetries of d M M ′ J ( β ) d^J_{MM'}(\beta) d M M ′ J ( β ) and D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
(a) In accordance with Eqs. (568) –(571) the functions d M M ′ J ( β ) d^J_{MM'}(\beta) d M M ′ J ( β ) are real and satisfy the relations
d M M ′ J ( β ) = ( − 1 ) M − M ′ d − M − M ′ J ( β ) = ( − 1 ) M − M ′ d M ′ M J ( β ) = d − M ′ − M J ( β ) , d^J_{MM'}(\beta)=(-1)^{M-M'}d^J_{-M\,-M'}(\beta)=(-1)^{M-M'}d^J_{M'M}(\beta)=d^J_{-M'\,-M}(\beta), d M M ′ J ( β ) = ( − 1 ) M − M ′ d − M − M ′ J ( β ) = ( − 1 ) M − M ′ d M ′ M J ( β ) = d − M ′ − M J ( β ) ,
d M M ′ J ( − β ) = ( − 1 ) M − M ′ d M M ′ J ( β ) = d M ′ M J ( β ) , d^J_{MM'}(-\beta)=(-1)^{M-M'}d^J_{MM'}(\beta)=d^J_{M'M}(\beta), d M M ′ J ( − β ) = ( − 1 ) M − M ′ d M M ′ J ( β ) = d M ′ M J ( β ) ,
d M M ′ J ( π − β ) = ( − 1 ) J − M ′ d − M M ′ J ( β ) = ( − 1 ) J + M d M − M ′ J ( β ) , d^J_{MM'}(\pi-\beta)=(-1)^{J-M'}d^J_{-M\,M'}(\beta)=(-1)^{J+M}d^J_{M\,-M'}(\beta), d M M ′ J ( π − β ) = ( − 1 ) J − M ′ d − M M ′ J ( β ) = ( − 1 ) J + M d M − M ′ J ( β ) ,
d M M ′ J ( β ± 2 π n ) = ( − 1 ) 2 J n d M M ′ J ( β ) , n is integer , d^J_{MM'}(\beta\pm2\pi n)=(-1)^{2Jn}d^J_{MM'}(\beta),\qquad n\text{ is integer}, d M M ′ J ( β ± 2 π n ) = ( − 1 ) 2 J n d M M ′ J ( β ) , n is integer ,
d M M ′ J ( β ± ( 2 n + 1 ) π ) = ( − 1 ) ± ( 2 n + 1 ) J − M ′ d M − M ′ J ( β ) , n is integer . (590) d^J_{MM'}(\beta\pm(2n+1)\pi)=(-1)^{\pm(2n+1)J-M'}d^J_{M\,-M'}(\beta),\qquad n\text{ is integer}. \tag*{(590)} d M M ′ J ( β ± ( 2 n + 1 ) π ) = ( − 1 ) ± ( 2 n + 1 ) J − M ′ d M − M ′ J ( β ) , n is integer . (590)
(b) Equations (4.4.1) imply the following symmetry properties of D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
D M M ′ J ( α , β , γ ) = ε η D − M − M ′ J ( α , β , γ ) = η D M M ′ J ∗ ( α , β , γ ) = ε D − M − M ′ J ∗ ( α , β , γ ) = ε D M ′ M J ( γ , β , α ) = η D − M ′ − M J ( γ , β , α ) = ε η D M ′ M J ∗ ( γ , β , α ) = D − M ′ − M J ∗ ( γ , β , α ) = ε D M M ′ J ( α , − β , γ ) = η D − M − M ′ J ( α , − β , γ ) = ε η D M M ′ J ∗ ( α , − β , γ ) = D − M − M ′ J ∗ ( α , − β , γ ) = D M ′ M J ( γ , − β , α ) = ε η D − M ′ − M J ( γ , − β , α ) = η D M ′ M J ∗ ( γ , − β , α ) = ε D − M ′ − M J ∗ ( γ , − β , α ) = ε D − M − M ′ J ( − α , β , − γ ) = η D M M ′ J ( − α , β , − γ ) = ε η D − M − M ′ J ∗ ( − α , β , − γ ) = D M M ′ J ∗ ( − α , β , − γ ) = D − M ′ − M J ( − γ , β , − α ) = ε η D M ′ M J ( − γ , β , − α ) = η D − M ′ − M J ∗ ( − γ , β , − α ) = ε D M ′ M J ∗ ( − γ , β , − α ) = D − M − M ′ J ( − α , − β , − γ ) = ε η D M M ′ J ( − α , − β , − γ ) = η D − M − M ′ J ∗ ( − α , − β , − γ ) = ε D M M ′ J ∗ ( − α , − β , − γ ) = ε D − M ′ − M J ( − γ , − β , − α ) = η D M ′ M J ( − γ , − β , − α ) = ε η D − M ′ − M J ∗ ( − γ , − β , − α ) = D M ′ M J ∗ ( − γ , − β , − α ) . (591) \begin{aligned}
D^J_{MM'}(\alpha,\beta,\gamma)&=\varepsilon\eta D^J_{-M\,-M'}(\alpha,\beta,\gamma)=\eta D^{J*}_{MM'}(\alpha,\beta,\gamma)=\varepsilon D^{J*}_{-M\,-M'}(\alpha,\beta,\gamma)\\
&=\varepsilon D^J_{M'M}(\gamma,\beta,\alpha)=\eta D^J_{-M'\,-M}(\gamma,\beta,\alpha)=\varepsilon\eta D^{J*}_{M'M}(\gamma,\beta,\alpha)=D^{J*}_{-M'\,-M}(\gamma,\beta,\alpha)\\
&=\varepsilon D^J_{MM'}(\alpha,-\beta,\gamma)=\eta D^J_{-M\,-M'}(\alpha,-\beta,\gamma)=\varepsilon\eta D^{J*}_{MM'}(\alpha,-\beta,\gamma)=D^{J*}_{-M\,-M'}(\alpha,-\beta,\gamma)\\
&=D^J_{M'M}(\gamma,-\beta,\alpha)=\varepsilon\eta D^J_{-M'\,-M}(\gamma,-\beta,\alpha)=\eta D^{J*}_{M'M}(\gamma,-\beta,\alpha)=\varepsilon D^{J*}_{-M'\,-M}(\gamma,-\beta,\alpha)\\
&=\varepsilon D^J_{-M\,-M'}(-\alpha,\beta,-\gamma)=\eta D^J_{MM'}(-\alpha,\beta,-\gamma)=\varepsilon\eta D^{J*}_{-M\,-M'}(-\alpha,\beta,-\gamma)=D^{J*}_{MM'}(-\alpha,\beta,-\gamma)\\
&=D^J_{-M'\,-M}(-\gamma,\beta,-\alpha)=\varepsilon\eta D^J_{M'M}(-\gamma,\beta,-\alpha)=\eta D^{J*}_{-M'\,-M}(-\gamma,\beta,-\alpha)=\varepsilon D^{J*}_{M'M}(-\gamma,\beta,-\alpha)\\
&=D^J_{-M\,-M'}(-\alpha,-\beta,-\gamma)=\varepsilon\eta D^J_{MM'}(-\alpha,-\beta,-\gamma)=\eta D^{J*}_{-M\,-M'}(-\alpha,-\beta,-\gamma)=\varepsilon D^{J*}_{MM'}(-\alpha,-\beta,-\gamma)\\
&=\varepsilon D^J_{-M'\,-M}(-\gamma,-\beta,-\alpha)=\eta D^J_{M'M}(-\gamma,-\beta,-\alpha)=\varepsilon\eta D^{J*}_{-M'\,-M}(-\gamma,-\beta,-\alpha)=D^{J*}_{M'M}(-\gamma,-\beta,-\alpha). \tag*{(591)}
\end{aligned} D M M ′ J ( α , β , γ ) = ε η D − M − M ′ J ( α , β , γ ) = η D M M ′ J ∗ ( α , β , γ ) = ε D − M − M ′ J ∗ ( α , β , γ ) = ε D M ′ M J ( γ , β , α ) = η D − M ′ − M J ( γ , β , α ) = ε η D M ′ M J ∗ ( γ , β , α ) = D − M ′ − M J ∗ ( γ , β , α ) = ε D M M ′ J ( α , − β , γ ) = η D − M − M ′ J ( α , − β , γ ) = ε η D M M ′ J ∗ ( α , − β , γ ) = D − M − M ′ J ∗ ( α , − β , γ ) = D M ′ M J ( γ , − β , α ) = ε η D − M ′ − M J ( γ , − β , α ) = η D M ′ M J ∗ ( γ , − β , α ) = ε D − M ′ − M J ∗ ( γ , − β , α ) = ε D − M − M ′ J ( − α , β , − γ ) = η D M M ′ J ( − α , β , − γ ) = ε η D − M − M ′ J ∗ ( − α , β , − γ ) = D M M ′ J ∗ ( − α , β , − γ ) = D − M ′ − M J ( − γ , β , − α ) = ε η D M ′ M J ( − γ , β , − α ) = η D − M ′ − M J ∗ ( − γ , β , − α ) = ε D M ′ M J ∗ ( − γ , β , − α ) = D − M − M ′ J ( − α , − β , − γ ) = ε η D M M ′ J ( − α , − β , − γ ) = η D − M − M ′ J ∗ ( − α , − β , − γ ) = ε D M M ′ J ∗ ( − α , − β , − γ ) = ε D − M ′ − M J ( − γ , − β , − α ) = η D M ′ M J ( − γ , − β , − α ) = ε η D − M ′ − M J ∗ ( − γ , − β , − α ) = D M ′ M J ∗ ( − γ , − β , − α ) . (591)
where
ε = ( − 1 ) M ′ − M , η = e − i 2 M α − i 2 M ′ γ . (592) \varepsilon=(-1)^{M'-M},\qquad\eta=e^{-i2M\alpha-i2M'\gamma}. \tag*{(592)} ε = ( − 1 ) M ′ − M , η = e − i 2 M α − i 2 M ′ γ . (592)
The periodicity conditions for D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) are
D M M ′ J ( α , β ± 2 n π , γ ) = ( − 1 ) 2 n J D M M ′ J ( α , β , γ ) , D^J_{MM'}(\alpha,\beta\pm2n\pi,\gamma)=(-1)^{2nJ}D^J_{MM'}(\alpha,\beta,\gamma), D M M ′ J ( α , β ± 2 nπ , γ ) = ( − 1 ) 2 n J D M M ′ J ( α , β , γ ) ,
D M M ′ J ( α , β ± ( 2 n + 1 ) π , γ ) = ( − 1 ) ± ( 2 n + 1 ) J − M ′ D M − M ′ J ( α , β , − γ ) , D^J_{MM'}(\alpha,\beta\pm(2n+1)\pi,\gamma)=(-1)^{\pm(2n+1)J-M'}D^J_{M\,-M'}(\alpha,\beta,-\gamma), D M M ′ J ( α , β ± ( 2 n + 1 ) π , γ ) = ( − 1 ) ± ( 2 n + 1 ) J − M ′ D M − M ′ J ( α , β , − γ ) ,
D M M ′ J ( α ± n π , β , γ ) = ( − i ) ± 2 n M D M M ′ J ( α , β , γ ) , D^J_{MM'}(\alpha\pm n\pi,\beta,\gamma)=(-i)^{\pm2nM}D^J_{MM'}(\alpha,\beta,\gamma), D M M ′ J ( α ± nπ , β , γ ) = ( − i ) ± 2 n M D M M ′ J ( α , β , γ ) ,
D M M ′ J ( α , β , γ ± n π ) = ( − i ) ± 2 n M ′ D M M ′ J ( α , β , γ ) , (593) D^J_{MM'}(\alpha,\beta,\gamma\pm n\pi)=(-i)^{\pm2nM'}D^J_{MM'}(\alpha,\beta,\gamma), \tag*{(593)} D M M ′ J ( α , β , γ ± nπ ) = ( − i ) ± 2 n M ′ D M M ′ J ( α , β , γ ) , (593)
where n n n is integer. Note also that
D M M ′ J ( α ~ , β , γ ~ ) = e i M ( α − α ~ ) D M M ′ J ( α , β , γ ) e i M ′ ( γ − γ ~ ) . (594) D^J_{MM'}(\tilde{\alpha},\beta,\tilde{\gamma})=e^{iM(\alpha-\tilde{\alpha})}D^J_{MM'}(\alpha,\beta,\gamma)e^{iM'(\gamma-\tilde{\gamma})}. \tag*{(594)} D M M ′ J ( α ~ , β , γ ~ ) = e i M ( α − α ~ ) D M M ′ J ( α , β , γ ) e i M ′ ( γ − γ ~ ) . (594)
Some properties of D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) follow from the addition theorem (Sec. a). Let us consider some special cases.
(i) The matrix of the transformation S { x , y , z } → S ′ ′ { x ′ , − y ′ , − z ′ } S\{x,y,z\}\to S''\{x',-y',-z'\} S { x , y , z } → S ′′ { x ′ , − y ′ , − z ′ } may be obtained from the matrix of the transformation S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } by substituting ( α + π , π − β , − γ ) (\alpha+\pi,\pi-\beta,-\gamma) ( α + π , π − β , − γ ) for ( α , β , γ ) (\alpha,\beta,\gamma) ( α , β , γ ) . On the other hand, this substitution corresponds to an additional rotation R ^ x ′ \hat{R}_{x'} R ^ x ′ about the x ′ x' x ′ -axis through an angle − π -\pi − π . Hence,
D M M ′ J ( α + π , π − β , − γ ) = R ^ x ′ D M M ′ J ( α , β , γ ) = ∑ M ′ ′ D M M ′ ′ J ( α , β , γ ) D M ′ ′ M ′ J ( − π 2 , − π , π 2 ) = ( − 1 ) J D M − M ′ J ( α , β , γ ) . (595) \begin{aligned}
D^J_{MM'}(\alpha+\pi,\pi-\beta,-\gamma)=\hat{R}_{x'}D^J_{MM'}(\alpha,\beta,\gamma)=\sum_{M''}D^J_{MM''}(\alpha,\beta,\gamma)D^J_{M''M'}\left(-\frac{\pi}{2},-\pi,\frac{\pi}{2}\right) \\
&=(-1)^JD^J_{M\,-M'}(\alpha,\beta,\gamma). \tag*{(595)}
\end{aligned} D M M ′ J ( α + π , π − β , − γ ) = R ^ x ′ D M M ′ J ( α , β , γ ) = M ′′ ∑ D M M ′′ J ( α , β , γ ) D M ′′ M ′ J ( − 2 π , − π , 2 π ) = ( − 1 ) J D M − M ′ J ( α , β , γ ) . (595)
(ii) The matrix of the transformation S { x , y , z } → S ′ ′ ′ { − x ′ , y ′ , − z ′ } S\{x,y,z\}\to S'''\{-x',y',-z'\} S { x , y , z } → S ′′′ { − x ′ , y ′ , − z ′ } may be obtained from the matrix of the transformation S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } by substituting ( α − π , π − β , π − γ ) (\alpha-\pi,\pi-\beta,\pi-\gamma) ( α − π , π − β , π − γ ) for ( α , β , γ ) (\alpha,\beta,\gamma) ( α , β , γ ) . This substitution corresponds to an additional rotation R ^ y ′ \hat{R}_{y'} R ^ y ′ about the y ′ y' y ′ -axis through an angle − π -\pi − π . Hence,
D M M ′ J ( α − π , π − β , π − γ ) = R ^ y ′ D M M ′ J ( α , β , γ ) = ∑ M ′ ′ D M M ′ ′ J ( α , β , γ ) D M ′ ′ M ′ J ( 0 , − π , 0 ) = ( − 1 ) J + M ′ D M − M ′ J ( α , β , γ ) . (596) \begin{aligned}
D^J_{MM'}(\alpha-\pi,\pi-\beta,\pi-\gamma)=\hat{R}_{y'}D^J_{MM'}(\alpha,\beta,\gamma)=\sum_{M''}D^J_{MM''}(\alpha,\beta,\gamma)D^J_{M''M'}(0,-\pi,0) \\
&=(-1)^{J+M'}D^J_{M\,-M'}(\alpha,\beta,\gamma). \tag*{(596)}
\end{aligned} D M M ′ J ( α − π , π − β , π − γ ) = R ^ y ′ D M M ′ J ( α , β , γ ) = M ′′ ∑ D M M ′′ J ( α , β , γ ) D M ′′ M ′ J ( 0 , − π , 0 ) = ( − 1 ) J + M ′ D M − M ′ J ( α , β , γ ) . (596)
(iii) The matrix of the transformation S { x , y , z } → S ′ ′ ′ ′ { − x ′ , − y ′ , z ′ } S\{x,y,z\}\to S''''\{-x',-y',z'\} S { x , y , z } → S ′′′′ { − x ′ , − y ′ , z ′ } may be derived from the matrix of the transformation S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } by replacing ( α , β , γ ) → ( α , β , γ − π ) (\alpha,\beta,\gamma)\to(\alpha,\beta,\gamma-\pi) ( α , β , γ ) → ( α , β , γ − π ) . This corresponds to an additional rotation R ^ z ′ \hat{R}_{z'} R ^ z ′ about the z ′ z' z ′ -axis through an angle − π -\pi − π . Thus,
D M M ′ J ( α , β , γ − π ) = R ^ z ′ D M M ′ J ( α , β , γ ) = ∑ M ′ ′ D M M ′ ′ J ( α , β , γ ) D M ′ ′ M ′ J ( 0 , 0 , − π ) = ( − 1 ) M ′ D M M ′ J ( α , β , γ ) . (597) D^J_{MM'}(\alpha,\beta,\gamma-\pi)=\hat{R}_{z'}D^J_{MM'}(\alpha,\beta,\gamma)=\sum_{M''}D^J_{MM''}(\alpha,\beta,\gamma)D^J_{M''M'}(0,0,-\pi)=(-1)^{M'}D^J_{MM'}(\alpha,\beta,\gamma). \tag*{(597)} D M M ′ J ( α , β , γ − π ) = R ^ z ′ D M M ′ J ( α , β , γ ) = M ′′ ∑ D M M ′′ J ( α , β , γ ) D M ′′ M ′ J ( 0 , 0 , − π ) = ( − 1 ) M ′ D M M ′ J ( α , β , γ ) . (597)
Note that three successive rotations through angles − π -\pi − π about the axes x ′ x' x ′ , y ′ y' y ′ and z ′ z' z ′ return the coordinate system to its initial position.
Rotation matrix U M M ′ J U^J_{MM'} U M M ′ J , in terms of angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ
Definition
In some cases the description of rotations in terms of ω \omega ω , Θ \Theta Θ , Φ \Phi Φ (where ω \omega ω is the angle of rotation and Θ \Theta Θ , Φ \Phi Φ are the angles which determine the rotation axis, see § 9.9 ) is more convenient than that in terms of the Euler angles α \alpha α , β \beta β , γ \gamma γ . Matrix elements of the rotation operator in terms of variables ω \omega ω , Θ \Theta Θ , Φ \Phi Φ will be denoted by U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) :
U M M ′ J ( ω ; Θ , Φ ) = ⟨ J M ∣ e − i ω n ⋅ J ^ ∣ J M ′ ⟩ . (598) U^J_{MM'}(\omega;\Theta,\Phi)=\langle JM|e^{-i\omega\mathbf{n}\cdot\hat{\mathbf{J}}}|JM'\rangle. \tag*{(598)} U M M ′ J ( ω ; Θ , Φ ) = ⟨ J M ∣ e − iω n ⋅ J ^ ∣ J M ′ ⟩ . (598)
Therefore, under a rotation specified by ω \omega ω , Θ \Theta Θ , Φ \Phi Φ the components of an irreducible tensor of rank J J J transform as
M J M ′ ( ϑ ′ , φ ′ ) = ∑ M M J M ( ϑ , φ ) U M M ′ J ( ω ; Θ , Φ ) . (599) \mathcal{M}_{JM'}(\vartheta',\varphi')=\sum_M\mathcal{M}_{JM}(\vartheta,\varphi)U^J_{MM'}(\omega;\Theta,\Phi). \tag*{(599)} M J M ′ ( ϑ ′ , φ ′ ) = M ∑ M J M ( ϑ , φ ) U M M ′ J ( ω ; Θ , Φ ) . (599)
The polar angles ϑ \vartheta ϑ , φ \varphi φ and ϑ ′ \vartheta' ϑ ′ , φ ′ \varphi' φ ′ which specify a direction of an arbitrary vector in the initial and rotated coordinate systems in terms ω \omega ω , Θ \Theta Θ , Φ \Phi Φ are related by Eqs. (184) and (1.4.56).
Explicit form
(a) Arbitrary rotation as product of three rotations
An arbitrary rotation specified by angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ may be considered as a result of three successive rotations of coordinate system:
R 1 R_1 R 1 (α 1 = Φ \alpha_1=\Phi α 1 = Φ , β 1 = Θ \beta_1=\Theta β 1 = Θ , γ 1 = − Φ \gamma_1=-\Phi γ 1 = − Φ ), i.e., the rotation which turns the z z z -axis to the direction of n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) ;
R 2 R_2 R 2 (α 2 = ω \alpha_2=\omega α 2 = ω , β 2 = 0 \beta_2=0 β 2 = 0 , γ 2 = 0 \gamma_2=0 γ 2 = 0 ), i.e., the rotation about n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) through an angle ω \omega ω ;
R 3 R_3 R 3 (α 3 = Φ \alpha_3=\Phi α 3 = Φ , β 3 = − Θ \beta_3=-\Theta β 3 = − Θ , γ 3 = − Φ \gamma_3=-\Phi γ 3 = − Φ ), i.e., the rotation which is inverse to R 1 R_1 R 1 .
The result of these three rotations yields the relation between U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) and the Wigner D D D -functions
U M M ′ J ( ω ; Θ , Φ ) = ∑ M ′ ′ D M M ′ ′ J ( Φ , Θ , − Φ ) e − i M ′ ′ ω D M ′ ′ M ′ J ( Φ , − Θ , − Φ ) . (600) U^J_{MM'}(\omega;\Theta,\Phi)=\sum_{M''}D^J_{MM''}(\Phi,\Theta,-\Phi)e^{-iM''\omega}D^J_{M''M'}(\Phi,-\Theta,-\Phi). \tag*{(600)} U M M ′ J ( ω ; Θ , Φ ) = M ′′ ∑ D M M ′′ J ( Φ , Θ , − Φ ) e − i M ′′ ω D M ′′ M ′ J ( Φ , − Θ , − Φ ) . (600)
Equation (4.5.3) enables one to find an explicit form for U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) for particular J J J , M M M and M ′ M' M ′ .
(b) Construction from Cayley-Klein parameters
According to Eq. (607) the functions U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) may be directly constructed from the matrix elements U m m ′ 1 / 2 U^{1/2}_{mm'} U m m ′ 1/2 which represent the Cayley-Klein parameters (see Eq. (609) ). This gives the expression
U M M ′ J ( ω ; Θ , Φ ) = { ( − i v ) 2 J ( u − i v ) M + M ′ e − i ( M − M ′ ) Φ ∑ s ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! s ! ( s + M + M ′ ) ! ( J − M − s ) ! ( J − M ′ − s ) ! ( 1 − v − 2 ) s , M + M ′ > 0 , ( − i v ) 2 J ( u ∗ − i v ) − ( M + M ′ ) e − i ( M − M ′ ) Φ ∑ s ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! s ! ( s − M − M ′ ) ! ( J + M − s ) ! ( J + M ′ − s ) ! ( 1 − v − 2 ) s , M + M ′ ≤ 0. (601) U^J_{MM'}(\omega;\Theta,\Phi)=
\begin{cases}
(-iv)^{2J}\left(\dfrac{u}{-iv}\right)^{M+M'}e^{-i(M-M')\Phi}\sum_s\dfrac{\sqrt{(J+M)!(J-M)!(J+M')!(J-M')!}}{s!(s+M+M')!(J-M-s)!(J-M'-s)!}\left(1-v^{-2}\right)^s, & M+M'>0,\\[1.2em]
(-iv)^{2J}\left(\dfrac{u^*}{-iv}\right)^{-(M+M')}e^{-i(M-M')\Phi}\sum_s\dfrac{\sqrt{(J+M)!(J-M)!(J+M')!(J-M')!}}{s!(s-M-M')!(J+M-s)!(J+M'-s)!}\left(1-v^{-2}\right)^s, & M+M'\leq0.
\end{cases} \tag*{(601)} U M M ′ J ( ω ; Θ , Φ ) = ⎩ ⎨ ⎧ ( − i v ) 2 J ( − i v u ) M + M ′ e − i ( M − M ′ ) Φ ∑ s s ! ( s + M + M ′ )! ( J − M − s )! ( J − M ′ − s )! ( J + M )! ( J − M )! ( J + M ′ )! ( J − M ′ )! ( 1 − v − 2 ) s , ( − i v ) 2 J ( − i v u ∗ ) − ( M + M ′ ) e − i ( M − M ′ ) Φ ∑ s s ! ( s − M − M ′ )! ( J + M − s )! ( J + M ′ − s )! ( J + M )! ( J − M )! ( J + M ′ )! ( J − M ′ )! ( 1 − v − 2 ) s , M + M ′ > 0 , M + M ′ ≤ 0. (601)
In this case Eqs. (155) have been used, and the following notations are introduced
v = sin ω 2 sin Θ , u = cos ω 2 − i sin ω 2 cos Θ . (602) v=\sin\frac{\omega}{2}\sin\Theta,\qquad u=\cos\frac{\omega}{2}-i\sin\frac{\omega}{2}\cos\Theta. \tag*{(602)} v = sin 2 ω sin Θ , u = cos 2 ω − i sin 2 ω cos Θ. (602)
In Eqs. (601) the summation index s s s runs over all integer values which do not lead to negative factorial arguments.
*(c) Alternative expression for U U U
Another explicit form of U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) can be obtained directly from D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) by changing variables ( ω , Θ , Φ ) → ( α , β , γ ) (\omega,\Theta,\Phi)\to(\alpha,\beta,\gamma) ( ω , Θ , Φ ) → ( α , β , γ ) with the aid of Eqs. (145) and (1.4.17)
U M M ′ J ( ω ; Θ , Φ ) = i M − M ′ e − i ( M − M ′ ) Φ ( 1 − i tan ω 2 cos Θ 1 + tan 2 ω 2 cos 2 Θ ) M + M ′ d M M ′ J ( ξ ) . (603) U^J_{MM'}(\omega;\Theta,\Phi)=i^{M-M'}e^{-i(M-M')\Phi}\left(\frac{1-i\tan\frac{\omega}{2}\cos\Theta}{\sqrt{1+\tan^2\frac{\omega}{2}\cos^2\Theta}}\right)^{M+M'}d^J_{MM'}(\xi). \tag*{(603)} U M M ′ J ( ω ; Θ , Φ ) = i M − M ′ e − i ( M − M ′ ) Φ 1 + tan 2 2 ω cos 2 Θ 1 − i tan 2 ω cos Θ M + M ′ d M M ′ J ( ξ ) . (603)
The functions d M M ′ J ( ξ ) d^{J}_{MM'}(\xi) d M M ′ J ( ξ ) are defined in Sec. 4.3, and the angle ξ \xi ξ is determined by
sin ξ 2 = sin ω 2 sin Θ . (604) \sin\frac{\xi}{2}=\sin\frac{\omega}{2}\sin\Theta. \tag*{(604)} sin 2 ξ = sin 2 ω sin Θ. (604)
(d) Expansion in spherical harmonics
The function U M M ′ J ( ω ; Θ , Φ ) U^{J}_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) may be expanded in a series of the spherical harmonics which depend on polar angles Θ \Theta Θ and Φ \Phi Φ ,
U M M ′ J ( ω ; Θ , Φ ) = ∑ λ μ ( − i ) λ 2 λ + 1 2 J + 1 χ λ J ( ω ) C J M λ μ J M ′ 4 π 2 λ + 1 Y λ μ ( Θ , Φ ) , (605) U^{J}_{MM'}(\omega;\Theta,\Phi)=\sum_{\lambda\mu}(-i)^{\lambda}\frac{2\lambda+1}{2J+1}\chi^{J}_{\lambda}(\omega)C^{JM'}_{JM\lambda\mu}\sqrt{\frac{4\pi}{2\lambda+1}}Y_{\lambda\mu}(\Theta,\Phi), \tag*{(605)} U M M ′ J ( ω ; Θ , Φ ) = λ μ ∑ ( − i ) λ 2 J + 1 2 λ + 1 χ λ J ( ω ) C J M λ μ J M ′ 2 λ + 1 4 π Y λ μ ( Θ , Φ ) , (605)
where χ λ J ( ω ) \chi^{J}_{\lambda}(\omega) χ λ J ( ω ) is a generalized character (of order λ \lambda λ ) of the irreducible representation of rank J J J . Explicit forms and properties of χ λ J ( ω ) \chi^{J}_{\lambda}(\omega) χ λ J ( ω ) will be given below in Sec. 4.15. Equation (4.5.8) shows that U M M ′ J ( ω ; Θ , Φ ) U^{J}_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) depends on M M M and M ′ M' M ′ only through the Clebsch-Gordan coefficients C J M λ μ J M ′ C^{JM'}_{JM\lambda\mu} C J M λ μ J M ′ .
Differential Equations
U M M ′ J ( ω ; Θ , Φ ) U^{J}_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) are eigenfunctions of three operators J ^ z \hat{J}_{z} J ^ z , J ^ z ′ \hat{J}'_{z} J ^ z ′ and J ^ 2 \hat{J}^{2} J ^ 2 whose eigenvalues equal − M -M − M , − M ′ -M' − M ′ and J ( J + 1 ) J(J+1) J ( J + 1 ) , respectively. In terms of ω , Θ , Φ \omega,\Theta,\Phi ω , Θ , Φ these operators have the forms
J ^ z = − i [ cos Θ ∂ ∂ ω − 1 2 cot ω 2 sin Θ ∂ ∂ Θ + 1 2 ∂ ∂ Φ ] , (606) \hat{J}_{z}=-i\left[\cos\Theta\frac{\partial}{\partial\omega}-\frac{1}{2}\cot\frac{\omega}{2}\sin\Theta\frac{\partial}{\partial\Theta}+\frac{1}{2}\frac{\partial}{\partial\Phi}\right], \tag*{(606)} J ^ z = − i [ cos Θ ∂ ω ∂ − 2 1 cot 2 ω sin Θ ∂ Θ ∂ + 2 1 ∂ Φ ∂ ] , (606)
J ^ z ′ = − i [ cos Θ ∂ ∂ ω − 1 2 cot ω 2 sin Θ ∂ ∂ Θ − 1 2 ∂ ∂ Φ ] , (607) \hat{J}'_{z}=-i\left[\cos\Theta\frac{\partial}{\partial\omega}-\frac{1}{2}\cot\frac{\omega}{2}\sin\Theta\frac{\partial}{\partial\Theta}-\frac{1}{2}\frac{\partial}{\partial\Phi}\right], \tag*{(607)} J ^ z ′ = − i [ cos Θ ∂ ω ∂ − 2 1 cot 2 ω sin Θ ∂ Θ ∂ − 2 1 ∂ Φ ∂ ] , (607)
J ^ 2 = − [ ∂ 2 ∂ ω 2 + cot ω 2 ∂ ∂ ω + 1 4 sin 2 ω 2 ( ∂ 2 ∂ Θ 2 + cot Θ ∂ ∂ Θ + 1 sin 2 Θ ∂ 2 ∂ Φ 2 ) ] . (608) \hat{J}^{2}=-\left[\frac{\partial^{2}}{\partial\omega^{2}}+\cot\frac{\omega}{2}\frac{\partial}{\partial\omega}+\frac{1}{4\sin^{2}\frac{\omega}{2}}\left(\frac{\partial^{2}}{\partial\Theta^{2}}+\cot\Theta\frac{\partial}{\partial\Theta}+\frac{1}{\sin^{2}\Theta}\frac{\partial^{2}}{\partial\Phi^{2}}\right)\right]. \tag*{(608)} J ^ 2 = − [ ∂ ω 2 ∂ 2 + cot 2 ω ∂ ω ∂ + 4 sin 2 2 ω 1 ( ∂ Θ 2 ∂ 2 + cot Θ ∂ Θ ∂ + sin 2 Θ 1 ∂ Φ 2 ∂ 2 ) ] . (608)
Thus, U M M ′ J ( ω ; Θ , Φ ) U^{J}_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) is a solution of the differential equations
[ J ^ 2 − J ( J + 1 ) ] U M M ′ J ( ω ; Θ , Φ ) = 0 \left[\hat{J}^{2}-J(J+1)\right]U^{J}_{MM'}(\omega;\Theta,\Phi)=0 [ J ^ 2 − J ( J + 1 ) ] U M M ′ J ( ω ; Θ , Φ ) = 0
[ J ^ z + M ] U M M ′ J ( ω ; Θ , Φ ) = 0 , \left[\hat{J}_{z}+M\right]U^{J}_{MM'}(\omega;\Theta,\Phi)=0, [ J ^ z + M ] U M M ′ J ( ω ; Θ , Φ ) = 0 ,
[ J ^ z ′ + M ′ ] U M M ′ J ( ω ; Θ , Φ ) = 0. (609) \left[\hat{J}'_{z}+M'\right]U^{J}_{MM'}(\omega;\Theta,\Phi)=0. \tag*{(609)} [ J ^ z ′ + M ′ ] U M M ′ J ( ω ; Θ , Φ ) = 0. (609)
with the boundary conditions
U M M ′ J ( 0 ; Θ , Φ ) = δ M M ′ U^{J}_{MM'}(0;\Theta,\Phi)=\delta_{MM'} U M M ′ J ( 0 ; Θ , Φ ) = δ M M ′
∂ ∂ Φ U M M ′ J ( ω ; Θ , Φ ) ∣ Θ = 0 = ∂ ∂ Φ U M M ′ J ( ω ; Θ , Φ ) ∣ Θ = π = 0. (610) \left.\frac{\partial}{\partial\Phi}U^{J}_{MM'}(\omega;\Theta,\Phi)\right|_{\Theta=0}
=
\left.\frac{\partial}{\partial\Phi}U^{J}_{MM'}(\omega;\Theta,\Phi)\right|_{\Theta=\pi}
=0. \tag*{(610)} ∂ Φ ∂ U M M ′ J ( ω ; Θ , Φ ) Θ = 0 = ∂ Φ ∂ U M M ′ J ( ω ; Θ , Φ ) Θ = π = 0. (610)
Orthogonality and Completeness
A collection of the functions U M M ′ J ( ω ; Θ , Φ ) U^{J}_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) with all possible integer and half-integer J ≥ 0 J\geq0 J ≥ 0 constitutes a complete set of orthogonal functions of three variables ω , Θ , Φ \omega,\Theta,\Phi ω , Θ , Φ defined in the domain
0 ≤ θ ≤ π , 0 ≤ Φ < 2 π , 0 ≤ ω < 2 π , (611) 0\leq\theta\leq\pi,\qquad0\leq\Phi<2\pi,\qquad0\leq\omega<2\pi, \tag*{(611)} 0 ≤ θ ≤ π , 0 ≤ Φ < 2 π , 0 ≤ ω < 2 π , (611)
whose total volume is equal to 16 π 2 16\pi^{2} 16 π 2 .
(a) Orthogonality and Normalization
4 ∫ 0 2 π d ω sin 2 ω 2 ∫ 0 π d Θ sin Θ ∫ 0 2 π d Φ U M 1 M 1 ′ J 1 ∗ ( ω ; Θ , Φ ) U M 2 M 2 ′ J 2 ( ω ; Θ , Φ ) = 16 π 2 2 J 1 + 1 δ J 1 J 2 δ M 1 M 2 δ M 1 ′ M 2 ′ . (612) 4\int_{0}^{2\pi}d\omega\sin^{2}\frac{\omega}{2}\int_{0}^{\pi}d\Theta\sin\Theta\int_{0}^{2\pi}d\Phi\,U^{J_{1}*}_{M_{1}M'_{1}}(\omega;\Theta,\Phi)U^{J_{2}}_{M_{2}M'_{2}}(\omega;\Theta,\Phi)=\frac{16\pi^{2}}{2J_{1}+1}\delta_{J_{1}J_{2}}\delta_{M_{1}M_{2}}\delta_{M'_{1}M'_{2}}. \tag*{(612)} 4 ∫ 0 2 π d ω sin 2 2 ω ∫ 0 π d Θ sin Θ ∫ 0 2 π d Φ U M 1 M 1 ′ J 1 ∗ ( ω ; Θ , Φ ) U M 2 M 2 ′ J 2 ( ω ; Θ , Φ ) = 2 J 1 + 1 16 π 2 δ J 1 J 2 δ M 1 M 2 δ M 1 ′ M 2 ′ . (612)
(b) Completeness
∑ J = 0 , 1 / 2 , 1 … ∞ 2 J + 1 16 π 2 ∑ M M ′ U M M ′ J ( ω ; Θ , Φ ) U M M ′ J ∗ ( ω ~ ; Θ ~ , Φ ~ ) = δ ( Φ − Φ ~ ) δ ( Θ − Θ ~ ) δ ( ω − ω ~ ) 4 sin Θ sin 2 ω 2 . (613) \sum_{J=0,1/2,1\ldots}^{\infty}\frac{2J+1}{16\pi^{2}}\sum_{MM'}U^{J}_{MM'}(\omega;\Theta,\Phi)U^{J*}_{MM'}(\widetilde{\omega};\widetilde{\Theta},\widetilde{\Phi})=\frac{\delta(\Phi-\widetilde{\Phi})\delta(\Theta-\widetilde{\Theta})\delta(\omega-\widetilde{\omega})}{4\sin\Theta\sin^{2}\frac{\omega}{2}}. \tag*{(613)} J = 0 , 1/2 , 1 … ∑ ∞ 16 π 2 2 J + 1 M M ′ ∑ U M M ′ J ( ω ; Θ , Φ ) U M M ′ J ∗ ( ω ; Θ , Φ ) = 4 sin Θ sin 2 2 ω δ ( Φ − Φ ) δ ( Θ − Θ ) δ ( ω − ω ) . (613)
Principal Properties
[ U − 1 ( ω ; Θ , Φ ) ] M M ′ J = U M M ′ J ( − ω ; Θ , Φ ) = U M M ′ J ( ω ; π − Θ , π + Φ ) . (614) \left[U^{-1}(\omega;\Theta,\Phi)\right]^J_{MM'}=U^J_{MM'}(-\omega;\Theta,\Phi)=U^J_{MM'}(\omega;\pi-\Theta,\pi+\Phi). \tag*{(614)} [ U − 1 ( ω ; Θ , Φ ) ] M M ′ J = U M M ′ J ( − ω ; Θ , Φ ) = U M M ′ J ( ω ; π − Θ , π + Φ ) . (614)
U M M ′ J ∗ ( ω ; Θ , Φ ) = U M ′ M J ( ω ; π − Θ , π + Φ ) = U M ′ M J ( − ω ; Θ , Φ ) . (615) U^{J*}_{MM'}(\omega;\Theta,\Phi)=U^J_{M'M}(\omega;\pi-\Theta,\pi+\Phi)=U^J_{M'M}(-\omega;\Theta,\Phi). \tag*{(615)} U M M ′ J ∗ ( ω ; Θ , Φ ) = U M ′ M J ( ω ; π − Θ , π + Φ ) = U M ′ M J ( − ω ; Θ , Φ ) . (615)
U M M ′ J ( − ω ; Θ , Φ ) = ( − 1 ) M − M ′ U − M ′ − M J ( ω ; Θ , Φ ) , U M M ′ J ( ω ; − Θ , Φ ) = ( − 1 ) M − M ′ U M M ′ J ( ω ; Θ , Φ ) , U M M ′ J ( ω ; Θ , − Φ ) = U M ′ M J ( ω ; Θ , Φ ) . (616) \begin{aligned}
U^J_{MM'}(-\omega;\Theta,\Phi)&=(-1)^{M-M'}U^J_{-M'-M}(\omega;\Theta,\Phi),\\
U^J_{MM'}(\omega;-\Theta,\Phi)&=(-1)^{M-M'}U^J_{MM'}(\omega;\Theta,\Phi),\\
U^J_{MM'}(\omega;\Theta,-\Phi)&=U^J_{M'M}(\omega;\Theta,\Phi). \tag*{(616)}
\end{aligned} U M M ′ J ( − ω ; Θ , Φ ) U M M ′ J ( ω ; − Θ , Φ ) U M M ′ J ( ω ; Θ , − Φ ) = ( − 1 ) M − M ′ U − M ′ − M J ( ω ; Θ , Φ ) , = ( − 1 ) M − M ′ U M M ′ J ( ω ; Θ , Φ ) , = U M ′ M J ( ω ; Θ , Φ ) . (616)
U M M ′ J ( ω + 4 π ; Θ , Φ ) = U M M ′ J ( ω ; Θ , Φ ) , U M M ′ J ( ω ; Θ + 2 π , Φ ) = U M M ′ J ( ω ; Θ , Φ ) , U M M ′ J ( ω ; Θ , Φ + 2 π ) = U M M ′ J ( ω ; Θ , Φ ) . (617) \begin{aligned}
U^J_{MM'}(\omega+4\pi;\Theta,\Phi)&=U^J_{MM'}(\omega;\Theta,\Phi),\\
U^J_{MM'}(\omega;\Theta+2\pi,\Phi)&=U^J_{MM'}(\omega;\Theta,\Phi),\\
U^J_{MM'}(\omega;\Theta,\Phi+2\pi)&=U^J_{MM'}(\omega;\Theta,\Phi). \tag*{(617)}
\end{aligned} U M M ′ J ( ω + 4 π ; Θ , Φ ) U M M ′ J ( ω ; Θ + 2 π , Φ ) U M M ′ J ( ω ; Θ , Φ + 2 π ) = U M M ′ J ( ω ; Θ , Φ ) , = U M M ′ J ( ω ; Θ , Φ ) , = U M M ′ J ( ω ; Θ , Φ ) . (617)
U M M ′ J ( ω + 2 π ; Θ , Φ ) = ( − 1 ) 2 J U M M ′ J ( ω ; Θ , Φ ) , U M M ′ J ( 2 π − ω ; Θ , Φ ) = ( − 1 ) M + M ′ U − M ′ − M J ( ω ; Θ , Φ ) , U M M ′ J ( ω ; Θ + π , Φ ) = ( − 1 ) M − M ′ U − M ′ − M J ( ω ; Θ , Φ ) , U M M ′ J ( ω ; π − Θ , Φ ) = U − M ′ − M J ( ω ; Θ , Φ ) , U M M ′ J ( ω ; Θ , Φ + π ) = ( − 1 ) M − M ′ U M M ′ J ( ω ; Θ , Φ ) , U M M ′ J ( ω ; Θ , π − Φ ) = ( − 1 ) M − M ′ U M ′ M J ( ω ; Θ , Φ ) . (618) \begin{aligned}
U^J_{MM'}(\omega+2\pi;\Theta,\Phi)&=(-1)^{2J}U^J_{MM'}(\omega;\Theta,\Phi),&
U^J_{MM'}(2\pi-\omega;\Theta,\Phi)&=(-1)^{M+M'}U^J_{-M'-M}(\omega;\Theta,\Phi),\\
U^J_{MM'}(\omega;\Theta+\pi,\Phi)&=(-1)^{M-M'}U^J_{-M'-M}(\omega;\Theta,\Phi),&
U^J_{MM'}(\omega;\pi-\Theta,\Phi)&=U^J_{-M'-M}(\omega;\Theta,\Phi),\\
U^J_{MM'}(\omega;\Theta,\Phi+\pi)&=(-1)^{M-M'}U^J_{MM'}(\omega;\Theta,\Phi),&
U^J_{MM'}(\omega;\Theta,\pi-\Phi)&=(-1)^{M-M'}U^J_{M'M}(\omega;\Theta,\Phi). \tag*{(618)}
\end{aligned} U M M ′ J ( ω + 2 π ; Θ , Φ ) U M M ′ J ( ω ; Θ + π , Φ ) U M M ′ J ( ω ; Θ , Φ + π ) = ( − 1 ) 2 J U M M ′ J ( ω ; Θ , Φ ) , = ( − 1 ) M − M ′ U − M ′ − M J ( ω ; Θ , Φ ) , = ( − 1 ) M − M ′ U M M ′ J ( ω ; Θ , Φ ) , U M M ′ J ( 2 π − ω ; Θ , Φ ) U M M ′ J ( ω ; π − Θ , Φ ) U M M ′ J ( ω ; Θ , π − Φ ) = ( − 1 ) M + M ′ U − M ′ − M J ( ω ; Θ , Φ ) , = U − M ′ − M J ( ω ; Θ , Φ ) , = ( − 1 ) M − M ′ U M ′ M J ( ω ; Θ , Φ ) . (618)
U M ′ M J ( ω ; Θ , Φ ) = U M M ′ J ( ω ; Θ , − Φ ) = U M M ′ J ∗ ( ω ; π − Θ , π + Φ ) , U − M − M ′ J ( ω ; Θ , Φ ) = ( − 1 ) M − M ′ U M M ′ J ( ω ; π − Θ , π − Φ ) , U M M ′ J ( ω ; Θ , Φ ) = ( − 1 ) J − M ′ U − M ′ − M J ( ω 1 ; Θ 1 , Φ 1 ) = ( − 1 ) J + M U M − M ′ J ( ω 1 ; Θ 1 , Φ 1 ) . (619) \begin{aligned}
U^J_{M'M}(\omega;\Theta,\Phi)&=U^J_{MM'}(\omega;\Theta,-\Phi)=U^{J*}_{MM'}(\omega;\pi-\Theta,\pi+\Phi),\\
U^J_{-M-M'}(\omega;\Theta,\Phi)&=(-1)^{M-M'}U^J_{MM'}(\omega;\pi-\Theta,\pi-\Phi),\\
U^J_{MM'}(\omega;\Theta,\Phi)&=(-1)^{J-M'}U^J_{-M'-M}(\omega_1;\Theta_1,\Phi_1)=(-1)^{J+M}U^J_{M-M'}(\omega_1;\Theta_1,\Phi_1). \tag*{(619)}
\end{aligned} U M ′ M J ( ω ; Θ , Φ ) U − M − M ′ J ( ω ; Θ , Φ ) U M M ′ J ( ω ; Θ , Φ ) = U M M ′ J ( ω ; Θ , − Φ ) = U M M ′ J ∗ ( ω ; π − Θ , π + Φ ) , = ( − 1 ) M − M ′ U M M ′ J ( ω ; π − Θ , π − Φ ) , = ( − 1 ) J − M ′ U − M ′ − M J ( ω 1 ; Θ 1 , Φ 1 ) = ( − 1 ) J + M U M − M ′ J ( ω 1 ; Θ 1 , Φ 1 ) . (619)
where angles ω 1 \omega_1 ω 1 , Θ 1 \Theta_1 Θ 1 , Φ 1 \Phi_1 Φ 1 and ω \omega ω , Θ \Theta Θ , Φ \Phi Φ are related by the equations
cos ω 1 2 = sin ω 2 sin Θ sin Φ , sin ω 1 2 cos Θ 1 = − sin ω 2 sin Θ cos Φ , cot Φ 1 = − tan ω 2 cos Θ . (620) \begin{aligned}
\cos\frac{\omega_1}{2}&=\sin\frac{\omega}{2}\sin\Theta\sin\Phi,\\
\sin\frac{\omega_1}{2}\cos\Theta_1&=-\sin\frac{\omega}{2}\sin\Theta\cos\Phi,\\
\cot\Phi_1&=-\tan\frac{\omega}{2}\cos\Theta. \tag*{(620)}
\end{aligned} cos 2 ω 1 sin 2 ω 1 cos Θ 1 cot Φ 1 = sin 2 ω sin Θ sin Φ , = − sin 2 ω sin Θ cos Φ , = − tan 2 ω cos Θ. (620)
cos ω 2 = sin ω 1 2 sin Θ 1 sin Φ 1 , sin ω 2 cos Θ = − sin ω 1 2 sin Θ 1 cos Φ 1 , cot Φ = − tan ω 1 2 cos Θ 1 . (621) \begin{aligned}
\cos\frac{\omega}{2}&=\sin\frac{\omega_1}{2}\sin\Theta_1\sin\Phi_1,\\
\sin\frac{\omega}{2}\cos\Theta&=-\sin\frac{\omega_1}{2}\sin\Theta_1\cos\Phi_1,\\
\cot\Phi&=-\tan\frac{\omega_1}{2}\cos\Theta_1. \tag*{(621)}
\end{aligned} cos 2 ω sin 2 ω cos Θ cot Φ = sin 2 ω 1 sin Θ 1 sin Φ 1 , = − sin 2 ω 1 sin Θ 1 cos Φ 1 , = − tan 2 ω 1 cos Θ 1 . (621)
sin 2 ω 1 2 sin 2 Θ 1 + sin 2 ω 2 sin 2 Θ = 1. (622) \sin^2\frac{\omega_1}{2}\sin^2\Theta_1+\sin^2\frac{\omega}{2}\sin^2\Theta=1. \tag*{(622)} sin 2 2 ω 1 sin 2 Θ 1 + sin 2 2 ω sin 2 Θ = 1. (622)
Some relations which are valid for D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) remain valid for U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) . In particular, the Clebsch-Gordon expansion
U M 1 M 1 ′ J 1 ( ω ; Θ , Φ ) U M 2 M 2 ′ J 2 ( ω ; Θ , Φ ) = ∑ J M C J 1 M 1 J 2 M 2 J M U M M ′ J ( ω ; Θ , Φ ) C J 1 M 1 ′ J 2 M 2 ′ J M ′ . (623) U^{J_1}_{M_1M'_1}(\omega;\Theta,\Phi)U^{J_2}_{M_2M'_2}(\omega;\Theta,\Phi)=\sum_{JM}C^{JM}_{J_1M_1J_2M_2}U^J_{MM'}(\omega;\Theta,\Phi)C^{JM'}_{J_1M'_1J_2M'_2}. \tag*{(623)} U M 1 M 1 ′ J 1 ( ω ; Θ , Φ ) U M 2 M 2 ′ J 2 ( ω ; Θ , Φ ) = J M ∑ C J 1 M 1 J 2 M 2 J M U M M ′ J ( ω ; Θ , Φ ) C J 1 M 1 ′ J 2 M 2 ′ J M ′ . (623)
is equivalent to Eq. (630) , and the addition theorem
∑ M ′ ′ U M M ′ ′ J ( ω 2 ; Θ 2 , Φ 2 ) U M ′ ′ M ′ J ( ω 1 ; Θ 1 , Φ 1 ) = U M M ′ J ( ω ; Θ , Φ ) (624) \sum_{M''} U^{J}_{MM''}(\omega_{2};\Theta_{2},\Phi_{2})U^{J}_{M''M'}(\omega_{1};\Theta_{1},\Phi_{1})=U^{J}_{MM'}(\omega;\Theta,\Phi) \tag*{(624)} M ′′ ∑ U M M ′′ J ( ω 2 ; Θ 2 , Φ 2 ) U M ′′ M ′ J ( ω 1 ; Θ 1 , Φ 1 ) = U M M ′ J ( ω ; Θ , Φ ) (624)
is equivalent to Eq. (645) . The angles ω 1 \omega_{1} ω 1 , Θ 1 \Theta_{1} Θ 1 , Φ 1 \Phi_{1} Φ 1 , ω 2 \omega_{2} ω 2 , Θ 2 \Theta_{2} Θ 2 , Φ 2 \Phi_{2} Φ 2 and ω \omega ω , Θ \Theta Θ , Φ \Phi Φ are related by Eqs. (205) .
Special Cases
(a) The rotation axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) coincides with one of the coordinate axes
(i) The x x x -axis ( Θ = π 2 , Φ = 0 ) (\Theta=\frac{\pi}{2},\Phi=0) ( Θ = 2 π , Φ = 0 )
U M M ′ J ( ω ; π 2 , 0 ) = D M M ′ J ( − π 2 , ω , π 2 ) = ( i ) M − M ′ d M M ′ J ( ω ) . (625) U^{J}_{MM'}\left(\omega;\frac{\pi}{2},0\right)=D^{J}_{MM'}\left(-\frac{\pi}{2},\omega,\frac{\pi}{2}\right)=(i)^{M-M'}d^{J}_{MM'}(\omega). \tag*{(625)} U M M ′ J ( ω ; 2 π , 0 ) = D M M ′ J ( − 2 π , ω , 2 π ) = ( i ) M − M ′ d M M ′ J ( ω ) . (625)
(ii) The y y y -axis ( Θ = π 2 , Φ = π 2 ) (\Theta=\frac{\pi}{2},\Phi=\frac{\pi}{2}) ( Θ = 2 π , Φ = 2 π )
U M M ′ J ( ω ; π 2 , π 2 ) = D M M ′ J ( 0 , ω , 0 ) = d M M ′ J ( ω ) . (626) U^{J}_{MM'}\left(\omega;\frac{\pi}{2},\frac{\pi}{2}\right)=D^{J}_{MM'}(0,\omega,0)=d^{J}_{MM'}(\omega). \tag*{(626)} U M M ′ J ( ω ; 2 π , 2 π ) = D M M ′ J ( 0 , ω , 0 ) = d M M ′ J ( ω ) . (626)
(iii) The z z z -axis ( Θ = 0 ) (\Theta=0) ( Θ = 0 )
U M M ′ J ( ω ; 0 , Φ ) = δ M M ′ e − i M ω . (627) U^{J}_{MM'}(\omega;0,\Phi)=\delta_{MM'}e^{-iM\omega}. \tag*{(627)} U M M ′ J ( ω ; 0 , Φ ) = δ M M ′ e − i M ω . (627)
(b) Rotation about n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) through a small angle ω ≪ π 2 \omega\ll\frac{\pi}{2} ω ≪ 2 π
U M M ′ J ( ω ; Θ , Φ ) = δ M M ′ ( 1 − i ω M cos Θ ) − i ω 2 e − i Φ sin Θ ( J − M ′ ) ( J + M ) δ M , M ′ + 1 − i ω 2 e i Φ ( J − M ) ( J + M ′ ) δ M , M ′ − 1 . (628) U^{J}_{MM'}(\omega;\Theta,\Phi)=\delta_{MM'}(1-i\omega M\cos\Theta)-\frac{i\omega}{2}e^{-i\Phi}\sin\Theta\sqrt{(J-M')(J+M)}\delta_{M,M'+1}
-\frac{i\omega}{2}e^{i\Phi}\sqrt{(J-M)(J+M')}\delta_{M,M'-1}. \tag*{(628)} U M M ′ J ( ω ; Θ , Φ ) = δ M M ′ ( 1 − iω M cos Θ ) − 2 iω e − i Φ sin Θ ( J − M ′ ) ( J + M ) δ M , M ′ + 1 − 2 iω e i Φ ( J − M ) ( J + M ′ ) δ M , M ′ − 1 . (628)
(c) Explicit forms of U M M ′ J ( ω ; Θ , Φ ) U^{J}_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) at M = ± J M=\pm J M = ± J and/or M ′ = ± J M'=\pm J M ′ = ± J
U J J J ( ω ; θ , Φ ) = ( cos ω 2 − i sin ω 2 cos Θ ) 2 J , U − J , − J J ( ω ; θ , Φ ) = ( cos ω 2 + i sin ω 2 cos Θ ) 2 J , U^{J}_{JJ}(\omega;\theta,\Phi)=\left(\cos\frac{\omega}{2}-i\sin\frac{\omega}{2}\cos\Theta\right)^{2J},\qquad U^{J}_{-J,-J}(\omega;\theta,\Phi)=\left(\cos\frac{\omega}{2}+i\sin\frac{\omega}{2}\cos\Theta\right)^{2J}, U J J J ( ω ; θ , Φ ) = ( cos 2 ω − i sin 2 ω cos Θ ) 2 J , U − J , − J J ( ω ; θ , Φ ) = ( cos 2 ω + i sin 2 ω cos Θ ) 2 J ,
U J , − J J ( ω ; Θ , Φ ) = ( − i sin ω 2 sin Θ e − i Φ ) 2 J , U − J , J J ( ω ; Θ , Φ ) = ( − i sin ω 2 sin Θ e i Φ ) 2 J . (629) U^{J}_{J,-J}(\omega;\Theta,\Phi)=\left(-i\sin\frac{\omega}{2}\sin\Theta e^{-i\Phi}\right)^{2J},\qquad U^{J}_{-J,J}(\omega;\Theta,\Phi)=\left(-i\sin\frac{\omega}{2}\sin\Theta e^{i\Phi}\right)^{2J}. \tag*{(629)} U J , − J J ( ω ; Θ , Φ ) = ( − i sin 2 ω sin Θ e − i Φ ) 2 J , U − J , J J ( ω ; Θ , Φ ) = ( − i sin 2 ω sin Θ e i Φ ) 2 J . (629)
(d) Explicit forms of U M M ′ J ( ω ; θ , Φ ) U^{J}_{MM'}(\omega;\theta,\Phi) U M M ′ J ( ω ; θ , Φ ) for J = 1 2 , 1 , 3 2 , 2 J=\frac{1}{2},1,\frac{3}{2},2 J = 2 1 , 1 , 2 3 , 2 are given below in Tables 4.23–4.26.
Sums involving D-functions
The Clebsch-Gordan Series
The product of two D D D -functions with the same arguments may be expanded in the following series
D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) = ∑ J = ∣ J 1 − J 2 ∣ J 1 + J 2 ∑ M N C J 1 M 1 J 2 M 2 J M D M N J ( α , β , γ ) C J 1 N 1 J 2 N 2 J N , (630) D^{J_{1}}_{M_{1}N_{1}}(\alpha,\beta,\gamma)D^{J_{2}}_{M_{2}N_{2}}(\alpha,\beta,\gamma)=\sum_{J=\lvert J_{1}-J_{2}\rvert}^{J_{1}+J_{2}}\sum_{MN}C^{J M}_{J_{1}M_{1}\,J_{2}M_{2}}D^{J}_{MN}(\alpha,\beta,\gamma)C^{J N}_{J_{1}N_{1}\,J_{2}N_{2}}, \tag*{(630)} D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) = J = ∣ J 1 − J 2 ∣ ∑ J 1 + J 2 M N ∑ C J 1 M 1 J 2 M 2 J M D M N J ( α , β , γ ) C J 1 N 1 J 2 N 2 J N , (630)
Here C J 1 M 1 J 2 M 2 J M C^{J M}_{J_{1}M_{1}\,J_{2}M_{2}} C J 1 M 1 J 2 M 2 J M is a Clebsch-Gordan coefficient (Chap. 8). The sum in Eq. (630) has 2 j + 1 2j+1 2 j + 1 terms, where j = min ( J 1 , J 2 ) j=\min(J_{1},J_{2}) j = min ( J 1 , J 2 ) . Equation (4.6.1) may be regarded as a particular case of the expansion of an arbitrary function in a series of the D D D -functions (Sec. 4.10).
Some Applications of the Clebsch-Gordan Expansion
The Clebsch-Gordan expansion, Eq. (630) , together with the orthogonality condition of the Clebsch-Gordan coefficients, Eq. (8.1.8), enable one to calculate sums of products of the D D D -functions with identical arguments. Hereafter we introduce the 3 j 3j 3 j -symbol { j 1 j 2 j 3 } \left\{\begin{matrix}j_{1}&j_{2}&j_{3}\end{matrix}\right\} { j 1 j 2 j 3 } defined by
{ j 1 j 2 j 3 } = { 1 if j 1 + j 2 + j 3 is integer and ∣ j 1 − j 2 ∣ ≤ j 3 ≤ j 1 + j 2 , 0 otherwise , (631) \left\{\begin{matrix}j_{1}&j_{2}&j_{3}\end{matrix}\right\}=
\begin{cases}
1 & \text{if }j_{1}+j_{2}+j_{3}\text{ is integer and }\lvert j_{1}-j_{2}\rvert\le j_{3}\le j_{1}+j_{2},\\
0 & \text{otherwise},
\end{cases} \tag*{(631)} { j 1 j 2 j 3 } = { 1 0 if j 1 + j 2 + j 3 is integer and ∣ j 1 − j 2 ∣ ≤ j 3 ≤ j 1 + j 2 , otherwise , (631)
{ j 1 j 2 j 3 } \{j_1j_2j_3\} { j 1 j 2 j 3 } is invariant with respect to permutations of j 1 , j 2 , j 3 j_1,j_2,j_3 j 1 , j 2 , j 3 . Using Eqs. (630) and (8.1.8), one obtains
∑ M 1 M 2 N 1 N 2 C J 1 M 1 J 2 M 2 J M D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) C J 1 N 1 J 2 N 2 J ′ N ′ = δ J J ′ { J 1 J 2 J } D M N J ( α , β , γ ) . (632) \sum_{\substack{M_1M_2\\N_1N_2}} C^{JM}_{J_1M_1J_2M_2}D^{J_1}_{M_1N_1}(\alpha,\beta,\gamma)D^{J_2}_{M_2N_2}(\alpha,\beta,\gamma)C^{J'N'}_{J_1N_1J_2N_2}=\delta_{JJ'}\{J_1J_2J\}D^J_{MN}(\alpha,\beta,\gamma). \tag*{(632)} M 1 M 2 N 1 N 2 ∑ C J 1 M 1 J 2 M 2 J M D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) C J 1 N 1 J 2 N 2 J ′ N ′ = δ J J ′ { J 1 J 2 J } D M N J ( α , β , γ ) . (632)
∑ J = ∣ J 1 − J 2 ∣ J 1 + J 2 ∑ M 1 M 2 C J 1 M 1 J 2 M 2 J M D M 1 N 1 J 1 ∗ ( α , β , γ ) D M N J ( α , β , γ ) C J 1 N 1 J 2 N 2 J ′ N ′ = δ N 1 N 1 ′ D M 2 N 2 J 2 ( α , β , γ ) . (633) \sum_{J=\lvert J_1-J_2\rvert}^{J_1+J_2}\sum_{M_1M_2}C^{JM}_{J_1M_1J_2M_2}D^{J_1*}_{M_1N_1}(\alpha,\beta,\gamma)D^J_{MN}(\alpha,\beta,\gamma)C^{J'N'}_{J_1N_1J_2N_2}=\delta_{N_1N'_1}D^{J_2}_{M_2N_2}(\alpha,\beta,\gamma). \tag*{(633)} J = ∣ J 1 − J 2 ∣ ∑ J 1 + J 2 M 1 M 2 ∑ C J 1 M 1 J 2 M 2 J M D M 1 N 1 J 1 ∗ ( α , β , γ ) D M N J ( α , β , γ ) C J 1 N 1 J 2 N 2 J ′ N ′ = δ N 1 N 1 ′ D M 2 N 2 J 2 ( α , β , γ ) . (633)
∑ N 1 N 2 D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) C J 1 N 1 J 2 N 2 J N = C J 1 M 1 J 2 M 2 J M D M N J ( α , β , γ ) . (634) \sum_{N_1N_2}D^{J_1}_{M_1N_1}(\alpha,\beta,\gamma)D^{J_2}_{M_2N_2}(\alpha,\beta,\gamma)C^{JN}_{J_1N_1J_2N_2}=C^{JM}_{J_1M_1J_2M_2}D^J_{MN}(\alpha,\beta,\gamma). \tag*{(634)} N 1 N 2 ∑ D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) C J 1 N 1 J 2 N 2 J N = C J 1 M 1 J 2 M 2 J M D M N J ( α , β , γ ) . (634)
∑ N 1 N 2 N D M N J ∗ ( α , β , γ ) D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) C J 1 N 1 J 2 N 2 J N = C J 1 M 1 J 2 M 2 J M . (635) \sum_{N_1N_2N}D^{J*}_{MN}(\alpha,\beta,\gamma)D^{J_1}_{M_1N_1}(\alpha,\beta,\gamma)D^{J_2}_{M_2N_2}(\alpha,\beta,\gamma)C^{JN}_{J_1N_1J_2N_2}=C^{JM}_{J_1M_1J_2M_2}. \tag*{(635)} N 1 N 2 N ∑ D M N J ∗ ( α , β , γ ) D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) C J 1 N 1 J 2 N 2 J N = C J 1 M 1 J 2 M 2 J M . (635)
∑ M 1 M 2 M N 1 N 2 N C J 1 M 1 J 2 M 2 J M D M N J ∗ ( α , β , γ ) D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) C J 1 N 1 J 2 N 2 J ′ N ′ = δ J J ′ δ N N ′ { J 1 J 2 J } . (636) \sum_{M_1M_2M\atop N_1N_2N}C^{JM}_{J_1M_1J_2M_2}D^{J*}_{MN}(\alpha,\beta,\gamma)D^{J_1}_{M_1N_1}(\alpha,\beta,\gamma)D^{J_2}_{M_2N_2}(\alpha,\beta,\gamma)C^{J'N'}_{J_1N_1J_2N_2}=\delta_{JJ'}\delta_{NN'}\{J_1J_2J\}. \tag*{(636)} N 1 N 2 N M 1 M 2 M ∑ C J 1 M 1 J 2 M 2 J M D M N J ∗ ( α , β , γ ) D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) C J 1 N 1 J 2 N 2 J ′ N ′ = δ J J ′ δ N N ′ { J 1 J 2 J } . (636)
Generalization of the Clebsch-Gordan Expansion
The Clebsch-Gordan expansion can be generalized to the case of an arbitrary number of D D D -functions of identical arguments by successive use of Eq. (632) . For instance, the summation of products of three D D D -functions yields
∑ M 1 M 2 M 3 N 1 N 2 N 3 C J 12 M 12 J 3 M 3 J M C J 1 M 1 J 2 M 2 J 12 M 12 D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) D M 3 N 3 J 3 ( α , β , γ ) C J 1 N 1 J 2 N 2 J 12 ′ N 12 C J 12 ′ N 12 J 3 N 3 J ′ N = δ J J ′ δ J 12 J 12 ′ { J 1 J 2 J 12 } { J 12 J 3 J } D M N J ( α , β , γ ) . (637) \sum_{\substack{M_1M_2M_3\\N_1N_2N_3}}C^{JM}_{J_{12}M_{12}J_3M_3}C^{J_{12}M_{12}}_{J_1M_1J_2M_2}D^{J_1}_{M_1N_1}(\alpha,\beta,\gamma)D^{J_2}_{M_2N_2}(\alpha,\beta,\gamma)D^{J_3}_{M_3N_3}(\alpha,\beta,\gamma)C^{J'_{12}N_{12}}_{J_1N_1J_2N_2}C^{J'N}_{J'_{12}N_{12}J_3N_3}
=\delta_{JJ'}\delta_{J_{12}J'_{12}}\{J_1J_2J_{12}\}\{J_{12}J_3J\}D^J_{MN}(\alpha,\beta,\gamma). \tag*{(637)} M 1 M 2 M 3 N 1 N 2 N 3 ∑ C J 12 M 12 J 3 M 3 J M C J 1 M 1 J 2 M 2 J 12 M 12 D M 1 N 1 J 1 ( α , β , γ ) D M 2 N 2 J 2 ( α , β , γ ) D M 3 N 3 J 3 ( α , β , γ ) C J 1 N 1 J 2 N 2 J 12 ′ N 12 C J 12 ′ N 12 J 3 N 3 J ′ N = δ J J ′ δ J 12 J 12 ′ { J 1 J 2 J 12 } { J 12 J 3 J } D M N J ( α , β , γ ) . (637)
In particular, for the case when J 3 = 0 J_3=0 J 3 = 0 , Eq. (637) is reduced to Eq. (632) .
In general, the sum of products of k D kD k D -functions is given by
∑ m 1 , … , m k n 1 , … , n k ∏ i = 1 k C J i − 1 M i − 1 j i m i J i M i D m i n i j i ( α , β , γ ) C J i − 1 ′ N i − 1 j i n i J i ′ N i = D M k N k J k ( α , β , γ ) ∏ i = 1 k δ J i J i ′ { j i J i − 1 J i } . (638) \sum_{\substack{m_1,\ldots,m_k\\n_1,\ldots,n_k}}\prod_{i=1}^{k}C^{J_iM_i}_{J_{i-1}M_{i-1}j_im_i}D^{j_i}_{m_in_i}(\alpha,\beta,\gamma)C^{J'_iN_i}_{J'_{i-1}N_{i-1}j_in_i}=D^{J_k}_{M_kN_k}(\alpha,\beta,\gamma)\prod_{i=1}^{k}\delta_{J_iJ'_i}\{j_iJ_{i-1}J_i\}. \tag*{(638)} m 1 , … , m k n 1 , … , n k ∑ i = 1 ∏ k C J i − 1 M i − 1 j i m i J i M i D m i n i j i ( α , β , γ ) C J i − 1 ′ N i − 1 j i n i J i ′ N i = D M k N k J k ( α , β , γ ) i = 1 ∏ k δ J i J i ′ { j i J i − 1 J i } . (638)
where J i J_i J i (i = 1 , 2 , … , k i=1,2,\ldots,k i = 1 , 2 , … , k ) is any angular momentum consistent with the vector addition rule,
J i = j 1 + j 2 + ⋯ + j i , M i = m 1 + m 2 + ⋯ + m i , N i = n 1 + n 2 + ⋯ + n i , J_i=j_1+j_2+\cdots+j_i,\qquad M_i=m_1+m_2+\cdots+m_i,\qquad N_i=n_1+n_2+\cdots+n_i, J i = j 1 + j 2 + ⋯ + j i , M i = m 1 + m 2 + ⋯ + m i , N i = n 1 + n 2 + ⋯ + n i ,
It is assumed in Eq. (638) that J 0 = J 0 ′ = M 0 = N 0 = 0 J_0=J'_0=M_0=N_0=0 J 0 = J 0 ′ = M 0 = N 0 = 0 .
In particular, for j 1 = j 2 = … = j k = 1 2 j_1=j_2=\ldots=j_k=\frac{1}{2} j 1 = j 2 = … = j k = 2 1 and J i + 1 = J i + 1 2 J_{i+1}=J_i+\frac{1}{2} J i + 1 = J i + 2 1 one has
∑ m 1 + ⋯ + m k = M n 1 + ⋯ + n k = N D m 1 n 1 1 2 ( α , β , γ ) D m 2 n 2 1 2 ( α , β , γ ) ⋯ D m k n k 1 2 ( α , β , γ ) = ( 2 J ) ! ( J + M ) ! ( J − M ) ! ( J + N ) ! ( J − N ) ! D M N J ( α , β , γ ) . (639) \sum_{\substack{m_1+\cdots+m_k=M\\n_1+\cdots+n_k=N}}D^{\frac{1}{2}}_{m_1n_1}(\alpha,\beta,\gamma)D^{\frac{1}{2}}_{m_2n_2}(\alpha,\beta,\gamma)\cdots D^{\frac{1}{2}}_{m_kn_k}(\alpha,\beta,\gamma)
=\frac{(2J)!}{\sqrt{(J+M)!(J-M)!(J+N)!(J-N)!}}D^J_{MN}(\alpha,\beta,\gamma). \tag*{(639)} m 1 + ⋯ + m k = M n 1 + ⋯ + n k = N ∑ D m 1 n 1 2 1 ( α , β , γ ) D m 2 n 2 2 1 ( α , β , γ ) ⋯ D m k n k 2 1 ( α , β , γ ) = ( J + M )! ( J − M )! ( J + N )! ( J − N )! ( 2 J )! D M N J ( α , β , γ ) . (639)
where J ≡ J k = k / 2 J\equiv J_k=k/2 J ≡ J k = k /2 is either integer or half-integer. Similarly, for j 1 = j 2 = … = j k = 1 j_1=j_2=\ldots=j_k=1 j 1 = j 2 = … = j k = 1 and J i + 1 = J i + 1 J_{i+1}=J_i+1 J i + 1 = J i + 1 ,
∑ m 1 + ⋯ + m k = M n 1 + ⋯ + n k = N ∏ i , j = 1 k ( 1 + δ m i 0 ) ( 1 + δ n i 0 ) D m 1 n 1 1 ( α , β , γ ) D m 2 n 2 1 ( α , β , γ ) ⋯ D m k n k 1 ( α , β , γ ) = ( 2 J ) ! ( J + M ) ! ( J − M ) ! ( J + N ) ! ( J − N ) ! D M N J ( α , β , γ ) . (640) \sum_{\substack{m_1+\cdots+m_k=M\\n_1+\cdots+n_k=N}}\sqrt{\prod_{i,j=1}^{k}(1+\delta_{m_i0})(1+\delta_{n_i0})}D^1_{m_1n_1}(\alpha,\beta,\gamma)D^1_{m_2n_2}(\alpha,\beta,\gamma)\cdots D^1_{m_kn_k}(\alpha,\beta,\gamma)
=\frac{(2J)!}{\sqrt{(J+M)!(J-M)!(J+N)!(J-N)!}}D^J_{MN}(\alpha,\beta,\gamma). \tag*{(640)} m 1 + ⋯ + m k = M n 1 + ⋯ + n k = N ∑ i , j = 1 ∏ k ( 1 + δ m i 0 ) ( 1 + δ n i 0 ) D m 1 n 1 1 ( α , β , γ ) D m 2 n 2 1 ( α , β , γ ) ⋯ D m k n k 1 ( α , β , γ ) = ( J + M )! ( J − M )! ( J + N )! ( J − N )! ( 2 J )! D M N J ( α , β , γ ) . (640)
where J ≡ J k = k J\equiv J_k=k J ≡ J k = k is integer.
Equations (4.6.10) and (4.6.11) are useful for evaluating the Wigner D D D -functions. For example, Eq. (639) gives an explicit form of the D D D -functions in terms of the Cayley-Klein parameters (§ 9.9.3 .) defined by
D 1 2 1 2 1 2 ( α , β , γ ) ≡ a , D 1 2 − 1 2 1 2 ( α , β , γ ) ≡ − b ∗ , D − 1 2 1 2 1 2 ( α , β , γ ) ≡ b , D − 1 2 − 1 2 1 2 ( α , β , γ ) ≡ a ∗ (641) \begin{aligned}
D^{\frac{1}{2}}_{\frac{1}{2}\frac{1}{2}}(\alpha,\beta,\gamma)&\equiv a, & D^{\frac{1}{2}}_{\frac{1}{2}-\frac{1}{2}}(\alpha,\beta,\gamma)&\equiv-b^*,\\
D^{\frac{1}{2}}_{-\frac{1}{2}\frac{1}{2}}(\alpha,\beta,\gamma)&\equiv b, & D^{\frac{1}{2}}_{-\frac{1}{2}-\frac{1}{2}}(\alpha,\beta,\gamma)&\equiv a^* \tag*{(641)}
\end{aligned} D 2 1 2 1 2 1 ( α , β , γ ) D − 2 1 2 1 2 1 ( α , β , γ ) ≡ a , ≡ b , D 2 1 − 2 1 2 1 ( α , β , γ ) D − 2 1 − 2 1 2 1 ( α , β , γ ) ≡ − b ∗ , ≡ a ∗ (641)
The appropriate expression for the D D D -functions has the form
D M N J ( α , β , γ ) = ( J + M ) ! ( J − M ) ! ( J + N ) ! ( J − N ) ! ∑ p , q , r , s ( a ) p ( b ) q ( a ∗ ) r ( − b ∗ ) s p ! q ! r ! s ! (642) D^J_{MN}(\alpha,\beta,\gamma)=\sqrt{(J+M)!(J-M)!(J+N)!(J-N)!}\sum_{p,q,r,s}\frac{(a)^p(b)^q(a^*)^r(-b^*)^s}{p!q!r!s!} \tag*{(642)} D M N J ( α , β , γ ) = ( J + M )! ( J − M )! ( J + N )! ( J − N )! p , q , r , s ∑ p ! q ! r ! s ! ( a ) p ( b ) q ( a ∗ ) r ( − b ∗ ) s (642)
Here the summation indices p , q , r p,q,r p , q , r and s s s run over all integer values consistent with the condition
p + q + r + s = 2 J , p − q − r + s = 2 M , p + q − r − s = 2 N . (643) \begin{aligned}
p+q+r+s&=2J,\\
p-q-r+s&=2M,\\
p+q-r-s&=2N. \tag*{(643)}
\end{aligned} p + q + r + s p − q − r + s p + q − r − s = 2 J , = 2 M , = 2 N . (643)
According to Eqs. (643) , only one parameter from p , q , r p,q,r p , q , r and s s s is independent, i.e., in fact, Eq. (642) represents a single sum. The independent summation index may be chosen in different ways. This yields different explicit forms for the D D D -functions, Eqs. (568) –(4.3.5). For example, if r r r is taken to be independent, Eq. (642) reduces to Eq. (568) .
Determinant of Matrix D M M ′ J D^J_{MM'} D M M ′ J
The determinant ∥ D M M ′ J ∥ \lVert D^J_{MM'}\rVert ∥ D M M ′ J ∥ of the rotation matrix is an invariant sum of products of 2 J + 1 2J+1 2 J + 1 D D D -functions. According to Eq. (543) , this matrix is unimodular, i.e.,
∑ P ( − 1 ) P D − J M 1 J ( α , β , γ ) D − J + 1 M 2 J ( α , β , γ ) ⋯ D J M 2 J + 1 J ( α , β , γ ) = 1 , (644) \sum_P(-1)^P D^J_{-J M_1}(\alpha,\beta,\gamma)D^J_{-J+1 M_2}(\alpha,\beta,\gamma)\cdots D^J_{J M_{2J+1}}(\alpha,\beta,\gamma)=1, \tag*{(644)} P ∑ ( − 1 ) P D − J M 1 J ( α , β , γ ) D − J + 1 M 2 J ( α , β , γ ) ⋯ D J M 2 J + 1 J ( α , β , γ ) = 1 , (644)
where M 1 , M 2 , … , M 2 J + 1 M_1,M_2,\ldots,M_{2J+1} M 1 , M 2 , … , M 2 J + 1 represent all possible permutations P P P of J , J − 1 , J − 2 , … , − J J,J-1,J-2,\ldots,-J J , J − 1 , J − 2 , … , − J . The phase factor ( − 1 ) P (-1)^P ( − 1 ) P equals + 1 +1 + 1 for even permutations, and − 1 -1 − 1 for odd ones.
Addition of rotations
The Addition Theorem for D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
Let two successive rotations of the coordinate system, S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\}\to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } and S ′ { x ′ , y ′ , z ′ } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S'\{x',y',z'\}\to S''\{x'',y'',z''\} S ′ { x ′ , y ′ , z ′ } → S ′′ { x ′′ , y ′′ , z ′′ } , be described by the Euler angles α 1 , β 1 , γ 1 \alpha_1,\beta_1,\gamma_1 α 1 , β 1 , γ 1 and α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 , respectively, and the resultant rotation S { x , y , z } → S ′ ′ { x ′ ′ , y ′ ′ , z ′ ′ } S\{x,y,z\}\to S''\{x'',y'',z''\} S { x , y , z } → S ′′ { x ′′ , y ′′ , z ′′ } be described by the angles α , β , γ \alpha,\beta,\gamma α , β , γ . In accordance with Sec. a, there are two alternative forms of the rotation addition.
(a) The operator of the resultant rotation, D ^ ( α , β , γ ) \hat{D}(\alpha,\beta,\gamma) D ^ ( α , β , γ ) , is given by Eq. (193) , if all rotations are performed according to scheme B (§ 9.9.1 ) and the Euler angles α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 ; α 1 , β 1 , γ 1 \alpha_1,\beta_1,\gamma_1 α 1 , β 1 , γ 1 and α , β , γ \alpha,\beta,\gamma α , β , γ are defined with respect to the initial system S { x , y , z } S\{x,y,z\} S { x , y , z } . Then the addition theorem reads
∑ M ′ ′ = − J J D M M ′ ′ J ( α 2 , β 2 , γ 2 ) D M ′ ′ M ′ J ( α 1 , β 1 , γ 1 ) = D M M ′ J ( α , β , γ ) , (645) \sum_{M''=-J}^{J}D^J_{M M''}(\alpha_2,\beta_2,\gamma_2)D^J_{M'' M'}(\alpha_1,\beta_1,\gamma_1)=D^J_{M M'}(\alpha,\beta,\gamma), \tag*{(645)} M ′′ = − J ∑ J D M M ′′ J ( α 2 , β 2 , γ 2 ) D M ′′ M ′ J ( α 1 , β 1 , γ 1 ) = D M M ′ J ( α , β , γ ) , (645)
where α , β , γ \alpha,\beta,\gamma α , β , γ are related to α 1 , β 1 , γ 1 \alpha_1,\beta_1,\gamma_1 α 1 , β 1 , γ 1 and α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 by Eqs. (195) –(1.4.70).
(b) The operator of the resultant rotation D ^ ( α , β , γ ) \hat{D}(\alpha,\beta,\gamma) D ^ ( α , β , γ ) is given by Eq. (202) , if α 1 , β 1 , γ 1 \alpha_1,\beta_1,\gamma_1 α 1 , β 1 , γ 1 and α , β , γ \alpha,\beta,\gamma α , β , γ are defined with respect to the initial system S { x , y , z } S\{x,y,z\} S { x , y , z } , but α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 are defined with respect to the intermediate system S ′ { x ′ , y ′ , z ′ } S'\{x',y',z'\} S ′ { x ′ , y ′ , z ′ } (scheme B), or if successive rotations are performed according to scheme A. In these cases the addition theorem reads
∑ M ′ ′ = − J J D M M ′ ′ J ( α 1 , β 1 , γ 1 ) D M ′ ′ M ′ J ( α 2 , β 2 , γ 2 ) = D M M ′ J ( α , β , γ ) , (646) \sum_{M''=-J}^{J}D^J_{M M''}(\alpha_1,\beta_1,\gamma_1)D^J_{M'' M'}(\alpha_2,\beta_2,\gamma_2)=D^J_{M M'}(\alpha,\beta,\gamma), \tag*{(646)} M ′′ = − J ∑ J D M M ′′ J ( α 1 , β 1 , γ 1 ) D M ′′ M ′ J ( α 2 , β 2 , γ 2 ) = D M M ′ J ( α , β , γ ) , (646)
and α , β , γ \alpha,\beta,\gamma α , β , γ are related to α 1 , β 1 , γ 1 \alpha_1,\beta_1,\gamma_1 α 1 , β 1 , γ 1 and α 2 , β 2 , γ 2 \alpha_2,\beta_2,\gamma_2 α 2 , β 2 , γ 2 by the equations which may be obtained from Eqs. (195) –(1.4.70) by replacing ( α 1 , β 1 , γ 1 ) ⇌ ( α 2 , β 2 , γ 2 ) (\alpha_1,\beta_1,\gamma_1)\rightleftharpoons(\alpha_2,\beta_2,\gamma_2) ( α 1 , β 1 , γ 1 ) ⇌ ( α 2 , β 2 , γ 2 ) .
In particular,
∑ M ′ ′ = − J J D M M ′ ′ J ( α , β 1 , φ ) D M ′ ′ M ′ J ( − φ , β 2 , γ ) = D M M ′ J ( α , β 1 + β 2 , γ ) , (647) \sum_{M''=-J}^{J}D^J_{M M''}(\alpha,\beta_1,\varphi)D^J_{M'' M'}(-\varphi,\beta_2,\gamma)=D^J_{M M'}(\alpha,\beta_1+\beta_2,\gamma), \tag*{(647)} M ′′ = − J ∑ J D M M ′′ J ( α , β 1 , φ ) D M ′′ M ′ J ( − φ , β 2 , γ ) = D M M ′ J ( α , β 1 + β 2 , γ ) , (647)
where φ \varphi φ is arbitrary, and
∑ M ′ ′ = − J J D M M ′ ′ J ( α , β , γ ) D M ′ M ′ ′ J ∗ ( α , β , γ ) = ∑ M ′ ′ = − J J D M M ′ ′ J ( α , β , γ ) D M ′ ′ M ′ J ( − γ , − β , − α ) = D M M ′ J ( 0 , 0 , 0 ) = δ M M ′ . (648) \begin{aligned}
\sum_{M''=-J}^{J}D^J_{M M''}(\alpha,\beta,\gamma)D^{J*}_{M' M''}(\alpha,\beta,\gamma)=\sum_{M''=-J}^{J}D^J_{M M''}(\alpha,\beta,\gamma)D^J_{M'' M'}(-\gamma,-\beta,-\alpha) \\
&=D^J_{M M'}(0,0,0)=\delta_{M M'}. \tag*{(648)}
\end{aligned} M ′′ = − J ∑ J D M M ′′ J ( α , β , γ ) D M ′ M ′′ J ∗ ( α , β , γ ) = M ′′ = − J ∑ J D M M ′′ J ( α , β , γ ) D M ′′ M ′ J ( − γ , − β , − α ) = D M M ′ J ( 0 , 0 , 0 ) = δ M M ′ . (648)
The Addition Theorem for d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β )
Equation (4.7.2) may be rewritten as
∑ M ′ ′ = − J J d M M ′ ′ J ( β 1 ) d M ′ ′ M ′ J ( β 2 ) e − i M ′ ′ ϕ = e − i M α d M M ′ J ( β ) e − i M ′ γ . (649) \sum_{M''=-J}^{J} d^{J}_{MM''}(\beta_{1})d^{J}_{M''M'}(\beta_{2})e^{-iM''\phi}=e^{-iM\alpha}d^{J}_{MM'}(\beta)e^{-iM'\gamma}. \tag*{(649)} M ′′ = − J ∑ J d M M ′′ J ( β 1 ) d M ′′ M ′ J ( β 2 ) e − i M ′′ ϕ = e − i M α d M M ′ J ( β ) e − i M ′ γ . (649)
Here
cot α = cos β 1 cot ϕ + cot β 2 sin β 1 sin ϕ , \cot\alpha=\cos\beta_{1}\cot\phi+\cot\beta_{2}\frac{\sin\beta_{1}}{\sin\phi}, cot α = cos β 1 cot ϕ + cot β 2 sin ϕ sin β 1 ,
cos β = cos β 1 cos β 2 − sin β 1 sin β 2 cos ϕ , (650) \cos\beta=\cos\beta_{1}\cos\beta_{2}-\sin\beta_{1}\sin\beta_{2}\cos\phi, \tag*{(650)} cos β = cos β 1 cos β 2 − sin β 1 sin β 2 cos ϕ , (650)
cot γ = cos β 2 cot ϕ + cot β 1 sin β 2 sin ϕ . \cot\gamma=\cos\beta_{2}\cot\phi+\cot\beta_{1}\frac{\sin\beta_{2}}{\sin\phi}. cot γ = cos β 2 cot ϕ + cot β 1 sin ϕ sin β 2 .
Equation (4.7.5) is simplified in the following particular cases. If ϕ = 0 \phi=0 ϕ = 0 and β 1 + β 2 ≤ π \beta_{1}+\beta_{2}\leq\pi β 1 + β 2 ≤ π , then α = 0 \alpha=0 α = 0 , β = β 1 + β 2 \beta=\beta_{1}+\beta_{2} β = β 1 + β 2 and γ = 0 \gamma=0 γ = 0 ,
∑ M ′ ′ = − J J d M M ′ ′ J ( β 1 ) d M ′ ′ M ′ J ( β 2 ) = d M M ′ J ( β 1 + β 2 ) . (651) \sum_{M''=-J}^{J}d^{J}_{MM''}(\beta_{1})d^{J}_{M''M'}(\beta_{2})=d^{J}_{MM'}(\beta_{1}+\beta_{2}). \tag*{(651)} M ′′ = − J ∑ J d M M ′′ J ( β 1 ) d M ′′ M ′ J ( β 2 ) = d M M ′ J ( β 1 + β 2 ) . (651)
If ϕ = 0 \phi=0 ϕ = 0 and β 1 + β 2 > π \beta_{1}+\beta_{2}>\pi β 1 + β 2 > π , then α = π \alpha=\pi α = π , β = 2 π − β 1 − β 2 \beta=2\pi-\beta_{1}-\beta_{2} β = 2 π − β 1 − β 2 and γ = π \gamma=\pi γ = π ,
∑ M ′ ′ = − J J d M M ′ ′ J ( β 1 ) d M ′ ′ M ′ J ( β 2 ) = ( − 1 ) M + M ′ d M M ′ J ( 2 π − β 1 − β 2 ) . (652) \sum_{M''=-J}^{J}d^{J}_{MM''}(\beta_{1})d^{J}_{M''M'}(\beta_{2})=(-1)^{M+M'}d^{J}_{MM'}(2\pi-\beta_{1}-\beta_{2}). \tag*{(652)} M ′′ = − J ∑ J d M M ′′ J ( β 1 ) d M ′′ M ′ J ( β 2 ) = ( − 1 ) M + M ′ d M M ′ J ( 2 π − β 1 − β 2 ) . (652)
If ϕ = π \phi=\pi ϕ = π and β 1 ≥ β 2 \beta_{1}\geq\beta_{2} β 1 ≥ β 2 , then α = 0 \alpha=0 α = 0 , β = β 1 − β 2 \beta=\beta_{1}-\beta_{2} β = β 1 − β 2 and γ = π \gamma=\pi γ = π ,
∑ M ′ ′ = − J J ( − 1 ) M ′ ′ − M ′ d M M ′ ′ J ( β 1 ) d M ′ ′ M ′ J ( β 2 ) = d M M ′ J ( β 1 − β 2 ) (653) \sum_{M''=-J}^{J}(-1)^{M''-M'}d^{J}_{MM''}(\beta_{1})d^{J}_{M''M'}(\beta_{2})=d^{J}_{MM'}(\beta_{1}-\beta_{2}) \tag*{(653)} M ′′ = − J ∑ J ( − 1 ) M ′′ − M ′ d M M ′′ J ( β 1 ) d M ′′ M ′ J ( β 2 ) = d M M ′ J ( β 1 − β 2 ) (653)
In particular, for β 1 = β 2 \beta_{1}=\beta_{2} β 1 = β 2 ,
∑ M ′ ′ = − J J ( − 1 ) M ′ ′ − M ′ d M M ′ ′ J ( β ) d M ′ ′ M ′ J ( β ) = δ M M ′ . (654) \sum_{M''=-J}^{J}(-1)^{M''-M'}d^{J}_{MM''}(\beta)d^{J}_{M''M'}(\beta)=\delta_{MM'}. \tag*{(654)} M ′′ = − J ∑ J ( − 1 ) M ′′ − M ′ d M M ′′ J ( β ) d M ′′ M ′ J ( β ) = δ M M ′ . (654)
If ϕ = π / 2 \phi=\pi/2 ϕ = π /2 , then
∑ M ′ ′ = − J J ( − i ) M ′ ′ d M M ′ ′ J ( β 1 ) d M ′ ′ M ′ J ( β 2 ) = e − i M α d M M ′ J ( β ) e − i M ′ γ . (655) \sum_{M''=-J}^{J}(-i)^{M''}d^{J}_{MM''}(\beta_{1})d^{J}_{M''M'}(\beta_{2})=e^{-iM\alpha}d^{J}_{MM'}(\beta)e^{-iM'\gamma}. \tag*{(655)} M ′′ = − J ∑ J ( − i ) M ′′ d M M ′′ J ( β 1 ) d M ′′ M ′ J ( β 2 ) = e − i M α d M M ′ J ( β ) e − i M ′ γ . (655)
where
cot α = cot β 2 sin β 1 , \cot\alpha=\cot\beta_{2}\sin\beta_{1}, cot α = cot β 2 sin β 1 ,
cos β = cos β 1 cos β 2 , (656) \cos\beta=\cos\beta_{1}\cos\beta_{2}, \tag*{(656)} cos β = cos β 1 cos β 2 , (656)
cot γ = cot β 1 sin β 2 . \cot\gamma=\cot\beta_{1}\sin\beta_{2}. cot γ = cot β 1 sin β 2 .
Addition of Two Identical Rotations
When α 1 = α 2 \alpha_{1}=\alpha_{2} α 1 = α 2 , β 1 = β 2 \beta_{1}=\beta_{2} β 1 = β 2 and γ 1 = γ 2 \gamma_{1}=\gamma_{2} γ 1 = γ 2 the addition theorem, Eq. (645) , together with the Clebsch-Gordan expansion, Eq. (630) , yield
∑ J = 0 , 1 , … 2 j ∑ m ′ ′ C j m j m ′ ′ J m + m ′ ′ D m + m ′ ′ m ′ ′ + m ′ J ( α , β , γ ) C j m ′ ′ j m ′ J m ′ ′ + m ′ = D m m ′ j ( α ˉ , β ˉ , γ ˉ ) , (657) \sum_{J=0,1,\ldots}^{2j}\sum_{m''}C^{J\,m+m''}_{j\,m\,j\,m''}D^{J}_{m+m''\,m''+m'}(\alpha,\beta,\gamma)C^{J\,m''+m'}_{j\,m''\,j\,m'}=D^{j}_{mm'}(\bar{\alpha},\bar{\beta},\bar{\gamma}), \tag*{(657)} J = 0 , 1 , … ∑ 2 j m ′′ ∑ C j m j m ′′ J m + m ′′ D m + m ′′ m ′′ + m ′ J ( α , β , γ ) C j m ′′ j m ′ J m ′′ + m ′ = D m m ′ j ( α ˉ , β ˉ , γ ˉ ) , (657)
where
cos β ˉ = cos 2 β − sin 2 β cos ( α + γ ) , \cos\bar{\beta}=\cos^{2}\beta-\sin^{2}\beta\cos(\alpha+\gamma), cos β ˉ = cos 2 β − sin 2 β cos ( α + γ ) ,
tan ( α ˉ − α ) = tan ( γ ˉ − γ ) = tan α + γ 2 cos β ˉ . (658) \tan(\bar{\alpha}-\alpha)=\tan(\bar{\gamma}-\gamma)=\frac{\tan\frac{\alpha+\gamma}{2}}{\cos\bar{\beta}}. \tag*{(658)} tan ( α ˉ − α ) = tan ( γ ˉ − γ ) = cos β ˉ tan 2 α + γ . (658)
In particular, if α = γ = 0 \alpha=\gamma=0 α = γ = 0 ,
∑ J = 0 , 1 , … 2 j ∑ m ′ ′ C j m j m ′ ′ J m + m ′ ′ d m + m ′ ′ m ′ ′ + m ′ J ( β ) C j m ′ ′ j m ′ J m ′ ′ + m ′ = d m m ′ j ( 2 β ) . (659) \sum_{J=0,1,\ldots}^{2j}\sum_{m''}C^{J\,m+m''}_{j\,m\,j\,m''}d^{J}_{m+m''\,m''+m'}(\beta)C^{J\,m''+m'}_{j\,m''\,j\,m'}=d^{j}_{mm'}(2\beta). \tag*{(659)} J = 0 , 1 , … ∑ 2 j m ′′ ∑ C j m j m ′′ J m + m ′′ d m + m ′′ m ′′ + m ′ J ( β ) C j m ′′ j m ′ J m ′′ + m ′ = d m m ′ j ( 2 β ) . (659)
The Multiplication Theorem for d M M ′ J ( β ) d^J_{MM'}(\beta) d M M ′ J ( β )
Equation (4.7.5) leads to a representation of products of two d M M ′ J ( β ) d^J_{MM'}(\beta) d M M ′ J ( β ) functions
d M M ′ ′ J ( β 1 ) d M ′ ′ M ′ J ( β 2 ) = 1 4 π ∫ − 2 π 2 π e i ( M ′ ′ φ − M α − M ′ γ ) d M M ′ J ( β ) d φ . (660) d^J_{MM''}(\beta_1)d^J_{M''M'}(\beta_2)=\frac{1}{4\pi}\int_{-2\pi}^{2\pi}e^{i(M''\varphi-M\alpha-M'\gamma)}d^J_{MM'}(\beta)\,d\varphi. \tag*{(660)} d M M ′′ J ( β 1 ) d M ′′ M ′ J ( β 2 ) = 4 π 1 ∫ − 2 π 2 π e i ( M ′′ φ − M α − M ′ γ ) d M M ′ J ( β ) d φ . (660)
Here α \alpha α , β \beta β , γ \gamma γ are related to β 1 \beta_1 β 1 , β 2 \beta_2 β 2 and φ \varphi φ by means of Eq. (4.7.6).
Sums Involving the D-Functions of Different Arguments
In addition to Eqs. (4.7.1) or (4.7.2) one can derive the following invariant sums which are equivalent to the characters of irreducible representations of rotation group for two rotations (Sec. 4.14):
(a)
∑ M M ′ D M M ′ J ( α 1 , β 1 , γ 1 ) D M ′ M J ( α 2 , β 2 , γ 2 ) ≡ χ J ( R 1 R 2 ) = χ J ( R 2 R 1 ) = sin [ ( 2 J + 1 ) ω 2 ] sin ω 2 , (661) \sum_{MM'}D^J_{MM'}(\alpha_1,\beta_1,\gamma_1)D^J_{M'M}(\alpha_2,\beta_2,\gamma_2)\equiv\chi^J(R_1R_2)=\chi^J(R_2R_1)=\frac{\sin\left[(2J+1)\frac{\omega}{2}\right]}{\sin\frac{\omega}{2}}, \tag*{(661)} M M ′ ∑ D M M ′ J ( α 1 , β 1 , γ 1 ) D M ′ M J ( α 2 , β 2 , γ 2 ) ≡ χ J ( R 1 R 2 ) = χ J ( R 2 R 1 ) = sin 2 ω sin [ ( 2 J + 1 ) 2 ω ] , (661)
where
cos ω 2 = cos β 1 2 cos β 2 2 cos α 1 + γ 1 + α 2 + γ 2 2 − sin β 1 2 sin β 2 2 cos α 1 − γ 1 − α 2 + γ 2 2 = cos β 1 + β 2 2 cos α 1 + γ 2 2 cos γ 1 + α 2 2 − cos β 1 − β 2 2 sin α 1 + γ 2 2 sin γ 1 + α 2 2 . (662) \begin{aligned}
\cos\frac{\omega}{2}=\cos\frac{\beta_1}{2}\cos\frac{\beta_2}{2}\cos\frac{\alpha_1+\gamma_1+\alpha_2+\gamma_2}{2}-\sin\frac{\beta_1}{2}\sin\frac{\beta_2}{2}\cos\frac{\alpha_1-\gamma_1-\alpha_2+\gamma_2}{2} \\
&=\cos\frac{\beta_1+\beta_2}{2}\cos\frac{\alpha_1+\gamma_2}{2}\cos\frac{\gamma_1+\alpha_2}{2}-\cos\frac{\beta_1-\beta_2}{2}\sin\frac{\alpha_1+\gamma_2}{2}\sin\frac{\gamma_1+\alpha_2}{2}. \tag*{(662)}
\end{aligned} cos 2 ω = cos 2 β 1 cos 2 β 2 cos 2 α 1 + γ 1 + α 2 + γ 2 − sin 2 β 1 sin 2 β 2 cos 2 α 1 − γ 1 − α 2 + γ 2 = cos 2 β 1 + β 2 cos 2 α 1 + γ 2 cos 2 γ 1 + α 2 − cos 2 β 1 − β 2 sin 2 α 1 + γ 2 sin 2 γ 1 + α 2 . (662)
(b)
∑ M M ′ D M M ′ J ( α 1 , β 1 , γ 1 ) D M M ′ J ∗ ( α 2 , β 2 , γ 2 ) ≡ χ J ( R 1 R 2 − 1 ) = χ J ( R 1 − 1 R 2 ) = sin [ ( 2 J + 1 ) ω ′ 2 ] sin ω ′ 2 (663) \sum_{MM'}D^J_{MM'}(\alpha_1,\beta_1,\gamma_1)D^{J*}_{MM'}(\alpha_2,\beta_2,\gamma_2)\equiv\chi^J(R_1R_2^{-1})=\chi^J(R_1^{-1}R_2)=\frac{\sin\left[(2J+1)\frac{\omega'}{2}\right]}{\sin\frac{\omega'}{2}} \tag*{(663)} M M ′ ∑ D M M ′ J ( α 1 , β 1 , γ 1 ) D M M ′ J ∗ ( α 2 , β 2 , γ 2 ) ≡ χ J ( R 1 R 2 − 1 ) = χ J ( R 1 − 1 R 2 ) = sin 2 ω ′ sin [ ( 2 J + 1 ) 2 ω ′ ] (663)
where
cos ω ′ 2 = cos β 1 2 cos β 2 2 cos α 1 + γ 1 − α 2 − γ 2 2 + sin β 1 2 sin β 2 2 cos α 1 − γ 1 − α 2 + γ 2 2 = cos β 1 − β 2 2 cos α 1 − α 2 2 cos γ 1 − γ 2 2 − cos β 1 + β 2 2 sin α 1 − α 2 2 sin γ 1 − γ 2 2 . (664) \begin{aligned}
\cos\frac{\omega'}{2}=\cos\frac{\beta_1}{2}\cos\frac{\beta_2}{2}\cos\frac{\alpha_1+\gamma_1-\alpha_2-\gamma_2}{2}+\sin\frac{\beta_1}{2}\sin\frac{\beta_2}{2}\cos\frac{\alpha_1-\gamma_1-\alpha_2+\gamma_2}{2} \\
&=\cos\frac{\beta_1-\beta_2}{2}\cos\frac{\alpha_1-\alpha_2}{2}\cos\frac{\gamma_1-\gamma_2}{2}-\cos\frac{\beta_1+\beta_2}{2}\sin\frac{\alpha_1-\alpha_2}{2}\sin\frac{\gamma_1-\gamma_2}{2}. \tag*{(664)}
\end{aligned} cos 2 ω ′ = cos 2 β 1 cos 2 β 2 cos 2 α 1 + γ 1 − α 2 − γ 2 + sin 2 β 1 sin 2 β 2 cos 2 α 1 − γ 1 − α 2 + γ 2 = cos 2 β 1 − β 2 cos 2 α 1 − α 2 cos 2 γ 1 − γ 2 − cos 2 β 1 + β 2 sin 2 α 1 − α 2 sin 2 γ 1 − γ 2 . (664)
Equation (4.7.20) may be obtained from Eq. (4.7.18) by replacing ( α 2 , β 2 , γ 2 ) → ( − γ 2 , − β 2 , − α 2 ) (\alpha_2,\beta_2,\gamma_2)\to(-\gamma_2,-\beta_2,-\alpha_2) ( α 2 , β 2 , γ 2 ) → ( − γ 2 , − β 2 , − α 2 ) .
Similarly, one can arrive at invariant sums which may be reduced to the characters involving three and more rotations, i.e., χ J ( R 1 R 2 R 3 ) \chi^J(R_1R_2R_3) χ J ( R 1 R 2 R 3 ) etc.
The Ponzano-Regge sum
Note the following sum involving products of three d M M ′ J ( β ) d^J_{MM'}(\beta) d M M ′ J ( β ) functions with the same J J J but different arguments (Ref. [46])
∑ J = J min ∞ ( 2 J + 1 ) d M 1 M 2 J ( β 3 ) d M 2 M 3 J ( β 1 ) d M 3 M 1 J ( β 2 ) = 2 Θ ( B ) π B cos ( ∑ i = 1 3 M i δ i ) . (665) \sum_{J=J_{\min}}^{\infty}(2J+1)d^J_{M_1M_2}(\beta_3)d^J_{M_2M_3}(\beta_1)d^J_{M_3M_1}(\beta_2)=\frac{2\Theta(B)}{\pi\sqrt{B}}\cos\left(\sum_{i=1}^{3}M_i\delta_i\right). \tag*{(665)} J = J m i n ∑ ∞ ( 2 J + 1 ) d M 1 M 2 J ( β 3 ) d M 2 M 3 J ( β 1 ) d M 3 M 1 J ( β 2 ) = π B 2Θ ( B ) cos ( i = 1 ∑ 3 M i δ i ) . (665)
Here the summation index J J J runs from J min = max { ∣ M 1 ∣ , ∣ M 2 ∣ , ∣ M 3 ∣ } J_{\min}=\max\{|M_1|,|M_2|,|M_3|\} J m i n = max { ∣ M 1 ∣ , ∣ M 2 ∣ , ∣ M 3 ∣ } to infinity, M 1 M_1 M 1 , M 2 M_2 M 2 , M 3 M_3 M 3 being fixed, and
B = ∣ 1 cos β 3 cos β 2 cos β 3 1 cos β 1 cos β 2 cos β 1 1 ∣ , (666) B=\left|\begin{array}{ccc}
1 & \cos\beta_3 & \cos\beta_2\\
\cos\beta_3 & 1 & \cos\beta_1\\
\cos\beta_2 & \cos\beta_1 & 1
\end{array}\right|, \tag*{(666)} B = 1 cos β 3 cos β 2 cos β 3 1 cos β 1 cos β 2 cos β 1 1 , (666)
Θ ( x ) = { 1 x > 0 0 x < 0 (667) \Theta(x)=\begin{cases}
1 & x>0\\
0 & x<0
\end{cases} \tag*{(667)} Θ ( x ) = { 1 0 x > 0 x < 0 (667)
The relations between the angles δ 1 \delta_1 δ 1 , δ 2 \delta_2 δ 2 , δ 3 \delta_3 δ 3 and β 1 \beta_1 β 1 , β 2 \beta_2 β 2 , β 3 \beta_3 β 3 are
cos δ i = cos β j cos β k − cos β i sin β j sin β k , ( i ≠ j ≠ k ) , (668) \cos\delta_i=\frac{\cos\beta_j\cos\beta_k-\cos\beta_i}{\sin\beta_j\sin\beta_k},\qquad(i\ne j\ne k), \tag*{(668)} cos δ i = sin β j sin β k cos β j cos β k − cos β i , ( i = j = k ) , (668)
sin β i sin δ i = sin β j sin δ j , ( i , j = 1 , 2 , 3 ) . (669) \frac{\sin\beta_i}{\sin\delta_i}=\frac{\sin\beta_j}{\sin\delta_j},\qquad(i,j=1,2,3). \tag*{(669)} sin δ i sin β i = sin δ j sin β j , ( i , j = 1 , 2 , 3 ) . (669)
These relations may be easily obtained by using spherical trigonometry (see Figure 4.1 ).
Figure 4.1. Geometrical interpretation of the angles in Eq. (665) . The arguments β 1 \beta_1 β 1 , β 2 \beta_2 β 2 , β 3 \beta_3 β 3 are the sides of a spherical triangle, i.e. the angles they subtend at the centre of the sphere, whose three rays are shown emanating from the centre. The interior angle of the triangle at the vertex opposite the side β i \beta_i β i equals π − δ i \pi-\delta_i π − δ i . The δ i \delta_i δ i are thereby determined by β 1 , β 2 , β 3 \beta_1,\beta_2,\beta_3 β 1 , β 2 , β 3 through the spherical cosine and sine rules {(4.7.24)}–{(4.7.25)}.
Recursion relations for D M M ′ J D^J_{MM'} D M M ′ J
Relations between D J D^J D J and D J ± 1 D^{J\pm1} D J ± 1
cos β D M M ′ J ( α , β , γ ) = ( J 2 − M 2 ) ( J 2 − M ′ 2 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) + M M ′ J ( J + 1 ) D M M ′ J ( α , β , γ ) + ( ( J + 1 ) 2 − M 2 ) [ ( J + 1 ) 2 − M ′ 2 ] ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (670) \cos\beta D^J_{MM'}(\alpha,\beta,\gamma)=\frac{\sqrt{(J^2-M^2)(J^2-M'^2)}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)+\frac{MM'}{J(J+1)}D^J_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{((J+1)^2-M^2)[(J+1)^2-M'^2]}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(670)} cos β D M M ′ J ( α , β , γ ) = J ( 2 J + 1 ) ( J 2 − M 2 ) ( J 2 − M ′2 ) D M M ′ J − 1 ( α , β , γ ) + J ( J + 1 ) M M ′ D M M ′ J ( α , β , γ ) + ( J + 1 ) ( 2 J + 1 ) (( J + 1 ) 2 − M 2 ) [( J + 1 ) 2 − M ′2 ] D M M ′ J + 1 ( α , β , γ ) (670)
sin β e i α D M + 1 M ′ J ( α , β , γ ) = − ( J + M ) ( J + M + 1 ) ( J 2 − M ′ 2 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) + M ′ ( J − M ) ( J + M + 1 ) J ( J + 1 ) D M M ′ J ( α , β , γ ) + ( J − M ) ( J − M + 1 ) [ ( J + 1 ) 2 − M ′ 2 ] ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (671) \sin\beta e^{i\alpha}D^J_{M+1M'}(\alpha,\beta,\gamma)=-\frac{\sqrt{(J+M)(J+M+1)(J^2-M'^2)}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)+\frac{M'\sqrt{(J-M)(J+M+1)}}{J(J+1)}D^J_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{(J-M)(J-M+1)[(J+1)^2-M'^2]}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(671)} sin β e i α D M + 1 M ′ J ( α , β , γ ) = − J ( 2 J + 1 ) ( J + M ) ( J + M + 1 ) ( J 2 − M ′2 ) D M M ′ J − 1 ( α , β , γ ) + J ( J + 1 ) M ′ ( J − M ) ( J + M + 1 ) D M M ′ J ( α , β , γ ) + ( J + 1 ) ( 2 J + 1 ) ( J − M ) ( J − M + 1 ) [( J + 1 ) 2 − M ′2 ] D M M ′ J + 1 ( α , β , γ ) (671)
sin β e − i α D M − 1 M ′ J ( α , β , γ ) = ( J − M ) ( J − M + 1 ) ( J 2 − M ′ 2 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) + M ′ ( J + M ) ( J − M + 1 ) J ( J + 1 ) D M M ′ J ( α , β , γ ) − ( J + M ) ( J + M + 1 ) [ ( J + 1 ) 2 − M ′ 2 ] ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (672) \sin\beta e^{-i\alpha}D^J_{M-1M'}(\alpha,\beta,\gamma)=\frac{\sqrt{(J-M)(J-M+1)(J^2-M'^2)}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)+\frac{M'\sqrt{(J+M)(J-M+1)}}{J(J+1)}D^J_{MM'}(\alpha,\beta,\gamma)-\frac{\sqrt{(J+M)(J+M+1)[(J+1)^2-M'^2]}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(672)} sin β e − i α D M − 1 M ′ J ( α , β , γ ) = J ( 2 J + 1 ) ( J − M ) ( J − M + 1 ) ( J 2 − M ′2 ) D M M ′ J − 1 ( α , β , γ ) + J ( J + 1 ) M ′ ( J + M ) ( J − M + 1 ) D M M ′ J ( α , β , γ ) − ( J + 1 ) ( 2 J + 1 ) ( J + M ) ( J + M + 1 ) [( J + 1 ) 2 − M ′2 ] D M M ′ J + 1 ( α , β , γ ) (672)
sin β e i γ D M M ′ + 1 J ( α , β , γ ) = ( J 2 − M 2 ) ( J + M ′ ) ( J + M ′ + 1 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) − M ( J − M ′ ) ( J + M ′ + 1 ) J ( J + 1 ) D M M ′ J ( α , β , γ ) − ( ( J + 1 ) 2 − M 2 ) ( J − M ′ ) ( J − M ′ + 1 ) ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (673) \sin\beta e^{i\gamma}D^J_{MM'+1}(\alpha,\beta,\gamma)=\frac{\sqrt{(J^2-M^2)(J+M')(J+M'+1)}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)-\frac{M\sqrt{(J-M')(J+M'+1)}}{J(J+1)}D^J_{MM'}(\alpha,\beta,\gamma)-\frac{\sqrt{((J+1)^2-M^2)(J-M')(J-M'+1)}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(673)} sin β e iγ D M M ′ + 1 J ( α , β , γ ) = J ( 2 J + 1 ) ( J 2 − M 2 ) ( J + M ′ ) ( J + M ′ + 1 ) D M M ′ J − 1 ( α , β , γ ) − J ( J + 1 ) M ( J − M ′ ) ( J + M ′ + 1 ) D M M ′ J ( α , β , γ ) − ( J + 1 ) ( 2 J + 1 ) (( J + 1 ) 2 − M 2 ) ( J − M ′ ) ( J − M ′ + 1 ) D M M ′ J + 1 ( α , β , γ ) (673)
sin β e − i γ D M M ′ − 1 J ( α , β , γ ) = − ( J 2 − M 2 ) ( J − M ′ ) ( J − M ′ + 1 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) − M ( J + M ′ ) ( J − M ′ + 1 ) J ( J + 1 ) D M M ′ J ( α , β , γ ) + [ ( J + 1 ) 2 − M 2 ] ( J + M ′ ) ( J + M ′ + 1 ) ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (674) \sin\beta e^{-i\gamma}D^{J}_{M\,M'-1}(\alpha,\beta,\gamma)=-\frac{\sqrt{(J^{2}-M^{2})(J-M')(J-M'+1)}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)-\frac{M\sqrt{(J+M')(J-M'+1)}}{J(J+1)}D^{J}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{[(J+1)^{2}-M^{2}](J+M')(J+M'+1)}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(674)} sin β e − iγ D M M ′ − 1 J ( α , β , γ ) = − J ( 2 J + 1 ) ( J 2 − M 2 ) ( J − M ′ ) ( J − M ′ + 1 ) D M M ′ J − 1 ( α , β , γ ) − J ( J + 1 ) M ( J + M ′ ) ( J − M ′ + 1 ) D M M ′ J ( α , β , γ ) + ( J + 1 ) ( 2 J + 1 ) [( J + 1 ) 2 − M 2 ] ( J + M ′ ) ( J + M ′ + 1 ) D M M ′ J + 1 ( α , β , γ ) (674)
( 1 + cos β ) e i ( α + γ ) D M + 1 M ′ + 1 J ( α , β , γ ) = ( J + M + 1 ) ( J + M ) ( J + M ′ + 1 ) ( J + M ′ ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) + ( J − M ) ( J + M + 1 ) ( J − M ′ ) ( J + M ′ + 1 ) J ( J + 1 ) D M M ′ J ( α , β , γ ) + ( J − M ) ( J − M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (675) (1+\cos\beta)e^{i(\alpha+\gamma)}D^{J}_{M+1\,M'+1}(\alpha,\beta,\gamma)=\frac{\sqrt{(J+M+1)(J+M)(J+M'+1)(J+M')}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{(J-M)(J+M+1)(J-M')(J+M'+1)}}{J(J+1)}D^{J}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{(J-M)(J-M+1)(J-M')(J-M'+1)}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(675)} ( 1 + cos β ) e i ( α + γ ) D M + 1 M ′ + 1 J ( α , β , γ ) = J ( 2 J + 1 ) ( J + M + 1 ) ( J + M ) ( J + M ′ + 1 ) ( J + M ′ ) D M M ′ J − 1 ( α , β , γ ) + J ( J + 1 ) ( J − M ) ( J + M + 1 ) ( J − M ′ ) ( J + M ′ + 1 ) D M M ′ J ( α , β , γ ) + ( J + 1 ) ( 2 J + 1 ) ( J − M ) ( J − M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) D M M ′ J + 1 ( α , β , γ ) (675)
( 1 + cos β ) e − i ( α + γ ) D M − 1 M ′ − 1 J ( α , β , γ ) = ( J − M ) ( J − M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) + ( J + M ) ( J − M + 1 ) ( J + M ′ ) ( J − M ′ + 1 ) J ( J + 1 ) D M M ′ J ( α , β , γ ) + ( J + M ) ( J + M + 1 ) ( J + M ′ ) ( J + M ′ + 1 ) ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (676) (1+\cos\beta)e^{-i(\alpha+\gamma)}D^{J}_{M-1\,M'-1}(\alpha,\beta,\gamma)=\frac{\sqrt{(J-M)(J-M+1)(J-M')(J-M'+1)}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{(J+M)(J-M+1)(J+M')(J-M'+1)}}{J(J+1)}D^{J}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{(J+M)(J+M+1)(J+M')(J+M'+1)}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(676)} ( 1 + cos β ) e − i ( α + γ ) D M − 1 M ′ − 1 J ( α , β , γ ) = J ( 2 J + 1 ) ( J − M ) ( J − M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) D M M ′ J − 1 ( α , β , γ ) + J ( J + 1 ) ( J + M ) ( J − M + 1 ) ( J + M ′ ) ( J − M ′ + 1 ) D M M ′ J ( α , β , γ ) + ( J + 1 ) ( 2 J + 1 ) ( J + M ) ( J + M + 1 ) ( J + M ′ ) ( J + M ′ + 1 ) D M M ′ J + 1 ( α , β , γ ) (676)
( 1 − cos β ) e i ( α − γ ) D M + 1 M ′ − 1 J ( α , β , γ ) = ( J + M ) ( J + M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) − ( J − M ) ( J + M + 1 ) ( J + M ′ ) ( J − M ′ + 1 ) J ( J + 1 ) D M M ′ J ( α , β , γ ) + ( J − M ) ( J − M + 1 ) ( J + M ′ ) ( J + M ′ + 1 ) ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (677) (1-\cos\beta)e^{i(\alpha-\gamma)}D^{J}_{M+1\,M'-1}(\alpha,\beta,\gamma)=\frac{\sqrt{(J+M)(J+M+1)(J-M')(J-M'+1)}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)-\frac{\sqrt{(J-M)(J+M+1)(J+M')(J-M'+1)}}{J(J+1)}D^{J}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{(J-M)(J-M+1)(J+M')(J+M'+1)}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(677)} ( 1 − cos β ) e i ( α − γ ) D M + 1 M ′ − 1 J ( α , β , γ ) = J ( 2 J + 1 ) ( J + M ) ( J + M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) D M M ′ J − 1 ( α , β , γ ) − J ( J + 1 ) ( J − M ) ( J + M + 1 ) ( J + M ′ ) ( J − M ′ + 1 ) D M M ′ J ( α , β , γ ) + ( J + 1 ) ( 2 J + 1 ) ( J − M ) ( J − M + 1 ) ( J + M ′ ) ( J + M ′ + 1 ) D M M ′ J + 1 ( α , β , γ ) (677)
( 1 − cos β ) e − i ( α − γ ) D M − 1 M ′ + 1 J ( α , β , γ ) = ( J − M ) ( J − M + 1 ) ( J + M ′ ) ( J + M ′ + 1 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) − ( J + M ) ( J − M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) J ( J + 1 ) D M M ′ J ( α , β , γ ) + ( J + M ) ( J + M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) (678) (1-\cos\beta)e^{-i(\alpha-\gamma)}D^{J}_{M-1\,M'+1}(\alpha,\beta,\gamma)=\frac{\sqrt{(J-M)(J-M+1)(J+M')(J+M'+1)}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)-\frac{\sqrt{(J+M)(J-M+1)(J-M')(J-M'+1)}}{J(J+1)}D^{J}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{(J+M)(J+M+1)(J-M')(J-M'+1)}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma) \tag*{(678)} ( 1 − cos β ) e − i ( α − γ ) D M − 1 M ′ + 1 J ( α , β , γ ) = J ( 2 J + 1 ) ( J − M ) ( J − M + 1 ) ( J + M ′ ) ( J + M ′ + 1 ) D M M ′ J − 1 ( α , β , γ ) − J ( J + 1 ) ( J + M ) ( J − M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) D M M ′ J ( α , β , γ ) + ( J + 1 ) ( 2 J + 1 ) ( J + M ) ( J + M + 1 ) ( J − M ′ ) ( J − M ′ + 1 ) D M M ′ J + 1 ( α , β , γ ) (678)
Equations (4.8.1)–(4.8.9) may be obtained from the Clebsch-Gordan series, Eq. (630) with J 1 = 1 J_{1}=1 J 1 = 1 .
Relations Between D J D^{J} D J and D J ± 1 / 2 D^{J\pm1/2} D J ± 1/2
cos β 2 e i ( α + γ ) / 2 D M + 1 2 M ′ + 1 2 J ( α , β , γ ) = ( J + M + 1 2 ) ( J + M ′ + 1 2 ) 2 J + 1 D M M ′ J − 1 2 ( α , β , γ ) + ( J − M + 1 2 ) ( J − M ′ + 1 2 ) 2 J + 1 D M M ′ J + 1 2 ( α , β , γ ) (679) \cos\frac{\beta}{2}e^{i(\alpha+\gamma)/2}D^{J}_{M+\frac{1}{2}\,M'+\frac{1}{2}}(\alpha,\beta,\gamma)=\frac{\sqrt{\left(J+M+\frac{1}{2}\right)\left(J+M'+\frac{1}{2}\right)}}{2J+1}D^{J-\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{\left(J-M+\frac{1}{2}\right)\left(J-M'+\frac{1}{2}\right)}}{2J+1}D^{J+\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma) \tag*{(679)} cos 2 β e i ( α + γ ) /2 D M + 2 1 M ′ + 2 1 J ( α , β , γ ) = 2 J + 1 ( J + M + 2 1 ) ( J + M ′ + 2 1 ) D M M ′ J − 2 1 ( α , β , γ ) + 2 J + 1 ( J − M + 2 1 ) ( J − M ′ + 2 1 ) D M M ′ J + 2 1 ( α , β , γ ) (679)
sin β 2 e i ( α − γ ) / 2 D M + 1 2 M ′ − 1 2 J ( α , β , γ ) = − ( J + M + 1 2 ) ( J − M ′ + 1 2 ) 2 J + 1 D M M ′ J − 1 2 ( α , β , γ ) + ( J − M + 1 2 ) ( J + M ′ + 1 2 ) 2 J + 1 D M M ′ J + 1 2 ( α , β , γ ) (680) \sin\frac{\beta}{2}e^{i(\alpha-\gamma)/2}D^{J}_{M+\frac{1}{2}\,M'-\frac{1}{2}}(\alpha,\beta,\gamma)=-\frac{\sqrt{\left(J+M+\frac{1}{2}\right)\left(J-M'+\frac{1}{2}\right)}}{2J+1}D^{J-\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma)+\frac{\sqrt{\left(J-M+\frac{1}{2}\right)\left(J+M'+\frac{1}{2}\right)}}{2J+1}D^{J+\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma) \tag*{(680)} sin 2 β e i ( α − γ ) /2 D M + 2 1 M ′ − 2 1 J ( α , β , γ ) = − 2 J + 1 ( J + M + 2 1 ) ( J − M ′ + 2 1 ) D M M ′ J − 2 1 ( α , β , γ ) + 2 J + 1 ( J − M + 2 1 ) ( J + M ′ + 2 1 ) D M M ′ J + 2 1 ( α , β , γ ) (680)
cos β 2 e − i α + γ 2 D M − 1 2 , M ′ − 1 2 J ( α , β , γ ) = ( J − M + 1 2 ) ( J − M ′ + 1 2 ) 2 J + 1 D M M ′ J − 1 2 ( α , β , γ ) + ( J + M + 1 2 ) ( J + M ′ + 1 2 ) 2 J + 1 D M M ′ J + 1 2 ( α , β , γ ) , (681) \cos\frac{\beta}{2}e^{-i\frac{\alpha+\gamma}{2}}D^{J}_{M-\frac{1}{2},M'-\frac{1}{2}}(\alpha,\beta,\gamma)=\frac{\sqrt{\left(J-M+\frac{1}{2}\right)\left(J-M'+\frac{1}{2}\right)}}{2J+1}D^{J-\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma)
+\frac{\sqrt{\left(J+M+\frac{1}{2}\right)\left(J+M'+\frac{1}{2}\right)}}{2J+1}D^{J+\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma), \tag*{(681)} cos 2 β e − i 2 α + γ D M − 2 1 , M ′ − 2 1 J ( α , β , γ ) = 2 J + 1 ( J − M + 2 1 ) ( J − M ′ + 2 1 ) D M M ′ J − 2 1 ( α , β , γ ) + 2 J + 1 ( J + M + 2 1 ) ( J + M ′ + 2 1 ) D M M ′ J + 2 1 ( α , β , γ ) , (681)
sin β 2 e − i α − γ 2 D M − 1 2 , M ′ + 1 2 J ( α , β , γ ) = ( J − M + 1 2 ) ( J + M ′ + 1 2 ) 2 J + 1 D M M ′ J − 1 2 ( α , β , γ ) − ( J + M + 1 2 ) ( J − M ′ + 1 2 ) 2 J + 1 D M M ′ J + 1 2 ( α , β , γ ) . (682) \sin\frac{\beta}{2}e^{-i\frac{\alpha-\gamma}{2}}D^{J}_{M-\frac{1}{2},M'+\frac{1}{2}}(\alpha,\beta,\gamma)=\frac{\sqrt{\left(J-M+\frac{1}{2}\right)\left(J+M'+\frac{1}{2}\right)}}{2J+1}D^{J-\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma)
-\frac{\sqrt{\left(J+M+\frac{1}{2}\right)\left(J-M'+\frac{1}{2}\right)}}{2J+1}D^{J+\frac{1}{2}}_{MM'}(\alpha,\beta,\gamma). \tag*{(682)} sin 2 β e − i 2 α − γ D M − 2 1 , M ′ + 2 1 J ( α , β , γ ) = 2 J + 1 ( J − M + 2 1 ) ( J + M ′ + 2 1 ) D M M ′ J − 2 1 ( α , β , γ ) − 2 J + 1 ( J + M + 2 1 ) ( J − M ′ + 2 1 ) D M M ′ J + 2 1 ( α , β , γ ) . (682)
Equations (4.8.10)–(4.8.13) may be obtained from Eq. (630) with J 1 = 1 2 J_1=\frac{1}{2} J 1 = 2 1 . They yield
D M M ′ J ( α , β , γ ) = J − M J − M ′ cos β 2 e i α + γ 2 D M + 1 2 , M ′ + 1 2 J − 1 2 ( α , β , γ ) − J + M J − M ′ sin β 2 e − i α − γ 2 D M − 1 2 , M ′ + 1 2 J − 1 2 ( α , β , γ ) , ( M ′ ≠ J ) . (683) D^{J}_{MM'}(\alpha,\beta,\gamma)=\sqrt{\frac{J-M}{J-M'}}\cos\frac{\beta}{2}e^{i\frac{\alpha+\gamma}{2}}D^{J-\frac{1}{2}}_{M+\frac{1}{2},M'+\frac{1}{2}}(\alpha,\beta,\gamma)
-\sqrt{\frac{J+M}{J-M'}}\sin\frac{\beta}{2}e^{-i\frac{\alpha-\gamma}{2}}D^{J-\frac{1}{2}}_{M-\frac{1}{2},M'+\frac{1}{2}}(\alpha,\beta,\gamma),\quad(M'\ne J). \tag*{(683)} D M M ′ J ( α , β , γ ) = J − M ′ J − M cos 2 β e i 2 α + γ D M + 2 1 , M ′ + 2 1 J − 2 1 ( α , β , γ ) − J − M ′ J + M sin 2 β e − i 2 α − γ D M − 2 1 , M ′ + 2 1 J − 2 1 ( α , β , γ ) , ( M ′ = J ) . (683)
D M M ′ J ( α , β , γ ) = J − M J + M ′ sin β 2 e i α − γ 2 D M + 1 2 , M ′ − 1 2 J − 1 2 ( α , β , γ ) + J + M J + M ′ cos β 2 e − i α + γ 2 D M − 1 2 , M ′ − 1 2 J − 1 2 ( α , β , γ ) , ( M ′ ≠ − J ) . (684) D^{J}_{MM'}(\alpha,\beta,\gamma)=\sqrt{\frac{J-M}{J+M'}}\sin\frac{\beta}{2}e^{i\frac{\alpha-\gamma}{2}}D^{J-\frac{1}{2}}_{M+\frac{1}{2},M'-\frac{1}{2}}(\alpha,\beta,\gamma)
+\sqrt{\frac{J+M}{J+M'}}\cos\frac{\beta}{2}e^{-i\frac{\alpha+\gamma}{2}}D^{J-\frac{1}{2}}_{M-\frac{1}{2},M'-\frac{1}{2}}(\alpha,\beta,\gamma),\quad(M'\ne-J). \tag*{(684)} D M M ′ J ( α , β , γ ) = J + M ′ J − M sin 2 β e i 2 α − γ D M + 2 1 , M ′ − 2 1 J − 2 1 ( α , β , γ ) + J + M ′ J + M cos 2 β e − i 2 α + γ D M − 2 1 , M ′ − 2 1 J − 2 1 ( α , β , γ ) , ( M ′ = − J ) . (684)
Relations Between DM MJ and DM 1 M 1J
− M + M ′ cos β sin β D M M ′ J ( α , β , γ ) = 1 2 ( J + M ′ ) ( J − M ′ + 1 ) D M M ′ − 1 J ( α , β , γ ) e − i γ + 1 2 ( J − M ′ ) ( J + M ′ + 1 ) D M M ′ + 1 J ( α , β , γ ) e i γ , (685) \frac{-M+M'\cos\beta}{\sin\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=\frac{1}{2}\sqrt{(J+M')(J-M'+1)}D^{J}_{MM'-1}(\alpha,\beta,\gamma)e^{-i\gamma}
+\frac{1}{2}\sqrt{(J-M')(J+M'+1)}D^{J}_{MM'+1}(\alpha,\beta,\gamma)e^{i\gamma}, \tag*{(685)} sin β − M + M ′ cos β D M M ′ J ( α , β , γ ) = 2 1 ( J + M ′ ) ( J − M ′ + 1 ) D M M ′ − 1 J ( α , β , γ ) e − iγ + 2 1 ( J − M ′ ) ( J + M ′ + 1 ) D M M ′ + 1 J ( α , β , γ ) e iγ , (685)
M ′ − M cos β sin β D M M ′ J ( α , β , γ ) = 1 2 ( J + M ) ( J − M + 1 ) D M − 1 , M ′ J ( α , β , γ ) e − i α + 1 2 ( J − M ) ( J + M + 1 ) D M + 1 , M ′ J ( α , β , γ ) e i α , (686) \frac{M'-M\cos\beta}{\sin\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=\frac{1}{2}\sqrt{(J+M)(J-M+1)}D^{J}_{M-1,M'}(\alpha,\beta,\gamma)e^{-i\alpha}
+\frac{1}{2}\sqrt{(J-M)(J+M+1)}D^{J}_{M+1,M'}(\alpha,\beta,\gamma)e^{i\alpha}, \tag*{(686)} sin β M ′ − M cos β D M M ′ J ( α , β , γ ) = 2 1 ( J + M ) ( J − M + 1 ) D M − 1 , M ′ J ( α , β , γ ) e − i α + 2 1 ( J − M ) ( J + M + 1 ) D M + 1 , M ′ J ( α , β , γ ) e i α , (686)
D M + 1 , M ′ J ( α , β , γ ) e i α = ( J + M ′ ) ( J − M ′ + 1 ) ( J − M ) ( J + M + 1 ) 1 + cos β 2 e − i γ D M M ′ − 1 J ( α , β , γ ) + M ′ sin β ( J − M ) ( J + M + 1 ) D M M ′ J ( α , β , γ ) − ( J − M ′ ) ( J + M ′ + 1 ) ( J − M ) ( J + M + 1 ) 1 − cos β 2 e i γ D M M ′ + 1 J ( α , β , γ ) , (687) D^{J}_{M+1,M'}(\alpha,\beta,\gamma)e^{i\alpha}
=\sqrt{\frac{(J+M')(J-M'+1)}{(J-M)(J+M+1)}}\frac{1+\cos\beta}{2}e^{-i\gamma}D^{J}_{MM'-1}(\alpha,\beta,\gamma)
+\frac{M'\sin\beta}{\sqrt{(J-M)(J+M+1)}}D^{J}_{MM'}(\alpha,\beta,\gamma)
-\sqrt{\frac{(J-M')(J+M'+1)}{(J-M)(J+M+1)}}\frac{1-\cos\beta}{2}e^{i\gamma}D^{J}_{MM'+1}(\alpha,\beta,\gamma), \tag*{(687)} D M + 1 , M ′ J ( α , β , γ ) e i α = ( J − M ) ( J + M + 1 ) ( J + M ′ ) ( J − M ′ + 1 ) 2 1 + cos β e − iγ D M M ′ − 1 J ( α , β , γ ) + ( J − M ) ( J + M + 1 ) M ′ sin β D M M ′ J ( α , β , γ ) − ( J − M ) ( J + M + 1 ) ( J − M ′ ) ( J + M ′ + 1 ) 2 1 − cos β e iγ D M M ′ + 1 J ( α , β , γ ) , (687)
D M − 1 , M ′ J ( α , β , γ ) e − i α = − ( J + M ′ ) ( J − M ′ + 1 ) ( J + M ) ( J − M + 1 ) 1 − cos β 2 e − i γ D M M ′ − 1 J ( α , β , γ ) + M ′ sin β ( J + M ) ( J − M + 1 ) D M M ′ J ( α , β , γ ) + ( J − M ′ ) ( J + M ′ + 1 ) ( J + M ) ( J − M + 1 ) 1 + cos β 2 e i γ D M M ′ + 1 J ( α , β , γ ) , (688) D^{J}_{M-1,M'}(\alpha,\beta,\gamma)e^{-i\alpha}
=-\sqrt{\frac{(J+M')(J-M'+1)}{(J+M)(J-M+1)}}\frac{1-\cos\beta}{2}e^{-i\gamma}D^{J}_{MM'-1}(\alpha,\beta,\gamma)
+\frac{M'\sin\beta}{\sqrt{(J+M)(J-M+1)}}D^{J}_{MM'}(\alpha,\beta,\gamma)
+\sqrt{\frac{(J-M')(J+M'+1)}{(J+M)(J-M+1)}}\frac{1+\cos\beta}{2}e^{i\gamma}D^{J}_{MM'+1}(\alpha,\beta,\gamma), \tag*{(688)} D M − 1 , M ′ J ( α , β , γ ) e − i α = − ( J + M ) ( J − M + 1 ) ( J + M ′ ) ( J − M ′ + 1 ) 2 1 − cos β e − iγ D M M ′ − 1 J ( α , β , γ ) + ( J + M ) ( J − M + 1 ) M ′ sin β D M M ′ J ( α , β , γ ) + ( J + M ) ( J − M + 1 ) ( J − M ′ ) ( J + M ′ + 1 ) 2 1 + cos β e iγ D M M ′ + 1 J ( α , β , γ ) , (688)
D M M ′ + 1 J ( α , β , γ ) e i γ = ( J + M ) ( J − M + 1 ) ( J − M ′ ) ( J + M ′ + 1 ) 1 + cos β 2 e − i α D M − 1 M ′ J ( α , β , γ ) − M sin β ( J − M ′ ) ( J + M ′ + 1 ) D M M ′ J ( α , β , γ ) − ( J − M ) ( J + M + 1 ) ( J − M ′ ) ( J + M ′ + 1 ) 1 − cos β 2 e i α D M + 1 M ′ J ( α , β , γ ) , (689) D^{J}_{MM'+1}(\alpha,\beta,\gamma)e^{i\gamma}=\sqrt{\frac{(J+M)(J-M+1)}{(J-M')(J+M'+1)}}\frac{1+\cos\beta}{2}e^{-i\alpha}D^{J}_{M-1M'}(\alpha,\beta,\gamma)-\frac{M\sin\beta}{\sqrt{(J-M')(J+M'+1)}}D^{J}_{MM'}(\alpha,\beta,\gamma)-\sqrt{\frac{(J-M)(J+M+1)}{(J-M')(J+M'+1)}}\frac{1-\cos\beta}{2}e^{i\alpha}D^{J}_{M+1M'}(\alpha,\beta,\gamma), \tag*{(689)} D M M ′ + 1 J ( α , β , γ ) e iγ = ( J − M ′ ) ( J + M ′ + 1 ) ( J + M ) ( J − M + 1 ) 2 1 + cos β e − i α D M − 1 M ′ J ( α , β , γ ) − ( J − M ′ ) ( J + M ′ + 1 ) M sin β D M M ′ J ( α , β , γ ) − ( J − M ′ ) ( J + M ′ + 1 ) ( J − M ) ( J + M + 1 ) 2 1 − cos β e i α D M + 1 M ′ J ( α , β , γ ) , (689)
D M M ′ − 1 J ( α , β , γ ) e − i γ = − ( J + M ) ( J − M + 1 ) ( J + M ′ ) ( J − M ′ + 1 ) 1 − cos β 2 e − i α D M − 1 M ′ J ( α , β , γ ) − M sin β ( J + M ′ ) ( J − M ′ + 1 ) D M M ′ J ( α , β , γ ) + ( J − M ) ( J + M + 1 ) ( J + M ′ ) ( J − M ′ + 1 ) 1 + cos β 2 e i α D M + 1 M ′ J ( α , β , γ ) . (690) D^{J}_{MM'-1}(\alpha,\beta,\gamma)e^{-i\gamma}=-\sqrt{\frac{(J+M)(J-M+1)}{(J+M')(J-M'+1)}}\frac{1-\cos\beta}{2}e^{-i\alpha}D^{J}_{M-1M'}(\alpha,\beta,\gamma)-\frac{M\sin\beta}{\sqrt{(J+M')(J-M'+1)}}D^{J}_{MM'}(\alpha,\beta,\gamma)+\sqrt{\frac{(J-M)(J+M+1)}{(J+M')(J-M'+1)}}\frac{1+\cos\beta}{2}e^{i\alpha}D^{J}_{M+1M'}(\alpha,\beta,\gamma). \tag*{(690)} D M M ′ − 1 J ( α , β , γ ) e − iγ = − ( J + M ′ ) ( J − M ′ + 1 ) ( J + M ) ( J − M + 1 ) 2 1 − cos β e − i α D M − 1 M ′ J ( α , β , γ ) − ( J + M ′ ) ( J − M ′ + 1 ) M sin β D M M ′ J ( α , β , γ ) + ( J + M ′ ) ( J − M ′ + 1 ) ( J − M ) ( J + M + 1 ) 2 1 + cos β e i α D M + 1 M ′ J ( α , β , γ ) . (690)
Differential relations for D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
Derivatives of D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) may be expressed in terms of the D D D -functions with different M M M , M ′ M' M ′ and J J J
sin β ∂ ∂ β D M M ′ J ( α , β , γ ) = − ( J + 1 ) ( J 2 − M 2 ) ( J 2 − M ′ 2 ) J ( 2 J + 1 ) D M M ′ J − 1 ( α , β , γ ) − M M ′ J ( J + 1 ) D M M ′ J ( α , β , γ ) + J [ ( J + 1 ) 2 − M 2 ] [ ( J + 1 ) 2 − M ′ 2 ] ( J + 1 ) ( 2 J + 1 ) D M M ′ J + 1 ( α , β , γ ) , (691) \sin\beta\frac{\partial}{\partial\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=-\frac{(J+1)\sqrt{(J^{2}-M^{2})(J^{2}-M'^{2})}}{J(2J+1)}D^{J-1}_{MM'}(\alpha,\beta,\gamma)-\frac{MM'}{J(J+1)}D^{J}_{MM'}(\alpha,\beta,\gamma)+\frac{J\sqrt{[(J+1)^{2}-M^{2}][(J+1)^{2}-M'^{2}]}}{(J+1)(2J+1)}D^{J+1}_{MM'}(\alpha,\beta,\gamma), \tag*{(691)} sin β ∂ β ∂ D M M ′ J ( α , β , γ ) = − J ( 2 J + 1 ) ( J + 1 ) ( J 2 − M 2 ) ( J 2 − M ′ 2 ) D M M ′ J − 1 ( α , β , γ ) − J ( J + 1 ) M M ′ D M M ′ J ( α , β , γ ) + ( J + 1 ) ( 2 J + 1 ) J [( J + 1 ) 2 − M 2 ] [( J + 1 ) 2 − M ′ 2 ] D M M ′ J + 1 ( α , β , γ ) , (691)
∂ ∂ β D M M ′ J ( α , β , γ ) = − 1 2 ( J + M ) ( J − M + 1 ) e − i α D M − 1 M ′ J ( α , β , γ ) + 1 2 ( J − M ) ( J + M + 1 ) e i α D M + 1 M ′ J ( α , β , γ ) , (692) \frac{\partial}{\partial\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=-\frac{1}{2}\sqrt{(J+M)(J-M+1)}e^{-i\alpha}D^{J}_{M-1M'}(\alpha,\beta,\gamma)+\frac{1}{2}\sqrt{(J-M)(J+M+1)}e^{i\alpha}D^{J}_{M+1M'}(\alpha,\beta,\gamma), \tag*{(692)} ∂ β ∂ D M M ′ J ( α , β , γ ) = − 2 1 ( J + M ) ( J − M + 1 ) e − i α D M − 1 M ′ J ( α , β , γ ) + 2 1 ( J − M ) ( J + M + 1 ) e i α D M + 1 M ′ J ( α , β , γ ) , (692)
∂ ∂ β D M M ′ J ( α , β , γ ) = 1 2 ( J + M ′ ) ( J − M ′ + 1 ) e − i γ D M M ′ − 1 J ( α , β , γ ) − 1 2 ( J − M ′ ) ( J + M ′ + 1 ) e − i γ D M M ′ + 1 J ( α , β , γ ) , (693) \frac{\partial}{\partial\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=\frac{1}{2}\sqrt{(J+M')(J-M'+1)}e^{-i\gamma}D^{J}_{MM'-1}(\alpha,\beta,\gamma)-\frac{1}{2}\sqrt{(J-M')(J+M'+1)}e^{-i\gamma}D^{J}_{MM'+1}(\alpha,\beta,\gamma), \tag*{(693)} ∂ β ∂ D M M ′ J ( α , β , γ ) = 2 1 ( J + M ′ ) ( J − M ′ + 1 ) e − iγ D M M ′ − 1 J ( α , β , γ ) − 2 1 ( J − M ′ ) ( J + M ′ + 1 ) e − iγ D M M ′ + 1 J ( α , β , γ ) , (693)
∂ ∂ β D M M ′ J ( α , β , γ ) = M ′ − M cos β sin β D M M ′ J ( α , β , γ ) − ( J + M ) ( J − M + 1 ) e − i α D M − 1 M ′ J ( α , β , γ ) , (694) \frac{\partial}{\partial\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=\frac{M'-M\cos\beta}{\sin\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)-\sqrt{(J+M)(J-M+1)}e^{-i\alpha}D^{J}_{M-1M'}(\alpha,\beta,\gamma), \tag*{(694)} ∂ β ∂ D M M ′ J ( α , β , γ ) = sin β M ′ − M cos β D M M ′ J ( α , β , γ ) − ( J + M ) ( J − M + 1 ) e − i α D M − 1 M ′ J ( α , β , γ ) , (694)
∂ ∂ β D M M ′ J ( α , β , γ ) = − M ′ − M cos β sin β D M M ′ J ( α , β , γ ) + ( J − M ) ( J + M + 1 ) e i α D M + 1 M ′ J ( α , β , γ ) , (695) \frac{\partial}{\partial\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=-\frac{M'-M\cos\beta}{\sin\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)+\sqrt{(J-M)(J+M+1)}e^{i\alpha}D^{J}_{M+1M'}(\alpha,\beta,\gamma), \tag*{(695)} ∂ β ∂ D M M ′ J ( α , β , γ ) = − sin β M ′ − M cos β D M M ′ J ( α , β , γ ) + ( J − M ) ( J + M + 1 ) e i α D M + 1 M ′ J ( α , β , γ ) , (695)
∂ ∂ β D M M ′ J ( α , β , γ ) = M − M ′ cos β sin β D M M ′ J ( α , β , γ ) + ( J + M ′ ) ( J − M ′ + 1 ) e − i γ D M M ′ − 1 J ( α , β , γ ) , (696) \frac{\partial}{\partial\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=\frac{M-M'\cos\beta}{\sin\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)+\sqrt{(J+M')(J-M'+1)}e^{-i\gamma}D^{J}_{MM'-1}(\alpha,\beta,\gamma), \tag*{(696)} ∂ β ∂ D M M ′ J ( α , β , γ ) = sin β M − M ′ cos β D M M ′ J ( α , β , γ ) + ( J + M ′ ) ( J − M ′ + 1 ) e − iγ D M M ′ − 1 J ( α , β , γ ) , (696)
∂ ∂ β D M M ′ J ( α , β , γ ) = − M − M ′ cos β sin β D M M ′ J ( α , β , γ ) − ( J − M ′ ) ( J + M ′ + 1 ) e i γ D M M ′ + 1 J ( α , β , γ ) , (697) \frac{\partial}{\partial\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)=-\frac{M-M'\cos\beta}{\sin\beta}D^{J}_{MM'}(\alpha,\beta,\gamma)-\sqrt{(J-M')(J+M'+1)}e^{i\gamma}D^{J}_{MM'+1}(\alpha,\beta,\gamma), \tag*{(697)} ∂ β ∂ D M M ′ J ( α , β , γ ) = − sin β M − M ′ cos β D M M ′ J ( α , β , γ ) − ( J − M ′ ) ( J + M ′ + 1 ) e iγ D M M ′ + 1 J ( α , β , γ ) , (697)
∂ ∂ α D M M ′ J ( α , β , γ ) = − i M D M M ′ J ( α , β , γ ) , (698) \frac{\partial}{\partial\alpha}D^{J}_{MM'}(\alpha,\beta,\gamma)=-iMD^{J}_{MM'}(\alpha,\beta,\gamma), \tag*{(698)} ∂ α ∂ D M M ′ J ( α , β , γ ) = − i M D M M ′ J ( α , β , γ ) , (698)
∂ ∂ γ D M M ′ J ( α , β , γ ) = − i M ′ D M M ′ J ( α , β , γ ) , (699) \frac{\partial}{\partial\gamma}D^{J}_{MM'}(\alpha,\beta,\gamma)=-iM'D^{J}_{MM'}(\alpha,\beta,\gamma), \tag*{(699)} ∂ γ ∂ D M M ′ J ( α , β , γ ) = − i M ′ D M M ′ J ( α , β , γ ) , (699)
See also Eqs. (551) and (552) .
Orthogonality and completeness of the D D D -functions
The functions D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) with different J J J (integer and half-integer) are mutually orthogonal with respect to integration over a double volume of the 3-dimensional rotation group (i.e., over the volume of SU2 _2 2 group) because the period of D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) with half-integer J J J is 4 π 4\pi 4 π rather than 2 π 2\pi 2 π . The double domain may be chosen in two ways:
V 1 : 0 ≤ α < 4 π , 0 ≤ β ≤ π , 0 ≤ γ < 2 π , (700) V_1:\quad0\leq\alpha<4\pi,\quad0\leq\beta\leq\pi,\quad0\leq\gamma<2\pi, \tag*{(700)} V 1 : 0 ≤ α < 4 π , 0 ≤ β ≤ π , 0 ≤ γ < 2 π , (700)
V 2 : 0 ≤ α < 2 π , 0 ≤ β ≤ π , 0 ≤ γ < 4 π . (701) V_2:\quad0\leq\alpha<2\pi,\quad0\leq\beta\leq\pi,\quad0\leq\gamma<4\pi. \tag*{(701)} V 2 : 0 ≤ α < 2 π , 0 ≤ β ≤ π , 0 ≤ γ < 4 π . (701)
The total volume of the double domain V 1 V_1 V 1 or V 2 V_2 V 2 is equal to 16 π 2 16\pi^2 16 π 2 . Orthogonality and normalization condition is
∫ 0 4 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 2 M 2 ′ J 2 ∗ ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 4 π d γ D M 2 M 2 ′ J 2 ∗ ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = 16 π 2 2 J 1 + 1 δ J 1 J 2 δ M 1 M 2 δ M 1 ′ M 2 ′ . (702) \begin{aligned}
&\int_{0}^{4\pi}d\alpha\int_{0}^{\pi}d\beta\sin\beta\int_{0}^{2\pi}d\gamma\,D^{J_2*}_{M_2M'_2}(\alpha,\beta,\gamma)D^{J_1}_{M_1M'_1}(\alpha,\beta,\gamma)\\
&=\int_{0}^{2\pi}d\alpha\int_{0}^{\pi}d\beta\sin\beta\int_{0}^{4\pi}d\gamma\,D^{J_2*}_{M_2M'_2}(\alpha,\beta,\gamma)D^{J_1}_{M_1M'_1}(\alpha,\beta,\gamma)\\
&=\frac{16\pi^2}{2J_1+1}\delta_{J_1J_2}\delta_{M_1M_2}\delta_{M'_1M'_2}. \tag*{(702)}
\end{aligned} ∫ 0 4 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 2 M 2 ′ J 2 ∗ ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 4 π d γ D M 2 M 2 ′ J 2 ∗ ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = 2 J 1 + 1 16 π 2 δ J 1 J 2 δ M 1 M 2 δ M 1 ′ M 2 ′ . (702)
If J 1 J_1 J 1 and J 2 J_2 J 2 are both either integer or half-integer, orthogonality of the D D D -functions takes place at integration over the single volume of rotation group which corresponds to the domain
V : 0 ≤ α < 2 π , 0 ≤ β ≤ π , 0 ≤ γ < 2 π . (703) V:\quad0\leq\alpha<2\pi,\quad0\leq\beta\leq\pi,\quad0\leq\gamma<2\pi. \tag*{(703)} V : 0 ≤ α < 2 π , 0 ≤ β ≤ π , 0 ≤ γ < 2 π . (703)
The total volume of the domain V V V is equal to 8 π 2 8\pi^2 8 π 2 . Hence, the orthogonality condition (4.10.3) is reduced to
∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 2 M 2 ′ J 2 ∗ ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = 8 π 2 2 J 1 + 1 δ J 1 J 2 δ M 1 M 2 δ M 1 ′ M 2 ′ . (704) \int_{0}^{2\pi}d\alpha\int_{0}^{\pi}d\beta\sin\beta\int_{0}^{2\pi}d\gamma\,D^{J_2*}_{M_2M'_2}(\alpha,\beta,\gamma)D^{J_1}_{M_1M'_1}(\alpha,\beta,\gamma)=\frac{8\pi^2}{2J_1+1}\delta_{J_1J_2}\delta_{M_1M_2}\delta_{M'_1M'_2}. \tag*{(704)} ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 2 M 2 ′ J 2 ∗ ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = 2 J 1 + 1 8 π 2 δ J 1 J 2 δ M 1 M 2 δ M 1 ′ M 2 ′ . (704)
In physical applications Eq. (704) is used more often than Eq. (702) . The orthonormality of D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) may be rewritten in terms of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β )
∫ 0 π d β sin β d M M ′ J ( β ) d M M ′ J ′ ( β ) = 2 2 J + 1 δ J J ′ . (705) \int_{0}^{\pi}d\beta\sin\beta\,d^{J}_{MM'}(\beta)d^{J'}_{MM'}(\beta)=\frac{2}{2J+1}\delta_{JJ'}. \tag*{(705)} ∫ 0 π d β sin β d M M ′ J ( β ) d M M ′ J ′ ( β ) = 2 J + 1 2 δ J J ′ . (705)
Thus, the collection of functions D M M J ( α , β , γ ) D^{J}_{MM}(\alpha,\beta,\gamma) D M M J ( α , β , γ ) with integer and half-integer J J J constitutes a complete set of orthonormalized functions. The completeness condition is
∑ J = 0 , 1 2 , 1 , … ∞ ∑ M = − J J ∑ M ′ = − J J 2 J + 1 16 π 2 D M M ′ J ∗ ( α 1 , β 1 , γ 1 ) D M M ′ J ( α 2 , β 2 , γ 2 ) = δ ( α 1 − α 2 ) δ ( cos β 1 − cos β 2 ) δ ( γ 1 − γ 2 ) . (706) \sum_{J=0,\frac{1}{2},1,\ldots}^{\infty}\sum_{M=-J}^{J}\sum_{M'=-J}^{J}\frac{2J+1}{16\pi^2}D^{J*}_{MM'}(\alpha_1,\beta_1,\gamma_1)D^{J}_{MM'}(\alpha_2,\beta_2,\gamma_2)
=\delta(\alpha_1-\alpha_2)\delta(\cos\beta_1-\cos\beta_2)\delta(\gamma_1-\gamma_2). \tag*{(706)} J = 0 , 2 1 , 1 , … ∑ ∞ M = − J ∑ J M ′ = − J ∑ J 16 π 2 2 J + 1 D M M ′ J ∗ ( α 1 , β 1 , γ 1 ) D M M ′ J ( α 2 , β 2 , γ 2 ) = δ ( α 1 − α 2 ) δ ( cos β 1 − cos β 2 ) δ ( γ 1 − γ 2 ) . (706)
Any function f ( α , β , γ ) f(\alpha,\beta,\gamma) f ( α , β , γ ) which is defined in the domain V 1 V_1 V 1 (or V 2 V_2 V 2 ) and satisfies the condition
∭ V 1 ( V 2 ) d α d β sin β d γ ∣ f ( α , β , γ ) ∣ 2 < ∞ , (707) \iiint_{V_1(V_2)}d\alpha\,d\beta\sin\beta\,d\gamma\,\lvert f(\alpha,\beta,\gamma)\rvert^2<\infty, \tag*{(707)} ∭ V 1 ( V 2 ) d α d β sin β d γ ∣ f ( α , β , γ ) ∣ 2 < ∞ , (707)
may be expanded in a series of the D D D -functions
f ( α , β , γ ) = ∑ J = 0 , 1 2 , 1 , … ∞ ∑ M = − J J ∑ M ′ = − J J a M M ′ J D M M ′ J ( α , β , γ ) . (708) f(\alpha,\beta,\gamma)=\sum_{J=0,\frac{1}{2},1,\ldots}^{\infty}\sum_{M=-J}^{J}\sum_{M'=-J}^{J}a^{J}_{MM'}D^{J}_{MM'}(\alpha,\beta,\gamma). \tag*{(708)} f ( α , β , γ ) = J = 0 , 2 1 , 1 , … ∑ ∞ M = − J ∑ J M ′ = − J ∑ J a M M ′ J D M M ′ J ( α , β , γ ) . (708)
The expansion coefficients a M M ′ J a^{J}_{MM'} a M M ′ J are determined by
a M M ′ J = 2 J + 1 16 π 2 ∭ V 1 ( V 2 ) d α d β sin β d γ f ( α , β , γ ) D M M ′ J ∗ ( α , β , γ ) , (709) a^{J}_{MM'}=\frac{2J+1}{16\pi^2}\iiint_{V_1(V_2)}d\alpha\,d\beta\sin\beta\,d\gamma\,f(\alpha,\beta,\gamma)D^{J*}_{MM'}(\alpha,\beta,\gamma), \tag*{(709)} a M M ′ J = 16 π 2 2 J + 1 ∭ V 1 ( V 2 ) d α d β sin β d γ f ( α , β , γ ) D M M ′ J ∗ ( α , β , γ ) , (709)
and obey the relation
∑ J = 0 , 1 2 , 1 , … ∞ 16 π 2 2 J + 1 ∑ M = − J J ∑ M ′ = − J J ∣ a M M ′ J ∣ 2 = ∭ V 1 ( V 2 ) d α d β sin β d γ ∣ f ( α , β , γ ) ∣ 2 . (710) \sum_{J=0,\frac{1}{2},1,\ldots}^{\infty}\frac{16\pi^2}{2J+1}\sum_{M=-J}^{J}\sum_{M'=-J}^{J}\lvert a^{J}_{MM'}\rvert^2=\iiint_{V_1(V_2)}d\alpha\,d\beta\sin\beta\,d\gamma\,\lvert f(\alpha,\beta,\gamma)\rvert^2. \tag*{(710)} J = 0 , 2 1 , 1 , … ∑ ∞ 2 J + 1 16 π 2 M = − J ∑ J M ′ = − J ∑ J ∣ a M M ′ J ∣ 2 = ∭ V 1 ( V 2 ) d α d β sin β d γ ∣ f ( α , β , γ ) ∣ 2 . (710)
If a function f ( α , β , γ ) f(\alpha,\beta,\gamma) f ( α , β , γ ) is defined in the domain V V V , it may be expanded in a series of the D D D -functions with only integer or only half-integer J J J
f ( α , β , γ ) = ∑ J i n t e g e r o r h a l f - i n t e g e r ∑ M = − J J ∑ M ′ = − J J b M M ′ J D M M ′ J ( α , β , γ ) . (711) f(\alpha,\beta,\gamma)=\sum_{\substack{J\ \mathrm{integer}\\\mathrm{or\ half\text{-}integer}}\ }\sum_{M=-J}^{J}\sum_{M'=-J}^{J}b^J_{MM'}D^J_{MM'}(\alpha,\beta,\gamma). \tag*{(711)} f ( α , β , γ ) = J integer or half - integer ∑ M = − J ∑ J M ′ = − J ∑ J b M M ′ J D M M ′ J ( α , β , γ ) . (711)
The expansion coefficients b M M ′ J b^J_{MM'} b M M ′ J are determined by
b M M ′ J = 2 J + 1 8 π 2 ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ f ( α , β , γ ) D M M ′ J ∗ ( α , β , γ ) (712) b^J_{MM'}=\frac{2J+1}{8\pi^2}\int_0^{2\pi}d\alpha\int_0^\pi d\beta\sin\beta\int_0^{2\pi}d\gamma f(\alpha,\beta,\gamma)D^{J*}_{MM'}(\alpha,\beta,\gamma) \tag*{(712)} b M M ′ J = 8 π 2 2 J + 1 ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ f ( α , β , γ ) D M M ′ J ∗ ( α , β , γ ) (712)
for both cases of integer and half-integer J J J . A difference between the expansions of f ( α , β , γ ) f(\alpha,\beta,\gamma) f ( α , β , γ ) in these two cases occurs if one uses the expansions outside of the domain V V V . The expansion coefficients in Eq. (711) satisfy the relation
∑ J i n t e g e r o r h a l f - i n t e g e r 8 π 2 2 J + 1 ∑ M = − J J ∑ M ′ = − J J ∣ b M M ′ J ∣ 2 = ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ ∣ f ( α , β , γ ) ∣ 2 . (713) \sum_{\substack{J\ \mathrm{integer}\\\mathrm{or\ half\text{-}integer}}\ }\frac{8\pi^2}{2J+1}\sum_{M=-J}^{J}\sum_{M'=-J}^{J}\left|b^J_{MM'}\right|^2=\int_0^{2\pi}d\alpha\int_0^\pi d\beta\sin\beta\int_0^{2\pi}d\gamma\left|f(\alpha,\beta,\gamma)\right|^2. \tag*{(713)} J integer or half - integer ∑ 2 J + 1 8 π 2 M = − J ∑ J M ′ = − J ∑ J b M M ′ J 2 = ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ ∣ f ( α , β , γ ) ∣ 2 . (713)
Integrals involving the D D D -functions
Integration of Products of D M M ′ J D^J_{MM'} D M M ′ J
∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M M ′ J ( α , β , γ ) = δ J 0 δ M 0 δ M ′ 0 8 π 2 , ( J i s i n t e g e r ) , (714) \int_0^{2\pi}d\alpha\int_0^\pi d\beta\sin\beta\int_0^{2\pi}d\gamma D^J_{MM'}(\alpha,\beta,\gamma)=\delta_{J0}\delta_{M0}\delta_{M'0}8\pi^2,\quad(J\ \mathrm{is\ integer}), \tag*{(714)} ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M M ′ J ( α , β , γ ) = δ J 0 δ M 0 δ M ′ 0 8 π 2 , ( J is integer ) , (714)
∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 1 M 1 ′ J 1 ( α , β , γ ) D M 2 M 2 ′ J 2 ( α , β , γ ) = ( − 1 ) M 2 − M 2 ′ 8 π 2 2 J 2 + 1 δ J 1 J 2 δ − M 1 M 2 δ − M 1 ′ M 2 ′ , \int_0^{2\pi}d\alpha\int_0^\pi d\beta\sin\beta\int_0^{2\pi}d\gamma D^{J_1}_{M_1M'_1}(\alpha,\beta,\gamma)D^{J_2}_{M_2M'_2}(\alpha,\beta,\gamma)=(-1)^{M_2-M'_2}\frac{8\pi^2}{2J_2+1}\delta_{J_1J_2}\delta_{-M_1M_2}\delta_{-M'_1M'_2}, ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 1 M 1 ′ J 1 ( α , β , γ ) D M 2 M 2 ′ J 2 ( α , β , γ ) = ( − 1 ) M 2 − M 2 ′ 2 J 2 + 1 8 π 2 δ J 1 J 2 δ − M 1 M 2 δ − M 1 ′ M 2 ′ ,
( J 1 + J 2 i s i n t e g e r ) (715) (J_1+J_2\ \mathrm{is\ integer}) \tag*{(715)} ( J 1 + J 2 is integer ) (715)
∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 2 M 2 ′ J 2 ∗ ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = 8 π 2 2 J 2 + 1 δ J 1 J 2 δ M 1 M 2 δ M 1 ′ M 2 ′ , \int_0^{2\pi}d\alpha\int_0^\pi d\beta\sin\beta\int_0^{2\pi}d\gamma D^{J_2*}_{M_2M'_2}(\alpha,\beta,\gamma)D^{J_1}_{M_1M'_1}(\alpha,\beta,\gamma)=\frac{8\pi^2}{2J_2+1}\delta_{J_1J_2}\delta_{M_1M_2}\delta_{M'_1M'_2}, ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 2 M 2 ′ J 2 ∗ ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = 2 J 2 + 1 8 π 2 δ J 1 J 2 δ M 1 M 2 δ M 1 ′ M 2 ′ ,
( J 1 + J 2 i s i n t e g e r ) , (716) (J_1+J_2\ \mathrm{is\ integer}), \tag*{(716)} ( J 1 + J 2 is integer ) , (716)
∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 3 M 3 ′ J 3 ( α , β , γ ) D M 2 M 2 ′ J 2 ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = ( − 1 ) M 3 − M 3 ′ 8 π 2 2 J 3 + 1 C J 1 M 1 , J 2 M 2 J 3 , − M 3 C J 1 M 1 ′ , J 2 M 2 ′ J 3 , − M 3 ′ , \begin{aligned}
&\int_0^{2\pi}d\alpha\int_0^\pi d\beta\sin\beta\int_0^{2\pi}d\gamma D^{J_3}_{M_3M'_3}(\alpha,\beta,\gamma)D^{J_2}_{M_2M'_2}(\alpha,\beta,\gamma)D^{J_1}_{M_1M'_1}(\alpha,\beta,\gamma)\\
&=(-1)^{M_3-M'_3}\frac{8\pi^2}{2J_3+1}C^{J_3,-M_3}_{J_1M_1,J_2M_2}C^{J_3,-M'_3}_{J_1M'_1,J_2M'_2},
\end{aligned} ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 3 M 3 ′ J 3 ( α , β , γ ) D M 2 M 2 ′ J 2 ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = ( − 1 ) M 3 − M 3 ′ 2 J 3 + 1 8 π 2 C J 1 M 1 , J 2 M 2 J 3 , − M 3 C J 1 M 1 ′ , J 2 M 2 ′ J 3 , − M 3 ′ ,
( J 1 + J 2 + J 3 i s i n t e g e r ) , (717) (J_1+J_2+J_3\ \mathrm{is\ integer}), \tag*{(717)} ( J 1 + J 2 + J 3 is integer ) , (717)
∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 3 M 3 ′ J 3 ∗ ( α , β , γ ) D M 2 M 2 ′ J 2 ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = 8 π 2 2 J 3 + 1 C J 1 M 1 , J 2 M 2 J 3 M 3 C J 1 M 1 ′ , J 2 M 2 ′ J 3 M 3 ′ , \begin{aligned}
&\int_0^{2\pi}d\alpha\int_0^\pi d\beta\sin\beta\int_0^{2\pi}d\gamma D^{J_3*}_{M_3M'_3}(\alpha,\beta,\gamma)D^{J_2}_{M_2M'_2}(\alpha,\beta,\gamma)D^{J_1}_{M_1M'_1}(\alpha,\beta,\gamma)\\
&=\frac{8\pi^2}{2J_3+1}C^{J_3M_3}_{J_1M_1,J_2M_2}C^{J_3M'_3}_{J_1M'_1,J_2M'_2},
\end{aligned} ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ D M 3 M 3 ′ J 3 ∗ ( α , β , γ ) D M 2 M 2 ′ J 2 ( α , β , γ ) D M 1 M 1 ′ J 1 ( α , β , γ ) = 2 J 3 + 1 8 π 2 C J 1 M 1 , J 2 M 2 J 3 M 3 C J 1 M 1 ′ , J 2 M 2 ′ J 3 M 3 ′ ,
(J_1+J_2+J_3\ \mathrm{is\ integer})$ {#eq:4.11.5}
Equations (4.11.1)–(4.11.5) are valid, if the conditions in parentheses are satisfied. These conditions may be omitted, if the integration domain is V 1 V_1 V 1 or V 2 V_2 V 2 instead of V V V ; in the latter case the factor 8 π 2 8\pi^2 8 π 2 on the right-hand side of Eqs. (4.11.1)–(4.11.5) must be replaced by 16 π 2 16\pi^2 16 π 2 (see § 18.2 ).
Integrals involving products of four or more D D D -functions may be reduced to Eqs. (4.11.4)–(4.11.5) by using the Clebsch–Gordan expansion (see § 17.1.1 ).
Integrals Involving d M M ′ J ( β ) d^J_{MM'}(\beta) d M M ′ J ( β )
From Eqs. (4.11.1)–(4.11.5) one may obtain
∫ 0 π d β sin β d 00 J ( β ) = 2 δ J 0 , (718) \int_0^\pi d\beta\sin\beta d^J_{00}(\beta)=2\delta_{J0}, \tag*{(718)} ∫ 0 π d β sin β d 00 J ( β ) = 2 δ J 0 , (718)
∫ 0 π d β sin β d M M ′ J ( β ) d M M ′ J ′ ( β ) = 2 2 J + 1 δ J J ′ , (719) \int_0^\pi d\beta\sin\beta d^J_{MM'}(\beta)d^{J'}_{MM'}(\beta)=\frac{2}{2J+1}\delta_{JJ'}, \tag*{(719)} ∫ 0 π d β sin β d M M ′ J ( β ) d M M ′ J ′ ( β ) = 2 J + 1 2 δ J J ′ , (719)
∫ 0 π d β sin β d M 1 M 1 ′ J 1 ( β ) d M 2 M 2 ′ J 2 ( β ) d M 3 M 3 ′ J 3 ( β ) δ M 1 + M 2 M 3 δ M 1 ′ + M 2 ′ M 3 ′ = 2 2 J 3 + 1 C J 1 M 1 , J 2 M 2 J 3 M 3 C J 1 M 1 ′ , J 2 M 2 ′ J 3 M 3 ′ . (720) \int_0^\pi d\beta\sin\beta d^{J_1}_{M_1M'_1}(\beta)d^{J_2}_{M_2M'_2}(\beta)d^{J_3}_{M_3M'_3}(\beta)\delta_{M_1+M_2M_3}\delta_{M'_1+M'_2M'_3}=\frac{2}{2J_3+1}C^{J_3M_3}_{J_1M_1,J_2M_2}C^{J_3M'_3}_{J_1M'_1,J_2M'_2}. \tag*{(720)} ∫ 0 π d β sin β d M 1 M 1 ′ J 1 ( β ) d M 2 M 2 ′ J 2 ( β ) d M 3 M 3 ′ J 3 ( β ) δ M 1 + M 2 M 3 δ M 1 ′ + M 2 ′ M 3 ′ = 2 J 3 + 1 2 C J 1 M 1 , J 2 M 2 J 3 M 3 C J 1 M 1 ′ , J 2 M 2 ′ J 3 M 3 ′ . (720)
Note also the following relation
∫ 0 β d β ( sin β 2 ) M − M ′ + 1 ( cos β 2 ) M + M ′ + 1 d M M ′ J ( β ) = − 1 J ( J + 1 ) − M ( M + 1 ) ( sin β 2 ) M − M ′ + 1 ( cos β 2 ) M + M ′ + 1 d M + 1 M ′ J ( β ) . (721) \int_{0}^{\beta}d\beta\left(\sin\frac{\beta}{2}\right)^{M-M'+1}\left(\cos\frac{\beta}{2}\right)^{M+M'+1}d^{J}_{MM'}(\beta)
=
\frac{-1}{\sqrt{J(J+1)-M(M+1)}}\left(\sin\frac{\beta}{2}\right)^{M-M'+1}\left(\cos\frac{\beta}{2}\right)^{M+M'+1}d^{J}_{M+1M'}(\beta). \tag*{(721)} ∫ 0 β d β ( sin 2 β ) M − M ′ + 1 ( cos 2 β ) M + M ′ + 1 d M M ′ J ( β ) = J ( J + 1 ) − M ( M + 1 ) − 1 ( sin 2 β ) M − M ′ + 1 ( cos 2 β ) M + M ′ + 1 d M + 1 M ′ J ( β ) . (721)
Equation (4.11.9) is valid, if M ≥ M ′ ≥ 0 M \ge M' \ge0 M ≥ M ′ ≥ 0 . Other cases may be deduced from Eq. (721) using the symmetries of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) (Eq. (590) ).
Invariant summation of integrals involving D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
Hereafter R R R will denote the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ and we also will set d R = sin β d β d α d γ dR=\sin\beta d\beta d\alpha d\gamma d R = sin β d β d α d γ :
∫ d R f ( R ) ≡ ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ f ( α , β , γ ) . (722) \int dR f(R)\equiv\int_{0}^{2\pi}d\alpha\int_{0}^{\pi}d\beta\sin\beta\int_{0}^{2\pi}d\gamma f(\alpha,\beta,\gamma). \tag*{(722)} ∫ d R f ( R ) ≡ ∫ 0 2 π d α ∫ 0 π d β sin β ∫ 0 2 π d γ f ( α , β , γ ) . (722)
Making use of this notation, one gets
∑ M M ′ ∫ d R D M M J ∗ ( R ) D M ′ M ′ J ′ ( R ) = 8 π 2 δ J J ′ , (723) \sum_{MM'}\int dR D^{J*}_{MM}(R)D^{J'}_{M'M'}(R)=8\pi^{2}\delta_{JJ'}, \tag*{(723)} M M ′ ∑ ∫ d R D M M J ∗ ( R ) D M ′ M ′ J ′ ( R ) = 8 π 2 δ J J ′ , (723)
∑ M M ′ ∫ d R D M M ′ J ( R ) D M ′ M J ′ ( R ) = ( − 1 ) 2 J 8 π 2 δ J J ′ , (724) \sum_{MM'}\int dR D^{J}_{MM'}(R)D^{J'}_{M'M}(R)=(-1)^{2J}8\pi^{2}\delta_{JJ'}, \tag*{(724)} M M ′ ∑ ∫ d R D M M ′ J ( R ) D M ′ M J ′ ( R ) = ( − 1 ) 2 J 8 π 2 δ J J ′ , (724)
∑ M 1 M 2 M 3 ∫ d R D M 3 M 3 J 3 ( R ) D M 2 M 2 J 2 ( R ) D M 1 M 1 J 1 ( R ) = 8 π 2 { J 1 J 2 J 3 0 0 0 } , (725) \sum_{M_{1}M_{2}M_{3}}\int dR D^{J_{3}}_{M_{3}M_{3}}(R)D^{J_{2}}_{M_{2}M_{2}}(R)D^{J_{1}}_{M_{1}M_{1}}(R)=8\pi^{2}\left\{\begin{matrix}J_{1}&J_{2}&J_{3}\\0&0&0\end{matrix}\right\}, \tag*{(725)} M 1 M 2 M 3 ∑ ∫ d R D M 3 M 3 J 3 ( R ) D M 2 M 2 J 2 ( R ) D M 1 M 1 J 1 ( R ) = 8 π 2 { J 1 0 J 2 0 J 3 0 } , (725)
∑ A l l M , N ∫ d R 1 D M 1 N 1 J 1 ( R 1 ) D M 2 N 2 J 2 ( R 1 ) D M 3 M 3 J 3 ( R 1 ) ∫ d R 2 D N 1 M 1 J 1 ( R 2 ) D M 2 N 2 J 2 ∗ ( R 2 ) D M 3 ′ M 3 ′ J ′ ( R 2 ) = ( − 1 ) 2 J 3 ′ ( 8 π 2 ) 2 { J 1 J 2 J 3 J 1 J 2 J 3 ′ } , (726) \sum_{\mathrm{All}\ M,N}\int dR_{1}D^{J_{1}}_{M_{1}N_{1}}(R_{1})D^{J_{2}}_{M_{2}N_{2}}(R_{1})D^{J_{3}}_{M_{3}M_{3}}(R_{1})\int dR_{2}D^{J_{1}}_{N_{1}M_{1}}(R_{2})D^{J_{2}*}_{M_{2}N_{2}}(R_{2})D^{J'}_{M'_{3}M'_{3}}(R_{2})
=(-1)^{2J'_{3}}(8\pi^{2})^{2}
\left\{\begin{matrix}J_{1}&J_{2}&J_{3}\\J_{1}&J_{2}&J'_{3}\end{matrix}\right\}, \tag*{(726)} All M , N ∑ ∫ d R 1 D M 1 N 1 J 1 ( R 1 ) D M 2 N 2 J 2 ( R 1 ) D M 3 M 3 J 3 ( R 1 ) ∫ d R 2 D N 1 M 1 J 1 ( R 2 ) D M 2 N 2 J 2 ∗ ( R 2 ) D M 3 ′ M 3 ′ J ′ ( R 2 ) = ( − 1 ) 2 J 3 ′ ( 8 π 2 ) 2 { J 1 J 1 J 2 J 2 J 3 J 3 ′ } , (726)
∑ A l l M , N ∫ d R 1 D M 1 ′ N 1 ′ J 1 ′ ∗ ( R 1 ) D M 2 ′ N 2 ′ J 2 ′ ( R 1 ) D M 3 M 3 J 3 ( R 1 ) ∫ d R 2 D M 2 ′ N 2 ′ J 2 ′ ∗ ( R 2 ) D M 3 ′ N 3 ′ J 3 ′ ( R 2 ) D M 1 M 1 J 1 ( R 2 ) × ∫ d R 3 D M 3 ′ N 3 ′ J 3 ′ ∗ ( R 3 ) D M 1 ′ N 1 ′ J 1 ′ ( R 3 ) D M 2 M 2 J 2 ( R 3 ) = ( 8 π 2 ) 3 { J 1 J 2 J 3 J 1 ′ J 2 ′ J 3 ′ } 2 , (727) \sum_{\mathrm{All}\ M,N}\int dR_{1}D^{J'_{1}*}_{M'_{1}N'_{1}}(R_{1})D^{J'_{2}}_{M'_{2}N'_{2}}(R_{1})D^{J_{3}}_{M_{3}M_{3}}(R_{1})\int dR_{2}D^{J'_{2}*}_{M'_{2}N'_{2}}(R_{2})D^{J'_{3}}_{M'_{3}N'_{3}}(R_{2})D^{J_{1}}_{M_{1}M_{1}}(R_{2})
\times\int dR_{3}D^{J'_{3}*}_{M'_{3}N'_{3}}(R_{3})D^{J'_{1}}_{M'_{1}N'_{1}}(R_{3})D^{J_{2}}_{M_{2}M_{2}}(R_{3})
=(8\pi^{2})^{3}\left\{\begin{matrix}J_{1}&J_{2}&J_{3}\\J'_{1}&J'_{2}&J'_{3}\end{matrix}\right\}^{2}, \tag*{(727)} All M , N ∑ ∫ d R 1 D M 1 ′ N 1 ′ J 1 ′ ∗ ( R 1 ) D M 2 ′ N 2 ′ J 2 ′ ( R 1 ) D M 3 M 3 J 3 ( R 1 ) ∫ d R 2 D M 2 ′ N 2 ′ J 2 ′ ∗ ( R 2 ) D M 3 ′ N 3 ′ J 3 ′ ( R 2 ) D M 1 M 1 J 1 ( R 2 ) × ∫ d R 3 D M 3 ′ N 3 ′ J 3 ′ ∗ ( R 3 ) D M 1 ′ N 1 ′ J 1 ′ ( R 3 ) D M 2 M 2 J 2 ( R 3 ) = ( 8 π 2 ) 3 { J 1 J 1 ′ J 2 J 2 ′ J 3 J 3 ′ } 2 , (727)
∑ A l l M , N ∫ d R 1 D M M J ( R 1 ) D N 2 N 2 ′ J 2 ( R 1 ) D N 3 N 3 ′ J 3 ( R 1 ) ∫ d R 2 D M ′ M ′ J ′ ( R 2 ) D N 1 M 1 J 1 ( R 2 ) D N 2 ′ M 2 ′ J 2 ( R 2 ) × ∫ d R 3 D M ′ ′ M ′ ′ J ′ ′ ( R 3 ) D M 2 ′ M 2 ′ J 2 ( R 3 ) D N 2 ′ M 3 J 3 ( R 3 ) ∫ d R 4 D M 1 N 1 J 1 ( R 4 ) D M 2 N 2 J 2 ( R 4 ) D M 2 ′ ′ M 2 ′ ′ J 2 ( R 4 ) D M 3 N 3 J 3 ( R 4 ) = ( − 1 ) 2 J 1 ( 8 π 2 ) 4 { J 1 J 2 J ′ J 3 J 2 J } { J 1 J 2 J ′ J 3 J 2 J ′ } { J 1 J 2 J ′ ′ J 3 J 2 J ′ ′ } , (728) \sum_{\mathrm{All}\ M,N}\int dR_{1}D^{J}_{MM}(R_{1})D^{J_{2}}_{N_{2}N'_{2}}(R_{1})D^{J_{3}}_{N_{3}N'_{3}}(R_{1})\int dR_{2}D^{J'}_{M'M'}(R_{2})D^{J_{1}}_{N_{1}M_{1}}(R_{2})D^{J_{2}}_{N'_{2}M'_{2}}(R_{2})
\times\int dR_{3}D^{J''}_{M''M''}(R_{3})D^{J_{2}}_{M'_{2}M'_{2}}(R_{3})D^{J_{3}}_{N'_{2}M_{3}}(R_{3})\int dR_{4}D^{J_{1}}_{M_{1}N_{1}}(R_{4})D^{J_{2}}_{M_{2}N_{2}}(R_{4})D^{J_{2}}_{M''_{2}M''_{2}}(R_{4})D^{J_{3}}_{M_{3}N_{3}}(R_{4})
=(-1)^{2J_{1}}(8\pi^{2})^{4}
\left\{\begin{matrix}J_{1}&J_{2}&J'\\J_{3}&J_{2}&J\end{matrix}\right\}
\left\{\begin{matrix}J_{1}&J_{2}&J'\\J_{3}&J_{2}&J'\end{matrix}\right\}
\left\{\begin{matrix}J_{1}&J_{2}&J''\\J_{3}&J_{2}&J''\end{matrix}\right\}, \tag*{(728)} All M , N ∑ ∫ d R 1 D M M J ( R 1 ) D N 2 N 2 ′ J 2 ( R 1 ) D N 3 N 3 ′ J 3 ( R 1 ) ∫ d R 2 D M ′ M ′ J ′ ( R 2 ) D N 1 M 1 J 1 ( R 2 ) D N 2 ′ M 2 ′ J 2 ( R 2 ) × ∫ d R 3 D M ′′ M ′′ J ′′ ( R 3 ) D M 2 ′ M 2 ′ J 2 ( R 3 ) D N 2 ′ M 3 J 3 ( R 3 ) ∫ d R 4 D M 1 N 1 J 1 ( R 4 ) D M 2 N 2 J 2 ( R 4 ) D M 2 ′′ M 2 ′′ J 2 ( R 4 ) D M 3 N 3 J 3 ( R 4 ) = ( − 1 ) 2 J 1 ( 8 π 2 ) 4 { J 1 J 3 J 2 J 2 J ′ J } { J 1 J 3 J 2 J 2 J ′ J ′ } { J 1 J 3 J 2 J 2 J ′′ J ′′ } , (728)
∑ A l l M , N ∫ d R 1 D M M J ( R 1 ) D M ′ N ′ J ( R 1 ) D M 1 N 1 J 1 ∗ ( R 1 ) ∫ d R 2 D N ′ M ′ J ′ ( R 2 ) D N N J ( R 2 ) D M 2 N 2 J 2 ∗ ( R 2 ) × ∫ d R 3 D M 1 N 1 J 1 ( R 3 ) D M 2 N 2 J 2 ( R 3 ) D M 3 M 3 J ∗ 3 ( R 3 ) = ( 8 π 2 ) 3 { J J J 1 J J J 2 J 1 J 2 J 3 } , (729) \sum_{\mathrm{All}\ M,N}\int dR_{1}D^{J}_{MM}(R_{1})D^{J}_{M'N'}(R_{1})D^{J_{1}*}_{M_{1}N_{1}}(R_{1})\int dR_{2}D^{J'}_{N'M'}(R_{2})D^{J}_{NN}(R_{2})D^{J_{2}*}_{M_{2}N_{2}}(R_{2})
\times\int dR_{3}D^{J_{1}}_{M_{1}N_{1}}(R_{3})D^{J_{2}}_{M_{2}N_{2}}(R_{3})D^{J_{*3}}_{M_{3}M_{3}}(R_{3})=(8\pi^{2})^{3}
\left\{\begin{matrix}J&J&J_{1}\\J&J&J_{2}\\J_{1}&J_{2}&J_{3}\end{matrix}\right\}, \tag*{(729)} All M , N ∑ ∫ d R 1 D M M J ( R 1 ) D M ′ N ′ J ( R 1 ) D M 1 N 1 J 1 ∗ ( R 1 ) ∫ d R 2 D N ′ M ′ J ′ ( R 2 ) D N N J ( R 2 ) D M 2 N 2 J 2 ∗ ( R 2 ) × ∫ d R 3 D M 1 N 1 J 1 ( R 3 ) D M 2 N 2 J 2 ( R 3 ) D M 3 M 3 J ∗ 3 ( R 3 ) = ( 8 π 2 ) 3 ⎩ ⎨ ⎧ J J J 1 J J J 2 J 1 J 2 J 3 ⎭ ⎬ ⎫ , (729)
∑ A l l M , N ∫ d R 1 D M 1 N 1 J 1 ( R 1 ) D M 2 N 2 J 2 ( R 1 ) D M 12 M 12 J 12 ( R 1 ) ∫ d R 2 D M 3 N 3 J 3 ( R 2 ) D M 4 N 4 J 4 ( R 2 ) D M 34 M 34 J 34 ( R 2 ) × ∫ d R 3 D M 13 N 13 J 13 ( R 3 ) D M 24 N 24 J 24 ( R 3 ) D M M J ( R 3 ) ∫ d R 4 D N 1 M 1 J 1 ( R 4 ) D N 3 M 3 J 3 ( R 4 ) D N 13 M 13 J 13 ( R 4 ) × ∫ d R 5 D N 2 M 2 J 2 ( R 5 ) D N 4 M 4 J 4 ( R 5 ) D N 24 M 24 J 24 ( R 5 ) = ( 8 π 2 ) 5 { J 1 J 2 J 12 J 3 J 4 J 34 J 13 J 24 J } 2 . (730) \sum_{\mathrm{All}\ M,N}\int dR_1 D^{J_1}_{M_1N_1}(R_1)D^{J_2}_{M_2N_2}(R_1)D^{J_{12}}_{M_{12}M_{12}}(R_1)\int dR_2 D^{J_3}_{M_3N_3}(R_2)D^{J_4}_{M_4N_4}(R_2)D^{J_{34}}_{M_{34}M_{34}}(R_2)
\\
{}\times\int dR_3 D^{J_{13}}_{M_{13}N_{13}}(R_3)D^{J_{24}}_{M_{24}N_{24}}(R_3)D^J_{MM}(R_3)\int dR_4 D^{J_1}_{N_1M_1}(R_4)D^{J_3}_{N_3M_3}(R_4)D^{J_{13}}_{N_{13}M_{13}}(R_4)
\\
{}\times\int dR_5 D^{J_2}_{N_2M_2}(R_5)D^{J_4}_{N_4M_4}(R_5)D^{J_{24}}_{N_{24}M_{24}}(R_5)
=(8\pi^2)^5
\begin{Bmatrix}
J_1 & J_2 & J_{12}\\
J_3 & J_4 & J_{34}\\
J_{13} & J_{24} & J
\end{Bmatrix}^{2}. \tag*{(730)} All M , N ∑ ∫ d R 1 D M 1 N 1 J 1 ( R 1 ) D M 2 N 2 J 2 ( R 1 ) D M 12 M 12 J 12 ( R 1 ) ∫ d R 2 D M 3 N 3 J 3 ( R 2 ) D M 4 N 4 J 4 ( R 2 ) D M 34 M 34 J 34 ( R 2 ) × ∫ d R 3 D M 13 N 13 J 13 ( R 3 ) D M 24 N 24 J 24 ( R 3 ) D M M J ( R 3 ) ∫ d R 4 D N 1 M 1 J 1 ( R 4 ) D N 3 M 3 J 3 ( R 4 ) D N 13 M 13 J 13 ( R 4 ) × ∫ d R 5 D N 2 M 2 J 2 ( R 5 ) D N 4 M 4 J 4 ( R 5 ) D N 24 M 24 J 24 ( R 5 ) = ( 8 π 2 ) 5 ⎩ ⎨ ⎧ J 1 J 3 J 13 J 2 J 4 J 24 J 12 J 34 J ⎭ ⎬ ⎫ 2 . (730)
The left-hand sides of Eqs. (726) –(730) include integrations as well as summations. If the integrations are carried out before the summations (with the aid of Eqs. (717) –(4.11.5)), then Eqs. (726) –(730) are reduced to sums of the Clebsch–Gordan coefficients (Sec. 8.7). On the other hand, if the summations are preformed before the integrations, one obtains the integral representations of 6 j 6j 6 j - and 9 j 9j 9 j -symbols (see § 1.3 and § 2.3 ).
Generating functions for dM MJ()
The functions d M M ′ J ( β ) d^J_{MM'}(\beta) d M M ′ J ( β ) may be defined as coefficients of expansions of various generating functions.
(a) If J J J and M ′ M' M ′ are fixed, then
( cos β 2 e i φ / 2 + i sin β 2 e − i φ / 2 ) J + M ′ ( i sin β 2 e i φ / 2 + cos β 2 e − i φ / 2 ) J − M ′ = ∑ M = − J J ( J − M ′ ) ! ( J + M ′ ) ! ( J − M ) ! ( J + M ) ! ( − i ) M − M ′ e i M φ d M M ′ J ( β ) (731) \left(\cos\frac{\beta}{2}e^{i\varphi/2}+i\sin\frac{\beta}{2}e^{-i\varphi/2}\right)^{J+M'}
\left(i\sin\frac{\beta}{2}e^{i\varphi/2}+\cos\frac{\beta}{2}e^{-i\varphi/2}\right)^{J-M'}
=
\sum_{M=-J}^{J}
\sqrt{\frac{(J-M')!(J+M')!}{(J-M)!(J+M)!}}(-i)^{M-M'}e^{iM\varphi}d^J_{MM'}(\beta) \tag*{(731)} ( cos 2 β e i φ /2 + i sin 2 β e − i φ /2 ) J + M ′ ( i sin 2 β e i φ /2 + cos 2 β e − i φ /2 ) J − M ′ = M = − J ∑ J ( J − M )! ( J + M )! ( J − M ′ )! ( J + M ′ )! ( − i ) M − M ′ e i M φ d M M ′ J ( β ) (731)
In particular, for φ = 0 \varphi=0 φ = 0
e i J β = ∑ M = − J J ( J − M ′ ) ! ( J + M ′ ) ! ( J − M ) ! ( J + M ) ! ( − i ) M − M ′ d M M ′ J ( β ) (732) e^{iJ\beta}
=
\sum_{M=-J}^{J}
\sqrt{\frac{(J-M')!(J+M')!}{(J-M)!(J+M)!}}(-i)^{M-M'}d^J_{MM'}(\beta) \tag*{(732)} e i J β = M = − J ∑ J ( J − M )! ( J + M )! ( J − M ′ )! ( J + M ′ )! ( − i ) M − M ′ d M M ′ J ( β ) (732)
and for φ = π \varphi=\pi φ = π
e − i J β = ∑ M = − J J ( J − M ′ ) ! ( J + M ′ ) ! ( J − M ) ! ( J + M ) ! i M − M ′ d M M ′ J ( β ) (733) e^{-iJ\beta}
=
\sum_{M=-J}^{J}
\sqrt{\frac{(J-M')!(J+M')!}{(J-M)!(J+M)!}}i^{M-M'}d^J_{MM'}(\beta) \tag*{(733)} e − i J β = M = − J ∑ J ( J − M )! ( J + M )! ( J − M ′ )! ( J + M ′ )! i M − M ′ d M M ′ J ( β ) (733)
(b) If M M M and M ′ M' M ′ are fixed, one can obtain several expressions in which s , μ , ν s,\mu,\nu s , μ , ν are related to J , M , M ′ J,M,M' J , M , M ′ by Eqs. (580) , ξ M M ′ \xi_{MM'} ξ M M ′ is defined by Eq. (581) ; and R R R denotes
R = 1 − 2 t cos β + t 2 , ∣ t ∣ < 1. (734) R=\sqrt{1-2t\cos\beta+t^2},\qquad|t|<1. \tag*{(734)} R = 1 − 2 t cos β + t 2 , ∣ t ∣ < 1. (734)
ξ M M ′ R ( 2 sin β 2 1 + R − t ) μ ( 2 cos β 2 1 + R + t ) ν = ∑ s = 0 ∞ ( s + μ ) ! ( s + ν ) ! s ! ( s + μ + ν ) ! t s d M M ′ s + μ + ν 2 ( β ) . (735) \frac{\xi_{MM'}}{R}
\left(\frac{2\sin\frac{\beta}{2}}{1+R-t}\right)^\mu
\left(\frac{2\cos\frac{\beta}{2}}{1+R+t}\right)^\nu
=
\sum_{s=0}^{\infty}
\sqrt{\frac{(s+\mu)!(s+\nu)!}{s!(s+\mu+\nu)!}}
t^s d^{s+\frac{\mu+\nu}{2}}_{MM'}(\beta). \tag*{(735)} R ξ M M ′ ( 1 + R − t 2 sin 2 β ) μ ( 1 + R + t 2 cos 2 β ) ν = s = 0 ∑ ∞ s ! ( s + μ + ν )! ( s + μ )! ( s + ν )! t s d M M ′ s + 2 μ + ν ( β ) . (735)
ξ M M ′ ( sin β 2 ) μ ( cos β 2 ) ν 2 F 1 ( ; 1 + μ ; − t sin 2 β 2 ) 2 F 1 ( ; 1 + ν ; t cos 2 β 2 ) = ∑ s = 0 ∞ μ ! ν ! s ! ( s + μ ) ! ( s + ν ) ! ( s + μ + ν ) ! t s d M M ′ s + μ + ν 2 ( β ) (736) \xi_{MM'}
\left(\sin\frac{\beta}{2}\right)^\mu
\left(\cos\frac{\beta}{2}\right)^\nu
{}_2F_1\left(;1+\mu;-t\sin^2\frac{\beta}{2}\right)
{}_2F_1\left(;1+\nu;t\cos^2\frac{\beta}{2}\right)
=
\sum_{s=0}^{\infty}
\frac{\mu!\nu!}{\sqrt{s!(s+\mu)!(s+\nu)!(s+\mu+\nu)!}}
t^s d^{s+\frac{\mu+\nu}{2}}_{MM'}(\beta) \tag*{(736)} ξ M M ′ ( sin 2 β ) μ ( cos 2 β ) ν 2 F 1 ( ; 1 + μ ; − t sin 2 2 β ) 2 F 1 ( ; 1 + ν ; t cos 2 2 β ) = s = 0 ∑ ∞ s ! ( s + μ )! ( s + ν )! ( s + μ + ν )! μ ! ν ! t s d M M ′ s + 2 μ + ν ( β ) (736)
ξ M M ′ ( 1 − t ) 1 − μ − ν ( sin β 2 ) μ ( cos β 2 ) ν 2 F 1 ( μ + ν + 1 2 , μ + ν + 2 2 ; 1 + μ ; − 4 t sin 2 β 2 ( 1 − t ) 2 ) = ∑ s = 0 ∞ ( s + ν ) ! ( s + μ + ν ) ! s ! ( s + μ ) ! μ ! ( μ + ν ) ! t s d M M ′ s + μ + ν 2 ( β ) (737) \xi_{MM'}(1-t)^{1-\mu-\nu}
\left(\sin\frac{\beta}{2}\right)^\mu
\left(\cos\frac{\beta}{2}\right)^\nu
{}_2F_1\left(\frac{\mu+\nu+1}{2},\frac{\mu+\nu+2}{2};1+\mu;-\frac{4t\sin^2\frac{\beta}{2}}{(1-t)^2}\right)
=
\sum_{s=0}^{\infty}
\sqrt{\frac{(s+\nu)!(s+\mu+\nu)!}{s!(s+\mu)!}}
\frac{\mu!}{(\mu+\nu)!}
t^s d^{s+\frac{\mu+\nu}{2}}_{MM'}(\beta) \tag*{(737)} ξ M M ′ ( 1 − t ) 1 − μ − ν ( sin 2 β ) μ ( cos 2 β ) ν 2 F 1 ( 2 μ + ν + 1 , 2 μ + ν + 2 ; 1 + μ ; − ( 1 − t ) 2 4 t sin 2 2 β ) = s = 0 ∑ ∞ s ! ( s + μ )! ( s + ν )! ( s + μ + ν )! ( μ + ν )! μ ! t s d M M ′ s + 2 μ + ν ( β ) (737)
ξ M M ′ ( sin β 2 ) μ ( cos β 2 ) ν 2 F 1 ( λ , μ + ν + 1 − λ ; 1 + μ ; 1 − t − R 2 ) 2 F 1 ( λ , μ + ν + 1 − λ ; 1 + ν ; 1 + t − R 2 ) = ∑ s = 0 ∞ ( μ + ν + 1 − λ ) s λ s μ ! ν ! s ! ( s + μ + ν ) ! ( s + μ ) ! ( s + ν ) ! t s d M M ′ s + μ + ν 2 ( β ) . (738) \xi_{MM'}\left(\sin\frac{\beta}{2}\right)^\mu\left(\cos\frac{\beta}{2}\right)^\nu{}_2F_1\left(\lambda,\mu+\nu+1-\lambda;1+\mu;\frac{1-t-R}{2}\right){}_2F_1\left(\lambda,\mu+\nu+1-\lambda;1+\nu;\frac{1+t-R}{2}\right)
= \sum_{s=0}^{\infty}\frac{(\mu+\nu+1-\lambda)_s\lambda_s\mu!\nu!}{\sqrt{s!(s+\mu+\nu)!(s+\mu)!(s+\nu)!}}t^s d^{s+\frac{\mu+\nu}{2}}_{MM'}(\beta). \tag*{(738)} ξ M M ′ ( sin 2 β ) μ ( cos 2 β ) ν 2 F 1 ( λ , μ + ν + 1 − λ ; 1 + μ ; 2 1 − t − R ) 2 F 1 ( λ , μ + ν + 1 − λ ; 1 + ν ; 2 1 + t − R ) = s = 0 ∑ ∞ s ! ( s + μ + ν )! ( s + μ )! ( s + ν )! ( μ + ν + 1 − λ ) s λ s μ ! ν ! t s d M M ′ s + 2 μ + ν ( β ) . (738)
In Eq. (738) λ \lambda λ is an arbitrary integer.
Characters χ J ( R ) \chi^J(R) χ J ( R ) of irreducible representations of rotation group
Definition
The trace of the finite rotation matrix in the JM-representation
χ J ( R ) = ∑ M = − J J D M M J ( R ) , (739) \chi^J(R)=\sum_{M=-J}^{J}D^J_{MM}(R), \tag*{(739)} χ J ( R ) = M = − J ∑ J D M M J ( R ) , (739)
is called the characteristic function, or simply, the character of the irreducible representation of rank J J J .
In contrast to D M M ′ J ( R ) D^J_{MM'}(R) D M M ′ J ( R ) , the function χ J ( R ) \chi^J(R) χ J ( R ) is invariant under rotations of the coordinate systems. Explicit forms of χ J ( R ) \chi^J(R) χ J ( R ) are simpler when R R R is specified by ω , Θ , Φ \omega,\Theta,\Phi ω , Θ , Φ rather than by α , β , γ \alpha,\beta,\gamma α , β , γ (§ 9.9.2 ). With variables ω , Θ , Φ \omega,\Theta,\Phi ω , Θ , Φ in use, χ J ( R ) \chi^J(R) χ J ( R ) is entirely determined by the rotation angles ω \omega ω and is independent of the rotation axis n ( θ , Φ ) \mathbf{n}(\theta,\Phi) n ( θ , Φ ) :
χ J ( R ) = χ J ( ω ) . (740) \chi^J(R)=\chi^J(\omega). \tag*{(740)} χ J ( R ) = χ J ( ω ) . (740)
Explicit Forms
(a) Trigonometric formulas
χ J ( ω ) = sin [ ( 2 J + 1 ) ω 2 ] sin ω 2 , (741) \chi^J(\omega)=\frac{\sin\left[(2J+1)\frac{\omega}{2}\right]}{\sin\frac{\omega}{2}}, \tag*{(741)} χ J ( ω ) = sin 2 ω sin [ ( 2 J + 1 ) 2 ω ] , (741)
χ J ( ω ) = ∑ M = − J J e − i M ω = ∑ M = − J J cos M ω , (742) \chi^J(\omega)=\sum_{M=-J}^{J}e^{-iM\omega}=\sum_{M=-J}^{J}\cos M\omega, \tag*{(742)} χ J ( ω ) = M = − J ∑ J e − i M ω = M = − J ∑ J cos M ω , (742)
χ J ( ω ) = ∑ n = 0 [ J ] ( − 1 ) n ( 2 J − n ) ! ( 2 J − 2 n ) ! n ! ( 2 cos ω 2 ) 2 J − 2 n , (743) \chi^J(\omega)=\sum_{n=0}^{[J]}(-1)^n\frac{(2J-n)!}{(2J-2n)!n!}\left(2\cos\frac{\omega}{2}\right)^{2J-2n}, \tag*{(743)} χ J ( ω ) = n = 0 ∑ [ J ] ( − 1 ) n ( 2 J − 2 n )! n ! ( 2 J − n )! ( 2 cos 2 ω ) 2 J − 2 n , (743)
χ J ( ω ) = ∑ n = 0 [ J ] ( − 1 ) n ( 2 J + 1 ) ! ( 2 n + 1 ) ! ( 2 J − 2 n ) ! ( cos ω 2 ) 2 J − 2 n ( sin ω 2 ) 2 n , (744) \chi^J(\omega)=\sum_{n=0}^{[J]}(-1)^n\frac{(2J+1)!}{(2n+1)!(2J-2n)!}\left(\cos\frac{\omega}{2}\right)^{2J-2n}\left(\sin\frac{\omega}{2}\right)^{2n}, \tag*{(744)} χ J ( ω ) = n = 0 ∑ [ J ] ( − 1 ) n ( 2 n + 1 )! ( 2 J − 2 n )! ( 2 J + 1 )! ( cos 2 ω ) 2 J − 2 n ( sin 2 ω ) 2 n , (744)
χ J ( ω ) = 1 2 J + 1 d d ( cos ω 2 ) cos [ ( 2 J + 1 ) ω 2 ] , (745) \chi^J(\omega)=\frac{1}{2J+1}\frac{d}{d\left(\cos\frac{\omega}{2}\right)}\cos\left[(2J+1)\frac{\omega}{2}\right], \tag*{(745)} χ J ( ω ) = 2 J + 1 1 d ( cos 2 ω ) d cos [ ( 2 J + 1 ) 2 ω ] , (745)
χ J ( ω ) = ( 2 J + 1 ) ! 2 2 J ( 4 J + 1 ) ! sin ω 2 [ d d ( cos ω 2 ) ] 2 J ( sin ω 2 ) 4 J + 1 . (746) \chi^J(\omega)=\frac{(2J+1)!2^{2J}}{(4J+1)!\sin\frac{\omega}{2}}\left[\frac{d}{d\left(\cos\frac{\omega}{2}\right)}\right]^{2J}\left(\sin\frac{\omega}{2}\right)^{4J+1}. \tag*{(746)} χ J ( ω ) = ( 4 J + 1 )! sin 2 ω ( 2 J + 1 )! 2 2 J [ d ( cos 2 ω ) d ] 2 J ( sin 2 ω ) 4 J + 1 . (746)
(b) Relations to hypergeometric functions
χ J ( ω ) = ( 2 J + 1 ) F ( − J , J + 1 ; 3 / 2 ; sin 2 ω 2 ) , (747) \chi^J(\omega)=(2J+1)F\left(-J,J+1;3/2;\sin^2\frac{\omega}{2}\right), \tag*{(747)} χ J ( ω ) = ( 2 J + 1 ) F ( − J , J + 1 ; 3/2 ; sin 2 2 ω ) , (747)
χ J ( ω ) = ( 2 J + 1 ) F ( − 2 J , 2 ( J + 1 ) ; 3 / 2 ; sin 2 ω 4 ) . (748) \chi^J(\omega)=(2J+1)F\left(-2J,2(J+1);3/2;\sin^2\frac{\omega}{4}\right). \tag*{(748)} χ J ( ω ) = ( 2 J + 1 ) F ( − 2 J , 2 ( J + 1 ) ; 3/2 ; sin 2 4 ω ) . (748)
(c) Relation to the Chebyshev polynomials of the second kind
χ J ( ω ) = U 2 J ( cos ω 2 ) . (749) \chi^J(\omega)=U_{2J}\left(\cos\frac{\omega}{2}\right). \tag*{(749)} χ J ( ω ) = U 2 J ( cos 2 ω ) . (749)
(d) Relation to the Gegenbauer polynomials
χ J ( ω ) = C 2 J 1 ( cos ω 2 ) . (750) \chi^J(\omega)=C^1_{2J}\left(\cos\frac{\omega}{2}\right). \tag*{(750)} χ J ( ω ) = C 2 J 1 ( cos 2 ω ) . (750)
Relation to the Jacobi polynomials
χ J ( ω ) = ( 4 J + 2 ) ! ! 2 ( 4 J + 1 ) ! ! P 2 J ( 1 2 , 1 2 ) ( cos ω 2 ) (751) \chi^J(\omega)=\frac{(4J+2)!!}{2(4J+1)!!}P_{2J}^{(\frac{1}{2},\frac{1}{2})}\left(\cos\frac{\omega}{2}\right) \tag*{(751)} χ J ( ω ) = 2 ( 4 J + 1 )!! ( 4 J + 2 )!! P 2 J ( 2 1 , 2 1 ) ( cos 2 ω ) (751)
Integral representation
χ J ( ω ) = 2 J + 1 2 ∫ − 1 1 ( cos ω 2 + i x sin ω 2 ) 2 J d x . (752) \chi^J(\omega)=\frac{2J+1}{2}\int_{-1}^{1}\left(\cos\frac{\omega}{2}+ix\sin\frac{\omega}{2}\right)^{2J}\,dx. \tag*{(752)} χ J ( ω ) = 2 2 J + 1 ∫ − 1 1 ( cos 2 ω + i x sin 2 ω ) 2 J d x . (752)
To express χ J ( R ) \chi^J(R) χ J ( R ) in terms of the Euler angles α \alpha α , β \beta β , γ \gamma γ one can use the relation
cos ω 2 = cos β 2 cos α + γ 2 . (753) \cos\frac{\omega}{2}=\cos\frac{\beta}{2}\cos\frac{\alpha+\gamma}{2}. \tag*{(753)} cos 2 ω = cos 2 β cos 2 α + γ . (753)
Principal Properties
In contrast to D M M ′ J ( R ) D^J_{MM'}(R) D M M ′ J ( R ) , the characters χ J ( R ) \chi^J(R) χ J ( R ) are real.
( χ J ( R ) ) ∗ = χ J ( R ) . (754) \left(\chi^J(R)\right)^*=\chi^J(R). \tag*{(754)} ( χ J ( R ) ) ∗ = χ J ( R ) . (754)
The characters which correspond to direct and inverse rotations, R R R and R − 1 R^{-1} R − 1 , are equal
χ J ( R − 1 ) = χ J ( R ) . (755) \chi^J\left(R^{-1}\right)=\chi^J(R). \tag*{(755)} χ J ( R − 1 ) = χ J ( R ) . (755)
χ J ( R ) \chi^J(R) χ J ( R ) is invariant with respect to coordinate rotation and inversion
χ J ( U R U − 1 ) = χ J ( R ) , (756) \chi^J\left(URU^{-1}\right)=\chi^J(R), \tag*{(756)} χ J ( U R U − 1 ) = χ J ( R ) , (756)
where U U U is any orthogonal transformation of the coordinate system.
The characters which depend on the combined rotations R 1 R 2 … R n R_1R_2\ldots R_n R 1 R 2 … R n do not change under cyclic permutations of the rotations
χ J ( R 1 R 2 … R n ) = χ J ( R 2 … R n R 1 ) . (757) \chi^J(R_1R_2\ldots R_n)=\chi^J(R_2\ldots R_nR_1). \tag*{(757)} χ J ( R 1 R 2 … R n ) = χ J ( R 2 … R n R 1 ) . (757)
In particular,
χ J ( R 1 R 2 ) = χ J ( R 2 R 1 ) , (758) \chi^J(R_1R_2)=\chi^J(R_2R_1), \tag*{(758)} χ J ( R 1 R 2 ) = χ J ( R 2 R 1 ) , (758)
in spite of the non-commutativity of R 1 R_1 R 1 and R 2 R_2 R 2 .
Products of the characters may be expanded in a Clebsch-Gordan series
χ J 1 ( R ) χ J 2 ( R ) = ∑ J { J 1 J 2 J } χ J ( R ) (759) \chi^{J_1}(R)\chi^{J_2}(R)=\sum_J\{J_1J_2J\}\chi^J(R) \tag*{(759)} χ J 1 ( R ) χ J 2 ( R ) = J ∑ { J 1 J 2 J } χ J ( R ) (759)
where { J 1 J 2 J 3 } = 1 \{J_1J_2J_3\}=1 { J 1 J 2 J 3 } = 1 , if the triangle inequality (§ 17.1.2 ) is satisfied, and { J 1 J 2 J 3 } = 0 \{J_1J_2J_3\}=0 { J 1 J 2 J 3 } = 0 otherwise.
The transformation J = J ~ = − J − 1 J=\widetilde{J}=-J-1 J = J = − J − 1 reverses the character sign
χ J ~ ( R ) = − χ J ( R ) (760) \chi^{\widetilde{J}}(R)=-\chi^J(R) \tag*{(760)} χ J ( R ) = − χ J ( R ) (760)
The function χ J ( ω ) \chi^J(\omega) χ J ( ω ) is even and periodic
χ J ( − ω ) = χ J ( ω ) , χ J ( ω + 4 π ) = χ J ( ω ) , χ J ( ω + 2 π ) = ( − 1 ) 2 J χ J ( ω ) . (761) \begin{aligned}
\chi^J(-\omega)&=\chi^J(\omega),\\
\chi^J(\omega+4\pi)&=\chi^J(\omega),\\
\chi^J(\omega+2\pi)&=(-1)^{2J}\chi^J(\omega). \tag*{(761)}
\end{aligned} χ J ( − ω ) χ J ( ω + 4 π ) χ J ( ω + 2 π ) = χ J ( ω ) , = χ J ( ω ) , = ( − 1 ) 2 J χ J ( ω ) . (761)
The addition theorem
The character which depends on the combined rotation R 1 R 2 R_1R_2 R 1 R 2 may be represented as a superposition of products of the generalized characters which depend on R 1 R_1 R 1 and R 2 R_2 R 2 (Sec. 4.15):
χ J ( ω ) = ∑ λ = 0 2 J ( − 1 ) λ 2 λ + 1 2 J + 1 χ λ ( ω 1 ) χ λ ( ω 2 ) P λ ( cos Θ 12 ) , (762) \chi^J(\omega)=\sum_{\lambda=0}^{2J}(-1)^\lambda\frac{2\lambda+1}{2J+1}\chi^\lambda(\omega_1)\chi^\lambda(\omega_2)P_\lambda(\cos\Theta_{12}), \tag*{(762)} χ J ( ω ) = λ = 0 ∑ 2 J ( − 1 ) λ 2 J + 1 2 λ + 1 χ λ ( ω 1 ) χ λ ( ω 2 ) P λ ( cos Θ 12 ) , (762)
where ω 1 \omega_1 ω 1 , ω 2 \omega_2 ω 2 and ω \omega ω are the rotation angles corresponding to R 1 R_1 R 1 , R 2 R_2 R 2 and R = R 1 R 2 R=R_1R_2 R = R 1 R 2 , respectively. These angles are mutually related by means of Eqs. (204) ; Θ 12 \Theta_{12} Θ 12 is the angle between the rotation axes of R 1 R_1 R 1 and R 2 R_2 R 2
cos Θ 12 = ( n 1 ⋅ n 2 ) = cos Θ 1 cos Θ 2 + sin Θ 1 sin Θ 2 cos ( Φ 1 − Φ 2 ) . (763) \cos\Theta_{12}=(\boldsymbol{n}_1\cdot\boldsymbol{n}_2)=\cos\Theta_1\cos\Theta_2+\sin\Theta_1\sin\Theta_2\cos(\Phi_1-\Phi_2). \tag*{(763)} cos Θ 12 = ( n 1 ⋅ n 2 ) = cos Θ 1 cos Θ 2 + sin Θ 1 sin Θ 2 cos ( Φ 1 − Φ 2 ) . (763)
Differential Equation
The character χ J ( ω ) \chi^J(\omega) χ J ( ω ) satisfies the equation
d 2 d ω 2 χ J ( ω ) + cot ω 2 d d ω χ J ( ω ) + J ( J + 1 ) χ J ( ω ) = 0. (764) \frac{d^2}{d\omega^2}\chi^J(\omega)+\cot\frac{\omega}{2}\frac{d}{d\omega}\chi^J(\omega)+J(J+1)\chi^J(\omega)=0. \tag*{(764)} d ω 2 d 2 χ J ( ω ) + cot 2 ω d ω d χ J ( ω ) + J ( J + 1 ) χ J ( ω ) = 0. (764)
and the boundary conditions
χ J ( 0 ) = 2 J + 1 , \chi^J(0)=2J+1, χ J ( 0 ) = 2 J + 1 ,
χ J ( ω ± 4 π n ) = χ J ( ω ) , where n is integer . (765) \chi^J(\omega\pm4\pi n)=\chi^J(\omega),\qquad\text{where }n\text{ is integer}. \tag*{(765)} χ J ( ω ± 4 π n ) = χ J ( ω ) , where n is integer . (765)
Differential Relations
d d ω χ J ( ω ) = − J ( J + 1 ) χ 1 J ( ω ) ; (766) \frac{d}{d\omega}\chi^J(\omega)=-\sqrt{J(J+1)}\chi^J_1(\omega); \tag*{(766)} d ω d χ J ( ω ) = − J ( J + 1 ) χ 1 J ( ω ) ; (766)
( d d cos ω 2 ) k χ J ( ω ) = 1 2 J + 1 ( 2 J + k + 1 ) ! ( 2 J − k ) ! χ k J ( ω ) ( sin ω 2 ) k , (767) \left(\frac{d}{d\cos\frac{\omega}{2}}\right)^k\chi^J(\omega)=\frac{1}{\sqrt{2J+1}}\sqrt{\frac{(2J+k+1)!}{(2J-k)!}}\frac{\chi^J_k(\omega)}{\left(\sin\frac{\omega}{2}\right)^k}, \tag*{(767)} ( d cos 2 ω d ) k χ J ( ω ) = 2 J + 1 1 ( 2 J − k )! ( 2 J + k + 1 )! ( sin 2 ω ) k χ k J ( ω ) , (767)
where χ k J ( ω ) \chi^J_k(\omega) χ k J ( ω ) is the generalized character (Sec. 4.15:
sin ω 2 d d ω χ J ( ω ) = J cos ω 2 χ J ( ω ) − ( J + 1 2 ) χ J − 1 2 ( ω ) = ( J + 1 2 ) χ J + 1 2 ( ω ) − ( J + 1 ) cos ω 2 χ J ( ω ) . (768) \sin\frac{\omega}{2}\frac{d}{d\omega}\chi^J(\omega)=J\cos\frac{\omega}{2}\chi^J(\omega)-\left(J+\frac{1}{2}\right)\chi^{J-\frac{1}{2}}(\omega)=\left(J+\frac{1}{2}\right)\chi^{J+\frac{1}{2}}(\omega)-(J+1)\cos\frac{\omega}{2}\chi^J(\omega). \tag*{(768)} sin 2 ω d ω d χ J ( ω ) = J cos 2 ω χ J ( ω ) − ( J + 2 1 ) χ J − 2 1 ( ω ) = ( J + 2 1 ) χ J + 2 1 ( ω ) − ( J + 1 ) cos 2 ω χ J ( ω ) . (768)
Algebraic Relations
χ J + 1 2 ( ω ) = 2 cos ω 2 χ J ( ω ) − χ J − 1 2 ( ω ) , (769) \chi^{J+\frac{1}{2}}(\omega)=2\cos\frac{\omega}{2}\chi^J(\omega)-\chi^{J-\frac{1}{2}}(\omega), \tag*{(769)} χ J + 2 1 ( ω ) = 2 cos 2 ω χ J ( ω ) − χ J − 2 1 ( ω ) , (769)
χ J 1 ( ω ) − χ J 2 ( ω ) = 2 χ J 1 − J 2 − 1 2 ( ω ) cos [ ( J 1 + J 2 + 1 ) ω 2 ] , (770) \chi^{J_1}(\omega)-\chi^{J_2}(\omega)=2\chi^{\frac{J_1-J_2-1}{2}}(\omega)\cos\left[(J_1+J_2+1)\frac{\omega}{2}\right], \tag*{(770)} χ J 1 ( ω ) − χ J 2 ( ω ) = 2 χ 2 J 1 − J 2 − 1 ( ω ) cos [ ( J 1 + J 2 + 1 ) 2 ω ] , (770)
χ J 1 ( ω ) + χ J 2 ( ω ) = 2 χ J 1 + J 2 2 ( ω ) cos [ ( J 1 − J 2 ) ω 2 ] . (771) \chi^{J_1}(\omega)+\chi^{J_2}(\omega)=2\chi^{\frac{J_1+J_2}{2}}(\omega)\cos\left[(J_1-J_2)\frac{\omega}{2}\right]. \tag*{(771)} χ J 1 ( ω ) + χ J 2 ( ω ) = 2 χ 2 J 1 + J 2 ( ω ) cos [ ( J 1 − J 2 ) 2 ω ] . (771)
Equations (4.14.32) and (4.14.33) are valid, if J 1 + J 2 J_1+J_2 J 1 + J 2 is integer.
χ J − 1 2 ( ω ) = 2 cos J ω 2 χ J − 1 2 ( ω ) , (772) \chi^{J-\frac{1}{2}}(\omega)=2\cos\frac{J\omega}{2}\chi^{\frac{J-1}{2}}(\omega), \tag*{(772)} χ J − 2 1 ( ω ) = 2 cos 2 J ω χ 2 J − 1 ( ω ) , (772)
where J J J is integer and positive.
− 2 sin 2 ω 2 χ J 1 ( ω ) χ J 2 ( ω ) = cos [ ( J 1 + J 2 + 1 ) ω ] − cos [ ( J 1 − J 2 ) ω ] . (773) -2\sin^2\frac{\omega}{2}\chi^{J_1}(\omega)\chi^{J_2}(\omega)=\cos[(J_1+J_2+1)\omega]-\cos[(J_1-J_2)\omega]. \tag*{(773)} − 2 sin 2 2 ω χ J 1 ( ω ) χ J 2 ( ω ) = cos [( J 1 + J 2 + 1 ) ω ] − cos [( J 1 − J 2 ) ω ] . (773)
In particular,
2 sin 2 ω 2 [ χ J ( ω ) ] 2 = 1 − cos [ ( 2 J + 1 ) ω ] (774) 2\sin^2\frac{\omega}{2}\left[\chi^J(\omega)\right]^2=1-\cos[(2J+1)\omega] \tag*{(774)} 2 sin 2 2 ω [ χ J ( ω ) ] 2 = 1 − cos [( 2 J + 1 ) ω ] (774)
χ J ( ω ) = 2 2 J ∏ k = 1 2 J sin ( ω 2 + k π 2 J + 1 ) . (775) \chi^J(\omega)=2^{2J}\prod_{k=1}^{2J}\sin\left(\frac{\omega}{2}+\frac{k\pi}{2J+1}\right). \tag*{(775)} χ J ( ω ) = 2 2 J k = 1 ∏ 2 J sin ( 2 ω + 2 J + 1 k π ) . (775)
Orthogonality and Completeness
The collection of characters χ J ( ω ) \chi^J(\omega) χ J ( ω ) with different J J J constitute a complete set of orthogonal functions of argument ω \omega ω in the domain 0 ≤ ω < 2 π 0\leq\omega<2\pi 0 ≤ ω < 2 π .
The orthogonality and normalization condition for these functions reads
∫ 0 2 π χ J 1 ( ω ) χ J 2 ( ω ) sin 2 ω 2 d ω = π δ J 1 J 2 . (776) \int_0^{2\pi}\chi^{J_1}(\omega)\chi^{J_2}(\omega)\sin^2\frac{\omega}{2}\,d\omega=\pi\delta_{J_1J_2}. \tag*{(776)} ∫ 0 2 π χ J 1 ( ω ) χ J 2 ( ω ) sin 2 2 ω d ω = π δ J 1 J 2 . (776)
The completeness condition has the form
∑ J = 0 , 1 2 , 1 , … ∞ χ J ( ω 1 ) χ J ( ω 2 ) = π δ ( ω 1 − ω 2 ) sin 2 ω 1 2 . (777) \sum_{J=0,\frac{1}{2},1,\ldots}^{\infty}\chi^J(\omega_1)\chi^J(\omega_2)=\frac{\pi\delta(\omega_1-\omega_2)}{\sin^2\frac{\omega_1}{2}}. \tag*{(777)} J = 0 , 2 1 , 1 , … ∑ ∞ χ J ( ω 1 ) χ J ( ω 2 ) = sin 2 2 ω 1 π δ ( ω 1 − ω 2 ) . (777)
Integrals Involving J()
Integrals Involving χ J ( ω ) \chi^{J}(\omega) χ J ( ω )
∫ 0 2 π d ω sin 2 ω 2 χ J ( ω ) = π δ J 0 , (778) \int_{0}^{2\pi} d\omega\sin^{2}\frac{\omega}{2}\chi^{J}(\omega)=\pi\delta_{J0}, \tag*{(778)} ∫ 0 2 π d ω sin 2 2 ω χ J ( ω ) = π δ J 0 , (778)
∫ 0 2 π d ω sin 2 ω 2 χ J ( 2 ω ) = π ( − 1 ) 2 J , (779) \int_{0}^{2\pi} d\omega\sin^{2}\frac{\omega}{2}\chi^{J}(2\omega)=\pi(-1)^{2J}, \tag*{(779)} ∫ 0 2 π d ω sin 2 2 ω χ J ( 2 ω ) = π ( − 1 ) 2 J , (779)
∫ 0 2 π d ω sin 2 ω 2 cos ω 2 − cos Ω 2 χ J ( ω ) = − 2 π cos [ ( 2 J + 1 ) Ω 2 ] , (780) \int_{0}^{2\pi} d\omega\frac{\sin^{2}\frac{\omega}{2}}{\cos\frac{\omega}{2}-\cos\frac{\Omega}{2}}\chi^{J}(\omega)=-2\pi\cos\left[(2J+1)\frac{\Omega}{2}\right], \tag*{(780)} ∫ 0 2 π d ω cos 2 ω − cos 2 Ω sin 2 2 ω χ J ( ω ) = − 2 π cos [ ( 2 J + 1 ) 2 Ω ] , (780)
In Eq. (780) the Cauchy principal value of the integral is assumed.
∫ 0 2 π d ω sin 2 ω 2 χ J 1 ( ω ) χ J 2 ( ω ) = π δ J 1 J 2 , (781) \int_{0}^{2\pi} d\omega\sin^{2}\frac{\omega}{2}\chi^{J_{1}}(\omega)\chi^{J_{2}}(\omega)=\pi\delta_{J_{1}J_{2}}, \tag*{(781)} ∫ 0 2 π d ω sin 2 2 ω χ J 1 ( ω ) χ J 2 ( ω ) = π δ J 1 J 2 , (781)
∫ 0 2 π d ω sin 2 ω 2 χ J 1 ( ω ) χ J 2 ( ω ) χ J 3 ( ω ) = π { J 1 J 2 J 3 } . (782) \int_{0}^{2\pi} d\omega\sin^{2}\frac{\omega}{2}\chi^{J_{1}}(\omega)\chi^{J_{2}}(\omega)\chi^{J_{3}}(\omega)=\pi\left\{J_{1}J_{2}J_{3}\right\}. \tag*{(782)} ∫ 0 2 π d ω sin 2 2 ω χ J 1 ( ω ) χ J 2 ( ω ) χ J 3 ( ω ) = π { J 1 J 2 J 3 } . (782)
Sums Involving χ J ( ω ) \chi^{J}(\omega) χ J ( ω )
Finite sums
∑ J = J 1 , J 1 + 1 , … J 2 χ J ( ω ) = sin [ ( J 2 + J 1 + 1 ) ω 2 ] sin [ ( J 2 − J 1 + 1 ) ω 2 ] sin 2 ω 2 = χ J 2 + J 1 2 ( ω ) χ J 2 − J 1 2 ( ω ) , (783) \sum_{J=J_{1},J_{1}+1,\ldots}^{J_{2}}\chi^{J}(\omega)=\frac{\sin\left[(J_{2}+J_{1}+1)\frac{\omega}{2}\right]\sin\left[(J_{2}-J_{1}+1)\frac{\omega}{2}\right]}{\sin^{2}\frac{\omega}{2}}=\chi^{\frac{J_{2}+J_{1}}{2}}(\omega)\chi^{\frac{J_{2}-J_{1}}{2}}(\omega), \tag*{(783)} J = J 1 , J 1 + 1 , … ∑ J 2 χ J ( ω ) = sin 2 2 ω sin [ ( J 2 + J 1 + 1 ) 2 ω ] sin [ ( J 2 − J 1 + 1 ) 2 ω ] = χ 2 J 2 + J 1 ( ω ) χ 2 J 2 − J 1 ( ω ) , (783)
∑ J = 0 , 1 , 2 , … J 0 ( 2 J + 1 ) χ J ( ω ) = ( 2 J 0 + 3 ) sin [ ( 2 J 0 + 1 ) ω 2 ] − ( 2 J 0 + 1 ) sin [ ( 2 J 0 + 3 ) ω 2 ] 4 sin 3 ω 2 . (784) \sum_{J=0,1,2,\ldots}^{J_{0}}(2J+1)\chi^{J}(\omega)=\frac{(2J_{0}+3)\sin\left[(2J_{0}+1)\frac{\omega}{2}\right]-(2J_{0}+1)\sin\left[(2J_{0}+3)\frac{\omega}{2}\right]}{4\sin^{3}\frac{\omega}{2}}. \tag*{(784)} J = 0 , 1 , 2 , … ∑ J 0 ( 2 J + 1 ) χ J ( ω ) = 4 sin 3 2 ω ( 2 J 0 + 3 ) sin [ ( 2 J 0 + 1 ) 2 ω ] − ( 2 J 0 + 1 ) sin [ ( 2 J 0 + 3 ) 2 ω ] . (784)
The summation index in Eq. (783) runs over integer values, if J 1 J_{1} J 1 and J 2 J_{2} J 2 are integers, or half-integer values, if J 1 J_{1} J 1 and J 2 J_{2} J 2 are half-integers.
In the equations given below the summation indices run over both integer and half-integer values:
∑ J = J 1 , J 1 + 1 2 , J 1 + 1 , … J 2 χ J ( ω ) = sin [ ( J 2 + J 1 + 1 ) ω 2 ] sin [ ( J 2 − J 1 + 1 2 ) ω 2 ] sin ω 2 sin ω 4 , (785) \sum_{J=J_{1},J_{1}+\frac{1}{2},J_{1}+1,\ldots}^{J_{2}}\chi^{J}(\omega)=\frac{\sin\left[(J_{2}+J_{1}+1)\frac{\omega}{2}\right]\sin\left[(J_{2}-J_{1}+\frac{1}{2})\frac{\omega}{2}\right]}{\sin\frac{\omega}{2}\sin\frac{\omega}{4}}, \tag*{(785)} J = J 1 , J 1 + 2 1 , J 1 + 1 , … ∑ J 2 χ J ( ω ) = sin 2 ω sin 4 ω sin [ ( J 2 + J 1 + 1 ) 2 ω ] sin [ ( J 2 − J 1 + 2 1 ) 2 ω ] , (785)
∑ J = 0 , 1 2 , 1 , … J 0 ( 2 J + 1 ) χ J ( ω ) = ( 2 J 0 + 2 ) sin [ ( 2 J 0 + 1 ) ω 2 ] − ( 2 J 0 + 1 ) sin [ ( 2 J 0 + 2 ) ω 2 ] 4 sin ω 2 sin 2 ω 4 , (786) \sum_{J=0,\frac{1}{2},1,\ldots}^{J_{0}}(2J+1)\chi^{J}(\omega)=\frac{(2J_{0}+2)\sin\left[(2J_{0}+1)\frac{\omega}{2}\right]-(2J_{0}+1)\sin\left[(2J_{0}+2)\frac{\omega}{2}\right]}{4\sin\frac{\omega}{2}\sin^{2}\frac{\omega}{4}}, \tag*{(786)} J = 0 , 2 1 , 1 , … ∑ J 0 ( 2 J + 1 ) χ J ( ω ) = 4 sin 2 ω sin 2 4 ω ( 2 J 0 + 2 ) sin [ ( 2 J 0 + 1 ) 2 ω ] − ( 2 J 0 + 1 ) sin [ ( 2 J 0 + 2 ) 2 ω ] , (786)
∑ J = 0 , 1 2 , 1 , … J 0 [ χ J ( ω ) ] 2 = ( 4 J 0 + 3 ) sin ω 2 − sin [ ( 4 J 0 + 3 ) ω 2 ] 4 sin 3 ω 2 , (787) \sum_{J=0,\frac{1}{2},1,\ldots}^{J_{0}}\left[\chi^{J}(\omega)\right]^{2}=\frac{(4J_{0}+3)\sin\frac{\omega}{2}-\sin\left[(4J_{0}+3)\frac{\omega}{2}\right]}{4\sin^{3}\frac{\omega}{2}}, \tag*{(787)} J = 0 , 2 1 , 1 , … ∑ J 0 [ χ J ( ω ) ] 2 = 4 sin 3 2 ω ( 4 J 0 + 3 ) sin 2 ω − sin [ ( 4 J 0 + 3 ) 2 ω ] , (787)
∑ J = 0 , 1 2 , 1 , … J 0 χ J ( ω ) χ J ( ω ′ ) = χ J 0 + 1 2 ( ω ) χ J 0 ( ω ′ ) − χ J 0 ( ω ) χ J 0 + 1 2 ( ω ′ ) 2 ( cos ω 2 − cos ω ′ 2 ) , (788) \sum_{J=0,\frac{1}{2},1,\ldots}^{J_{0}}\chi^{J}(\omega)\chi^{J}(\omega')=\frac{\chi^{J_{0}+\frac{1}{2}}(\omega)\chi^{J_{0}}(\omega')-\chi^{J_{0}}(\omega)\chi^{J_{0}+\frac{1}{2}}(\omega')}{2\left(\cos\frac{\omega}{2}-\cos\frac{\omega'}{2}\right)}, \tag*{(788)} J = 0 , 2 1 , 1 , … ∑ J 0 χ J ( ω ) χ J ( ω ′ ) = 2 ( cos 2 ω − cos 2 ω ′ ) χ J 0 + 2 1 ( ω ) χ J 0 ( ω ′ ) − χ J 0 ( ω ) χ J 0 + 2 1 ( ω ′ ) , (788)
∑ J = 0 , 1 2 , 1 , … J 0 χ J ( ω ) cos [ ( 2 J + 1 ) ω ′ 2 ] = sin ω 2 − cos [ ( 2 J 0 + 1 ) ω ′ 2 ] sin [ ( J 0 + 1 ) ω ] 2 sin ω 2 ( cos ω ′ 2 − cos ω 2 ) + sin [ ( 2 J 0 + 1 ) ω 2 ] cos [ ( J 0 + 1 ) ω ′ ] 2 sin ω 2 ( cos ω ′ 2 − cos ω 2 ) . (789) \begin{aligned}
\sum_{J=0,\frac{1}{2},1,\ldots}^{J_{0}}\chi^{J}(\omega)\cos\left[(2J+1)\frac{\omega'}{2}\right]
={}&\frac{\sin\frac{\omega}{2}-\cos\left[(2J_{0}+1)\frac{\omega'}{2}\right]\sin[(J_{0}+1)\omega]}{2\sin\frac{\omega}{2}\left(\cos\frac{\omega'}{2}-\cos\frac{\omega}{2}\right)}\\
&+\frac{\sin\left[(2J_{0}+1)\frac{\omega}{2}\right]\cos[(J_{0}+1)\omega']}{2\sin\frac{\omega}{2}\left(\cos\frac{\omega'}{2}-\cos\frac{\omega}{2}\right)}. \tag*{(789)}
\end{aligned} J = 0 , 2 1 , 1 , … ∑ J 0 χ J ( ω ) cos [ ( 2 J + 1 ) 2 ω ′ ] = 2 sin 2 ω ( cos 2 ω ′ − cos 2 ω ) sin 2 ω − cos [ ( 2 J 0 + 1 ) 2 ω ′ ] sin [( J 0 + 1 ) ω ] + 2 sin 2 ω ( cos 2 ω ′ − cos 2 ω ) sin [ ( 2 J 0 + 1 ) 2 ω ] cos [( J 0 + 1 ) ω ′ ] . (789)
Infinite series
∑ J = 0 , 1 2 , 1 , … ∞ χ J ( ω ) = 1 4 sin 2 ω 4 , ( ω ≠ 0 ) , (790) \sum_{J=0,\frac{1}{2},1,\ldots}^{\infty}\chi^{J}(\omega)=\frac{1}{4\sin^{2}\frac{\omega}{4}},\quad(\omega\ne0), \tag*{(790)} J = 0 , 2 1 , 1 , … ∑ ∞ χ J ( ω ) = 4 sin 2 4 ω 1 , ( ω = 0 ) , (790)
∑ J = 0 , 1 2 , 1 , … ∞ ( 2 J + 1 ) χ J ( ω ) = π sin 2 ω 2 δ ( ω ) . (791) \sum_{J=0,\frac{1}{2},1,\ldots}^{\infty}(2J+1)\chi^{J}(\omega)=\frac{\pi}{\sin^{2}\frac{\omega}{2}}\delta(\omega). \tag*{(791)} J = 0 , 2 1 , 1 , … ∑ ∞ ( 2 J + 1 ) χ J ( ω ) = sin 2 2 ω π δ ( ω ) . (791)
Let us introduce the notation
R 2 ≡ 1 − 2 t cos ω 2 + t 2 , where ∣ t ∣ < 1 : (792) R^2 \equiv1-2t\cos\frac{\omega}{2}+t^2,\qquad\text{where } |t|<1: \tag*{(792)} R 2 ≡ 1 − 2 t cos 2 ω + t 2 , where ∣ t ∣ < 1 : (792)
∑ J = 0 , 1 2 , 1 , … ∞ t 2 J χ J ( ω ) = 1 R 2 , (793) \sum_{J=0,\frac12,1,\ldots}^{\infty} t^{2J}\chi^J(\omega)=\frac{1}{R^2}, \tag*{(793)} J = 0 , 2 1 , 1 , … ∑ ∞ t 2 J χ J ( ω ) = R 2 1 , (793)
∑ J = 0 , 1 2 , 1 , … ∞ ( 2 J + 1 ) t 2 J χ J ( ω ) = 1 − t 2 R 4 , (794) \sum_{J=0,\frac12,1,\ldots}^{\infty}(2J+1)t^{2J}\chi^J(\omega)=\frac{1-t^2}{R^4}, \tag*{(794)} J = 0 , 2 1 , 1 , … ∑ ∞ ( 2 J + 1 ) t 2 J χ J ( ω ) = R 4 1 − t 2 , (794)
∑ J = 0 , 1 2 , 1 , … ∞ ( 4 J + 1 ) ! ! ( 4 J + 2 ) ! ! t 2 J χ J ( ω ) = 1 R 2 ( 1 − t cos ω 2 + R ) , (795) \sum_{J=0,\frac12,1,\ldots}^{\infty}\frac{(4J+1)!!}{(4J+2)!!}t^{2J}\chi^J(\omega)=\frac{1}{R\sqrt{2\left(1-t\cos\frac{\omega}{2}+R\right)}}, \tag*{(795)} J = 0 , 2 1 , 1 , … ∑ ∞ ( 4 J + 2 )!! ( 4 J + 1 )!! t 2 J χ J ( ω ) = R 2 ( 1 − t cos 2 ω + R ) 1 , (795)
∑ J = 0 , 1 2 , 1 , … ∞ 1 ( 2 J + 1 ) ! t 2 J + 1 χ J ( ω ) = sin ( t sin ω 2 ) sin ω 2 e t cos ω 2 , (796) \sum_{J=0,\frac12,1,\ldots}^{\infty}\frac{1}{(2J+1)!}t^{2J+1}\chi^J(\omega)=\frac{\sin\left(t\sin\frac{\omega}{2}\right)}{\sin\frac{\omega}{2}}e^{t\cos\frac{\omega}{2}}, \tag*{(796)} J = 0 , 2 1 , 1 , … ∑ ∞ ( 2 J + 1 )! 1 t 2 J + 1 χ J ( ω ) = sin 2 ω sin ( t sin 2 ω ) e t c o s 2 ω , (796)
∑ J = 0 , 1 2 , 1 , … ∞ 1 2 J + 1 t 2 J χ J ( ω ) = 1 2 i t sin ω 2 ln ( 1 − t e − i ω / 2 1 − t e i ω / 2 ) , (797) \sum_{J=0,\frac12,1,\ldots}^{\infty}\frac{1}{2J+1}t^{2J}\chi^J(\omega)=\frac{1}{2it\sin\frac{\omega}{2}}\ln\left(\frac{1-te^{-i\omega/2}}{1-te^{i\omega/2}}\right), \tag*{(797)} J = 0 , 2 1 , 1 , … ∑ ∞ 2 J + 1 1 t 2 J χ J ( ω ) = 2 i t sin 2 ω 1 ln ( 1 − t e iω /2 1 − t e − iω /2 ) , (797)
∑ J = 0 , 1 2 , 1 , … ∞ ∑ γ ( ν ) Γ ( 2 J + ν ) Γ ( ν ) t 2 J χ J ( ω ) ( 2 J + 1 ) ! = ( 1 − t cos ω 2 ) − ν 2 F 1 ( ν 2 , ν + 1 2 ; 3 2 ; ( t sin ω 2 1 − t cos ω 2 ) 2 ) , (798) \sum_{J=0,\frac12,1,\ldots}^{\infty}\sum_{\gamma(\nu)}\frac{\Gamma(2J+\nu)}{\Gamma(\nu)}\frac{t^{2J}\chi^J(\omega)}{(2J+1)!}=\left(1-t\cos\frac{\omega}{2}\right)^{-\nu}\,{}_2F_1\left(\frac{\nu}{2},\frac{\nu+1}{2};\frac{3}{2};\left(\frac{t\sin\frac{\omega}{2}}{1-t\cos\frac{\omega}{2}}\right)^2\right), \tag*{(798)} J = 0 , 2 1 , 1 , … ∑ ∞ γ ( ν ) ∑ Γ ( ν ) Γ ( 2 J + ν ) ( 2 J + 1 )! t 2 J χ J ( ω ) = ( 1 − t cos 2 ω ) − ν 2 F 1 ( 2 ν , 2 ν + 1 ; 2 3 ; ( 1 − t cos 2 ω t sin 2 ω ) 2 ) , (798)
∑ J = 0 , 1 2 , 1 , … ∞ 1 ( 4 J + 2 ) ! ( 4 t ) 2 J χ J ( ω ) = 1 2 0 F 1 ( ; 3 2 ; − t sin 2 ω 4 ) 0 F 1 ( ; 3 2 ; t cos 2 ω 4 ) , (799) \sum_{J=0,\frac12,1,\ldots}^{\infty}\frac{1}{(4J+2)!}(4t)^{2J}\chi^J(\omega)=\frac{1}{2}{}_0F_1\left(; \frac{3}{2};-t\sin^2\frac{\omega}{4}\right){}_0F_1\left(; \frac{3}{2};t\cos^2\frac{\omega}{4}\right), \tag*{(799)} J = 0 , 2 1 , 1 , … ∑ ∞ ( 4 J + 2 )! 1 ( 4 t ) 2 J χ J ( ω ) = 2 1 0 F 1 ( ; 2 3 ; − t sin 2 4 ω ) 0 F 1 ( ; 2 3 ; t cos 2 4 ω ) , (799)
∑ J = 0 , 1 2 , 1 , … ∞ t 2 J χ J ( ω ) χ J ( ω ′ ) = 1 − t 2 ( 1 + t 2 ) 2 − 4 t ( 1 + t 2 ) cos ω 2 cos ω ′ 2 + 2 t 2 ( cos ω + cos ω ′ ) . (800) \sum_{J=0,\frac12,1,\ldots}^{\infty}t^{2J}\chi^J(\omega)\chi^J(\omega')=
\frac{1-t^2}{(1+t^2)^2-4t(1+t^2)\cos\frac{\omega}{2}\cos\frac{\omega'}{2}+2t^2(\cos\omega+\cos\omega')}. \tag*{(800)} J = 0 , 2 1 , 1 , … ∑ ∞ t 2 J χ J ( ω ) χ J ( ω ′ ) = ( 1 + t 2 ) 2 − 4 t ( 1 + t 2 ) cos 2 ω cos 2 ω ′ + 2 t 2 ( cos ω + cos ω ′ ) 1 − t 2 . (800)
χ J ( ω ) \chi^J(\omega) χ J ( ω ) for Particular Values of ω \omega ω
χ J ( 0 ) = 2 J + 1 , (801) \chi^J(0)=2J+1, \tag*{(801)} χ J ( 0 ) = 2 J + 1 , (801)
χ J ( 2 π ) = ( − 1 ) 2 J ( 2 J + 1 ) , (802) \chi^J(2\pi)=(-1)^{2J}(2J+1), \tag*{(802)} χ J ( 2 π ) = ( − 1 ) 2 J ( 2 J + 1 ) , (802)
χ J ( π ) = { 0 if J is half-integer ( − 1 ) J if J is integer (803) \chi^J(\pi)=
\begin{cases}
0 & \text{if } J \text{ is half-integer}\\
(-1)^J & \text{if } J \text{ is integer}
\end{cases} \tag*{(803)} χ J ( π ) = { 0 ( − 1 ) J if J is half-integer if J is integer (803)
χ J ( π 2 ) = { 2 J = 1 / 2 , 9 / 2 , 17 / 2 , … 1 J = 0 , 1 , 4 , 5 , 8 , 9 , … 0 J = 3 / 2 , 7 / 2 , 11 / 2 , 15 / 2 , … − 1 J = 2 , 3 , 6 , 7 , 10 , 11 , … − 2 J = 5 / 2 , 13 / 2 , 21 / 2 , … (804) \chi^J\left(\frac{\pi}{2}\right)=
\begin{cases}
\sqrt{2} & J=1/2,9/2,17/2,\ldots\\
1 & J=0,1,4,5,8,9,\ldots\\
0 & J=3/2,7/2,11/2,15/2,\ldots\\
-1 & J=2,3,6,7,10,11,\ldots\\
-\sqrt{2} & J=5/2,13/2,21/2,\ldots
\end{cases} \tag*{(804)} χ J ( 2 π ) = ⎩ ⎨ ⎧ 2 1 0 − 1 − 2 J = 1/2 , 9/2 , 17/2 , … J = 0 , 1 , 4 , 5 , 8 , 9 , … J = 3/2 , 7/2 , 11/2 , 15/2 , … J = 2 , 3 , 6 , 7 , 10 , 11 , … J = 5/2 , 13/2 , 21/2 , … (804)
Special Cases of χ J ( ω ) \chi^J(\omega) χ J ( ω )
χ 0 ( ω ) = 1 , (805) \chi^0(\omega)=1, \tag*{(805)} χ 0 ( ω ) = 1 , (805)
χ 1 / 2 ( ω ) = 2 cos ω 2 , (806) \chi^{1/2}(\omega)=2\cos\frac{\omega}{2}, \tag*{(806)} χ 1/2 ( ω ) = 2 cos 2 ω , (806)
χ 1 ( ω ) = 4 cos 2 ω 2 − 1 , (807) \chi^1(\omega)=4\cos^2\frac{\omega}{2}-1, \tag*{(807)} χ 1 ( ω ) = 4 cos 2 2 ω − 1 , (807)
χ 3 / 2 ( ω ) = 8 cos 3 ω 2 − 4 cos ω 2 , (808) \chi^{3/2}(\omega)=8\cos^3\frac{\omega}{2}-4\cos\frac{\omega}{2}, \tag*{(808)} χ 3/2 ( ω ) = 8 cos 3 2 ω − 4 cos 2 ω , (808)
χ 2 ( ω ) = 16 cos 4 ω 2 − 12 cos 2 ω 2 + 1 , (809) \chi^2(\omega)=16\cos^4\frac{\omega}{2}-12\cos^2\frac{\omega}{2}+1, \tag*{(809)} χ 2 ( ω ) = 16 cos 4 2 ω − 12 cos 2 2 ω + 1 , (809)
χ 5 / 2 ( ω ) = 32 cos 5 ω 2 − 32 cos 3 ω 2 + 6 cos ω 2 . (810) \chi^{5/2}(\omega)=32\cos^5\frac{\omega}{2}-32\cos^3\frac{\omega}{2}+6\cos\frac{\omega}{2}. \tag*{(810)} χ 5/2 ( ω ) = 32 cos 5 2 ω − 32 cos 3 2 ω + 6 cos 2 ω . (810)
Generalized characters, χ λ J ( R ) \chi_{\lambda}^{J}(R) χ λ J ( R ) , of irreducible representations of the rotation group
Definition
Let us introduce the function χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) associated with χ J ( ω ) \chi^{J}(\omega) χ J ( ω ) by the differential relation
χ λ J ( ω ) = 2 J + 1 ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ( sin ω 2 ) λ ( d d cos ω 2 ) λ χ J ( ω ) , (811) \chi_{\lambda}^{J}(\omega)=\sqrt{2J+1}\sqrt{\frac{(2J-\lambda)!}{(2J+\lambda+1)!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}\left(\frac{d}{d\cos\frac{\omega}{2}}\right)^{\lambda}\chi^{J}(\omega), \tag*{(811)} χ λ J ( ω ) = 2 J + 1 ( 2 J + λ + 1 )! ( 2 J − λ )! ( sin 2 ω ) λ ( d cos 2 ω d ) λ χ J ( ω ) , (811)
where λ \lambda λ is integer, 0 ≤ λ ≤ 2 J 0\leq\lambda\leq2J 0 ≤ λ ≤ 2 J , and χ J ( ω ) \chi^{J}(\omega) χ J ( ω ) is the character of the irreducible representation of rank J J J (Eq. (739) ).
The function χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) will be called the generalized character (of order λ \lambda λ ) of the irreducible representation of rank J J J . Note that the relations between χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) and χ J ( ω ) \chi^{J}(\omega) χ J ( ω ) are similar to those between the associated Legendre functions P λ J ( x ) P_{\lambda}^{J}(x) P λ J ( x ) and the Legendre polynomials P l ( x ) P_l(x) P l ( x ) .
At λ = 0 \lambda=0 λ = 0 one has χ λ J ( ω ) = χ J ( ω ) \chi_{\lambda}^{J}(\omega)=\chi^{J}(\omega) χ λ J ( ω ) = χ J ( ω ) .
Explicit Forms
(a) Trigonometric series
χ λ J ( ω ) = i λ ∑ M e − i M ω C J M J M λ 0 , (812) \chi_{\lambda}^{J}(\omega)=i^{\lambda}\sum_{M}e^{-iM\omega}C_{JM\,J M\,\lambda0}, \tag*{(812)} χ λ J ( ω ) = i λ M ∑ e − i M ω C J M J M λ 0 , (812)
χ λ J ( ω ) = ( sin ω 2 ) λ 2 J + 1 ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! 2 λ ∑ s = 0 ∣ J − λ / 2 ∣ ( − 1 ) s ( 2 J − s ) ! s ! ( 2 J − λ − 2 s ) ! ( 2 cos ω 2 ) 2 J − λ − 2 s , (813) \chi_{\lambda}^{J}(\omega)=\left(\sin\frac{\omega}{2}\right)^{\lambda}\sqrt{2J+1}\sqrt{\frac{(2J-\lambda)!}{(2J+\lambda+1)!}}2^{\lambda}\sum_{s=0}^{\lvert J-\lambda/2\rvert}\frac{(-1)^s(2J-s)!}{s!(2J-\lambda-2s)!}\left(2\cos\frac{\omega}{2}\right)^{2J-\lambda-2s}, \tag*{(813)} χ λ J ( ω ) = ( sin 2 ω ) λ 2 J + 1 ( 2 J + λ + 1 )! ( 2 J − λ )! 2 λ s = 0 ∑ ∣ J − λ /2 ∣ s ! ( 2 J − λ − 2 s )! ( − 1 ) s ( 2 J − s )! ( 2 cos 2 ω ) 2 J − λ − 2 s , (813)
χ λ J ( ω ) = ( 2 sin ω 2 ) λ λ ! 2 J + 1 ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ∑ s = 0 2 J − λ ( λ + s ) ! ( 2 J − s ) ! s ! ( 2 J − λ − s ) ! cos [ ( 2 J − λ − 2 s ) ω 2 ] . (814) \chi_{\lambda}^{J}(\omega)=\frac{\left(2\sin\frac{\omega}{2}\right)^{\lambda}}{\lambda!}\sqrt{2J+1}\sqrt{\frac{(2J-\lambda)!}{(2J+\lambda+1)!}}\sum_{s=0}^{2J-\lambda}\frac{(\lambda+s)!(2J-s)!}{s!(2J-\lambda-s)!}\cos\left[(2J-\lambda-2s)\frac{\omega}{2}\right]. \tag*{(814)} χ λ J ( ω ) = λ ! ( 2 sin 2 ω ) λ 2 J + 1 ( 2 J + λ + 1 )! ( 2 J − λ )! s = 0 ∑ 2 J − λ s ! ( 2 J − λ − s )! ( λ + s )! ( 2 J − s )! cos [ ( 2 J − λ − 2 s ) 2 ω ] . (814)
(b) Differential form
χ λ J ( ω ) = 1 2 J + 1 ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ( sin ω 2 ) λ ( d d cos ω 2 ) λ + 1 cos [ ( 2 J + 1 ) ω 2 ] . (815) \chi_{\lambda}^{J}(\omega)=\frac{1}{\sqrt{2J+1}}\sqrt{\frac{(2J-\lambda)!}{(2J+\lambda+1)!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}\left(\frac{d}{d\cos\frac{\omega}{2}}\right)^{\lambda+1}\cos\left[(2J+1)\frac{\omega}{2}\right]. \tag*{(815)} χ λ J ( ω ) = 2 J + 1 1 ( 2 J + λ + 1 )! ( 2 J − λ )! ( sin 2 ω ) λ ( d cos 2 ω d ) λ + 1 cos [ ( 2 J + 1 ) 2 ω ] . (815)
See also Eq. (811) .
(c) Relation between χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) and the Gegenbauer polynomials
χ λ J ( ω ) = ( 2 λ ) ! ! 2 J + 1 ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ( sin ω 2 ) λ C 2 J − λ λ + 1 ( cos ω 2 ) . (816) \chi_{\lambda}^{J}(\omega)=(2\lambda)!!\sqrt{2J+1}\sqrt{\frac{(2J-\lambda)!}{(2J+\lambda+1)!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}C_{2J-\lambda}^{\lambda+1}\left(\cos\frac{\omega}{2}\right). \tag*{(816)} χ λ J ( ω ) = ( 2 λ )!! 2 J + 1 ( 2 J + λ + 1 )! ( 2 J − λ )! ( sin 2 ω ) λ C 2 J − λ λ + 1 ( cos 2 ω ) . (816)
(d) Relations of χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) to the Jacobi polynomials
χ λ J ( ω ) = 2 J + 1 ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ( 4 J + 1 ) ! ! 2 2 J − λ ( sin ω 2 ) λ P 2 J − λ ( λ + 1 2 , λ + 1 2 ) ( cos ω 2 ) , (817) \chi_{\lambda}^{J}(\omega)=\frac{\sqrt{2J+1}\sqrt{(2J-\lambda)!(2J+\lambda+1)!}}{(4J+1)!!}2^{2J-\lambda}\left(\sin\frac{\omega}{2}\right)^{\lambda}P_{2J-\lambda}^{(\lambda+\frac{1}{2},\lambda+\frac{1}{2})}\left(\cos\frac{\omega}{2}\right), \tag*{(817)} χ λ J ( ω ) = ( 4 J + 1 )!! 2 J + 1 ( 2 J − λ )! ( 2 J + λ + 1 )! 2 2 J − λ ( sin 2 ω ) λ P 2 J − λ ( λ + 2 1 , λ + 2 1 ) ( cos 2 ω ) , (817)
χ λ J ( ω ) = { 2 J + 1 ( 2 J + λ ) ! ! ( 2 J − λ ) ! ! ( 2 J + λ + 1 ) ! ! ( 2 J − λ − 1 ) ! ! ( sin ω 2 ) λ P J − λ / 2 ( λ + 1 2 , − 1 2 ) ( cos ω ) , 2 J − λ is even , 2 J + 1 ( 2 J − λ − 1 ) ! ! ( 2 J + λ + 1 ) ! ! ( 2 J − λ ) ! ! ( 2 J + λ ) ! ! ( cos ω 2 ) ( sin ω 2 ) λ P J − λ + 1 2 ( λ + 1 2 , 1 2 ) ( cos ω ) , 2 J − λ is odd . (818) \chi_{\lambda}^{J}(\omega)=
\begin{cases}
\sqrt{2J+1}\sqrt{\dfrac{(2J+\lambda)!!(2J-\lambda)!!}{(2J+\lambda+1)!!(2J-\lambda-1)!!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}P_{J-\lambda/2}^{(\lambda+\frac{1}{2},-\frac{1}{2})}(\cos\omega), & 2J-\lambda\text{ is even},\\
\sqrt{2J+1}\sqrt{\dfrac{(2J-\lambda-1)!!(2J+\lambda+1)!!}{(2J-\lambda)!!(2J+\lambda)!!}}\left(\cos\frac{\omega}{2}\right)\left(\sin\frac{\omega}{2}\right)^{\lambda}P_{J-\frac{\lambda+1}{2}}^{(\lambda+\frac{1}{2},\frac{1}{2})}(\cos\omega), & 2J-\lambda\text{ is odd}.
\end{cases} \tag*{(818)} χ λ J ( ω ) = ⎩ ⎨ ⎧ 2 J + 1 ( 2 J + λ + 1 )!! ( 2 J − λ − 1 )!! ( 2 J + λ )!! ( 2 J − λ )!! ( sin 2 ω ) λ P J − λ /2 ( λ + 2 1 , − 2 1 ) ( cos ω ) , 2 J + 1 ( 2 J − λ )!! ( 2 J + λ )!! ( 2 J − λ − 1 )!! ( 2 J + λ + 1 )!! ( cos 2 ω ) ( sin 2 ω ) λ P J − 2 λ + 1 ( λ + 2 1 , 2 1 ) ( cos ω ) , 2 J − λ is even , 2 J − λ is odd . (818)
(e) Relations of χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) to the Hypergeometric functions
χ λ J ( ω ) = 2 J + 1 ( 2 λ + 1 ) ! ! ( 2 J + λ + 1 ) ! ( 2 J − λ ) ! ( sin ω 2 ) λ F ( − 2 J + λ , 2 J + λ + 2 ; λ + 3 2 ; sin 2 ω 4 ) (819) \chi_{\lambda}^{J}(\omega)=\frac{\sqrt{2J+1}}{(2\lambda+1)!!}\sqrt{\frac{(2J+\lambda+1)!}{(2J-\lambda)!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}F\left(-2J+\lambda,2J+\lambda+2;\lambda+\frac{3}{2};\sin^{2}\frac{\omega}{4}\right) \tag*{(819)} χ λ J ( ω ) = ( 2 λ + 1 )!! 2 J + 1 ( 2 J − λ )! ( 2 J + λ + 1 )! ( sin 2 ω ) λ F ( − 2 J + λ , 2 J + λ + 2 ; λ + 2 3 ; sin 2 4 ω ) (819)
χ λ J ( ω ) = ( − 1 ) 2 J − λ ( 2 λ ) ! ! 2 J + 1 ( 2 λ + 1 ) ! ( 2 J + λ + 1 ) ! ( 2 J − λ ) ! ( sin ω 2 ) λ F ( − 2 J + λ , 2 J + λ + 2 ; λ + 3 2 ; cos 2 ω 4 ) (820) \chi_{\lambda}^{J}(\omega)=\frac{(-1)^{2J-\lambda}(2\lambda)!!\sqrt{2J+1}}{(2\lambda+1)!}\sqrt{\frac{(2J+\lambda+1)!}{(2J-\lambda)!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}F\left(-2J+\lambda,2J+\lambda+2;\lambda+\frac{3}{2};\cos^{2}\frac{\omega}{4}\right) \tag*{(820)} χ λ J ( ω ) = ( 2 λ + 1 )! ( − 1 ) 2 J − λ ( 2 λ )!! 2 J + 1 ( 2 J − λ )! ( 2 J + λ + 1 )! ( sin 2 ω ) λ F ( − 2 J + λ , 2 J + λ + 2 ; λ + 2 3 ; cos 2 4 ω ) (820)
χ λ J ( ω ) = ( − 1 ) 2 J − λ 2 J + 1 2 4 J − λ ( 2 J ) ! ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ( sin ω 2 ) λ ( sin ω 4 ) 4 J − 2 λ F ( − 2 J + λ , − 2 J − 1 2 ; − 4 J − 1 ; 1 sin 2 ω 2 ) (821) \chi_{\lambda}^{J}(\omega)=\frac{(-1)^{2J-\lambda}\sqrt{2J+1}2^{4J-\lambda}(2J)!}{\sqrt{(2J-\lambda)!(2J+\lambda+1)!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}\left(\sin\frac{\omega}{4}\right)^{4J-2\lambda}F\left(-2J+\lambda,-2J-\frac{1}{2};-4J-1;\frac{1}{\sin^{2}\frac{\omega}{2}}\right) \tag*{(821)} χ λ J ( ω ) = ( 2 J − λ )! ( 2 J + λ + 1 )! ( − 1 ) 2 J − λ 2 J + 1 2 4 J − λ ( 2 J )! ( sin 2 ω ) λ ( sin 4 ω ) 4 J − 2 λ F ( − 2 J + λ , − 2 J − 2 1 ; − 4 J − 1 ; sin 2 2 ω 1 ) (821)
χ λ J ( ω ) = 2 J + 1 ( 4 J ) ! ! ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ( cos ω 2 ) 2 J − λ ( sin ω 2 ) λ F ( − J + 1 2 , − 2 J + λ + 1 2 ; − 2 J ; 1 cos 2 ω 2 ) (822) \chi_{\lambda}^{J}(\omega)=\frac{\sqrt{2J+1}(4J)!!}{\sqrt{(2J-\lambda)!(2J+\lambda+1)!}}\left(\cos\frac{\omega}{2}\right)^{2J-\lambda}\left(\sin\frac{\omega}{2}\right)^{\lambda}F\left(-J+\frac{1}{2},\frac{-2J+\lambda+1}{2};-2J;\frac{1}{\cos^{2}\frac{\omega}{2}}\right) \tag*{(822)} χ λ J ( ω ) = ( 2 J − λ )! ( 2 J + λ + 1 )! 2 J + 1 ( 4 J )!! ( cos 2 ω ) 2 J − λ ( sin 2 ω ) λ F ( − J + 2 1 , 2 − 2 J + λ + 1 ; − 2 J ; cos 2 2 ω 1 ) (822)
χ λ J ( ω ) = 2 J + 1 ( 2 λ + 1 ) ! ! ( 2 J + λ + 1 ) ! ( 2 J − λ ) ! ( sin ω 2 ) λ ( cos ω 4 ) 4 J − 2 λ F ( − 2 J + λ , − 2 J − 1 2 ; λ + 3 2 ; − tan 2 ω 4 ) (823) \chi_{\lambda}^{J}(\omega)=\frac{\sqrt{2J+1}}{(2\lambda+1)!!}\sqrt{\frac{(2J+\lambda+1)!}{(2J-\lambda)!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}\left(\cos\frac{\omega}{4}\right)^{4J-2\lambda}F\left(-2J+\lambda,-2J-\frac{1}{2};\lambda+\frac{3}{2};-\tan^{2}\frac{\omega}{4}\right) \tag*{(823)} χ λ J ( ω ) = ( 2 λ + 1 )!! 2 J + 1 ( 2 J − λ )! ( 2 J + λ + 1 )! ( sin 2 ω ) λ ( cos 4 ω ) 4 J − 2 λ F ( − 2 J + λ , − 2 J − 2 1 ; λ + 2 3 ; − tan 2 4 ω ) (823)
χ λ J ( ω ) = 2 J + 1 ( 2 J ) ! i λ ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ( e − i ω 2 − e i ω 2 ) λ e ± i ω 2 ( 2 J − λ ) F ( − 2 J + λ , λ + 1 ; − 2 J ; e ∓ i ω ) (824) \chi_{\lambda}^{J}(\omega)=\frac{\sqrt{2J+1}(2J)!i^{\lambda}}{\sqrt{(2J-\lambda)!(2J+\lambda+1)!}}\left(e^{-i\frac{\omega}{2}}-e^{i\frac{\omega}{2}}\right)^{\lambda}e^{\pm i\frac{\omega}{2}(2J-\lambda)}F\left(-2J+\lambda,\lambda+1;-2J;e^{\mp i\omega}\right) \tag*{(824)} χ λ J ( ω ) = ( 2 J − λ )! ( 2 J + λ + 1 )! 2 J + 1 ( 2 J )! i λ ( e − i 2 ω − e i 2 ω ) λ e ± i 2 ω ( 2 J − λ ) F ( − 2 J + λ , λ + 1 ; − 2 J ; e ∓ iω ) (824)
χ λ J ( ω ) = { ( − 1 ) J + 3 λ 2 2 J + 1 ( 2 J − λ − 1 ) ! ! ( 2 J + λ ) ! ! ( 2 J − λ ) ! ! ( 2 J + λ + 1 ) ! ! ( sin ω 2 ) λ F ( − J + λ 2 , J + 1 + λ 2 ; 1 2 ; cos 2 ω 2 ) , if 2 J − λ is even , ( − 1 ) J + 3 λ − 1 2 2 J + 1 ( 2 J − λ ) ! ! ( 2 J + λ + 1 ) ! ! ( 2 J − λ − 1 ) ! ! ( 2 J + λ ) ! ! cos ω 2 ( sin ω 2 ) λ F ( − J + λ + 1 2 , J + 1 + λ + 1 2 ; 3 2 ; cos 2 ω 2 ) , if 2 J − λ is odd . (825) \chi_{\lambda}^{J}(\omega)=
\begin{cases}
(-1)^{J+\frac{3\lambda}{2}}\sqrt{2J+1}\sqrt{\frac{(2J-\lambda-1)!!(2J+\lambda)!!}{(2J-\lambda)!!(2J+\lambda+1)!!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}F\left(-J+\frac{\lambda}{2},J+1+\frac{\lambda}{2};\frac{1}{2};\cos^{2}\frac{\omega}{2}\right), & \text{if }2J-\lambda\text{ is even},\\
(-1)^{J+\frac{3\lambda-1}{2}}\sqrt{2J+1}\sqrt{\frac{(2J-\lambda)!!(2J+\lambda+1)!!}{(2J-\lambda-1)!!(2J+\lambda)!!}}\cos\frac{\omega}{2}\left(\sin\frac{\omega}{2}\right)^{\lambda}F\left(-J+\frac{\lambda+1}{2},J+1+\frac{\lambda+1}{2};\frac{3}{2};\cos^{2}\frac{\omega}{2}\right), & \text{if }2J-\lambda\text{ is odd}.
\end{cases} \tag*{(825)} χ λ J ( ω ) = ⎩ ⎨ ⎧ ( − 1 ) J + 2 3 λ 2 J + 1 ( 2 J − λ )!! ( 2 J + λ + 1 )!! ( 2 J − λ − 1 )!! ( 2 J + λ )!! ( sin 2 ω ) λ F ( − J + 2 λ , J + 1 + 2 λ ; 2 1 ; cos 2 2 ω ) , ( − 1 ) J + 2 3 λ − 1 2 J + 1 ( 2 J − λ − 1 )!! ( 2 J + λ )!! ( 2 J − λ )!! ( 2 J + λ + 1 )!! cos 2 ω ( sin 2 ω ) λ F ( − J + 2 λ + 1 , J + 1 + 2 λ + 1 ; 2 3 ; cos 2 2 ω ) , if 2 J − λ is even , if 2 J − λ is odd . (825)
χ λ J ( ω ) = { 2 J + 1 ( 2 λ + 1 ) ! ! ( 2 J + λ + 1 ) ! ( 2 J − λ ) ! ( sin ω 2 ) λ F ( − J + λ 2 , J + 1 + λ 2 ; λ + 3 2 ; sin 2 ω 2 ) , if 2 J − λ is even , 2 J + 1 ( 2 λ + 1 ) ! ! ( 2 J + λ + 1 ) ! ( 2 J − λ ) ! cos ω 2 ( sin ω 2 ) λ F ( − J + λ + 1 2 , J + 1 + λ + 1 2 ; λ + 3 2 ; sin 2 ω 2 ) , if 2 J − λ is odd . (826) \chi_{\lambda}^{J}(\omega)=
\begin{cases}
\frac{\sqrt{2J+1}}{(2\lambda+1)!!}\sqrt{\frac{(2J+\lambda+1)!}{(2J-\lambda)!}}\left(\sin\frac{\omega}{2}\right)^{\lambda}F\left(-J+\frac{\lambda}{2},J+1+\frac{\lambda}{2};\lambda+\frac{3}{2};\sin^{2}\frac{\omega}{2}\right), & \text{if }2J-\lambda\text{ is even},\\
\frac{\sqrt{2J+1}}{(2\lambda+1)!!}\sqrt{\frac{(2J+\lambda+1)!}{(2J-\lambda)!}}\cos\frac{\omega}{2}\left(\sin\frac{\omega}{2}\right)^{\lambda}F\left(-J+\frac{\lambda+1}{2},J+1+\frac{\lambda+1}{2};\lambda+\frac{3}{2};\sin^{2}\frac{\omega}{2}\right), & \text{if }2J-\lambda\text{ is odd}.
\end{cases} \tag*{(826)} χ λ J ( ω ) = ⎩ ⎨ ⎧ ( 2 λ + 1 )!! 2 J + 1 ( 2 J − λ )! ( 2 J + λ + 1 )! ( sin 2 ω ) λ F ( − J + 2 λ , J + 1 + 2 λ ; λ + 2 3 ; sin 2 2 ω ) , ( 2 λ + 1 )!! 2 J + 1 ( 2 J − λ )! ( 2 J + λ + 1 )! cos 2 ω ( sin 2 ω ) λ F ( − J + 2 λ + 1 , J + 1 + 2 λ + 1 ; λ + 2 3 ; sin 2 2 ω ) , if 2 J − λ is even , if 2 J − λ is odd . (826)
(f) Integral representations
χ λ J ( ω ) = 1 2 λ ! ( 2 J + 1 ) ( 2 J + λ + 1 ) ! ( 2 J − λ ) ! 1 ( sin ω 2 ) λ + 1 ∫ 0 ω cos [ ( 2 J + 1 ) ψ 2 ] ( cos ψ 2 − cos ω 2 ) λ d ψ (827) \chi_{\lambda}^{J}(\omega)=\frac{1}{2\lambda!}\sqrt{\frac{(2J+1)(2J+\lambda+1)!}{(2J-\lambda)!}}\frac{1}{\left(\sin\frac{\omega}{2}\right)^{\lambda+1}}\int_{0}^{\omega}\cos\left[(2J+1)\frac{\psi}{2}\right]\left(\cos\frac{\psi}{2}-\cos\frac{\omega}{2}\right)^{\lambda}d\psi \tag*{(827)} χ λ J ( ω ) = 2 λ ! 1 ( 2 J − λ )! ( 2 J + 1 ) ( 2 J + λ + 1 )! ( sin 2 ω ) λ + 1 1 ∫ 0 ω cos [ ( 2 J + 1 ) 2 ψ ] ( cos 2 ψ − cos 2 ω ) λ d ψ (827)
χ λ J ( ω ) = ( − i ) λ ( 2 J + 1 ) ( 2 J + λ + 1 ) ! ( 2 J − λ ) ! 2 ( 2 J ) ! ∫ − 1 + 1 P λ ( x ) [ cos ω 2 + i x sin ω 2 ] 2 J d x (828) \chi_{\lambda}^{J}(\omega)=(-i)^{\lambda}\frac{\sqrt{(2J+1)(2J+\lambda+1)!(2J-\lambda)!}}{2(2J)!}\int_{-1}^{+1}P_{\lambda}(x)\left[\cos\frac{\omega}{2}+ix\sin\frac{\omega}{2}\right]^{2J}dx \tag*{(828)} χ λ J ( ω ) = ( − i ) λ 2 ( 2 J )! ( 2 J + 1 ) ( 2 J + λ + 1 )! ( 2 J − λ )! ∫ − 1 + 1 P λ ( x ) [ cos 2 ω + i x sin 2 ω ] 2 J d x (828)
Principal Properties
(a) Symmetries
χ λ J ∗ ( ω ) = χ λ J ( ω ) = ( − 1 ) λ χ λ J ( − ω ) (829) \chi_{\lambda}^{J*}(\omega)=\chi_{\lambda}^{J}(\omega)=(-1)^{\lambda}\chi_{\lambda}^{J}(-\omega) \tag*{(829)} χ λ J ∗ ( ω ) = χ λ J ( ω ) = ( − 1 ) λ χ λ J ( − ω ) (829)
χ λ J ( ω + 2 π n ) = ( − 1 ) 2 J n χ λ J ( ω ) , χ λ J ( 2 π − ω ) = ( − 1 ) 2 J − λ χ λ J ( ω ) . (830) \chi_{\lambda}^{J}(\omega+2\pi n)=(-1)^{2Jn}\chi_{\lambda}^{J}(\omega),\qquad\chi_{\lambda}^{J}(2\pi-\omega)=(-1)^{2J-\lambda}\chi_{\lambda}^{J}(\omega). \tag*{(830)} χ λ J ( ω + 2 π n ) = ( − 1 ) 2 J n χ λ J ( ω ) , χ λ J ( 2 π − ω ) = ( − 1 ) 2 J − λ χ λ J ( ω ) . (830)
(b) Particular values of ω \omega ω
χ λ J ( 0 ) = ( 2 J + 1 ) δ λ 0 , χ λ J ( 2 π ) = ( − 1 ) 2 J ( 2 J + 1 ) δ λ 0 , \chi_{\lambda}^{J}(0)=(2J+1)\delta_{\lambda0},\qquad\chi_{\lambda}^{J}(2\pi)=(-1)^{2J}(2J+1)\delta_{\lambda0}, χ λ J ( 0 ) = ( 2 J + 1 ) δ λ 0 , χ λ J ( 2 π ) = ( − 1 ) 2 J ( 2 J + 1 ) δ λ 0 ,
χ λ J ( π ) = { ( − 1 ) J − λ / 2 ( 2 J + λ ) ! ! ( 2 J − λ − 1 ) ! ! ( 2 J − λ ) ! ! ( 2 J + λ + 1 ) ! ! if 2 J − λ is even , 0 if 2 J − λ is odd . (831) \chi_{\lambda}^{J}(\pi)=
\begin{cases}
(-1)^{J-\lambda/2}\sqrt{\dfrac{(2J+\lambda)!!(2J-\lambda-1)!!}{(2J-\lambda)!!(2J+\lambda+1)!!}} & \text{if }2J-\lambda\text{ is even},\\
0 & \text{if }2J-\lambda\text{ is odd}.
\end{cases} \tag*{(831)} χ λ J ( π ) = ⎩ ⎨ ⎧ ( − 1 ) J − λ /2 ( 2 J − λ )!! ( 2 J + λ + 1 )!! ( 2 J + λ )!! ( 2 J − λ − 1 )!! 0 if 2 J − λ is even , if 2 J − λ is odd . (831)
(c) Recursion relations
2 d d ω χ λ J ( ω ) = λ 2 λ + 1 ( 2 J + 1 ) 2 − λ 2 χ λ − 1 J ( ω ) − λ + 1 2 λ + 1 ( 2 J + 1 ) 2 − ( λ + 1 ) 2 χ λ + 1 J ( ω ) , (832) 2\frac{d}{d\omega}\chi_{\lambda}^{J}(\omega)=\frac{\lambda}{2\lambda+1}\sqrt{(2J+1)^2-\lambda^2}\chi_{\lambda-1}^{J}(\omega)-\frac{\lambda+1}{2\lambda+1}\sqrt{(2J+1)^2-(\lambda+1)^2}\chi_{\lambda+1}^{J}(\omega), \tag*{(832)} 2 d ω d χ λ J ( ω ) = 2 λ + 1 λ ( 2 J + 1 ) 2 − λ 2 χ λ − 1 J ( ω ) − 2 λ + 1 λ + 1 ( 2 J + 1 ) 2 − ( λ + 1 ) 2 χ λ + 1 J ( ω ) , (832)
( 2 λ + 1 ) cot ω 2 χ λ J ( ω ) = ( 2 J + 1 ) 2 − λ 2 χ λ − 1 J ( ω ) + ( 2 J + 1 ) 2 − ( λ + 1 ) 2 χ λ + 1 J ( ω ) . (833) (2\lambda+1)\cot\frac{\omega}{2}\chi_{\lambda}^{J}(\omega)=\sqrt{(2J+1)^2-\lambda^2}\chi_{\lambda-1}^{J}(\omega)+\sqrt{(2J+1)^2-(\lambda+1)^2}\chi_{\lambda+1}^{J}(\omega). \tag*{(833)} ( 2 λ + 1 ) cot 2 ω χ λ J ( ω ) = ( 2 J + 1 ) 2 − λ 2 χ λ − 1 J ( ω ) + ( 2 J + 1 ) 2 − ( λ + 1 ) 2 χ λ + 1 J ( ω ) . (833)
(d) Asymptotics
If ω → 0 \omega\to0 ω → 0 and J → ∞ J\to\infty J → ∞ , while J ω ≡ x < ∞ J\omega\equiv x<\infty J ω ≡ x < ∞ , then
lim J → ∞ ω → 0 1 2 J + 1 χ λ J ( ω ) = j λ ( x ) , (834) \lim_{\substack{J\to\infty\\\omega\to0}}\frac{1}{2J+1}\chi_{\lambda}^{J}(\omega)=j_{\lambda}(x), \tag*{(834)} J → ∞ ω → 0 lim 2 J + 1 1 χ λ J ( ω ) = j λ ( x ) , (834)
where j λ ( x ) j_{\lambda}(x) j λ ( x ) is a spherical Bessel function. If ω → 0 \omega\to0 ω → 0 and J J J is fixed, then
χ λ J ( ω ) ≈ ( ω 2 ) λ 1 ( 2 λ + 1 ) ! ! ( 2 J + 1 ) ( 2 J + λ + 1 ) ! ( 2 J − λ ) ! . (835) \chi_{\lambda}^{J}(\omega)\approx\left(\frac{\omega}{2}\right)^{\lambda}\frac{1}{(2\lambda+1)!!}\sqrt{(2J+1)\frac{(2J+\lambda+1)!}{(2J-\lambda)!}}. \tag*{(835)} χ λ J ( ω ) ≈ ( 2 ω ) λ ( 2 λ + 1 )!! 1 ( 2 J + 1 ) ( 2 J − λ )! ( 2 J + λ + 1 )! . (835)
Differential Equation
The functions χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) are solutions of the linear differential equation
d 2 d ω 2 χ λ J ( ω ) + cot ω 2 d d ω χ λ J ( ω ) + [ J ( J + 1 ) − λ ( λ + 1 ) 4 sin 2 ω 2 ] χ λ J ( ω ) = 0 (836) \frac{d^2}{d\omega^2}\chi_{\lambda}^{J}(\omega)+\cot\frac{\omega}{2}\frac{d}{d\omega}\chi_{\lambda}^{J}(\omega)+\left[J(J+1)-\frac{\lambda(\lambda+1)}{4\sin^2\frac{\omega}{2}}\right]\chi_{\lambda}^{J}(\omega)=0 \tag*{(836)} d ω 2 d 2 χ λ J ( ω ) + cot 2 ω d ω d χ λ J ( ω ) + [ J ( J + 1 ) − 4 sin 2 2 ω λ ( λ + 1 ) ] χ λ J ( ω ) = 0 (836)
with the boundary conditions
χ λ J ( 0 ) = ( 2 J + 1 ) δ λ 0 , χ λ J ( 2 π ) = ( − 1 ) 2 J ( 2 J + 1 ) δ λ 0 . (837) \chi_{\lambda}^{J}(0)=(2J+1)\delta_{\lambda0},\qquad\chi_{\lambda}^{J}(2\pi)=(-1)^{2J}(2J+1)\delta_{\lambda0}. \tag*{(837)} χ λ J ( 0 ) = ( 2 J + 1 ) δ λ 0 , χ λ J ( 2 π ) = ( − 1 ) 2 J ( 2 J + 1 ) δ λ 0 . (837)
Orthogonality and Completeness
The collection of functions χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) with different J ≥ λ / 2 J\ge\lambda/2 J ≥ λ /2 and fixed λ \lambda λ constitutes a complete set of orthogonal functions of argument ω \omega ω defined in the domain 0 ≤ ω < 2 π 0\le\omega<2\pi 0 ≤ ω < 2 π .
These functions are orthogonal,
∫ 0 2 π d ω sin 2 ω 2 χ λ J 1 ( ω ) χ λ J 2 ( ω ) = π δ J 1 J 2 , (838) \int_{0}^{2\pi}d\omega\sin^2\frac{\omega}{2}\chi_{\lambda}^{J_1}(\omega)\chi_{\lambda}^{J_2}(\omega)=\pi\delta_{J_1J_2}, \tag*{(838)} ∫ 0 2 π d ω sin 2 2 ω χ λ J 1 ( ω ) χ λ J 2 ( ω ) = π δ J 1 J 2 , (838)
and the completeness condition reads
∑ J = 0 , 1 2 , 1 , … ∞ χ λ J ( ω 1 ) χ λ J ( ω 2 ) = π δ ( ω 1 − ω 2 ) sin 2 ω 1 2 . (839) \sum_{J=0,\frac{1}{2},1,\ldots}^{\infty}\chi_{\lambda}^{J}(\omega_1)\chi_{\lambda}^{J}(\omega_2)=\frac{\pi\delta(\omega_1-\omega_2)}{\sin^2\frac{\omega_1}{2}}. \tag*{(839)} J = 0 , 2 1 , 1 , … ∑ ∞ χ λ J ( ω 1 ) χ λ J ( ω 2 ) = sin 2 2 ω 1 π δ ( ω 1 − ω 2 ) . (839)
The Addition Theorem for J ( ) J() J ( )
The generalized character χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω ) for the rotation R ( ω ; Θ , Φ ) R(\omega;\Theta,\Phi) R ( ω ; Θ , Φ ) , which resulted from two successive rotations R 1 ( ω 1 ; Θ 1 , Φ 1 ) R_1(\omega_1;\Theta_1,\Phi_1) R 1 ( ω 1 ; Θ 1 , Φ 1 ) and R 2 ( ω 2 ; Θ 2 , Φ 2 ) R_2(\omega_2;\Theta_2,\Phi_2) R 2 ( ω 2 ; Θ 2 , Φ 2 ) may be expanded in terms of χ λ J ( R 1 ) \chi_{\lambda}^{J}(R_1) χ λ J ( R 1 ) and χ λ J ( R 2 ) \chi_{\lambda}^{J}(R_2) χ λ J ( R 2 ) as
χ λ J ( ω ) = ( 2 J − λ ) ! ( 2 J + λ + 1 ) ! ( 2 J + 1 ) ( sin ω 2 sin ω 1 2 sin ω 2 2 sin χ ) λ ∑ l = λ 2 J ( − 1 ) l ( 2 l + 1 ) P l λ ( cos χ ) χ λ J ( ω 1 ) χ λ J ( ω 2 ) , (840) \chi_{\lambda}^{J}(\omega)=\sqrt{\frac{(2J-\lambda)!}{(2J+\lambda+1)!(2J+1)}}\left(\frac{\sin\frac{\omega}{2}}{\sin\frac{\omega_1}{2}\sin\frac{\omega_2}{2}\sin\chi}\right)^{\lambda}\sum_{l=\lambda}^{2J}(-1)^l(2l+1)P_l^{\lambda}(\cos\chi)\chi_{\lambda}^{J}(\omega_1)\chi_{\lambda}^{J}(\omega_2), \tag*{(840)} χ λ J ( ω ) = ( 2 J + λ + 1 )! ( 2 J + 1 ) ( 2 J − λ )! ( sin 2 ω 1 sin 2 ω 2 sin χ sin 2 ω ) λ l = λ ∑ 2 J ( − 1 ) l ( 2 l + 1 ) P l λ ( cos χ ) χ λ J ( ω 1 ) χ λ J ( ω 2 ) , (840)
where P l λ ( cos χ ) P_l^{\lambda}(\cos\chi) P l λ ( cos χ ) is an associated Legendre function, χ \chi χ is an angle between the rotation axes n 1 ( Θ 1 , Φ 1 ) \mathbf{n}_1(\Theta_1,\Phi_1) n 1 ( Θ 1 , Φ 1 ) and n 2 ( Θ 2 , Φ 2 ) \mathbf{n}_2(\Theta_2,\Phi_2) n 2 ( Θ 2 , Φ 2 )
cos χ = ( n 1 ⋅ n 2 ) = cos Θ 1 cos Θ 2 + sin Θ 1 sin Θ 2 cos ( Φ 1 − Φ 2 ) . (841) \cos\chi=(\mathbf{n}_1\cdot\mathbf{n}_2)=\cos\Theta_1\cos\Theta_2+\sin\Theta_1\sin\Theta_2\cos(\Phi_1-\Phi_2). \tag*{(841)} cos χ = ( n 1 ⋅ n 2 ) = cos Θ 1 cos Θ 2 + sin Θ 1 sin Θ 2 cos ( Φ 1 − Φ 2 ) . (841)
The angle of the resulting rotation, ω \omega ω , is determined in terms of ω 1 , ω 2 \omega_1,\omega_2 ω 1 , ω 2 and χ \chi χ by
cos ω 2 = cos ω 1 2 cos ω 2 2 − sin ω 1 2 sin ω 2 2 cos χ . (842) \cos\frac{\omega}{2}=\cos\frac{\omega_1}{2}\cos\frac{\omega_2}{2}-\sin\frac{\omega_1}{2}\sin\frac{\omega_2}{2}\cos\chi. \tag*{(842)} cos 2 ω = cos 2 ω 1 cos 2 ω 2 − sin 2 ω 1 sin 2 ω 2 cos χ . (842)
Special Cases of J() for Particular
Sums and Infinite Series Involving χ λ J ( ω ) \chi_{\lambda}^{J}(\omega) χ λ J ( ω )
(a) Summation over λ \lambda λ
∑ λ = 0 2 J ( ± 1 ) λ i λ 2 λ + 1 2 J + 1 χ λ J ( ω ) C J M λ 0 J M = e ± i M ω , ( ∣ M ∣ ≤ J ) , (843) \sum_{\lambda=0}^{2J}(\pm1)^{\lambda}i^{\lambda}\frac{2\lambda+1}{2J+1}\chi_{\lambda}^{J}(\omega)C_{JM\lambda0}^{J M}=e^{\pm iM\omega},\qquad(|M|\leq J), \tag*{(843)} λ = 0 ∑ 2 J ( ± 1 ) λ i λ 2 J + 1 2 λ + 1 χ λ J ( ω ) C J M λ 0 J M = e ± i M ω , ( ∣ M ∣ ≤ J ) , (843)
∑ λ = 0 , 2 , … λ ≤ 2 J ( − i ) λ 2 λ + 1 2 J + 1 χ λ J ( ω ) C J M λ 0 J M = cos M ω , ( ∣ M ∣ ≤ J ) , (844) \sum_{\substack{\lambda=0,2,\ldots\\ \lambda\leq2J}}(-i)^{\lambda}\frac{2\lambda+1}{2J+1}\chi_{\lambda}^{J}(\omega)C_{JM\lambda0}^{J M}=\cos M\omega,\qquad(|M|\leq J), \tag*{(844)} λ = 0 , 2 , … λ ≤ 2 J ∑ ( − i ) λ 2 J + 1 2 λ + 1 χ λ J ( ω ) C J M λ 0 J M = cos M ω , ( ∣ M ∣ ≤ J ) , (844)
∑ λ = 1 , 3 , … λ ≤ 2 J ( − i ) λ − 1 2 λ + 1 2 J + 1 χ λ J ( ω ) C J M λ 0 J M = sin M ω , ( ∣ M ∣ ≤ J ) . (845) \sum_{\substack{\lambda=1,3,\ldots\\ \lambda\leq2J}}(-i)^{\lambda-1}\frac{2\lambda+1}{2J+1}\chi_{\lambda}^{J}(\omega)C_{JM\lambda0}^{J M}=\sin M\omega,\qquad(|M|\leq J). \tag*{(845)} λ = 1 , 3 , … λ ≤ 2 J ∑ ( − i ) λ − 1 2 J + 1 2 λ + 1 χ λ J ( ω ) C J M λ 0 J M = sin M ω , ( ∣ M ∣ ≤ J ) . (845)
In Eqs. (4.15.33)–(4.15.35) J ≥ ∣ M ∣ J\geq|M| J ≥ ∣ M ∣ is an arbitrary integer or half-integer.
∑ λ = 0 , 2 , … λ ≤ 2 l 2 λ + 1 2 l + 1 ( λ − 1 ) ! ! λ ! ! χ λ l ( ϑ ) C l 0 λ 0 l 0 = P l ( cos ϑ ) , (846) \sum_{\substack{\lambda=0,2,\ldots\\ \lambda\leq2l}}\frac{2\lambda+1}{2l+1}\frac{(\lambda-1)!!}{\lambda!!}\chi_{\lambda}^{l}(\vartheta)C_{l0\lambda0}^{l0}=P_l(\cos\vartheta), \tag*{(846)} λ = 0 , 2 , … λ ≤ 2 l ∑ 2 l + 1 2 λ + 1 λ !! ( λ − 1 )!! χ λ l ( ϑ ) C l 0 λ 0 l 0 = P l ( cos ϑ ) , (846)
∑ λ = ∣ m ∣ 2 l 2 λ + 1 2 l + 1 ( λ + m − 1 ) ! ! ( λ − m − 1 ) ! ! ( λ + m ) ! ! ( λ − m ) ! ! χ λ l ( ϑ ) C l 0 λ m l m = 4 π 2 l + 1 Y l m ( ϑ , 0 ) , (847) \sum_{\lambda=|m|}^{2l}\frac{2\lambda+1}{2l+1}\sqrt{\frac{(\lambda+m-1)!!(\lambda-m-1)!!}{(\lambda+m)!!(\lambda-m)!!}}\chi_{\lambda}^{l}(\vartheta)C_{l0\lambda m}^{lm}=\sqrt{\frac{4\pi}{2l+1}}Y_{lm}(\vartheta,0), \tag*{(847)} λ = ∣ m ∣ ∑ 2 l 2 l + 1 2 λ + 1 ( λ + m )!! ( λ − m )!! ( λ + m − 1 )!! ( λ − m − 1 )!! χ λ l ( ϑ ) C l 0 λm l m = 2 l + 1 4 π Y l m ( ϑ , 0 ) , (847)
( λ + m (\lambda+m ( λ + m is even ),
∑ λ = ∣ M ′ − M ∣ 2 J ( − 1 ) λ 2 λ + 1 2 J + 1 ( λ + M ′ − M − 1 ) ! ! ( λ − M ′ + M − 1 ) ! ! ( λ + M ′ − M ) ! ! ( λ − M ′ + M ) ! ! χ λ J ( β ) C J M ′ λ M ′ − M J M = d M M ′ J ( β ) , (848) \sum_{\lambda=|M'-M|}^{2J}(-1)^{\lambda}\frac{2\lambda+1}{2J+1}\sqrt{\frac{(\lambda+M'-M-1)!!(\lambda-M'+M-1)!!}{(\lambda+M'-M)!!(\lambda-M'+M)!!}}\chi_{\lambda}^{J}(\beta)C_{JM'\lambda\,M'-M}^{JM}=d_{MM'}^{J}(\beta), \tag*{(848)} λ = ∣ M ′ − M ∣ ∑ 2 J ( − 1 ) λ 2 J + 1 2 λ + 1 ( λ + M ′ − M )!! ( λ − M ′ + M )!! ( λ + M ′ − M − 1 )!! ( λ − M ′ + M − 1 )!! χ λ J ( β ) C J M ′ λ M ′ − M J M = d M M ′ J ( β ) , (848)
( λ + M ′ − M (\lambda+M'-M ( λ + M ′ − M is even ).
(b) Summation over J J J
∑ J = ν / 2 ∞ ( 2 J + ν + 1 ) ! ( 2 J − ν ) ! χ ν J ( ω ) 2 J + 1 t 2 J = ( 2 ν ) ! ! ( − t sin ω 2 ) ν R 2 ( ν + 1 ) , (849) \sum_{J=\nu/2}^{\infty}\sqrt{\frac{(2J+\nu+1)!}{(2J-\nu)!}}\frac{\chi_{\nu}^{J}(\omega)}{\sqrt{2J+1}}t^{2J}=(2\nu)!!\frac{\left(-t\sin\frac{\omega}{2}\right)^{\nu}}{R^{2(\nu+1)}}, \tag*{(849)} J = ν /2 ∑ ∞ ( 2 J − ν )! ( 2 J + ν + 1 )! 2 J + 1 χ ν J ( ω ) t 2 J = ( 2 ν )!! R 2 ( ν + 1 ) ( − t sin 2 ω ) ν , (849)
∑ J = ν / 2 ∞ ( 4 J + 1 ) ! ! χ ν J ( ω ) ( t 2 ) 2 J ( 2 J + 1 ) ( 2 J − ν ) ! ( 2 J + ν + 1 ) ! = 2 ( − t sin ω 2 ) ν R ( 1 − t cos ω 2 + R ) ν + 1 2 , (850) \sum_{J=\nu/2}^{\infty}\frac{(4J+1)!!\chi_{\nu}^{J}(\omega)\left(\frac{t}{2}\right)^{2J}}{\sqrt{(2J+1)(2J-\nu)!(2J+\nu+1)!}}=\sqrt{2}\frac{\left(-t\sin\frac{\omega}{2}\right)^{\nu}}{R\left(1-t\cos\frac{\omega}{2}+R\right)^{\nu+\frac{1}{2}}}, \tag*{(850)} J = ν /2 ∑ ∞ ( 2 J + 1 ) ( 2 J − ν )! ( 2 J + ν + 1 )! ( 4 J + 1 )!! χ ν J ( ω ) ( 2 t ) 2 J = 2 R ( 1 − t cos 2 ω + R ) ν + 2 1 ( − t sin 2 ω ) ν , (850)
where R 2 = 1 − 2 t cos ( ω / 2 ) + t 2 R^{2}=1-2t\cos(\omega/2)+t^{2} R 2 = 1 − 2 t cos ( ω /2 ) + t 2 .
∑ J = ν / 2 ∞ i 2 J − ν ( 2 J + 1 ) ( 2 J + ν + 1 ) ! ( 2 J − ν ) ! χ ν J ( ω ) J 2 J + 1 ( y ) = y 2 ( y sin ω 2 ) ν e i y cos ω 2 , (851) \sum_{J=\nu/2}^{\infty}i^{2J-\nu}\sqrt{(2J+1)\frac{(2J+\nu+1)!}{(2J-\nu)!}}\chi_{\nu}^{J}(\omega)J_{2J+1}(y)=\frac{y}{2}\left(y\sin\frac{\omega}{2}\right)^{\nu}e^{iy\cos\frac{\omega}{2}}, \tag*{(851)} J = ν /2 ∑ ∞ i 2 J − ν ( 2 J + 1 ) ( 2 J − ν )! ( 2 J + ν + 1 )! χ ν J ( ω ) J 2 J + 1 ( y ) = 2 y ( y sin 2 ω ) ν e i y c o s 2 ω , (851)
∑ J = ν / 2 ∞ χ ν J ( ω ) t 2 J ( 2 J + 1 ) ( 2 J − ν ) ! ( 2 J + ν + 1 ) ! = j ν ( t sin ω 2 ) e t cos ω 2 , (852) \sum_{J=\nu/2}^{\infty}\frac{\chi_{\nu}^{J}(\omega)t^{2J}}{\sqrt{(2J+1)(2J-\nu)!(2J+\nu+1)!}}=j_{\nu}\left(t\sin\frac{\omega}{2}\right)e^{t\cos\frac{\omega}{2}}, \tag*{(852)} J = ν /2 ∑ ∞ ( 2 J + 1 ) ( 2 J − ν )! ( 2 J + ν + 1 )! χ ν J ( ω ) t 2 J = j ν ( t sin 2 ω ) e t c o s 2 ω , (852)
∑ J = ν / 2 ∞ Γ ( k + 2 J − ν ) Γ ( k ) χ ν J ( ω ) t 2 J ( 2 J + 1 ) ( 2 J − ν ) ! ( 2 J + ν + 1 ) ! = 1 ( 2 ν + 1 ) ! ! ( t sin ω 2 ) ν ( 1 − t cos ω 2 ) k F ( k 2 , k + 1 2 ; ν + 3 2 ; − ( t sin ω 2 1 − t cos ω 2 ) 2 ) . (853) \sum_{J=\nu/2}^{\infty}\frac{\Gamma(k+2J-\nu)}{\Gamma(k)}\frac{\chi_{\nu}^{J}(\omega)t^{2J}}{\sqrt{(2J+1)(2J-\nu)!(2J+\nu+1)!}}
=\frac{1}{(2\nu+1)!!}\frac{\left(t\sin\frac{\omega}{2}\right)^{\nu}}{\left(1-t\cos\frac{\omega}{2}\right)^{k}}F\left(\frac{k}{2},\frac{k+1}{2};\nu+\frac{3}{2};-\left(\frac{t\sin\frac{\omega}{2}}{1-t\cos\frac{\omega}{2}}\right)^{2}\right). \tag*{(853)} J = ν /2 ∑ ∞ Γ ( k ) Γ ( k + 2 J − ν ) ( 2 J + 1 ) ( 2 J − ν )! ( 2 J + ν + 1 )! χ ν J ( ω ) t 2 J = ( 2 ν + 1 )!! 1 ( 1 − t cos 2 ω ) k ( t sin 2 ω ) ν F ( 2 k , 2 k + 1 ; ν + 2 3 ; − ( 1 − t cos 2 ω t sin 2 ω ) 2 ) . (853)
Here k k k is an arbitrary integer.
∑ J = ν / 2 ∞ χ ν J ( ω ) ( 2 t ) 2 J ( 4 J + 1 ) ! ! ( 2 J + 1 ) ( 2 J − ν ) ! ( 2 J + ν + 1 ) ! = ( 2 t sin ω 2 ) ν [ ( 2 ν + 1 ) ! ! ] 2 0 F 1 ( ; ν + 3 2 ; − t sin 2 ω 4 ) 0 F 1 ( ; ν + 3 2 ; t cos 2 ω 4 ) . (854) \sum_{J=\nu/2}^{\infty}\frac{\chi_{\nu}^{J}(\omega)(2t)^{2J}}{(4J+1)!!\sqrt{(2J+1)(2J-\nu)!(2J+\nu+1)!}}
=\frac{\left(2t\sin\frac{\omega}{2}\right)^{\nu}}{[(2\nu+1)!!]^{2}}{}_0F_1\left(; \nu+\frac{3}{2};-t\sin^{2}\frac{\omega}{4}\right){}_0F_1\left(; \nu+\frac{3}{2};t\cos^{2}\frac{\omega}{4}\right). \tag*{(854)} J = ν /2 ∑ ∞ ( 4 J + 1 )!! ( 2 J + 1 ) ( 2 J − ν )! ( 2 J + ν + 1 )! χ ν J ( ω ) ( 2 t ) 2 J = [( 2 ν + 1 )!! ] 2 ( 2 t sin 2 ω ) ν 0 F 1 ( ; ν + 2 3 ; − t sin 2 4 ω ) 0 F 1 ( ; ν + 2 3 ; t cos 2 4 ω ) . (854)
The right-hand sides of Eqs. (4.15.39)–(4.15.44) may be treated as generating functions of χ ν J ( ω ) \chi_{\nu}^{J}(\omega) χ ν J ( ω ) .
Special Cases of J ( ) J() J ( ) for Particular J J J
(a) For λ = 2 J , 2 J − 1 , 2 J − 2 , 2 J − 3 \lambda= 2J, 2J - 1, 2J - 2, 2J - 3 λ = 2 J , 2 J − 1 , 2 J − 2 , 2 J − 3 the function χ λ J ( ω ) \chi_\lambda^J(\omega) χ λ J ( ω ) is given by
χ 2 J J ( ω ) = 2 J + 1 ( 4 J ) ! ! ( 4 J + 1 ) ! ! ( sin ω 2 ) 2 J , (855) \chi_{2J}^J(\omega)=\sqrt{2J+1}\sqrt{\frac{(4J)!!}{(4J+1)!!}}\left(\sin\frac{\omega}{2}\right)^{2J}, \tag*{(855)} χ 2 J J ( ω ) = 2 J + 1 ( 4 J + 1 )!! ( 4 J )!! ( sin 2 ω ) 2 J , (855)
χ 2 J − 1 J ( ω ) = 2 J + 1 ( 4 J ) ! ! ( 4 J − 1 ) ! ! ( sin ω 2 ) 2 J − 1 cos ω 2 , (856) \chi_{2J-1}^J(\omega)=\sqrt{2J+1}\sqrt{\frac{(4J)!!}{(4J-1)!!}}\left(\sin\frac{\omega}{2}\right)^{2J-1}\cos\frac{\omega}{2}, \tag*{(856)} χ 2 J − 1 J ( ω ) = 2 J + 1 ( 4 J − 1 )!! ( 4 J )!! ( sin 2 ω ) 2 J − 1 cos 2 ω , (856)
χ 2 J − 2 J ( ω ) = 2 J + 1 ( 4 J − 2 ) ! ! 2 ( 4 J − 1 ) ! ! ( sin ω 2 ) 2 J − 2 [ 4 J cos 2 ω 2 − 1 ] , (857) \chi_{2J-2}^J(\omega)=\sqrt{2J+1}\sqrt{\frac{(4J-2)!!}{2(4J-1)!!}}\left(\sin\frac{\omega}{2}\right)^{2J-2}\left[4J\cos^2\frac{\omega}{2}-1\right], \tag*{(857)} χ 2 J − 2 J ( ω ) = 2 J + 1 2 ( 4 J − 1 )!! ( 4 J − 2 )!! ( sin 2 ω ) 2 J − 2 [ 4 J cos 2 2 ω − 1 ] , (857)
χ 2 J − 3 J ( ω ) = 2 J + 1 ( 4 J − 2 ) ! ! 6 ( 4 J − 3 ) ! ! ( sin ω 2 ) 2 J − 3 [ 4 J cos 3 ω 2 − 3 cos ω 2 ] . (858) \chi_{2J-3}^J(\omega)=\sqrt{2J+1}\sqrt{\frac{(4J-2)!!}{6(4J-3)!!}}\left(\sin\frac{\omega}{2}\right)^{2J-3}\left[4J\cos^3\frac{\omega}{2}-3\cos\frac{\omega}{2}\right]. \tag*{(858)} χ 2 J − 3 J ( ω ) = 2 J + 1 6 ( 4 J − 3 )!! ( 4 J − 2 )!! ( sin 2 ω ) 2 J − 3 [ 4 J cos 3 2 ω − 3 cos 2 ω ] . (858)
(b) When λ = 0 , 1 , 2 \lambda= 0,1,2 λ = 0 , 1 , 2 one has
χ 0 J ( ω ) = sin ( 2 J + 1 ) ω 2 sin ω 2 = cos J ω − cos ( J + 1 ) ω 1 − cos ω , (859) \chi_0^J(\omega)=\frac{\sin(2J+1)\frac{\omega}{2}}{\sin\frac{\omega}{2}}=\frac{\cos J\omega-\cos(J+1)\omega}{1-\cos\omega}, \tag*{(859)} χ 0 J ( ω ) = sin 2 ω sin ( 2 J + 1 ) 2 ω = 1 − cos ω cos J ω − cos ( J + 1 ) ω , (859)
χ 1 J ( ω ) = − 1 J ( J + 1 ) 2 J cos ( 2 J + 1 ) ω 2 sin ω 2 − sin J ω 2 sin 2 ω 2 = − 1 J ( J + 1 ) J sin ( J + 1 ) ω − ( J + 1 ) sin J ω 1 − cos ω , (860) \chi_1^J(\omega)=\frac{-1}{\sqrt{J(J+1)}}\frac{2J\cos(2J+1)\frac{\omega}{2}\sin\frac{\omega}{2}-\sin J\omega}{2\sin^2\frac{\omega}{2}}=\frac{-1}{\sqrt{J(J+1)}}\frac{J\sin(J+1)\omega-(J+1)\sin J\omega}{1-\cos\omega}, \tag*{(860)} χ 1 J ( ω ) = J ( J + 1 ) − 1 2 sin 2 2 ω 2 J cos ( 2 J + 1 ) 2 ω sin 2 ω − sin J ω = J ( J + 1 ) − 1 1 − cos ω J sin ( J + 1 ) ω − ( J + 1 ) sin J ω , (860)
χ 2 J ( ω ) = 1 J ( J + 1 ) ( 2 J − 1 ) ( 2 J + 3 ) { [ 3 − 2 J ( 2 J − 1 ) sin 2 ω 2 ] sin ( 2 J + 1 ) ω 2 2 sin 3 ω 2 − 3 2 ( 2 J + 1 ) cos J ω sin 2 ω 2 } . (861) \chi_2^J(\omega)=\frac{1}{\sqrt{J(J+1)(2J-1)(2J+3)}}\left\{\frac{\left[3-2J(2J-1)\sin^2\frac{\omega}{2}\right]\sin(2J+1)\frac{\omega}{2}}{2\sin^3\frac{\omega}{2}}-\frac{3}{2}(2J+1)\frac{\cos J\omega}{\sin^2\frac{\omega}{2}}\right\}. \tag*{(861)} χ 2 J ( ω ) = J ( J + 1 ) ( 2 J − 1 ) ( 2 J + 3 ) 1 { 2 sin 3 2 ω [ 3 − 2 J ( 2 J − 1 ) sin 2 2 ω ] sin ( 2 J + 1 ) 2 ω − 2 3 ( 2 J + 1 ) sin 2 2 ω cos J ω } . (861)
(c) χ λ J ( ω ) \chi_\lambda^J(\omega) χ λ J ( ω ) may be expressed in terms of derivatives of χ J ( ω ) \chi^J(\omega) χ J ( ω )
χ 0 J ( ω ) = χ J ( ω ) , \chi_0^J(\omega)=\chi^J(\omega), χ 0 J ( ω ) = χ J ( ω ) ,
χ 1 J ( ω ) = − 1 J ( J + 1 ) d χ J ( ω ) d ω , \chi_1^J(\omega)=-\frac{1}{\sqrt{J(J+1)}}\frac{d\chi^J(\omega)}{d\omega}, χ 1 J ( ω ) = − J ( J + 1 ) 1 d ω d χ J ( ω ) ,
χ 2 J ( ω ) = 1 J ( J + 1 ) ( 2 J − 1 ) ( 2 J + 3 ) [ J ( J + 1 ) χ J ( ω ) + 3 d 2 χ J ( ω ) d ω 2 ] , \chi_2^J(\omega)=\frac{1}{\sqrt{J(J+1)(2J-1)(2J+3)}}\left[J(J+1)\chi^J(\omega)+3\frac{d^2\chi^J(\omega)}{d\omega^2}\right], χ 2 J ( ω ) = J ( J + 1 ) ( 2 J − 1 ) ( 2 J + 3 ) 1 [ J ( J + 1 ) χ J ( ω ) + 3 d ω 2 d 2 χ J ( ω ) ] ,
χ 3 J ( ω ) = − 4 2 J + 1 ( 2 J − 3 ) ! ( 2 J + 4 ) ! { [ 3 J ( J + 1 ) − 1 ] d χ J ( ω ) d ω + 5 d 3 χ J ( ω ) d ω 3 } , \chi_3^J(\omega)=-4\sqrt{2J+1}\sqrt{\frac{(2J-3)!}{(2J+4)!}}\left\{[3J(J+1)-1]\frac{d\chi^J(\omega)}{d\omega}+5\frac{d^3\chi^J(\omega)}{d\omega^3}\right\}, χ 3 J ( ω ) = − 4 2 J + 1 ( 2 J + 4 )! ( 2 J − 3 )! { [ 3 J ( J + 1 ) − 1 ] d ω d χ J ( ω ) + 5 d ω 3 d 3 χ J ( ω ) } ,
χ 4 J ( ω ) = 2 2 J + 1 ( 2 J − 4 ) ! ( 2 J + 5 ) ! { 3 ( J − 1 ) J ( J + 1 ) ( J + 2 ) χ J ( ω ) + 5 [ 6 J ( J + 1 ) − 5 ] d 2 χ J ( ω ) d ω 2 + 35 d 4 χ J ( ω ) d ω 4 } . (862) \chi_4^J(\omega)=2\sqrt{2J+1}\sqrt{\frac{(2J-4)!}{(2J+5)!}}\left\{3(J-1)J(J+1)(J+2)\chi^J(\omega)+5[6J(J+1)-5]\frac{d^2\chi^J(\omega)}{d\omega^2}+35\frac{d^4\chi^J(\omega)}{d\omega^4}\right\}. \tag*{(862)} χ 4 J ( ω ) = 2 2 J + 1 ( 2 J + 5 )! ( 2 J − 4 )! { 3 ( J − 1 ) J ( J + 1 ) ( J + 2 ) χ J ( ω ) + 5 [ 6 J ( J + 1 ) − 5 ] d ω 2 d 2 χ J ( ω ) + 35 d ω 4 d 4 χ J ( ω ) } . (862)
Special Cases of χ λ J ( ω ) \chi^{J}_{\lambda}(\omega) χ λ J ( ω ) for Particular J J J
J = 0 χ 0 0 ( ω ) = 1 ; J = 1 2 χ 0 1 2 ( ω ) = 2 cos ω 2 χ 1 1 2 ( ω ) = 2 3 sin ω 2 J = 1 χ 0 1 ( ω ) = 1 + 2 cos ω χ 1 1 ( ω ) = 2 sin ω χ 2 1 ( ω ) = 2 5 ( 1 − cos ω ) ; J = 3 2 χ 0 3 2 ( ω ) = 4 cos ω cos ω 2 = 4 cos ω 2 ( 2 cos 2 ω 2 − 1 ) , χ 1 3 2 ( ω ) = 4 3 ⋅ 5 ( 2 + 3 cos ω ) sin ω 2 = 4 3 ⋅ 5 sin ω 2 ( 6 cos 2 ω 2 − 1 ) , χ 2 3 2 ( ω ) = 8 5 ( sin ω 2 ) 2 cos ω 2 = 4 5 sin ω 2 sin ω , χ 3 3 2 ( ω ) = 4 5 ⋅ 7 sin ω 2 ( 1 − cos ω ) = 8 5 ⋅ 7 ( sin ω 2 ) 3 ; J = 2 χ 0 2 ( ω ) = 4 cos 2 ω + 2 cos ω − 1 , χ 1 2 ( ω ) = 2 3 sin ω ( 1 + 4 cos ω ) , χ 2 2 ( ω ) = 2 7 ( 3 + cos ω − 4 cos 2 ω ) , χ 3 2 ( ω ) = 2 2 7 sin ω ( 1 − cos ω ) , χ 4 2 ( ω ) = 2 3 2 7 ( 1 − cos ω ) 2 (863) \begin{aligned}
J&=0 && \chi^{0}_{0}(\omega)=1;\\
J&=\tfrac{1}{2} && \chi^{\frac{1}{2}}_{0}(\omega)=2\cos\frac{\omega}{2}\\
&&&\chi^{\frac{1}{2}}_{1}(\omega)=\frac{2}{\sqrt{3}}\sin\frac{\omega}{2}\\
J&=1 && \chi^{1}_{0}(\omega)=1+2\cos\omega\\
&&&\chi^{1}_{1}(\omega)=\sqrt{2}\sin\omega\\
&&&\chi^{1}_{2}(\omega)=\sqrt{\frac{2}{5}}(1-\cos\omega);\\
J&=\tfrac{3}{2} && \chi^{\frac{3}{2}}_{0}(\omega)=4\cos\omega\cos\frac{\omega}{2}=4\cos\frac{\omega}{2}\left(2\cos^{2}\frac{\omega}{2}-1\right),\\
&&&\chi^{\frac{3}{2}}_{1}(\omega)=\frac{4}{\sqrt{3\cdot5}}(2+3\cos\omega)\sin\frac{\omega}{2}
=\frac{4}{\sqrt{3\cdot5}}\sin\frac{\omega}{2}\left(6\cos^{2}\frac{\omega}{2}-1\right),\\
&&&\chi^{\frac{3}{2}}_{2}(\omega)=\frac{8}{\sqrt{5}}\left(\sin\frac{\omega}{2}\right)^{2}\cos\frac{\omega}{2}
=\frac{4}{\sqrt{5}}\sin\frac{\omega}{2}\sin\omega,\\
&&&\chi^{\frac{3}{2}}_{3}(\omega)=\frac{4}{\sqrt{5\cdot7}}\sin\frac{\omega}{2}(1-\cos\omega)
=\frac{8}{\sqrt{5\cdot7}}\left(\sin\frac{\omega}{2}\right)^{3};\\
J&=2 && \chi^{2}_{0}(\omega)=4\cos^{2}\omega+2\cos\omega-1,\\
&&&\chi^{2}_{1}(\omega)=\sqrt{\frac{2}{3}}\sin\omega(1+4\cos\omega),\\
&&&\chi^{2}_{2}(\omega)=\sqrt{\frac{2}{7}}(3+\cos\omega-4\cos2\omega),\\
&&&\chi^{2}_{3}(\omega)=2\sqrt{\frac{2}{7}}\sin\omega(1-\cos\omega),\\
&&&\chi^{2}_{4}(\omega)=\frac{2}{3}\sqrt{\frac{2}{7}}(1-\cos\omega)^{2} \tag*{(863)}
\end{aligned} J J J J J = 0 = 2 1 = 1 = 2 3 = 2 χ 0 0 ( ω ) = 1 ; χ 0 2 1 ( ω ) = 2 cos 2 ω χ 1 2 1 ( ω ) = 3 2 sin 2 ω χ 0 1 ( ω ) = 1 + 2 cos ω χ 1 1 ( ω ) = 2 sin ω χ 2 1 ( ω ) = 5 2 ( 1 − cos ω ) ; χ 0 2 3 ( ω ) = 4 cos ω cos 2 ω = 4 cos 2 ω ( 2 cos 2 2 ω − 1 ) , χ 1 2 3 ( ω ) = 3 ⋅ 5 4 ( 2 + 3 cos ω ) sin 2 ω = 3 ⋅ 5 4 sin 2 ω ( 6 cos 2 2 ω − 1 ) , χ 2 2 3 ( ω ) = 5 8 ( sin 2 ω ) 2 cos 2 ω = 5 4 sin 2 ω sin ω , χ 3 2 3 ( ω ) = 5 ⋅ 7 4 sin 2 ω ( 1 − cos ω ) = 5 ⋅ 7 8 ( sin 2 ω ) 3 ; χ 0 2 ( ω ) = 4 cos 2 ω + 2 cos ω − 1 , χ 1 2 ( ω ) = 3 2 sin ω ( 1 + 4 cos ω ) , χ 2 2 ( ω ) = 7 2 ( 3 + cos ω − 4 cos 2 ω ) , χ 3 2 ( ω ) = 2 7 2 sin ω ( 1 − cos ω ) , χ 4 2 ( ω ) = 3 2 7 2 ( 1 − cos ω ) 2 (863)
D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) for particular values of the arguments
Here k k k , l l l and n n n are integers.
D M M ′ J ( 0 , 0 , 0 ) = δ M M ′ , (864) D^{J}_{MM'}(0,0,0)=\delta_{MM'}, \tag*{(864)} D M M ′ J ( 0 , 0 , 0 ) = δ M M ′ , (864)
D M M ′ J ( α , 0 , γ ) = δ M M ′ e − i M ( α + γ ) , (865) D^{J}_{MM'}(\alpha,0,\gamma)=\delta_{MM'}e^{-iM(\alpha+\gamma)}, \tag*{(865)} D M M ′ J ( α , 0 , γ ) = δ M M ′ e − i M ( α + γ ) , (865)
D M M ′ J ( α , ± 2 n π , γ ) = δ M M ′ ( − 1 ) 2 n J e − i M ( α + γ ) , (866) D^{J}_{MM'}(\alpha,\pm2n\pi,\gamma)=\delta_{MM'}(-1)^{2nJ}e^{-iM(\alpha+\gamma)}, \tag*{(866)} D M M ′ J ( α , ± 2 nπ , γ ) = δ M M ′ ( − 1 ) 2 n J e − i M ( α + γ ) , (866)
D M M ′ J ( α , ± ( 2 n + 1 ) π , γ ) = δ − M M ′ ( − 1 ) ± ( 2 n + 1 ) J + M e − i M ( α − γ ) , (867) D^{J}_{MM'}(\alpha,\pm(2n+1)\pi,\gamma)=\delta_{-MM'}(-1)^{\pm(2n+1)J+M}e^{-iM(\alpha-\gamma)}, \tag*{(867)} D M M ′ J ( α , ± ( 2 n + 1 ) π , γ ) = δ − M M ′ ( − 1 ) ± ( 2 n + 1 ) J + M e − i M ( α − γ ) , (867)
D M M ′ J ( α , π 2 , γ ) = ( − 1 ) M − M ′ e − i α M − i γ M ′ 1 2 J ( J + M ) ! ( J − M ) ! ( J + M ′ ) ! ( J − M ′ ) ! ∑ k ( − 1 ) k ( J + M ′ k ) ( J − M ′ k + M − M ′ ) . (868) D^{J}_{MM'}\left(\alpha,\frac{\pi}{2},\gamma\right)=(-1)^{M-M'}e^{-i\alpha M-i\gamma M'}\frac{1}{2^{J}}\sqrt{\frac{(J+M)!(J-M)!}{(J+M')!(J-M')!}}\sum_{k}(-1)^{k}\binom{J+M'}{k}\binom{J-M'}{k+M-M'}. \tag*{(868)} D M M ′ J ( α , 2 π , γ ) = ( − 1 ) M − M ′ e − i α M − iγ M ′ 2 J 1 ( J + M ′ )! ( J − M ′ )! ( J + M )! ( J − M )! k ∑ ( − 1 ) k ( k J + M ′ ) ( k + M − M ′ J − M ′ ) . (868)
Some particular cases of Eq. (868) are given by
D m 0 l ( α , π 2 , γ ) = ( − 1 ) l + m 2 δ l − m , 2 n ( l − m ) ! ( l + m ) ! 2 l ( l + m 2 ) ! ( l − m 2 ) ! e − i m α , D^{l}_{m0}\left(\alpha,\frac{\pi}{2},\gamma\right)=(-1)^{\frac{l+m}{2}}\delta_{l-m,2n}\frac{\sqrt{(l-m)!(l+m)!}}{2^{l}\left(\frac{l+m}{2}\right)!\left(\frac{l-m}{2}\right)!}e^{-im\alpha}, D m 0 l ( α , 2 π , γ ) = ( − 1 ) 2 l + m δ l − m , 2 n 2 l ( 2 l + m ) ! ( 2 l − m ) ! ( l − m )! ( l + m )! e − im α ,
D 0 m l ( α , π 2 , γ ) = ( − 1 ) l − m 2 δ l − m , 2 n ( l − m ) ! ( l + m ) ! 2 l ( l + m 2 ) ! ( l − m 2 ) ! e − i m γ , (869) D^{l}_{0m}\left(\alpha,\frac{\pi}{2},\gamma\right)=(-1)^{\frac{l-m}{2}}\delta_{l-m,2n}\frac{\sqrt{(l-m)!(l+m)!}}{2^{l}\left(\frac{l+m}{2}\right)!\left(\frac{l-m}{2}\right)!}e^{-im\gamma}, \tag*{(869)} D 0 m l ( α , 2 π , γ ) = ( − 1 ) 2 l − m δ l − m , 2 n 2 l ( 2 l + m ) ! ( 2 l − m ) ! ( l − m )! ( l + m )! e − imγ , (869)
D 00 l ( α , π 2 , γ ) = P l ( 0 ) = δ l , 2 n ( − 1 ) l / 2 ( l − 1 ) ! ! l ! ! , (870) D^{l}_{00}\left(\alpha,\frac{\pi}{2},\gamma\right)=P_{l}(0)=\delta_{l,2n}(-1)^{l/2}\frac{(l-1)!!}{l!!}, \tag*{(870)} D 00 l ( α , 2 π , γ ) = P l ( 0 ) = δ l , 2 n ( − 1 ) l /2 l !! ( l − 1 )!! , (870)
D ± 1 m l ( α , π 2 , γ ) = ( l − m ) ! ( l + m ) ! l ( l + 1 ) { δ l + m , 2 k m ( − 1 ) l − m 2 2 l ( l − m 2 ) ! ( l + m 2 ) ! ∓ δ l + m , 2 k + 1 ( − 1 ) l − m − 1 2 2 l − 1 ( l − m − 1 2 ) ! ( l + m − 1 2 ) ! } e ± i α − i m γ . (871) D^{l}_{\pm1m}\left(\alpha,\frac{\pi}{2},\gamma\right)=\sqrt{\frac{(l-m)!(l+m)!}{l(l+1)}}\left\{\delta_{l+m,2k}\frac{m(-1)^{\frac{l-m}{2}}}{2^{l}\left(\frac{l-m}{2}\right)!\left(\frac{l+m}{2}\right)!}\mp\delta_{l+m,2k+1}\frac{(-1)^{\frac{l-m-1}{2}}}{2^{l-1}\left(\frac{l-m-1}{2}\right)!\left(\frac{l+m-1}{2}\right)!}\right\}e^{\pm i\alpha-im\gamma}. \tag*{(871)} D ± 1 m l ( α , 2 π , γ ) = l ( l + 1 ) ( l − m )! ( l + m )! { δ l + m , 2 k 2 l ( 2 l − m ) ! ( 2 l + m ) ! m ( − 1 ) 2 l − m ∓ δ l + m , 2 k + 1 2 l − 1 ( 2 l − m − 1 ) ! ( 2 l + m − 1 ) ! ( − 1 ) 2 l − m − 1 } e ± i α − imγ . (871)
Squares of the D D D -functions for β = π 2 \beta=\frac{\pi}{2} β = 2 π may be written as
[ D M M ′ J ( α , π 2 , γ ) ] 2 = e − i 2 M α − i 2 M ′ γ ( − 1 ) M − M ′ ∑ l = 0 , 2 , 4 , … ( − 1 ) l / 2 ( l − 1 ) ! ! l ! ! C J M J − M l 0 C J M ′ J − M ′ l 0 . (872) \left[D^{J}_{MM'}\left(\alpha,\frac{\pi}{2},\gamma\right)\right]^{2}=e^{-i2M\alpha-i2M'\gamma}(-1)^{M-M'}\sum_{l=0,2,4,\ldots}(-1)^{l/2}\frac{(l-1)!!}{l!!}C^{l0}_{JM\,J-M}C^{l0}_{JM'\,J-M'}. \tag*{(872)} [ D M M ′ J ( α , 2 π , γ ) ] 2 = e − i 2 M α − i 2 M ′ γ ( − 1 ) M − M ′ l = 0 , 2 , 4 , … ∑ ( − 1 ) l /2 l !! ( l − 1 )!! C J M J − M l 0 C J M ′ J − M ′ l 0 . (872)
Using d M M ′ J ( π / 2 ) d^J_{MM'}(\pi/2) d M M ′ J ( π /2 ) , one can evaluate D M M ′ J ( α , β , γ ) D^J_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) for arbitrary arguments from the relation
D M M ′ J ( α , β , γ ) = ∑ m i D M m 1 J ( α , 0 , 0 ) D m 1 m 2 J ( 0 , π 2 , 0 ) D m 2 m 3 J ( β , 0 , 0 ) D m 3 m 4 J ( 0 , π 2 , 0 ) D m 4 M ′ J ( 0 , 0 , γ ) = ∑ m e − i M α d M m J ( π 2 ) e − i m β d m M ′ J ( π 2 ) e − i M ′ γ (873) \begin{aligned}
D^J_{MM'}(\alpha,\beta,\gamma)
&=\sum_{m_i}D^J_{Mm_1}(\alpha,0,0)D^J_{m_1m_2}\left(0,\frac{\pi}{2},0\right)D^J_{m_2m_3}(\beta,0,0)D^J_{m_3m_4}\left(0,\frac{\pi}{2},0\right)D^J_{m_4M'}(0,0,\gamma) \\
&=\sum_m e^{-iM\alpha}d^J_{Mm}\left(\frac{\pi}{2}\right)e^{-im\beta}d^J_{mM'}\left(\frac{\pi}{2}\right)e^{-iM'\gamma} \tag*{(873)}
\end{aligned} D M M ′ J ( α , β , γ ) = m i ∑ D M m 1 J ( α , 0 , 0 ) D m 1 m 2 J ( 0 , 2 π , 0 ) D m 2 m 3 J ( β , 0 , 0 ) D m 3 m 4 J ( 0 , 2 π , 0 ) D m 4 M ′ J ( 0 , 0 , γ ) = m ∑ e − i M α d M m J ( 2 π ) e − im β d m M ′ J ( 2 π ) e − i M ′ γ (873)
Codes to generate numerical values of D M M ′ J ( 0 , π / 2 , 0 ) = d M M ′ J ( π / 2 ) D^J_{MM'}(0,\pi/2,0)=d^J_{MM'}(\pi/2) D M M ′ J ( 0 , π /2 , 0 ) = d M M ′ J ( π /2 ) are given in Sec. 4.21.
Special cases of D M M ′ J D^J_{MM'} D M M ′ J for particular M M M or M ′ M' M ′
M = 0 M=0 M = 0 and/or M ′ = 0 M'=0 M ′ = 0
D m 0 l ( α , β , γ ) = ( − 1 ) m 4 π 2 l + 1 Y l − m ( β , α ) = 4 π 2 l + 1 Y l m ∗ ( β , α ) , D^l_{m0}(\alpha,\beta,\gamma)=(-1)^m\sqrt{\frac{4\pi}{2l+1}}Y_{l-m}(\beta,\alpha)=\sqrt{\frac{4\pi}{2l+1}}Y^*_{lm}(\beta,\alpha), D m 0 l ( α , β , γ ) = ( − 1 ) m 2 l + 1 4 π Y l − m ( β , α ) = 2 l + 1 4 π Y l m ∗ ( β , α ) ,
D 0 m l ( α , β , γ ) = 4 π 2 l + 1 Y l − m ( β , γ ) = ( − 1 ) m 4 π 2 l + 1 Y l m ∗ ( β , γ ) . (874) D^l_{0m}(\alpha,\beta,\gamma)=\sqrt{\frac{4\pi}{2l+1}}Y_{l-m}(\beta,\gamma)=(-1)^m\sqrt{\frac{4\pi}{2l+1}}Y^*_{lm}(\beta,\gamma). \tag*{(874)} D 0 m l ( α , β , γ ) = 2 l + 1 4 π Y l − m ( β , γ ) = ( − 1 ) m 2 l + 1 4 π Y l m ∗ ( β , γ ) . (874)
In particular,
D 00 l ( α , β , γ ) = P l ( cos β ) , D^l_{00}(\alpha,\beta,\gamma)=P_l(\cos\beta), D 00 l ( α , β , γ ) = P l ( cos β ) ,
D ± 1 0 l ( α , β , γ ) = ∓ e ± i α sin β l ( l + 1 ) P l ′ ( cos β ) , D^l_{\pm1\,0}(\alpha,\beta,\gamma)=\mp e^{\pm i\alpha}\frac{\sin\beta}{\sqrt{l(l+1)}}P'_l(\cos\beta), D ± 1 0 l ( α , β , γ ) = ∓ e ± i α l ( l + 1 ) sin β P l ′ ( cos β ) ,
D 0 ± 1 l ( α , β , γ ) = ∓ e ± i γ sin β l ( l + 1 ) P l ′ ( cos β ) , D^l_{0\,\pm1}(\alpha,\beta,\gamma)=\mp e^{\pm i\gamma}\frac{\sin\beta}{\sqrt{l(l+1)}}P'_l(\cos\beta), D 0 ± 1 l ( α , β , γ ) = ∓ e ± iγ l ( l + 1 ) sin β P l ′ ( cos β ) ,
D ± 2 0 l ( α , β , γ ) = e ± i 2 α { − l ( l + 1 ) ( l − 1 ) ( l + 2 ) P l ( cos β ) + 2 cos β ( l − 1 ) l ( l + 1 ) ( l + 2 ) P l ′ ( cos β ) } , D^l_{\pm2\,0}(\alpha,\beta,\gamma)=e^{\pm i2\alpha}\left\{-\sqrt{\frac{l(l+1)}{(l-1)(l+2)}}P_l(\cos\beta)+\frac{2\cos\beta}{\sqrt{(l-1)l(l+1)(l+2)}}P'_l(\cos\beta)\right\}, D ± 2 0 l ( α , β , γ ) = e ± i 2 α { − ( l − 1 ) ( l + 2 ) l ( l + 1 ) P l ( cos β ) + ( l − 1 ) l ( l + 1 ) ( l + 2 ) 2 cos β P l ′ ( cos β ) } ,
D 0 ± 2 l ( α , β , γ ) = e ± i 2 γ { − l ( l + 1 ) ( l − 1 ) ( l + 2 ) P l ( cos β ) + 2 cos β ( l − 1 ) l ( l + 1 ) ( l + 2 ) P l ′ ( cos β ) } . (875) D^l_{0\,\pm2}(\alpha,\beta,\gamma)=e^{\pm i2\gamma}\left\{-\sqrt{\frac{l(l+1)}{(l-1)(l+2)}}P_l(\cos\beta)+\frac{2\cos\beta}{\sqrt{(l-1)l(l+1)(l+2)}}P'_l(\cos\beta)\right\}. \tag*{(875)} D 0 ± 2 l ( α , β , γ ) = e ± i 2 γ { − ( l − 1 ) ( l + 2 ) l ( l + 1 ) P l ( cos β ) + ( l − 1 ) l ( l + 1 ) ( l + 2 ) 2 cos β P l ′ ( cos β ) } . (875)
M = ± 1 / 2 M=\pm1/2 M = ± 1/2 and/or M ′ = ± 1 / 2 M'=\pm1/2 M ′ = ± 1/2
D 1 2 ± M ′ J ( α , β , γ ) = e ± i α − γ 2 π 2 J + 1 sin β 2 { ± J ± M ′ + 1 J + 1 Y J + 1 2 ∓ 1 2 − M ′ ( β , γ ) ∓ J ∓ M ′ J Y J − 1 2 ∓ 1 2 − M ′ ( β , γ ) } , D^J_{\frac{1}{2}\,\pm M'}(\alpha,\beta,\gamma)=e^{\pm i\frac{\alpha-\gamma}{2}}\frac{\sqrt{\pi}}{\sqrt{2J+1}\sin\frac{\beta}{2}}\left\{\pm\sqrt{\frac{J\pm M'+1}{J+1}}Y_{J+\frac{1}{2}\mp\frac{1}{2}-M'}(\beta,\gamma)\mp\sqrt{\frac{J\mp M'}{J}}Y_{J-\frac{1}{2}\mp\frac{1}{2}-M'}(\beta,\gamma)\right\}, D 2 1 ± M ′ J ( α , β , γ ) = e ± i 2 α − γ 2 J + 1 sin 2 β π { ± J + 1 J ± M ′ + 1 Y J + 2 1 ∓ 2 1 − M ′ ( β , γ ) ∓ J J ∓ M ′ Y J − 2 1 ∓ 2 1 − M ′ ( β , γ ) } ,
D M ± 1 2 J ( α , β , γ ) = ( − 1 ) ± 1 2 − M e ± i α − γ 2 π 2 J + 1 sin β 2 { ± J ± M + 1 J + 1 Y J + 1 2 ∓ 1 2 − M ( β , α ) ∓ J ∓ M J Y J − 1 2 ∓ 1 2 − M ( β , α ) } . (876) D^J_{M\,\pm\frac{1}{2}}(\alpha,\beta,\gamma)=(-1)^{\pm\frac{1}{2}-M}e^{\pm i\frac{\alpha-\gamma}{2}}\frac{\sqrt{\pi}}{\sqrt{2J+1}\sin\frac{\beta}{2}}\left\{\pm\sqrt{\frac{J\pm M+1}{J+1}}Y_{J+\frac{1}{2}\mp\frac{1}{2}-M}(\beta,\alpha)\mp\sqrt{\frac{J\mp M}{J}}Y_{J-\frac{1}{2}\mp\frac{1}{2}-M}(\beta,\alpha)\right\}. \tag*{(876)} D M ± 2 1 J ( α , β , γ ) = ( − 1 ) ± 2 1 − M e ± i 2 α − γ 2 J + 1 sin 2 β π { ± J + 1 J ± M + 1 Y J + 2 1 ∓ 2 1 − M ( β , α ) ∓ J J ∓ M Y J − 2 1 ∓ 2 1 − M ( β , α ) } . (876)
In particular
D 1 2 1 2 J ( α , β , γ ) = e − i α + γ 2 cos β 2 J + 1 2 { P J + 1 2 ′ ( cos β ) − P J − 1 2 ′ ( cos β ) } , D^J_{\frac{1}{2}\,\frac{1}{2}}(\alpha,\beta,\gamma)=e^{-i\frac{\alpha+\gamma}{2}}\frac{\cos\frac{\beta}{2}}{J+\frac{1}{2}}\left\{P'_{J+\frac{1}{2}}(\cos\beta)-P'_{J-\frac{1}{2}}(\cos\beta)\right\}, D 2 1 2 1 J ( α , β , γ ) = e − i 2 α + γ J + 2 1 cos 2 β { P J + 2 1 ′ ( cos β ) − P J − 2 1 ′ ( cos β ) } ,
D 1 2 − 1 2 J ( α , β , γ ) = − e − i α − γ 2 sin β 2 J + 1 2 { P J + 1 2 ′ ( cos β ) + P J − 1 2 ′ ( cos β ) } , D^J_{\frac{1}{2}\,-\frac{1}{2}}(\alpha,\beta,\gamma)=-e^{-i\frac{\alpha-\gamma}{2}}\frac{\sin\frac{\beta}{2}}{J+\frac{1}{2}}\left\{P'_{J+\frac{1}{2}}(\cos\beta)+P'_{J-\frac{1}{2}}(\cos\beta)\right\}, D 2 1 − 2 1 J ( α , β , γ ) = − e − i 2 α − γ J + 2 1 sin 2 β { P J + 2 1 ′ ( cos β ) + P J − 2 1 ′ ( cos β ) } ,
D − 1 2 1 2 J ( α , β , γ ) = e i α − γ 2 sin β 2 J + 1 2 { P J + 1 2 ′ ( cos β ) + P J − 1 2 ′ ( cos β ) } , D^J_{-\frac{1}{2}\,\frac{1}{2}}(\alpha,\beta,\gamma)=e^{i\frac{\alpha-\gamma}{2}}\frac{\sin\frac{\beta}{2}}{J+\frac{1}{2}}\left\{P'_{J+\frac{1}{2}}(\cos\beta)+P'_{J-\frac{1}{2}}(\cos\beta)\right\}, D − 2 1 2 1 J ( α , β , γ ) = e i 2 α − γ J + 2 1 sin 2 β { P J + 2 1 ′ ( cos β ) + P J − 2 1 ′ ( cos β ) } ,
D − 1 2 − 1 2 J ( α , β , γ ) = e i α + γ 2 cos β 2 J + 1 2 { P J + 1 2 ′ ( cos β ) − P J − 1 2 ′ ( cos β ) } . (877) D^J_{-\frac{1}{2}\,-\frac{1}{2}}(\alpha,\beta,\gamma)=e^{i\frac{\alpha+\gamma}{2}}\frac{\cos\frac{\beta}{2}}{J+\frac{1}{2}}\left\{P'_{J+\frac{1}{2}}(\cos\beta)-P'_{J-\frac{1}{2}}(\cos\beta)\right\}. \tag*{(877)} D − 2 1 − 2 1 J ( α , β , γ ) = e i 2 α + γ J + 2 1 cos 2 β { P J + 2 1 ′ ( cos β ) − P J − 2 1 ′ ( cos β ) } . (877)
Asymptotics of D M ′ M ( α , β , γ ) D^{M}_{M' }(\alpha,\beta,\gamma) D M ′ M ( α , β , γ )
(c) M = ± 1 M=\pm1 M = ± 1 and/or M ′ = ± 1 M'=\pm1 M ′ = ± 1
D ± 1 m l ( α , β , γ ) = e ∓ i α 4 π l ( l + 1 ) ( 2 l + 1 ) { ∓ ( l − m ) ( l + m + 1 ) 1 ∓ cos β 2 Y l − m − 1 ( β , γ ) e i γ + m sin β Y l − m ( β , γ ) ± ( l + m ) ( l − m + 1 ) 1 ± cos β 2 Y l − m + 1 ( β , γ ) e − i γ } D^{l}_{\pm1m}(\alpha,\beta,\gamma)=e^{\mp i\alpha}\sqrt{\frac{4\pi}{l(l+1)(2l+1)}}\left\{\mp\sqrt{(l-m)(l+m+1)}\frac{1\mp\cos\beta}{2}Y_{l-m-1}(\beta,\gamma)e^{i\gamma}+m\sin\beta Y_{l-m}(\beta,\gamma)\pm\sqrt{(l+m)(l-m+1)}\frac{1\pm\cos\beta}{2}Y_{l-m+1}(\beta,\gamma)e^{-i\gamma}\right\} D ± 1 m l ( α , β , γ ) = e ∓ i α l ( l + 1 ) ( 2 l + 1 ) 4 π { ∓ ( l − m ) ( l + m + 1 ) 2 1 ∓ cos β Y l − m − 1 ( β , γ ) e iγ + m sin β Y l − m ( β , γ ) ± ( l + m ) ( l − m + 1 ) 2 1 ± cos β Y l − m + 1 ( β , γ ) e − iγ }
D m ± 1 l ( α , β , γ ) = e ± i γ ( − 1 ) m + 1 4 π l ( l + 1 ) ( 2 l + 1 ) { ∓ ( l − m ) ( l + m + 1 ) 1 ∓ cos β 2 Y l − m − 1 ( β , α ) e i α + m sin β Y l − m ( β , α ) ± ( l + m ) ( l − m + 1 ) 1 ± cos β 2 Y l − m + 1 ( β , α ) e − i α } (878) D^{l}_{m\pm1}(\alpha,\beta,\gamma)=e^{\pm i\gamma}(-1)^{m+1}\sqrt{\frac{4\pi}{l(l+1)(2l+1)}}\left\{\mp\sqrt{(l-m)(l+m+1)}\frac{1\mp\cos\beta}{2}Y_{l-m-1}(\beta,\alpha)e^{i\alpha}+m\sin\beta Y_{l-m}(\beta,\alpha)\pm\sqrt{(l+m)(l-m+1)}\frac{1\pm\cos\beta}{2}Y_{l-m+1}(\beta,\alpha)e^{-i\alpha}\right\} \tag*{(878)} D m ± 1 l ( α , β , γ ) = e ± iγ ( − 1 ) m + 1 l ( l + 1 ) ( 2 l + 1 ) 4 π { ∓ ( l − m ) ( l + m + 1 ) 2 1 ∓ cos β Y l − m − 1 ( β , α ) e i α + m sin β Y l − m ( β , α ) ± ( l + m ) ( l − m + 1 ) 2 1 ± cos β Y l − m + 1 ( β , α ) e − i α } (878)
In particular,
D ± 11 l ( α , β , γ ) = e ∓ i α − i γ 1 ± cos β l ( l + 1 ) { P l ′ ( cos β ) ∓ ( 1 ∓ cos β ) P l ′ ′ ( cos β ) } D^{l}_{\pm11}(\alpha,\beta,\gamma)=e^{\mp i\alpha-i\gamma}\frac{1\pm\cos\beta}{l(l+1)}\left\{P'_l(\cos\beta)\mp(1\mp\cos\beta)P''_l(\cos\beta)\right\} D ± 11 l ( α , β , γ ) = e ∓ i α − iγ l ( l + 1 ) 1 ± cos β { P l ′ ( cos β ) ∓ ( 1 ∓ cos β ) P l ′′ ( cos β ) }
D ± 1 − 1 l ( α , β , γ ) = e ∓ i α + i γ 1 ∓ cos β l ( l + 1 ) { P l ′ ( cos β ) ± ( 1 ± cos β ) P l ′ ′ ( cos β ) } (879) D^{l}_{\pm1-1}(\alpha,\beta,\gamma)=e^{\mp i\alpha+i\gamma}\frac{1\mp\cos\beta}{l(l+1)}\left\{P'_l(\cos\beta)\pm(1\pm\cos\beta)P''_l(\cos\beta)\right\} \tag*{(879)} D ± 1 − 1 l ( α , β , γ ) = e ∓ i α + iγ l ( l + 1 ) 1 ∓ cos β { P l ′ ( cos β ) ± ( 1 ± cos β ) P l ′′ ( cos β ) } (879)
(d) M = ± ( J − 1 ) M=\pm(J-1) M = ± ( J − 1 ) and/or M ′ = ± ( J − 1 ) M'=\pm(J-1) M ′ = ± ( J − 1 )
D J − 1 m J ( α , β , γ ) = ( − 1 ) J − m − 1 e − i ( J − 1 ) α − i m γ ( 2 J − 1 ) ! ( J + m ) ! ( J − m ) ! ( cos β 2 ) J + m − 1 ( sin β 2 ) J − m − 1 × [ J cos β − m ] , D^{J}_{J-1m}(\alpha,\beta,\gamma)=(-1)^{J-m-1}e^{-i(J-1)\alpha-im\gamma}\sqrt{\frac{(2J-1)!}{(J+m)!(J-m)!}}\left(\cos\frac{\beta}{2}\right)^{J+m-1}\left(\sin\frac{\beta}{2}\right)^{J-m-1}\times[J\cos\beta-m], D J − 1 m J ( α , β , γ ) = ( − 1 ) J − m − 1 e − i ( J − 1 ) α − imγ ( J + m )! ( J − m )! ( 2 J − 1 )! ( cos 2 β ) J + m − 1 ( sin 2 β ) J − m − 1 × [ J cos β − m ] ,
D − J + 1 m J ( α , β , γ ) = e i ( J − 1 ) α − i m γ ( 2 J − 1 ) ! ( J + m ) ! ( J − m ) ! ( cos β 2 ) J − m − 1 ( sin β 2 ) J + m − 1 × [ J cos β + m ] , D^{J}_{-J+1m}(\alpha,\beta,\gamma)=e^{i(J-1)\alpha-im\gamma}\sqrt{\frac{(2J-1)!}{(J+m)!(J-m)!}}\left(\cos\frac{\beta}{2}\right)^{J-m-1}\left(\sin\frac{\beta}{2}\right)^{J+m-1}\times[J\cos\beta+m], D − J + 1 m J ( α , β , γ ) = e i ( J − 1 ) α − imγ ( J + m )! ( J − m )! ( 2 J − 1 )! ( cos 2 β ) J − m − 1 ( sin 2 β ) J + m − 1 × [ J cos β + m ] ,
D m J − 1 J ( α , β , γ ) = e − i ( J − 1 ) γ − i m α ( 2 J − 1 ) ! ( J + m ) ! ( J − m ) ! ( cos β 2 ) J + m − 1 ( sin β 2 ) J − m − 1 × [ J cos β − m ] , D^{J}_{mJ-1}(\alpha,\beta,\gamma)=e^{-i(J-1)\gamma-im\alpha}\sqrt{\frac{(2J-1)!}{(J+m)!(J-m)!}}\left(\cos\frac{\beta}{2}\right)^{J+m-1}\left(\sin\frac{\beta}{2}\right)^{J-m-1}\times[J\cos\beta-m], D m J − 1 J ( α , β , γ ) = e − i ( J − 1 ) γ − im α ( J + m )! ( J − m )! ( 2 J − 1 )! ( cos 2 β ) J + m − 1 ( sin 2 β ) J − m − 1 × [ J cos β − m ] ,
D m − J + 1 J ( α , β , γ ) = ( − 1 ) J + m − 1 e i ( J − 1 ) γ − i m α ( 2 J − 1 ) ! ( J + m ) ! ( J − m ) ! ( cos β 2 ) J − m − 1 ( sin β 2 ) J + m − 1 × [ J cos β + m ] . (880) D^{J}_{m-J+1}(\alpha,\beta,\gamma)=(-1)^{J+m-1}e^{i(J-1)\gamma-im\alpha}\sqrt{\frac{(2J-1)!}{(J+m)!(J-m)!}}\left(\cos\frac{\beta}{2}\right)^{J-m-1}\left(\sin\frac{\beta}{2}\right)^{J+m-1}\times[J\cos\beta+m]. \tag*{(880)} D m − J + 1 J ( α , β , γ ) = ( − 1 ) J + m − 1 e i ( J − 1 ) γ − im α ( J + m )! ( J − m )! ( 2 J − 1 )! ( cos 2 β ) J − m − 1 ( sin 2 β ) J + m − 1 × [ J cos β + m ] . (880)
(e) M = ± J M=\pm J M = ± J and/or M ′ = ± J M'=\pm J M ′ = ± J
D J M J ( α , β , γ ) = ( 2 J ) ! ( J + M ) ! ( J − M ) ! ( cos β 2 ) J + M ( − sin β 2 ) J − M e − i J α − i M γ D^{J}_{JM}(\alpha,\beta,\gamma)=\sqrt{\frac{(2J)!}{(J+M)!(J-M)!}}\left(\cos\frac{\beta}{2}\right)^{J+M}\left(-\sin\frac{\beta}{2}\right)^{J-M}e^{-iJ\alpha-iM\gamma} D J M J ( α , β , γ ) = ( J + M )! ( J − M )! ( 2 J )! ( cos 2 β ) J + M ( − sin 2 β ) J − M e − i J α − i M γ
D − J M J ( α , β , γ ) = ( 2 J ) ! ( J + M ) ! ( J − M ) ! ( cos β 2 ) J − M ( sin β 2 ) J + M e i J α − i M γ D^{J}_{-JM}(\alpha,\beta,\gamma)=\sqrt{\frac{(2J)!}{(J+M)!(J-M)!}}\left(\cos\frac{\beta}{2}\right)^{J-M}\left(\sin\frac{\beta}{2}\right)^{J+M}e^{iJ\alpha-iM\gamma} D − J M J ( α , β , γ ) = ( J + M )! ( J − M )! ( 2 J )! ( cos 2 β ) J − M ( sin 2 β ) J + M e i J α − i M γ
D M J J ( α , β , γ ) = ( 2 J ) ! ( J + M ) ! ( J − M ) ! ( cos β 2 ) J + M ( sin β 2 ) J − M e − i M α − i J γ D^{J}_{MJ}(\alpha,\beta,\gamma)=\sqrt{\frac{(2J)!}{(J+M)!(J-M)!}}\left(\cos\frac{\beta}{2}\right)^{J+M}\left(\sin\frac{\beta}{2}\right)^{J-M}e^{-iM\alpha-iJ\gamma} D M J J ( α , β , γ ) = ( J + M )! ( J − M )! ( 2 J )! ( cos 2 β ) J + M ( sin 2 β ) J − M e − i M α − i J γ
D M − J J ( α , β , γ ) = ( 2 J ) ! ( J + M ) ! ( J − M ) ! ( cos β 2 ) J − M ( − sin β 2 ) J + M e − i M α + i J γ (881) D^{J}_{M-J}(\alpha,\beta,\gamma)=\sqrt{\frac{(2J)!}{(J+M)!(J-M)!}}\left(\cos\frac{\beta}{2}\right)^{J-M}\left(-\sin\frac{\beta}{2}\right)^{J+M}e^{-iM\alpha+iJ\gamma} \tag*{(881)} D M − J J ( α , β , γ ) = ( J + M )! ( J − M )! ( 2 J )! ( cos 2 β ) J − M ( − sin 2 β ) J + M e − i M α + i J γ (881)
Explicit forms of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) for particular values of J J J (J = 1 / 2 , 1 , 3 / 2 , 2 , 5 / 2 , 3 , 7 / 2 , 4 , 9 / 2 , 5 J=1/2,1,3/2,2,5/2,3,7/2,4,9/2,5 J = 1/2 , 1 , 3/2 , 2 , 5/2 , 3 , 7/2 , 4 , 9/2 , 5 ) are presented in Sec. 4.20.
Asymptotics of D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ )
Large Angular Momentum
If J ≫ 1 J \gg1 J ≫ 1 , one has
D M M ′ J ( α , β , γ ) ≈ e − i M α − i M ′ γ ξ M M ′ s ! ( s + μ + ν ) ! ( s + μ ) ! ( s + ν ) ! × 2 π s cos [ ( s + μ + ν + 1 2 ) β − π 4 ( 2 μ + 1 ) ] sin β + O ( 1 J 3 / 2 ) (882) D^{J}_{MM'}(\alpha,\beta,\gamma) \approx e^{-iM\alpha-iM'\gamma}\xi_{MM'}\sqrt{\frac{s!(s+\mu+\nu)!}{(s+\mu)!(s+\nu)!}}\times\sqrt{\frac{2}{\pi s}}\frac{\cos\left[\left(s+\frac{\mu+\nu+1}{2}\right)\beta-\frac{\pi}{4}(2\mu+1)\right]}{\sqrt{\sin\beta}}+O\left(\frac{1}{J^{3/2}}\right) \tag*{(882)} D M M ′ J ( α , β , γ ) ≈ e − i M α − i M ′ γ ξ M M ′ ( s + μ )! ( s + ν )! s ! ( s + μ + ν )! × π s 2 sin β cos [ ( s + 2 μ + ν + 1 ) β − 4 π ( 2 μ + 1 ) ] + O ( J 3/2 1 ) (882)
where s , μ , ν s,\mu,\nu s , μ , ν are related to J , M , M ′ J,M,M' J , M , M ′ through Eqs. (580) , and ξ M M ′ \xi_{MM'} ξ M M ′ is defined by Eq. (581) .
If J → ∞ J\to\infty J → ∞ and β → 0 \beta\to0 β → 0 , while J β < ∞ J\beta<\infty J β < ∞ , then
D M M ′ J ( α , β , γ ) ≈ e − i M α − i M ′ γ J M ′ − M ( J β ) . (883) D^{J}_{MM'}(\alpha,\beta,\gamma)\approx e^{-iM\alpha-iM'\gamma}J_{M'-M}(J\beta). \tag*{(883)} D M M ′ J ( α , β , γ ) ≈ e − i M α − i M ′ γ J M ′ − M ( J β ) . (883)
Here J n ( x ) J_n(x) J n ( x ) is the Bessel function.
Small Variation of Rotation Axis
If β → 0 \beta\to0 β → 0 , we have
D M M ′ J ( α , β , γ ) ≈ e − i M α − i M ′ γ ξ M M ′ μ ! ( s + μ + ν ) ! ( s + μ ) ! s ! ( s + ν ) ! ( sin β 2 ) μ { 1 − 2 s ( s + μ + ν + 1 ) + ν ( μ + 1 ) 2 ( μ + 1 ) ( β 2 ) 2 + ⋯ } (884) D^{J}_{MM'}(\alpha,\beta,\gamma)\approx e^{-iM\alpha-iM'\gamma}\frac{\xi_{MM'}}{\mu!}\sqrt{\frac{(s+\mu+\nu)!(s+\mu)!}{s!(s+\nu)!}}\left(\sin\frac{\beta}{2}\right)^\mu\left\{1-\frac{2s(s+\mu+\nu+1)+\nu(\mu+1)}{2(\mu+1)}\left(\frac{\beta}{2}\right)^2+\cdots\right\} \tag*{(884)} D M M ′ J ( α , β , γ ) ≈ e − i M α − i M ′ γ μ ! ξ M M ′ s ! ( s + ν )! ( s + μ + ν )! ( s + μ )! ( sin 2 β ) μ { 1 − 2 ( μ + 1 ) 2 s ( s + μ + ν + 1 ) + ν ( μ + 1 ) ( 2 β ) 2 + ⋯ } (884)
If π − β → 0 \pi-\beta\to0 π − β → 0 , we have
D M M ′ J ( α , β , γ ) ≈ e − i M α − i M ′ γ ξ M M ′ ν ! ( − 1 ) s ( s + μ + ν ) ! ( s + ν ) ! s ! ( s + μ ) ! ( sin π − β 2 ) ν { 1 − 2 s ( s + μ + ν + 1 ) + μ ( ν + 1 ) 2 ( ν + 1 ) ( π − β 2 ) 2 + ⋯ } . (885) D^{J}_{MM'}(\alpha,\beta,\gamma)\approx e^{-iM\alpha-iM'\gamma}\frac{\xi_{MM'}}{\nu!}(-1)^s\sqrt{\frac{(s+\mu+\nu)!(s+\nu)!}{s!(s+\mu)!}}\left(\sin\frac{\pi-\beta}{2}\right)^\nu\left\{1-\frac{2s(s+\mu+\nu+1)+\mu(\nu+1)}{2(\nu+1)}\left(\frac{\pi-\beta}{2}\right)^2+\cdots\right\}. \tag*{(885)} D M M ′ J ( α , β , γ ) ≈ e − i M α − i M ′ γ ν ! ξ M M ′ ( − 1 ) s s ! ( s + μ )! ( s + μ + ν )! ( s + ν )! ( sin 2 π − β ) ν { 1 − 2 ( ν + 1 ) 2 s ( s + μ + ν + 1 ) + μ ( ν + 1 ) ( 2 π − β ) 2 + ⋯ } . (885)
Infinitesimal Rotations
(a) Rotation ε \varepsilon ε about the x x x -Axis
lim ε → 0 1 ε [ D M M ′ J ( − π 2 , ε , π 2 ) − δ M M ′ ] = − i ⟨ J M ∣ J ^ x ∣ J M ′ ⟩ = − i 2 δ M ′ , M + 1 ( J − M ′ ) ( J + M ′ + 1 ) − i 2 δ M ′ , M − 1 ( J + M ′ ) ( J − M ′ + 1 ) (886) \lim_{\varepsilon\to0}\frac{1}{\varepsilon}\left[D^{J}_{MM'}\left(-\frac{\pi}{2},\varepsilon,\frac{\pi}{2}\right)-\delta_{MM'}\right]=-i\langle JM|\hat{J}_{x}|JM'\rangle
=-\frac{i}{2}\delta_{M',M+1}\sqrt{(J-M')(J+M'+1)}-\frac{i}{2}\delta_{M',M-1}\sqrt{(J+M')(J-M'+1)} \tag*{(886)} ε → 0 lim ε 1 [ D M M ′ J ( − 2 π , ε , 2 π ) − δ M M ′ ] = − i ⟨ J M ∣ J ^ x ∣ J M ′ ⟩ = − 2 i δ M ′ , M + 1 ( J − M ′ ) ( J + M ′ + 1 ) − 2 i δ M ′ , M − 1 ( J + M ′ ) ( J − M ′ + 1 ) (886)
(b) Rotation ε \varepsilon ε about the y y y -Axis
lim ε → 0 1 ε [ D M M ′ J ( 0 , ε , 0 ) − δ M M ′ ] = − i ⟨ J M ∣ J ^ y ∣ J M ′ ⟩ = − 1 2 δ M ′ , M + 1 ( J − M ′ ) ( J + M ′ + 1 ) + 1 2 δ M ′ , M − 1 ( J + M ′ ) ( J − M ′ + 1 ) (887) \lim_{\varepsilon\to0}\frac{1}{\varepsilon}\left[D^{J}_{MM'}(0,\varepsilon,0)-\delta_{MM'}\right]=-i\langle JM|\hat{J}_{y}|JM'\rangle
=-\frac{1}{2}\delta_{M',M+1}\sqrt{(J-M')(J+M'+1)}+\frac{1}{2}\delta_{M',M-1}\sqrt{(J+M')(J-M'+1)} \tag*{(887)} ε → 0 lim ε 1 [ D M M ′ J ( 0 , ε , 0 ) − δ M M ′ ] = − i ⟨ J M ∣ J ^ y ∣ J M ′ ⟩ = − 2 1 δ M ′ , M + 1 ( J − M ′ ) ( J + M ′ + 1 ) + 2 1 δ M ′ , M − 1 ( J + M ′ ) ( J − M ′ + 1 ) (887)
(c) Rotation ε \varepsilon ε about the z z z -Axis
lim ε → 0 1 ε [ D M M ′ J ( ε , 0 , 0 ) − δ M M ′ ] = − i ⟨ J M ∣ J ^ z ∣ J M ′ ⟩ = − i M δ M M ′ (888) \lim_{\varepsilon\to0}\frac{1}{\varepsilon}\left[D^{J}_{MM'}(\varepsilon,0,0)-\delta_{MM'}\right]=-i\langle JM|\hat{J}_{z}|JM'\rangle=-iM\delta_{MM'} \tag*{(888)} ε → 0 lim ε 1 [ D M M ′ J ( ε , 0 , 0 ) − δ M M ′ ] = − i ⟨ J M ∣ J ^ z ∣ J M ′ ⟩ = − i M δ M M ′ (888)
(d) Rotation ε \varepsilon ε about an Arbitrary Axis n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ )
lim ε → 0 1 ε [ D M M ′ J ( α , β , γ ) − δ M M ′ ] = − i ⟨ J M ∣ J ^ ⋅ n ∣ J M ′ ⟩ = − i M cos Θ δ M M ′ − i 2 sin Θ e − i Φ ( J − M ′ ) ( J + M ′ + 1 ) δ M ′ , M + 1 − i 2 sin Θ e i Φ ( J + M ′ ) ( J − M ′ + 1 ) δ M ′ , M − 1 (889) \lim_{\varepsilon\to0}\frac{1}{\varepsilon}\left[D^{J}_{MM'}(\alpha,\beta,\gamma)-\delta_{MM'}\right]=-i\langle JM|\hat{\mathbf{J}}\cdot\mathbf{n}|JM'\rangle=-iM\cos\Theta\delta_{MM'}
-\frac{i}{2}\sin\Theta e^{-i\Phi}\sqrt{(J-M')(J+M'+1)}\delta_{M',M+1}-\frac{i}{2}\sin\Theta e^{i\Phi}\sqrt{(J+M')(J-M'+1)}\delta_{M',M-1} \tag*{(889)} ε → 0 lim ε 1 [ D M M ′ J ( α , β , γ ) − δ M M ′ ] = − i ⟨ J M ∣ J ^ ⋅ n ∣ J M ′ ⟩ = − i M cos Θ δ M M ′ − 2 i sin Θ e − i Φ ( J − M ′ ) ( J + M ′ + 1 ) δ M ′ , M + 1 − 2 i sin Θ e i Φ ( J + M ′ ) ( J − M ′ + 1 ) δ M ′ , M − 1 (889)
In Eqs. (886) –(889) J ^ \hat{\mathbf{J}} J ^ is the operator of angular momentum defined in Sec. 2.1.
Definitions of D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) by other authors
Different authors use somewhat different definitions of the rotation matrix. The main differences are in the following:
(a) in using right- or left-handed coordinate system;
(b) in rotating either coordinate system or physical body;
(c) in using various definitions of the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ , namely,
(i) choosing different rotation axes,
(ii) choosing a different order of rotations, and
(iii) defining rotation either in a right-handed or left-handed sense;
(d) in considering transformations either of covariant or of contravariant components;
(e) in choosing various transformation rules for irreducible tensors;
(f) in accepting different phases of non-diagonal elements in Eqs. (554) and (4.2.5).
The Wigner D D D -functions used in this book coincide with those defined by Edmonds (Ref. [47]), Rose (Ref. [4]), Newton (Ref. [5]) and some other authors. The relations between these D D D -functions and corresponding functions of other authors are listed in Table 4.2 .
Table 4.2. Definitions of the Rotation Matrix by Other Authors.
Special cases of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) for particular J J J , M M M and M ′ M' M ′
General expressions for the D D D -functions, Eqs. (568) –(571) , may be reduced to simple closed forms for particular values of J J J . Explicit forms of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) for J ≤ 1 J \le1 J ≤ 1 are presented in Table 4.3 –Table 4.4 . d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) are given in terms of either cos β \cos\beta cos β and sin β \sin\beta sin β (if J J J is integer), or cos ( β / 2 ) \cos(\beta/2) cos ( β /2 ) and sin ( β / 2 ) \sin(\beta/2) sin ( β /2 ) (if J J J is half-integer). Expressions for d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) for J ≥ 5 / 2 J \ge5/2 J ≥ 5/2 are presented for M ≥ 0 M \ge0 M ≥ 0 and ∣ M ′ ∣ ≤ M \lvert M'\rvert\le M ∣ M ′ ∣ ≤ M only. For M < 0 M < 0 M < 0 and ∣ M ′ ∣ > M \lvert M'\rvert> M ∣ M ′ ∣ > M one can obtain d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) using the symmetry properties (Sec. 4.4).
Table 4.3.
Table 4.4.
Explicit forms of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) with J ≤ 6 J \le6 J ≤ 6 are also given by Buckmaster (Ref. [113]), and Wolters (Ref. [114]). Extensive numerical tables of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) for J ≤ 13 J \le13 J ≤ 13 (integer and half-integer) may be found in Behkami (Ref. [115]).
Tables of d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) for β = π / 2 \beta= \pi/2 β = π /2
Values of the d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) -functions for β = π / 2 \beta= \pi/2 β = π /2 and J = 1 / 2 , 1 , 3 / 2 , 2 , 5 / 2 , 3 , 7 / 2 , 4 , 9 / 2 , 5 J = 1/2, 1, 3/2, 2, 5/2, 3, 7/2, 4, 9/2, 5 J = 1/2 , 1 , 3/2 , 2 , 5/2 , 3 , 7/2 , 4 , 9/2 , 5 are given in Tables 4.13–4.22.
The d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) -functions can also be generated in closed algebraic form with a computer-algebra system. The routines below — provided as scripts/wigner_d.py and scripts/wigner_d.wl — return d M M ′ J ( β ) d^{J}_{MM'}(\beta) d M M ′ J ( β ) for given J J J , M M M , M ′ M' M ′ in the convention of this book, each with a self-test against the explicit Wigner formula.
In SymPy (Python), whose Rotation.d already follows this book’s convention:
#!/usr/bin/env python3
"""Wigner (small) d-function d^J_{M M'}(beta) in algebraic form -- VMK convention.
Uses SymPy's ``sympy.physics.quantum.spin.Rotation.d``, whose convention matches
Varshalovich-Moskalev-Khersonskii (VMK):
d^{1/2}_{-1/2,-1/2}(beta) = -sin(beta/2)
and the d^1 matrix reproduces the VMK table. This has been checked against the
explicit Wigner sum formula for j = 1/2, 1, 3/2, 2, 5/2 and all m, m'
(proof-by-points at 15 angles, 30-digit precision). Run `--selftest` to redo it.
Requires: sympy. Usage:
python3 wigner_d.py # demo
python3 wigner_d.py --selftest # re-validate against the Wigner formula
"""
from sympy import (symbols, Rational, simplify, Matrix, nsimplify, cos, sin,
sqrt, factorial, S, pi)
from sympy.physics.quantum.spin import Rotation
beta = symbols('beta', real=True)
Tables of d M M ′ j ( β ) d^j_{MM'}(\beta) d M M ′ j ( β ) for β = π / 2 \beta= \pi/2 β = π /2
def wigner_d(j, m, mp, angle=beta):
"""Algebraic d^j_{m,mp}(angle) in VMK / Wigner convention.
j, m, mp may be integers or half-integers; pass half-integers as
``Rational(1, 2)`` (or the string ``'1/2'``) so the result stays symbolic.
"""
j, m, mp = nsimplify(j), nsimplify(m), nsimplify(mp)
return simplify(Rotation.d(j, m, mp, angle).doit())
def wigner_d_matrix(j, angle=beta):
"""Full d^j(angle) as a SymPy Matrix; rows m and cols m' run j, j-1, ..., -j."""
j = nsimplify(j)
order = [j - k for k in range(int(2 * j) + 1)]
return Matrix([[wigner_d(j, m, mp, angle) for mp in order] for m in order])
def _reference_d(j, m, mp):
"""Independent Wigner explicit sum formula (VMK eq. 4.3.1), for validation."""
j, m, mp = nsimplify(j), nsimplify(m), nsimplify(mp)
pref = sqrt(factorial(j + m) * factorial(j - m) *
factorial(j + mp) * factorial(j - mp))
tot = S.Zero
for k in range(int(max(0, m - mp)), int(min(j + m, j - mp)) + 1):
tot += ((-1)**k * cos(beta / 2)**(2*j - 2*k + m - mp)
* sin(beta / 2)**(2*k - m + mp)
/ (factorial(k) * factorial(j + m - k)
* factorial(j - mp - k) * factorial(k - m + mp)))
return pref * tot
def _selftest():
"""d^j is a degree-2j trig polynomial in beta/2, so vanishing of the
difference at > 2*(2j)+1 distinct points proves the identity."""
pts = [Rational(p, 17) * pi for p in range(1, 16)] # 15 distinct angles
worst = 0.0
for j2 in range(1, 6): # 2j = 1..5
j = Rational(j2, 2)
order = [j - k for k in range(int(2 * j) + 1)]
for m in order:
for mp in order:
e = Rotation.d(j, m, mp, beta).doit() - _reference_d(j, m, mp)
for p in pts:
worst = max(worst, abs(complex(e.subs(beta, p).evalf(30))))
ok = worst < 1e-25
print(f"self-test j=1/2..5/2 vs Wigner formula: "
f"{'IDENTICAL' if ok else 'FAILED'} (worst |diff| = {worst:.2e})")
return ok
if __name__ == '__main__':
import sys
if '--selftest' in sys.argv:
raise SystemExit(0 if _selftest() else 1)
# --- specific angular momentum / projection quantum numbers ---
print("d^{1/2}_{1/2,-1/2}(beta) =",
wigner_d(Rational(1, 2), Rational(1, 2), Rational(-1, 2)))
print("d^{2}_{1,0}(beta) =", wigner_d(2, 1, 0))
print()
print("d^1(beta) =")
print(wigner_d_matrix(1))
In Mathematica (Wolfram Language) the built-in WignerD is the transpose of the function used here, so the two projection indices must be swapped:
(* ::Package:: *)
(* Wigner (small) d-function d^J_{M M'}(beta) in algebraic form -- VMK convention.
CONVENTION. Mathematica's built-in WignerD[{j,m,mp},beta] is the TRANSPOSE of
the Varshalovich-Moskalev-Khersonskii (VMK) function:
WignerD[{j, m, mp}, beta] == d^J_{mp,m}(beta) (VMK)
== (-1)^(m-mp) d^J_{m,mp}(beta)
== d^J_{m,mp}(-beta).
e.g. WignerD[{1/2,1/2,-1/2},beta] = +Sin[beta/2], whereas VMK has -Sin[beta/2].
So VMK's d^J_{m,mp}(beta) is obtained by SWAPPING the two projections:
wignerD[j,m,mp] := WignerD[{j, mp, m}, beta].
Verified against the explicit Wigner sum formula for j=1/2,1,3/2,2,5/2 and all
m,m'. Run "Get" on this file, then use wignerD / wignerDMatrix below, or run
WolframKernel -noprompt -script wigner_d.wl
to see the demo and the self-test result.
Half-integers must be exact (1/2, 3/2, ...), not machine reals, to stay algebraic. *)
(* VMK-convention small-d, algebraic in the symbol beta (\[Beta]) *)
wignerD[j_, m_, mp_, b_ : \[Beta]] := FullSimplify[WignerD[{j, mp, m}, b]];
(* Full VMK d^j(beta); rows m and cols m' run j, j-1, ..., -j *)
wignerDMatrix[j_, b_ : \[Beta]] := Module[{o = Range[j, -j, -1]},
FullSimplify[Outer[wignerD[j, #1, #2, b] &, o, o]]];
(* --- independent Wigner explicit sum formula (VMK), used only for the self-test --- *)
refD[j_, m_, mp_, b_] := Sqrt[(j + m)! (j - m)! (j + mp)! (j - mp)!]*
Sum[(-1)^k Cos[b/2]^(2 j - 2 k + m - mp) Sin[b/2]^(2 k - m + mp)/
(k! (j + m - k)! (j - mp - k)! (k - m + mp)!),
{k, Max[0, m - mp], Min[j + m, j - mp]}];
selfTest[] := Module[{Js = {1/2, 1, 3/2, 2, 5/2}, vals},
vals[j_] := Range[j, -j, -1];
AllTrue[Js, Function[j, AllTrue[vals[j], Function[m, AllTrue[vals[j],
Function[mp, FullSimplify[wignerD[j, m, mp, \[Beta]] - refD[j, m, mp, \[Beta]]] === 0]]]]]]];
(* Demo / validation printed when this file is run as a script *)
Print["self-test wignerD == VMK formula (j=1/2..5/2): ", selfTest[]];
Print["d^{1/2}_{1/2,-1/2}(beta) = ", wignerD[1/2, 1/2, -1/2]];
Print["d^{2}_{1,0}(beta) = ", wignerD[2, 1, 0]];
Print["d^1(beta) = ", wignerDMatrix[1]];
Special cases of U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ )
Explicit forms of U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) for J = 1 / 2 , 1 , 3 / 2 , 2 J = 1/2, 1, 3/2, 2 J = 1/2 , 1 , 3/2 , 2 are given in Tables 4.23-4.26.
The U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) -functions can likewise be generated in closed algebraic form, from U M M ′ J ( ω ; Θ , Φ ) = ⟨ J M ∣ e − i ω n ( Θ , Φ ) ⋅ J ^ ∣ J M ′ ⟩ U^J_{MM'}(\omega;\Theta,\Phi) = \langle JM|e^{-i\omega\,\mathbf{n}(\Theta,\Phi)\cdot\hat{\mathbf{J}}}|JM'\rangle U M M ′ J ( ω ; Θ , Φ ) = ⟨ J M ∣ e − iω n ( Θ , Φ ) ⋅ J ^ ∣ J M ′ ⟩ . The routines below — provided as scripts/Ufunction.py and scripts/Ufunction.wl — return U M M ′ J ( ω ; Θ , Φ ) U^J_{MM'}(\omega;\Theta,\Phi) U M M ′ J ( ω ; Θ , Φ ) for given J , M , M ′ J,M,M' J , M , M ′ in the convention of this book; they reduce to d M M ′ J ( ω ) d^J_{MM'}(\omega) d M M ′ J ( ω ) for Θ = Φ = π / 2 \Theta=\Phi=\pi/2 Θ = Φ = π /2 , and each carries a self-test against the matrix exponential e − i ω n ⋅ J ^ e^{-i\omega\,\mathbf{n}\cdot\hat{\mathbf{J}}} e − iω n ⋅ J ^ .
In SymPy (Python):
#!/usr/bin/env python3
"""U-function U^J_{M M'}(omega; Theta, Phi) in algebraic form -- VMK convention.
U^J_{M M'}(omega; Theta, Phi) = <J M| exp(-i omega n.J) |J M'> is the matrix of a
rotation by the angle `omega` about the axis
n(Theta, Phi) = (sin Theta cos Phi, sin Theta sin Phi, cos Theta),
in the convention of Varshalovich-Moskalev-Khersonskii (VMK), Sec. 4.5. It
reduces to the small Wigner d-function for a rotation about the y-axis:
U^J_{M M'}(omega; pi/2, pi/2) == d^J_{M M'}(omega) (see wigner_d.py).
The closed form has a REMOVABLE singularity at Theta = 0 (there ve = 0 and
1/vw^2 terms cancel analytically); use the symbolic result, or keep Theta != 0
for purely numerical evaluation. Pass half-integer J, M, M' as `Rational(1, 2)`
etc. so the result stays algebraic.
This is the SymPy analogue of "Ufunction.wl". Verified against the matrix
exponential exp(-i omega n.J) (via numpy/scipy) for j = 1/2 .. 2, and against
Rotation.d for the y-rotation slice. Run `--selftest` to redo the checks.
Requires: sympy (numpy + scipy optional, only for the full self-test). Usage:
python3 Ufunction.py # demo
python3 Ufunction.py --selftest # re-validate
from sympy import (symbols, I, sqrt, factorial, exp, cos, sin, pi, simplify,
nsimplify, Matrix, Rational)
from sympy.physics.quantum.spin import Rotation
omega, Theta, Phi = symbols('omega Theta Phi', real=True)
def u_function(j, m, mp, w=omega, th=Theta, ph=Phi):
"""Algebraic U^J_{m,mp}(w; th, ph) in the VMK convention."""
j, m, mp = nsimplify(j), nsimplify(m), nsimplify(mp)
ampp = abs(m + mp)
if m + mp >= 0:
uu = cos(w/2) - I*sin(w/2)*cos(th)
maxs = min(j - m, j - mp)
term = lambda s: 1/(factorial(s)*factorial(s + ampp)
* factorial(j - m - s)*factorial(j - mp - s))
else:
uu = cos(w/2) + I*sin(w/2)*cos(th)
maxs = min(j + m, j + mp)
term = lambda s: 1/(factorial(s)*factorial(s + ampp)
* factorial(j + m - s)*factorial(j + mp - s))
v = sin(w/2)*sin(th)
pref = ((-I*v)**(2*j - ampp) * uu**ampp * exp(-I*(m - mp)*ph)
* sqrt(factorial(j + m)*factorial(j - m)
* factorial(j + mp)*factorial(j - mp)))
tot = sum(term(s)*(1 - v**-2)**s for s in range(int(maxs) + 1))
return simplify(pref*tot)
def u_function_matrix(j, w=omega, th=Theta, ph=Phi):
"""Full U^J(w; th, ph) as a SymPy Matrix; rows M and cols M' run J...-J."""
j = nsimplify(j)
order = [j - k for k in range(int(2*j) + 1)]
return Matrix([[u_function(j, m, mp, w, th, ph) for mp in order]
for m in order])
def _selftest():
js = [Rational(1, 2), 1, Rational(3, 2), 2]
worst = 0.0
try:
import numpy as np
from scipy.linalg import expm
def spin_mats(j):
ms = [j - k for k in range(int(2*j) + 1)]
n = len(ms)
Jp = np.zeros((n, n), complex)
for k in range(n):
if k - 1 >= 0:
Jp[k - 1, k] = np.sqrt(j*(j + 1) - ms[k]*(ms[k] + 1))
return ((Jp + Jp.conj().T)/2, (Jp - Jp.conj().T)/(2j),
np.diag(ms).astype(complex), ms)
def ref(j, m, mp, w, th, ph):
Jx, Jy, Jz, ms = spin_mats(j)
nn = [np.sin(th)*np.cos(ph), np.sin(th)*np.sin(ph), np.cos(th)]
U = expm(-1j*w*(nn[0]*Jx + nn[1]*Jy + nn[2]*Jz))
return U[ms.index(m), ms.index(mp)]
pts = [(0.7, 1.1, 0.3), (2.0, 0.4, 1.7), (1.3, 2.5, -0.9)]
for j in js:
order = [j - k for k in range(int(2*j) + 1)]
for m in order:
for mp in order:
e = u_function(j, m, mp)
for (w, th, ph) in pts:
val = complex(e.subs({omega: w, Theta: th, Phi: ph}).evalf(30))
worst = max(worst, abs(val - ref(float(j), float(m),
float(mp), w, th, ph)))
print(f"self-test vs scipy expm(-i w n.J), j=1/2..2: "
f"worst |diff| = {worst:.2e}")
except ImportError:
print("self-test: numpy/scipy not available, skipping matrix-exponential check")
# analytic cross-check (sympy only): U(w; pi/2, pi/2) == d^J(w)
beta = symbols('beta', real=True)
ok = True
for j in js:
order = [j - k for k in range(int(2*j) + 1)]
for m in order:
for mp in order:
du = u_function(j, m, mp).subs({Theta: pi/2, Phi: pi/2})
dd = Rotation.d(j, m, mp, omega).doit()
for w in (0.4, 1.3, 2.7):
if abs(complex(du.subs(omega, w))
- complex(dd.subs(beta, w).subs(omega, w))) > 1e-12:
ok = False
print("cross-check U(w; pi/2, pi/2) == d^J(w):", "OK" if ok else "FAIL")
return (worst < 1e-10) and ok
if __name__ == '__main__':
import sys
if '--selftest' in sys.argv:
raise SystemExit(0 if _selftest() else 1)
# --- specific quantum numbers ---
print("U^{1/2}_{1/2,1/2}(w; Th, Ph) =",
u_function(Rational(1, 2), Rational(1, 2), Rational(1, 2)))
print("U^{1}_{1,0}(w; Th, Ph) =", u_function(1, 1, 0))
print()
print("U(w; pi/2, pi/2) for j=1 (== d^1(w)):")
print(u_function_matrix(1).subs({Theta: pi/2, Phi: pi/2}).applyfunc(simplify))
In Mathematica (Wolfram Language):
(* ::Package:: *)
(* U-function U^J_{M M'}(omega; Theta, Phi) in algebraic form -- VMK convention.
U^J_{M M'}(omega; Theta, Phi) = <J M| Exp[-I omega n.J] |J M'> is the matrix of
a rotation by the angle omega about the axis
n(Theta, Phi) = (Sin[Theta] Cos[Phi], Sin[Theta] Sin[Phi], Cos[Theta]),
in the convention of Varshalovich-Moskalev-Khersonskii (VMK), Sec. 4.5.
It reduces to the small Wigner d-function for a rotation about the y-axis:
U^J_{M M'}(omega; Pi/2, Pi/2) == d^J_{M M'}(omega).
The closed form below has a REMOVABLE singularity at Theta = 0 (there v = 0
and the 1/v^2 terms cancel analytically): use the symbolic result, or keep
Theta != 0 for purely numerical evaluation. Pass half-integer J, M, M' as
exact rationals (1/2, 3/2, ...) so the result stays algebraic.
Verified against the matrix exponential Exp[-I omega n.J] for j = 1/2 .. 2
(see selfTest[] at the bottom; run this file as a script to print it). *)
Ufunction[j_, m_, mp_, w_ : \[Omega], th_ : \[CapitalTheta], ph_ : \[CapitalPhi]] :=
Block[{uu, v = Sin[w/2] Sin[th], ampp = Abs[m + mp], maxs},
uu = Cos[w/2] - If[m + mp >= 0, I, -I] Sin[w/2] Cos[th];
maxs = If[m + mp >= 0, Min[j - m, j - mp], Min[j + m, j + mp]];
FullSimplify[
(-I v)^(2 j - ampp) uu^ampp Exp[-I (m - mp) ph]*
Sqrt[(j + m)! (j - m)! (j + mp)! (j - mp)!]*
Sum[If[m + mp >= 0,
1/(s! (s + ampp)! (j - m - s)! (j - mp - s)!),
1/(s! (s + ampp)! (j + m - s)! (j + mp - s)!)] (1 - v^-2)^s,
{s, 0, maxs}]]];
(* Full U^J(omega; Theta, Phi); rows M and cols M' run J, J-1, ..., -J *)
UfunctionMatrix[j_, w_ : \[Omega], th_ : \[CapitalTheta], ph_ : \[CapitalPhi]] :=
Module[{o = Range[j, -j, -1]},
Table[Ufunction[j, mm, mmp, w, th, ph], {mm, o}, {mmp, o}]];
(* --- independent reference for the self-test: <J M| Exp[-I w n.J] | J M'> --- *)
Jz[j_] := DiagonalMatrix[Table[m, {m, j, -j, -1}]];
Jp[j_] := Module[{ms = Table[m, {m, j, -j, -1}], n = Length[ms]},
Table[If[i + 1 == k, Sqrt[(j - m) (j + m + 1)], 0], {i, n}, {k, n}]];
Jx[j_] := (Jp[j] + Transpose[Jp[j]])/2;
Jy[j_] := (Jp[j] - Transpose[Jp[j]])/(2 I);
refU[j_, m_, mp_, w_, th_, ph_] :=
Module[{nn = {Sin[th] Cos[ph], Sin[th] Sin[ph], Cos[th]}, U},
Special cases of U M M ′ J ( ω ; Θ , Φ ) U^{J}_{MM'}(\omega; \Theta,\Phi) U M M ′ J ( ω ; Θ , Φ )
U = MatrixExp[-I w (nn[[1]] Jx[j] + nn[[2]] Jy[j] + nn[[3]] Jz[j])];
U[[j - m + 1, j - mp + 1]]];
selfTest[] := Module[
{pts = {{0.7, 1.1, 0.3}, {2.0, 0.4, 1.7}, {1.3, 2.5, -0.9}}, worst = 0, vals},
vals[j_] := Range[j, -j, -1];
Do[Do[Do[Do[
worst = Max[worst, Abs[N[Ufunction[j, m, mp, p[[1]], p[[2]], p[[3]]] -
refU[j, m, mp, p[[1]], p[[2]], p[[3]]]]]],
{p, pts}], {mp, vals[j]}], {m, vals[j]}], {j, {1/2, 1, 3/2, 2}}];
worst];
(* Demo / validation printed when this file is run as a script *)
Print["self-test max|Ufunction - Exp[-I w n.J]| (j=1/2..2) = ", selfTest[]];
Print["U^{1/2}_{1/2,1/2}(w; Th, Ph) = ", Ufunction[1/2, 1/2, 1/2]];
Print["U(w; Pi/2, Pi/2) for j=1 (== d^1(w)) = ",
FullSimplify[UfunctionMatrix[1, \[Omega], Pi/2, Pi/2]]];
Chapter 5
Spherical harmonics
A spherical harmonic Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) is a single-valued, continuous, bounded complex function of two real arguments ϑ , φ \vartheta,\varphi ϑ , φ with 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π and 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π . It is characterized by two parameters l l l and m m m , which take values l = 0 , 1 , 2 , … l=0,1,2,\ldots l = 0 , 1 , 2 , … and m = l , l − 1 , l − 2 , … − l + 2 , − l + 1 , − l m=l,l-1,l-2,\ldots-l+2,-l+1,-l m = l , l − 1 , l − 2 , … − l + 2 , − l + 1 , − l . Therefore, for a given l l l there exist ( 2 l + 1 ) (2l+1) ( 2 l + 1 ) functions corresponding to different m m m ’s. All derivatives of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) are single-valued, continuous and finite functions.
The spherical harmonics play an important role in quantum mechanics. They are eigenfunctions of the operator of orbital angular momentum and describe the angular distribution of particles which move in a spherically-symmetric field with the orbital angular momentum l l l and projection m m m . Strictly speaking, l l l specifies the absolute value of orbital angular momentum because l ( l + 1 ) l(l+1) l ( l + 1 ) is the eigenvalue of the square of the orbital angular momentum operator, L ^ 2 \hat{L}^{2} L ^ 2 ; m m m is the eigenvalue of L ^ z \hat{L}_{z} L ^ z which is the projection of the orbital angular momentum operator on the quantization axis.
Definition
Commutation Relations
The spherical harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) are components of some irreducible tensor of rank l l l (Chap. 3). Owing to this circumstance they may be defined by the commutation relations
[ L ^ μ , Y l m ( ϑ , φ ) ] = l ( l + 1 ) C l m 1 μ l m + μ Y l m + μ ( ϑ , φ ) , (890) [\hat{L}_{\mu},Y_{lm}(\vartheta,\varphi)]=\sqrt{l(l+1)}C^{lm+\mu}_{lm\,1\mu}Y_{lm+\mu}(\vartheta,\varphi), \tag*{(890)} [ L ^ μ , Y l m ( ϑ , φ )] = l ( l + 1 ) C l m 1 μ l m + μ Y l m + μ ( ϑ , φ ) , (890)
where L ^ μ ( ϑ , φ ) \hat{L}_{\mu}(\vartheta,\varphi) L ^ μ ( ϑ , φ ) is a spherical component of the operator L ^ \hat{\mathbf{L}} L ^ (see Eq. (253) ). The three commutation relations (5.1.1) (for μ = 1 , 0 , − 1 \mu=1,0,-1 μ = 1 , 0 , − 1 ) generate the following three equations
L ^ ± 1 Y l m ( ϑ , φ ) = ∓ l ( l + 1 ) − m ( m ± 1 ) 2 Y l m ± 1 ( ϑ , φ ) , (891) \hat{L}_{\pm1}Y_{lm}(\vartheta,\varphi)=\mp\sqrt{\frac{l(l+1)-m(m\pm1)}{2}}Y_{lm\pm1}(\vartheta,\varphi), \tag*{(891)} L ^ ± 1 Y l m ( ϑ , φ ) = ∓ 2 l ( l + 1 ) − m ( m ± 1 ) Y l m ± 1 ( ϑ , φ ) , (891)
L ^ 0 Y l m ( ϑ , φ ) = m Y l m ( ϑ , φ ) . \hat{L}_{0}Y_{lm}(\vartheta,\varphi)=mY_{lm}(\vartheta,\varphi). L ^ 0 Y l m ( ϑ , φ ) = m Y l m ( ϑ , φ ) .
Differential Equations
According to the commutation relations (5.1.1), Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) is the eigenfunction of the operators L ^ 2 \hat{L}^{2} L ^ 2 and L ^ z \hat{L}_{z} L ^ z
L ^ 2 Y l m ( ϑ , φ ) = l ( l + 1 ) Y l m ( ϑ , φ ) , L ^ z Y l m ( ϑ , φ ) = m Y l m ( ϑ , φ ) . (892) \begin{aligned}
\hat{L}^{2}Y_{lm}(\vartheta,\varphi)&=l(l+1)Y_{lm}(\vartheta,\varphi),\\
\hat{L}_{z}Y_{lm}(\vartheta,\varphi)&=mY_{lm}(\vartheta,\varphi). \tag*{(892)}
\end{aligned} L ^ 2 Y l m ( ϑ , φ ) L ^ z Y l m ( ϑ , φ ) = l ( l + 1 ) Y l m ( ϑ , φ ) , = m Y l m ( ϑ , φ ) . (892)
or in an expanded form
[ 1 sin ϑ ∂ ∂ ϑ ( sin ϑ ∂ ∂ ϑ ) + 1 sin 2 ϑ ∂ 2 ∂ φ 2 + l ( l + 1 ) ] Y l m ( ϑ , φ ) = 0 , (893) \left[\frac{1}{\sin\vartheta}\frac{\partial}{\partial\vartheta}\left(\sin\vartheta\frac{\partial}{\partial\vartheta}\right)+\frac{1}{\sin^{2}\vartheta}\frac{\partial^{2}}{\partial\varphi^{2}}+l(l+1)\right]Y_{lm}(\vartheta,\varphi)=0, \tag*{(893)} [ sin ϑ 1 ∂ ϑ ∂ ( sin ϑ ∂ ϑ ∂ ) + sin 2 ϑ 1 ∂ φ 2 ∂ 2 + l ( l + 1 ) ] Y l m ( ϑ , φ ) = 0 , (893)
[ i ∂ ∂ φ + m ] Y l m ( ϑ , φ ) = 0. \left[i\frac{\partial}{\partial\varphi}+m\right]Y_{lm}(\vartheta,\varphi)=0. [ i ∂ φ ∂ + m ] Y l m ( ϑ , φ ) = 0.
Equations (5.1.4) are invariant under the following transformations
(a) l → l ˉ = − l − 1 l\to\bar{l}=-l-1 l → l ˉ = − l − 1 ;
(b) ϑ → − ϑ \vartheta\to-\vartheta ϑ → − ϑ (or ϑ → π − ϑ \vartheta\to\pi-\vartheta ϑ → π − ϑ );
(c) m → − m , φ → − φ m\to-m,\varphi\to-\varphi m → − m , φ → − φ .
Boundary Conditions
The first of Eqs. (893) is of the second order. For fixed l l l and m m m it has two linearly independent solutions. However, only one of them is regular, i.e., satisfies the condition ∣ Y l m ( ϑ , φ ) ∣ 2 < ∞ \lvert Y_{lm}(\vartheta,\varphi)\rvert^2 < \infty ∣ Y l m ( ϑ , φ ) ∣ 2 < ∞ while the other solution is singular at ϑ = 0 \vartheta= 0 ϑ = 0 and ϑ = π \vartheta= \pi ϑ = π . For quantum mechanical applications the regular solution is of major interest. This solution will be considered in the present chapter. The regular solution is selected by the following boundary conditions
Y l m ( ϑ , φ ± 2 π n ) = Y l m ( ϑ , φ ) , ∂ ∂ φ Y l m ( ϑ , φ ) ∣ ϑ = 0 = ∂ ∂ φ Y l m ( ϑ , φ ) ∣ ϑ = π = 0. (894) Y_{lm}(\vartheta,\varphi\pm2\pi n) = Y_{lm}(\vartheta,\varphi),
\qquad
\left.\frac{\partial}{\partial\varphi}Y_{lm}(\vartheta,\varphi)\right|_{\vartheta=0}
=
\left.\frac{\partial}{\partial\varphi}Y_{lm}(\vartheta,\varphi)\right|_{\vartheta=\pi}
= 0. \tag*{(894)} Y l m ( ϑ , φ ± 2 π n ) = Y l m ( ϑ , φ ) , ∂ φ ∂ Y l m ( ϑ , φ ) ϑ = 0 = ∂ φ ∂ Y l m ( ϑ , φ ) ϑ = π = 0. (894)
Below we shall consider the spherical harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) with integer l l l and m m m (with ∣ m ∣ ≤ l \lvert m\rvert\le l ∣ m ∣ ≤ l ) because the boundary conditions (5.1.5) are fulfilled only for such values of the parameters.
Normalization
(a) The differential equations (5.1.4) and the boundary conditions (5.1.5) are homogeneous. Hence, they determine the spherical harmonics only up to some arbitrary complex factor. The absolute value of this factor can be fixed by the normalization.
The normalization and orthogonality relation of the spherical harmonics is given by
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l m ∗ ( ϑ , φ ) Y l ′ m ′ ( ϑ , φ ) = δ l l ′ δ m m ′ . (895) \int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Y_{lm}^{*}(\vartheta,\varphi)Y_{l'm'}(\vartheta,\varphi)=\delta_{ll'}\delta_{mm'}. \tag*{(895)} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l m ∗ ( ϑ , φ ) Y l ′ m ′ ( ϑ , φ ) = δ l l ′ δ m m ′ . (895)
(b) Sometimes instead of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) it is more convenient to use the function C l m ( ϑ , φ ) C_{lm}(\vartheta,\varphi) C l m ( ϑ , φ ) (see, e.g., Refs. [7, 21]) which differs from Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) by the normalization factor,
C l m ( ϑ , φ ) = 4 π 2 l + 1 Y l m ( ϑ , φ ) . (896) C_{lm}(\vartheta,\varphi)=\sqrt{\frac{4\pi}{2l+1}}Y_{lm}(\vartheta,\varphi). \tag*{(896)} C l m ( ϑ , φ ) = 2 l + 1 4 π Y l m ( ϑ , φ ) . (896)
The function C l m ( ϑ , φ ) C_{lm}(\vartheta,\varphi) C l m ( ϑ , φ ) satisfies the following relations
∑ m C l m ( ϑ , φ ) C l , − m ( ϑ , φ ) ( − 1 ) m = 1 , C l m ( 0 , 0 ) = δ m 0 , (897) \sum_m C_{lm}(\vartheta,\varphi)C_{l,-m}(\vartheta,\varphi)(-1)^m=1,\qquad C_{lm}(0,0)=\delta_{m0}, \tag*{(897)} m ∑ C l m ( ϑ , φ ) C l , − m ( ϑ , φ ) ( − 1 ) m = 1 , C l m ( 0 , 0 ) = δ m 0 , (897)
C l 0 ( ϑ , φ ) = P l ( cos ϑ ) , C l m ( ϑ , φ ) = D 0 , − m l ( 0 , ϑ , φ ) . C_{l0}(\vartheta,\varphi)=P_l(\cos\vartheta),\qquad C_{lm}(\vartheta,\varphi)=D^l_{0,-m}(0,\vartheta,\varphi). C l 0 ( ϑ , φ ) = P l ( cos ϑ ) , C l m ( ϑ , φ ) = D 0 , − m l ( 0 , ϑ , φ ) .
The normalization and orthogonality relation for C l m ( ϑ , φ ) C_{lm}(\vartheta,\varphi) C l m ( ϑ , φ ) can be represented in the form
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ C l m ∗ ( ϑ , φ ) C l ′ m ′ ( ϑ , φ ) = 2 l + 1 4 π δ l l ′ δ m m ′ . (898) \int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta C_{lm}^{*}(\vartheta,\varphi)C_{l'm'}(\vartheta,\varphi)=\frac{2l+1}{4\pi}\delta_{ll'}\delta_{mm'}. \tag*{(898)} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ C l m ∗ ( ϑ , φ ) C l ′ m ′ ( ϑ , φ ) = 4 π 2 l + 1 δ l l ′ δ m m ′ . (898)
Choice of Phase
(a) The phase differences of the harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and Y l m ′ ( ϑ , φ ) Y_{lm'}(\vartheta,\varphi) Y l m ′ ( ϑ , φ ) with m = m ′ ± 1 m=m'\pm1 m = m ′ ± 1 are determined by the commutation relations (5.1.1). Using these relations, we may find relative phases of all ( 2 l + 1 ) (2l+1) ( 2 l + 1 ) harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) with different m m m for each l l l . An overall phase factor may be fixed by specifying the phase of one of the harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) for some given values of arguments, for example,
Y l 0 ( 0 , 0 ) = 2 l + 1 4 π . (899) Y_{l0}(0,0)=\sqrt{\frac{2l+1}{4\pi}}. \tag*{(899)} Y l 0 ( 0 , 0 ) = 4 π 2 l + 1 . (899)
In this case the following relations are valid for the complex conjugate function Y l m ∗ ( ϑ , φ ) Y_{lm}^{*}(\vartheta,\varphi) Y l m ∗ ( ϑ , φ )
Y l m ∗ ( ϑ , φ ) = Y l m ( ϑ , − φ ) = ( − 1 ) m Y l , − m ( ϑ , φ ) . (900) Y_{lm}^{*}(\vartheta,\varphi)=Y_{lm}(\vartheta,-\varphi)=(-1)^mY_{l,-m}(\vartheta,\varphi). \tag*{(900)} Y l m ∗ ( ϑ , φ ) = Y l m ( ϑ , − φ ) = ( − 1 ) m Y l , − m ( ϑ , φ ) . (900)
In particular, Eqs. (899) and (900) show that Y l 0 ( ϑ , φ ) Y_{l0}(\vartheta,\varphi) Y l 0 ( ϑ , φ ) is real for 0 ≤ ϑ ≤ π 0\le\vartheta\le\pi 0 ≤ ϑ ≤ π , and Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) with m ≠ 0 m\ne0 m = 0 is real only for φ = 0 , π / m , 2 π / m , 3 π / m , … \varphi=0,\pi/m,2\pi/m,3\pi/m,\ldots φ = 0 , π / m , 2 π / m , 3 π / m , … , etc.
The above choice of the phase is widely used (e.g., see, Condon and Shortley [22]).
(b) In the literature (see, e.g., Refs. [17, 16]) one can also find the spherical harmonics defined according to another phase convention, namely,
Y ~ l m ( ϑ , φ ) = i l Y l m ( ϑ , φ ) . (901) \widetilde{Y}_{lm}(\vartheta,\varphi)=i^lY_{lm}(\vartheta,\varphi). \tag*{(901)} Y l m ( ϑ , φ ) = i l Y l m ( ϑ , φ ) . (901)
We shall refer to these harmonics as the modified spherical harmonics. They satisfy the phase relation
Y ~ l m ∗ ( ϑ , φ ) = ( − 1 ) l + m Y ~ l , − m ( ϑ , φ ) . (902) \widetilde{Y}_{lm}^{*}(\vartheta,\varphi)=(-1)^{l+m}\widetilde{Y}_{l,-m}(\vartheta,\varphi). \tag*{(902)} Y l m ∗ ( ϑ , φ ) = ( − 1 ) l + m Y l , − m ( ϑ , φ ) . (902)
Equations (5.1.4) and the relations (5.1.5), (5.1.10) and (5.1.11) completely define the harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) . Since l l l and m m m are integers, the function Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) is single-valued.
Zonal, Sectorial and Tesseral Harmonics
These functions are linear combinations of the spherical harmonics for ∣ m ∣ ≤ l |m| \le l ∣ m ∣ ≤ l
u l m ( ϑ , φ ) = 1 2 [ Y l m ( ϑ , φ ) + Y l m ∗ ( ϑ , φ ) ] = 2 l + 1 4 π ( l − m ) ! ( l + m ) ! cos m φ P l m ( cos ϑ ) , u_{lm}(\vartheta,\varphi)=\frac{1}{2}\left[Y_{lm}(\vartheta,\varphi)+Y_{lm}^{*}(\vartheta,\varphi)\right]=\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}\cos m\varphi P_l^m(\cos\vartheta), u l m ( ϑ , φ ) = 2 1 [ Y l m ( ϑ , φ ) + Y l m ∗ ( ϑ , φ ) ] = 4 π 2 l + 1 ( l + m )! ( l − m )! cos m φ P l m ( cos ϑ ) ,
v l m ( ϑ , φ ) = 1 2 i [ Y l m ( ϑ , φ ) − Y l m ∗ ( ϑ , φ ) ] = 2 l + 1 4 π ( l − m ) ! ( l + m ) ! sin m φ P l m ( cos ϑ ) . (903) v_{lm}(\vartheta,\varphi)=\frac{1}{2i}\left[Y_{lm}(\vartheta,\varphi)-Y_{lm}^{*}(\vartheta,\varphi)\right]=\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}\sin m\varphi P_l^m(\cos\vartheta). \tag*{(903)} v l m ( ϑ , φ ) = 2 i 1 [ Y l m ( ϑ , φ ) − Y l m ∗ ( ϑ , φ ) ] = 4 π 2 l + 1 ( l + m )! ( l − m )! sin m φ P l m ( cos ϑ ) . (903)
The functions u l m ( ϑ , φ ) u_{lm}(\vartheta,\varphi) u l m ( ϑ , φ ) and v l m ( ϑ , φ ) v_{lm}(\vartheta,\varphi) v l m ( ϑ , φ ) are real, in contrast to Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) . The functions u l 0 ( ϑ , φ ) u_{l0}(\vartheta,\varphi) u l 0 ( ϑ , φ ) are called the zonal harmonics because parallels where u l 0 ( ϑ , φ ) = 0 u_{l0}(\vartheta,\varphi)=0 u l 0 ( ϑ , φ ) = 0 divide a sphere of unit radius into l + 1 l+1 l + 1 zones. The functions u l l ( ϑ , φ ) u_{ll}(\vartheta,\varphi) u l l ( ϑ , φ ) and v l l ( ϑ , φ ) v_{ll}(\vartheta,\varphi) v l l ( ϑ , φ ) are called the sectorial harmonics because meridians where u l l ( ϑ , φ ) = 0 u_{ll}(\vartheta,\varphi)=0 u l l ( ϑ , φ ) = 0 or v l l ( ϑ , φ ) = 0 v_{ll}(\vartheta,\varphi)=0 v l l ( ϑ , φ ) = 0 divide the unit sphere into 2 l 2l 2 l sectors. The functions u l m ( ϑ , φ ) u_{lm}(\vartheta,\varphi) u l m ( ϑ , φ ) and v l m ( ϑ , φ ) v_{lm}(\vartheta,\varphi) v l m ( ϑ , φ ) for m ≠ 0 m\ne0 m = 0 and m ≠ l m\ne l m = l are called the tesseral harmonics because a set of parallels and meridians where u l m ( ϑ , φ ) = 0 u_{lm}(\vartheta,\varphi)=0 u l m ( ϑ , φ ) = 0 or v l m ( ϑ , φ ) = 0 v_{lm}(\vartheta,\varphi)=0 v l m ( ϑ , φ ) = 0 divides the whole spherical surface into 2 m ( l − m + 1 ) 2m(l-m+1) 2 m ( l − m + 1 ) cells. The cells, which correspond to positive and negative signs of any function, are arranged in as on a chequer board.
Solutions of Some Differential Equations in Terms of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ )
(a) The solution of the Laplace equation
∇ 2 f ( r , ϑ , φ ) = 0 (904) \nabla^2 f(r,\vartheta,\varphi)=0 \tag*{(904)} ∇ 2 f ( r , ϑ , φ ) = 0 (904)
in polar coordinates is given by
Z l m ( r , ϑ , φ ) = r l Y l m ( ϑ , φ ) and R l m ( r , ϑ , φ ) = r − l − 1 Y l m ( ϑ , φ ) , \mathfrak{Z}_{lm}(r,\vartheta,\varphi)=r^lY_{lm}(\vartheta,\varphi)\quad\text{and}\quad\mathfrak{R}_{lm}(r,\vartheta,\varphi)=r^{-l-1}Y_{lm}(\vartheta,\varphi), Z l m ( r , ϑ , φ ) = r l Y l m ( ϑ , φ ) and R l m ( r , ϑ , φ ) = r − l − 1 Y l m ( ϑ , φ ) ,
where Z l m ( r , ϑ , φ ) \mathfrak{Z}_{lm}(r,\vartheta,\varphi) Z l m ( r , ϑ , φ ) is regular and R l m ( r , ϑ , φ ) \mathfrak{R}_{lm}(r,\vartheta,\varphi) R l m ( r , ϑ , φ ) is singular at r = 0 r=0 r = 0 . These functions are called the solid harmonics. In the cartesian coordinate representation, a function Z l m \mathfrak{Z}_{lm} Z l m is a homogeneous harmonic polynomial of degree l l l
r l Y l m ( ϑ , φ ) = 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ∑ p , q , r 1 p ! q ! r ! ( − x + i y 2 ) p ( x − i y 2 ) q z r . (905) r^lY_{lm}(\vartheta,\varphi)=\sqrt{\frac{2l+1}{4\pi}(l+m)!(l-m)!}\sum_{p,q,r}\frac{1}{p!q!r!}\left(\frac{-x+iy}{2}\right)^p\left(\frac{x-iy}{2}\right)^qz^r. \tag*{(905)} r l Y l m ( ϑ , φ ) = 4 π 2 l + 1 ( l + m )! ( l − m )! p , q , r ∑ p ! q ! r ! 1 ( 2 − x + i y ) p ( 2 x − i y ) q z r . (905)
Here p , q , r p,q,r p , q , r are positive integers which satisfy the conditions p + q + r = l p+q+r=l p + q + r = l , p − q = m p-q=m p − q = m .
(b) The solutions of the Helmholtz wave equation
[ ∇ 2 + k 2 ] f ( r , ϑ , φ ) = 0 (906) \left[\nabla^2+k^2\right]f(r,\vartheta,\varphi)=0 \tag*{(906)} [ ∇ 2 + k 2 ] f ( r , ϑ , φ ) = 0 (906)
in polar coordinates may be expressed in terms of the functions z l ( k r ) Y l m ( ϑ , φ ) z_l(kr)Y_{lm}(\vartheta,\varphi) z l ( k r ) Y l m ( ϑ , φ ) where z l ( k r ) = π 2 k r Z l + 1 2 ( k r ) z_l(kr)=\sqrt{\frac{\pi}{2kr}}Z_{l+\frac{1}{2}}(kr) z l ( k r ) = 2 k r π Z l + 2 1 ( k r ) , Z l + 1 2 ( x ) Z_{l+\frac{1}{2}}(x) Z l + 2 1 ( x ) being any of the Bessel functions.
The functions L l m ( r , ϑ , φ ) = i l j l ( k r ) Y l m ( ϑ , φ ) \mathfrak{L}_{lm}(r,\vartheta,\varphi)=i^l j_l(kr)Y_{lm}(\vartheta,\varphi) L l m ( r , ϑ , φ ) = i l j l ( k r ) Y l m ( ϑ , φ ) and N l m ( r , ϑ , φ ) = i l n l ( k r ) Y l m ( ϑ , φ ) \mathfrak{N}_{lm}(r,\vartheta,\varphi)=i^l n_l(kr)Y_{lm}(\vartheta,\varphi) N l m ( r , ϑ , φ ) = i l n l ( k r ) Y l m ( ϑ , φ ) are called the standing spherical waves; L l m \mathfrak{L}_{lm} L l m is regular at r = 0 r=0 r = 0 , whereas N l m \mathfrak{N}_{lm} N l m is irregular. The functions B l m ( 1 ) ( r , ϑ , φ ) = i l h l ( 1 ) ( k r ) Y l m ( ϑ , φ ) \mathfrak{B}_{lm}^{(1)}(r,\vartheta,\varphi)=i^l h_l^{(1)}(kr)Y_{lm}(\vartheta,\varphi) B l m ( 1 ) ( r , ϑ , φ ) = i l h l ( 1 ) ( k r ) Y l m ( ϑ , φ ) and B l m ( 2 ) ( r , ϑ , φ ) = i l h l ( 2 ) ( k r ) Y l m ( ϑ , φ ) \mathfrak{B}_{lm}^{(2)}(r,\vartheta,\varphi)=i^l h_l^{(2)}(kr)Y_{lm}(\vartheta,\varphi) B l m ( 2 ) ( r , ϑ , φ ) = i l h l ( 2 ) ( k r ) Y l m ( ϑ , φ ) are called the running spherical waves. The first of these corresponds to a spherical wave which converges to the origin, r = 0 r=0 r = 0 , while the second corresponds to an outgoing spherical wave. In the limit k → 0 k\to0 k → 0 Eq. (5.1.17) transforms into (5.1.15). In this case
L l m ( r , ϑ , φ ) → Z l m ( r , ϑ , φ ) , N l m ( r , ϑ , φ ) → R l m ( r , ϑ , φ ) . \mathfrak{L}_{lm}(r,\vartheta,\varphi)\to\mathfrak{Z}_{lm}(r,\vartheta,\varphi),\mathfrak{N}_{lm}(r,\vartheta,\varphi)\to\mathfrak{R}_{lm}(r,\vartheta,\varphi). L l m ( r , ϑ , φ ) → Z l m ( r , ϑ , φ ) , N l m ( r , ϑ , φ ) → R l m ( r , ϑ , φ ) .
(c) Solutions of the equation
[ ∇ 2 − ( n − l ) ( n + l + 1 ) r 2 ] f ( r , ϑ , φ ) = 0 (907) \left[\nabla^2-\frac{(n-l)(n+l+1)}{r^2}\right]f(r,\vartheta,\varphi)=0 \tag*{(907)} [ ∇ 2 − r 2 ( n − l ) ( n + l + 1 ) ] f ( r , ϑ , φ ) = 0 (907)
in polar coordinates are expressed in terms of the functions
f l m n ( r , ϑ , φ ) = r n Y l m ( ϑ , φ ) . f_{lm}^{n}(r,\vartheta,\varphi)=r^nY_{lm}(\vartheta,\varphi). f l m n ( r , ϑ , φ ) = r n Y l m ( ϑ , φ ) .
For r = 0 r=0 r = 0 a function f l m n ( r , ϑ , φ ) f_{lm}^{n}(r,\vartheta,\varphi) f l m n ( r , ϑ , φ ) is regular, if n ≥ 0 n\geq0 n ≥ 0 , but irregular if n < 0 n<0 n < 0 . When n = l n=l n = l or n = − l − 1 n=-l-1 n = − l − 1 , Eq. (5.1.18) transforms into (5.1.15), yielding
f l m l ( r , ϑ , φ ) = Z l m ( r , ϑ , φ ) , f l m − l − 1 ( r , ϑ , φ ) = R l m ( r , ϑ , φ ) . (908) f_{lm}^{l}(r,\vartheta,\varphi)=\mathfrak{Z}_{lm}(r,\vartheta,\varphi),f_{lm}^{-l-1}(r,\vartheta,\varphi)=\mathfrak{R}_{lm}(r,\vartheta,\varphi). \tag*{(908)} f l m l ( r , ϑ , φ ) = Z l m ( r , ϑ , φ ) , f l m − l − 1 ( r , ϑ , φ ) = R l m ( r , ϑ , φ ) . (908)
Note that along with the above solutions, which are regular at ϑ = 0 , π \vartheta=0,\pi ϑ = 0 , π , the equations under consideration have irregular solutions, which are not discussed here.
Explicit forms of the spherical harmonics and their relations to other functions
According to § 21.2 , Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) may be represented by a product of two functions, one of which depends only on φ \varphi φ while the other depends on ϑ \vartheta ϑ . The φ \varphi φ -dependence of the spherical harmonics is given by the factor e i m φ e^{im\varphi} e im φ . The ϑ \vartheta ϑ -dependence is determined by the associated Legendre polynomials P l m ( cos ϑ ) P_l^m(\cos\vartheta) P l m ( cos ϑ ) [4, 27]. Taking into account the normalization, we get
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! P l m ( cos ϑ ) . (909) Y_{lm}(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}P_l^m(\cos\vartheta). \tag*{(909)} Y l m ( ϑ , φ ) = e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! P l m ( cos ϑ ) . (909)
For the spherical harmonics with ∣ m ∣ ≤ l |m|\le l ∣ m ∣ ≤ l one gets the following expressions (see Refs. [23, 24, 25]).
Differential Expressions for Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ )
Y l m ( ϑ , φ ) = e i m φ 2 l l ! 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( sin ϑ ) − m d l − m ( d cos ϑ ) l − m ( cos 2 ϑ − 1 ) l , (910) Y_{lm}(\vartheta,\varphi)=\frac{e^{im\varphi}}{2^l l!}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!}}(\sin\vartheta)^{-m}\frac{d^{l-m}}{(d\cos\vartheta)^{l-m}}(\cos^2\vartheta-1)^l, \tag*{(910)} Y l m ( ϑ , φ ) = 2 l l ! e im φ 4 π 2 l + 1 ( l − m )! ( l + m )! ( sin ϑ ) − m ( d cos ϑ ) l − m d l − m ( cos 2 ϑ − 1 ) l , (910)
Y l m ( ϑ , φ ) = ( − 1 ) m e i m φ 2 l l ! 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( sin ϑ ) m d l + m ( d cos ϑ ) l + m ( cos 2 ϑ − 1 ) l , (911) Y_{lm}(\vartheta,\varphi)=(-1)^m\frac{e^{im\varphi}}{2^l l!}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}(\sin\vartheta)^m\frac{d^{l+m}}{(d\cos\vartheta)^{l+m}}(\cos^2\vartheta-1)^l, \tag*{(911)} Y l m ( ϑ , φ ) = ( − 1 ) m 2 l l ! e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( sin ϑ ) m ( d cos ϑ ) l + m d l + m ( cos 2 ϑ − 1 ) l , (911)
Y l m ( ϑ , φ ) = e i m φ 2 l 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( cot ϑ 2 ) m ( d d cos ϑ ) l [ ( 1 + cos ϑ ) l − m ( cos ϑ − 1 ) l + m ] , (912) Y_{lm}(\vartheta,\varphi)=\frac{e^{im\varphi}}{2^l}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}\left(\cot\frac{\vartheta}{2}\right)^m\left(\frac{d}{d\cos\vartheta}\right)^l\left[(1+\cos\vartheta)^{l-m}(\cos\vartheta-1)^{l+m}\right], \tag*{(912)} Y l m ( ϑ , φ ) = 2 l e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 ( cot 2 ϑ ) m ( d cos ϑ d ) l [ ( 1 + cos ϑ ) l − m ( cos ϑ − 1 ) l + m ] , (912)
Y l m ( ϑ , φ ) = ( − 1 ) m e i m φ 2 l 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( tan ϑ 2 ) m ( d d cos ϑ ) l [ ( cos ϑ + 1 ) l + m ( cos ϑ − 1 ) l − m ] . (913) Y_{lm}(\vartheta,\varphi)=(-1)^m\frac{e^{im\varphi}}{2^l}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}\left(\tan\frac{\vartheta}{2}\right)^m\left(\frac{d}{d\cos\vartheta}\right)^l\left[(\cos\vartheta+1)^{l+m}(\cos\vartheta-1)^{l-m}\right]. \tag*{(913)} Y l m ( ϑ , φ ) = ( − 1 ) m 2 l e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 ( tan 2 ϑ ) m ( d cos ϑ d ) l [ ( cos ϑ + 1 ) l + m ( cos ϑ − 1 ) l − m ] . (913)
The spherical harmonics may be expressed in terms of m m m th order derivatives of the Legendre polynomials
Y l m ( ϑ , φ ) = ( − 1 ) m e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( sin ϑ ) m d m ( d cos ϑ ) m P l ( cos ϑ ) . (914) Y_{lm}(\vartheta,\varphi)=(-1)^m e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}(\sin\vartheta)^m\frac{d^m}{(d\cos\vartheta)^m}P_l(\cos\vartheta). \tag*{(914)} Y l m ( ϑ , φ ) = ( − 1 ) m e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( sin ϑ ) m ( d cos ϑ ) m d m P l ( cos ϑ ) . (914)
( m ≥ 0 ) (m\ge0) ( m ≥ 0 )
Assuming that
( d d μ ) − ∣ m ∣ f ( μ ) = ∫ 1 μ ∫ 1 μ ⋯ ∫ 1 μ f ( μ ) d μ ⋯ d μ ⏟ ∣ m ∣ (915) \left(\frac{d}{d\mu}\right)^{-|m|}f(\mu)=\underbrace{\int_1^\mu\int_1^\mu\cdots\int_1^\mu f(\mu)\,d\mu\cdots d\mu}_{|m|} \tag*{(915)} ( d μ d ) − ∣ m ∣ f ( μ ) = ∣ m ∣ ∫ 1 μ ∫ 1 μ ⋯ ∫ 1 μ f ( μ ) d μ ⋯ d μ (915)
one can use Eq. (914) not only at m ≥ 0 m\ge0 m ≥ 0 but also at m < 0 m<0 m < 0 .
Representations of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) as a Power Series of Trigonometric Functions of ϑ / 2 \vartheta/2 ϑ /2
In the following equations sums are over all integer values of s s s so that no factorial in the denominator has negative argument. The quantity ξ m 0 \xi_{m0} ξ m 0 is defined as
ξ m 0 = { ( − 1 ) m if m > 0 , 1 if m ≤ 0. (916) \xi_{m0}=
\begin{cases}
(-1)^m & \text{if }m>0,\\
1 & \text{if }m\le0.
\end{cases} \tag*{(916)} ξ m 0 = { ( − 1 ) m 1 if m > 0 , if m ≤ 0. (916)
Y l m ( ϑ , φ ) = ( − 1 ) m e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( tan ϑ 2 ) m ∑ s ( − 1 ) s ( l + s ) ! ( l − s ) ! ( sin ϑ 2 ) 2 s s ! ( s + m ) ! , (917) Y_{lm}(\vartheta,\varphi)=(-1)^m e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!}}\left(\tan\frac{\vartheta}{2}\right)^m\sum_s(-1)^s\frac{(l+s)!}{(l-s)!}\frac{\left(\sin\frac{\vartheta}{2}\right)^{2s}}{s!(s+m)!}, \tag*{(917)} Y l m ( ϑ , φ ) = ( − 1 ) m e im φ 4 π 2 l + 1 ( l − m )! ( l + m )! ( tan 2 ϑ ) m s ∑ ( − 1 ) s ( l − s )! ( l + s )! s ! ( s + m )! ( sin 2 ϑ ) 2 s , (917)
Y l m ( ϑ , φ ) = ( − 1 ) m e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( sin ϑ 2 cos ϑ 2 ) m ∑ s ( − 1 ) s ( l + m + s ) ! ( l − m − s ) ! ( sin ϑ 2 ) 2 s s ! ( s + m ) ! , (918) Y_{lm}(\vartheta,\varphi)=(-1)^m e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}\left(\sin\frac{\vartheta}{2}\cos\frac{\vartheta}{2}\right)^m\sum_s(-1)^s\frac{(l+m+s)!}{(l-m-s)!}\frac{\left(\sin\frac{\vartheta}{2}\right)^{2s}}{s!(s+m)!}, \tag*{(918)} Y l m ( ϑ , φ ) = ( − 1 ) m e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( sin 2 ϑ cos 2 ϑ ) m s ∑ ( − 1 ) s ( l − m − s )! ( l + m + s )! s ! ( s + m )! ( sin 2 ϑ ) 2 s , (918)
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( cot ϑ 2 ) m ∑ s ( − 1 ) l − s ( 2 l − s ) ! s ! ( l − s ) ! ( sin ϑ 2 ) 2 ( l − s ) ( l − m − s ) ! , (919) Y_{lm}(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}\left(\cot\frac{\vartheta}{2}\right)^m\sum_s(-1)^{l-s}\frac{(2l-s)!}{s!(l-s)!}\frac{\left(\sin\frac{\vartheta}{2}\right)^{2(l-s)}}{(l-m-s)!}, \tag*{(919)} Y l m ( ϑ , φ ) = e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( cot 2 ϑ ) m s ∑ ( − 1 ) l − s s ! ( l − s )! ( 2 l − s )! ( l − m − s )! ( sin 2 ϑ ) 2 ( l − s ) , (919)
Y l m ( ϑ , φ ) = ( − 1 ) l e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( cot ϑ 2 ) m ∑ s ( − 1 ) s ( l + s ) ! ( l − s ) ! ( cos ϑ 2 ) 2 s s ! ( s + m ) ! , (920) Y_{lm}(\vartheta,\varphi)=(-1)^l e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!}}\left(\cot\frac{\vartheta}{2}\right)^m\sum_s(-1)^s\frac{(l+s)!}{(l-s)!}\frac{\left(\cos\frac{\vartheta}{2}\right)^{2s}}{s!(s+m)!}, \tag*{(920)} Y l m ( ϑ , φ ) = ( − 1 ) l e im φ 4 π 2 l + 1 ( l − m )! ( l + m )! ( cot 2 ϑ ) m s ∑ ( − 1 ) s ( l − s )! ( l + s )! s ! ( s + m )! ( cos 2 ϑ ) 2 s , (920)
Y l m ( ϑ , φ ) = ( − 1 ) l e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( sin ϑ 2 cos ϑ 2 ) m ∑ s ( − 1 ) s ( l + m + s ) ! ( l − m − s ) ! ( cos ϑ 2 ) 2 s s ! ( s + m ) ! , (921) Y_{lm}(\vartheta,\varphi)=(-1)^l e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}\left(\sin\frac{\vartheta}{2}\cos\frac{\vartheta}{2}\right)^m\sum_s(-1)^s\frac{(l+m+s)!}{(l-m-s)!}\frac{\left(\cos\frac{\vartheta}{2}\right)^{2s}}{s!(s+m)!}, \tag*{(921)} Y l m ( ϑ , φ ) = ( − 1 ) l e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( sin 2 ϑ cos 2 ϑ ) m s ∑ ( − 1 ) s ( l − m − s )! ( l + m + s )! s ! ( s + m )! ( cos 2 ϑ ) 2 s , (921)
Y l m ( ϑ , φ ) = ( − 1 ) l − m e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( tan ϑ 2 ) m ∑ s ( − 1 ) l − s ( 2 l − s ) ! s ! ( l − s ) ! ( cos ϑ 2 ) 2 ( l − s ) ( l − m − s ) ! , (922) Y_{lm}(\vartheta,\varphi)=(-1)^{l-m}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}\left(\tan\frac{\vartheta}{2}\right)^m\sum_s(-1)^{l-s}\frac{(2l-s)!}{s!(l-s)!}\frac{\left(\cos\frac{\vartheta}{2}\right)^{2(l-s)}}{(l-m-s)!}, \tag*{(922)} Y l m ( ϑ , φ ) = ( − 1 ) l − m e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( tan 2 ϑ ) m s ∑ ( − 1 ) l − s s ! ( l − s )! ( 2 l − s )! ( l − m − s )! ( cos 2 ϑ ) 2 ( l − s ) , (922)
Y l m ( ϑ , φ ) = ξ m 0 e i m φ 2 l + 1 4 π l ! ( l + m ) ! ( l − m ) ! ( cos ϑ 2 ) 2 l ∑ s ( − 1 ) s ( tan ϑ 2 ) 2 s + ∣ m ∣ s ! ( s + ∣ m ∣ ) ! ( l − s ) ! ( l − ∣ m ∣ − s ) ! , (923) Y_{lm}(\vartheta,\varphi)=\xi_{m0}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}}\,l!\sqrt{(l+m)!(l-m)!}\left(\cos\frac{\vartheta}{2}\right)^{2l}
\sum_s\frac{(-1)^s\left(\tan\frac{\vartheta}{2}\right)^{2s+|m|}}{s!(s+|m|)!(l-s)!(l-|m|-s)!}, \tag*{(923)} Y l m ( ϑ , φ ) = ξ m 0 e im φ 4 π 2 l + 1 l ! ( l + m )! ( l − m )! ( cos 2 ϑ ) 2 l s ∑ s ! ( s + ∣ m ∣ )! ( l − s )! ( l − ∣ m ∣ − s )! ( − 1 ) s ( tan 2 ϑ ) 2 s + ∣ m ∣ , (923)
Y l m ( ϑ , φ ) = ξ m 0 ( − 1 ) l − m e i m φ 2 l + 1 4 π l ! ( l + m ) ! ( l − m ) ! ( sin ϑ 2 ) 2 l ∑ s ( − 1 ) s ( cot ϑ 2 ) 2 s + ∣ m ∣ s ! ( s + ∣ m ∣ ) ! ( l − s ) ! ( l − ∣ m ∣ − s ) ! , (924) Y_{lm}(\vartheta,\varphi)=\xi_{m0}(-1)^{l-m}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}}\,l!\sqrt{(l+m)!(l-m)!}\left(\sin\frac{\vartheta}{2}\right)^{2l}
\sum_s\frac{(-1)^s\left(\cot\frac{\vartheta}{2}\right)^{2s+|m|}}{s!(s+|m|)!(l-s)!(l-|m|-s)!}, \tag*{(924)} Y l m ( ϑ , φ ) = ξ m 0 ( − 1 ) l − m e im φ 4 π 2 l + 1 l ! ( l + m )! ( l − m )! ( sin 2 ϑ ) 2 l s ∑ s ! ( s + ∣ m ∣ )! ( l − s )! ( l − ∣ m ∣ − s )! ( − 1 ) s ( cot 2 ϑ ) 2 s + ∣ m ∣ , (924)
Representations of Y l m ( ϑ , φ ) Y_l^m(\vartheta,\varphi) Y l m ( ϑ , φ ) as a Power Series of Trigonometric Functions of ϑ \vartheta ϑ
In the equations of this section an integer index s s s assumes either only even or only odd values as indicated under the summation symbols. Sums are over such s s s for which the factorial arguments are non-negative.
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! × { ∑ s = ∣ m ∣ , ∣ m ∣ + 2 , … ( − 1 ) s + m 2 ( l + s ) ! ( s + m ) ! ! ( s − m ) ! ! ( l + m ) ! ! ( l − m ) ! ! ( l + s ) ! ! ( l − s ) ! ! ( sin ϑ ) s , if l − m is even , cos ϑ ∑ s = ∣ m ∣ , ∣ m ∣ + 2 , … ( − 1 ) l + m 2 ( l + s ) ! ( s + m ) ! ! ( s − m ) ! ! ( l + m − 1 ) ! ! ( l − m − 1 ) ! ! ( l + s − 1 ) ! ! ( l − s − 1 ) ! ! ( sin ϑ ) s , if l − m is odd , (925) \begin{aligned}
Y_{lm}(\vartheta,\varphi)&=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}\\
&\quad\times\begin{cases}
\sum_{s=|m|,|m|+2,\ldots}(-1)^{\frac{s+m}{2}}\frac{(l+s)!}{(s+m)!!(s-m)!!}\frac{(l+m)!!(l-m)!!}{(l+s)!!(l-s)!!}(\sin\vartheta)^s,& \text{if }l-m\text{ is even},\\
\cos\vartheta\sum_{s=|m|,|m|+2,\ldots}(-1)^{\frac{l+m}{2}}\frac{(l+s)!}{(s+m)!!(s-m)!!}\frac{(l+m-1)!!(l-m-1)!!}{(l+s-1)!!(l-s-1)!!}(\sin\vartheta)^s,& \text{if }l-m\text{ is odd},
\end{cases} \tag*{(925)}
\end{aligned} Y l m ( ϑ , φ ) = e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 × { ∑ s = ∣ m ∣ , ∣ m ∣ + 2 , … ( − 1 ) 2 s + m ( s + m )!! ( s − m )!! ( l + s )! ( l + s )!! ( l − s )!! ( l + m )!! ( l − m )!! ( sin ϑ ) s , cos ϑ ∑ s = ∣ m ∣ , ∣ m ∣ + 2 , … ( − 1 ) 2 l + m ( s + m )!! ( s − m )!! ( l + s )! ( l + s − 1 )!! ( l − s − 1 )!! ( l + m − 1 )!! ( l − m − 1 )!! ( sin ϑ ) s , if l − m is even , if l − m is odd , (925)
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( sin ϑ ) l × { ∑ s = 0 , 2 , … l − ∣ m ∣ ( − 1 ) l + m − s 2 ( l + m ) ! ! ( l − m ) ! ! ( l + m − s ) ! ! ( l − m − s ) ! ! ( 2 l − s − 1 ) ! ! s ! ! ( sin ϑ ) − s , if l − m is even , cos ϑ ∑ s = 1 , 3 , … l − ∣ m ∣ ( − 1 ) l + m − s 2 ( l + m − 1 ) ! ! ( l − m − 1 ) ! ! ( l + m − s ) ! ! ( l − m − s ) ! ! ( 2 l − s ) ! ! ( s − 1 ) ! ! ( sin ϑ ) − s , if l − m is odd , (926) \begin{aligned}
Y_{lm}(\vartheta,\varphi)&=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}(\sin\vartheta)^l\\
&\quad\times\begin{cases}
\sum_{s=0,2,\ldots}^{l-|m|}(-1)^{\frac{l+m-s}{2}}\frac{(l+m)!!(l-m)!!}{(l+m-s)!!(l-m-s)!!}\frac{(2l-s-1)!!}{s!!}(\sin\vartheta)^{-s},& \text{if }l-m\text{ is even},\\
\cos\vartheta\sum_{s=1,3,\ldots}^{l-|m|}(-1)^{\frac{l+m-s}{2}}\frac{(l+m-1)!!(l-m-1)!!}{(l+m-s)!!(l-m-s)!!}\frac{(2l-s)!!}{(s-1)!!}(\sin\vartheta)^{-s},& \text{if }l-m\text{ is odd},
\end{cases} \tag*{(926)}
\end{aligned} Y l m ( ϑ , φ ) = e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 ( sin ϑ ) l × { ∑ s = 0 , 2 , … l − ∣ m ∣ ( − 1 ) 2 l + m − s ( l + m − s )!! ( l − m − s )!! ( l + m )!! ( l − m )!! s !! ( 2 l − s − 1 )!! ( sin ϑ ) − s , cos ϑ ∑ s = 1 , 3 , … l − ∣ m ∣ ( − 1 ) 2 l + m − s ( l + m − s )!! ( l − m − s )!! ( l + m − 1 )!! ( l − m − 1 )!! ( s − 1 )!! ( 2 l − s )!! ( sin ϑ ) − s , if l − m is even , if l − m is odd , (926)
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( sin ϑ ) m ∑ s ( − 1 ) l + m − s 2 ( l + m + s − 1 ) ! ! ( l − m − s ) ! ! ( cos ϑ ) s s ! , (927) Y_{lm}(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}(\sin\vartheta)^m\sum_s(-1)^{\frac{l+m-s}{2}}\frac{(l+m+s-1)!!}{(l-m-s)!!}\frac{(\cos\vartheta)^s}{s!}, \tag*{(927)} Y l m ( ϑ , φ ) = e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( sin ϑ ) m s ∑ ( − 1 ) 2 l + m − s ( l − m − s )!! ( l + m + s − 1 )!! s ! ( cos ϑ ) s , (927)
l + m − s l+m-s l + m − s is even,
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( cos ϑ ) l ∑ s = ∣ m ∣ , ∣ m ∣ + 2 , … ( − 1 ) s + m 2 1 ( s + m ) ! ! ( s − m ) ! ! ( tan ϑ ) s ( l − s ) ! , (928) Y_{lm}(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}(l+m)!(l-m)!}(\cos\vartheta)^l\sum_{s=|m|,|m|+2,\ldots}(-1)^{\frac{s+m}{2}}\frac{1}{(s+m)!!(s-m)!!}\frac{(\tan\vartheta)^s}{(l-s)!}, \tag*{(928)} Y l m ( ϑ , φ ) = e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( cos ϑ ) l s = ∣ m ∣ , ∣ m ∣ + 2 , … ∑ ( − 1 ) 2 s + m ( s + m )!! ( s − m )!! 1 ( l − s )! ( tan ϑ ) s , (928)
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( sin ϑ ) l ∑ s ( − 1 ) l + m − s 2 1 ( l + m − s ) ! ! ( l − m − s ) ! ! ( cot ϑ ) s s ! , (929) Y_{lm}(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}(l+m)!(l-m)!}(\sin\vartheta)^l\sum_s(-1)^{\frac{l+m-s}{2}}\frac{1}{(l+m-s)!!(l-m-s)!!}\frac{(\cot\vartheta)^s}{s!}, \tag*{(929)} Y l m ( ϑ , φ ) = e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( sin ϑ ) l s ∑ ( − 1 ) 2 l + m − s ( l + m − s )!! ( l − m − s )!! 1 s ! ( cot ϑ ) s , (929)
l + m − s l+m-s l + m − s is even.
The quantity ∣ Y l m ( ϑ , φ ) ∣ 2 |Y_{lm}(\vartheta,\varphi)|^2 ∣ Y l m ( ϑ , φ ) ∣ 2 may be written in the form
∣ Y l m ( ϑ , φ ) ∣ 2 = 2 l + 1 4 π ∑ s = ∣ m ∣ , ∣ m ∣ + 1 , … l ( − 1 ) s + m ( l + s ) ! ( l − s ) ! ( 2 s − 1 ) ! ! ( 2 s ) ! ! ( sin ϑ ) 2 s ( s − m ) ! ( s + m ) ! . (930) |Y_{lm}(\vartheta,\varphi)|^2=\frac{2l+1}{4\pi}\sum_{s=|m|,|m|+1,\ldots}^{l}(-1)^{s+m}\frac{(l+s)!}{(l-s)!}\frac{(2s-1)!!}{(2s)!!}\frac{(\sin\vartheta)^{2s}}{(s-m)!(s+m)!}. \tag*{(930)} ∣ Y l m ( ϑ , φ ) ∣ 2 = 4 π 2 l + 1 s = ∣ m ∣ , ∣ m ∣ + 1 , … ∑ l ( − 1 ) s + m ( l − s )! ( l + s )! ( 2 s )!! ( 2 s − 1 )!! ( s − m )! ( s + m )! ( sin ϑ ) 2 s . (930)
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and the Hypergeometric Functions with Arguments Expressed in Terms of Trigonometric Functions of ϑ / 2 \vartheta/2 ϑ /2
Y l m ( ϑ , φ ) = ξ m 0 e i m φ 2 l + 1 4 π ( l + ∣ m ∣ ) ! ( l − ∣ m ∣ ) ! ( sin ϑ ) ∣ m ∣ ∣ m ∣ ! 2 ∣ m ∣ F ( − l + ∣ m ∣ , l + ∣ m ∣ + 1 ; ∣ m ∣ + 1 ; sin 2 ϑ 2 ) (931) Y_{lm}(\vartheta,\varphi)=\xi_{m0}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+|m|)!}{(l-|m|)!}\frac{(\sin\vartheta)^{|m|}}{|m|!2^{|m|}}}F\left(-l+|m|,l+|m|+1;|m|+1;\sin^2\frac{\vartheta}{2}\right) \tag*{(931)} Y l m ( ϑ , φ ) = ξ m 0 e im φ 4 π 2 l + 1 ( l − ∣ m ∣ )! ( l + ∣ m ∣ )! ∣ m ∣ ! 2 ∣ m ∣ ( sin ϑ ) ∣ m ∣ F ( − l + ∣ m ∣ , l + ∣ m ∣ + 1 ; ∣ m ∣ + 1 ; sin 2 2 ϑ ) (931)
Y l m ( ϑ , φ ) = ( − 1 ) l − m ξ m 0 e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( 2 l ) ! l ! ( sin ϑ 2 ) 2 l ( cot ϑ 2 ) ∣ m ∣ × F ( − l , − l + ∣ m ∣ ; − 2 l ; 1 sin 2 ϑ 2 ) (932) Y_{lm}(\vartheta,\varphi)=(-1)^{l-m}\xi_{m0}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}\frac{(2l)!}{l!}\left(\sin\frac{\vartheta}{2}\right)^{2l}\left(\cot\frac{\vartheta}{2}\right)^{|m|}
\times F\left(-l,-l+|m|;-2l;\frac{1}{\sin^2\frac{\vartheta}{2}}\right) \tag*{(932)} Y l m ( ϑ , φ ) = ( − 1 ) l − m ξ m 0 e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 l ! ( 2 l )! ( sin 2 ϑ ) 2 l ( cot 2 ϑ ) ∣ m ∣ × F ( − l , − l + ∣ m ∣ ; − 2 l ; sin 2 2 ϑ 1 ) (932)
Y l m ( ϑ , φ ) = ( − 1 ) l − m ξ m 0 e i m φ 2 l + 1 4 π ( l + ∣ m ∣ ) ! ( l − ∣ m ∣ ) ! ( sin ϑ ) ∣ m ∣ ∣ m ∣ ! 2 ∣ m ∣ × F ( − l + ∣ m ∣ , l + ∣ m ∣ + 1 ; ∣ m ∣ + 1 ; cos 2 ϑ 2 ) (933) Y_{lm}(\vartheta,\varphi)=(-1)^{l-m}\xi_{m0}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+|m|)!}{(l-|m|)!}\frac{(\sin\vartheta)^{|m|}}{|m|!2^{|m|}}}
\times F\left(-l+|m|,l+|m|+1;|m|+1;\cos^2\frac{\vartheta}{2}\right) \tag*{(933)} Y l m ( ϑ , φ ) = ( − 1 ) l − m ξ m 0 e im φ 4 π 2 l + 1 ( l − ∣ m ∣ )! ( l + ∣ m ∣ )! ∣ m ∣ ! 2 ∣ m ∣ ( sin ϑ ) ∣ m ∣ × F ( − l + ∣ m ∣ , l + ∣ m ∣ + 1 ; ∣ m ∣ + 1 ; cos 2 2 ϑ ) (933)
Y l m ( ϑ , φ ) = ξ m 0 e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( 2 l ) ! l ! ( cos ϑ 2 ) 2 l ( tan ϑ 2 ) ∣ m ∣ × F ( − l , − l + ∣ m ∣ ; − 2 l ; 1 cos 2 ϑ 2 ) (934) Y_{lm}(\vartheta,\varphi)=\xi_{m0}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}\frac{(2l)!}{l!}\left(\cos\frac{\vartheta}{2}\right)^{2l}\left(\tan\frac{\vartheta}{2}\right)^{|m|}
\times F\left(-l,-l+|m|;-2l;\frac{1}{\cos^2\frac{\vartheta}{2}}\right) \tag*{(934)} Y l m ( ϑ , φ ) = ξ m 0 e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 l ! ( 2 l )! ( cos 2 ϑ ) 2 l ( tan 2 ϑ ) ∣ m ∣ × F ( − l , − l + ∣ m ∣ ; − 2 l ; cos 2 2 ϑ 1 ) (934)
Y l m ( ϑ , φ ) = ξ m 0 e i m φ 1 ∣ m ∣ ! 2 l + 1 4 π ( l + ∣ m ∣ ) ! ( l − ∣ m ∣ ) ! ( tan ϑ 2 ) ∣ m ∣ ( cos ϑ 2 ) 2 l F ( − l + ∣ m ∣ , − l ; ∣ m ∣ + 1 ; − tan 2 ϑ 2 ) (935) Y_{lm}(\vartheta,\varphi)=\xi_{m0}e^{im\varphi}\frac{1}{|m|!}\sqrt{\frac{2l+1}{4\pi}\frac{(l+|m|)!}{(l-|m|)!}}\left(\tan\frac{\vartheta}{2}\right)^{|m|}\left(\cos\frac{\vartheta}{2}\right)^{2l}F\left(-l+|m|,-l;|m|+1;-\tan^2\frac{\vartheta}{2}\right) \tag*{(935)} Y l m ( ϑ , φ ) = ξ m 0 e im φ ∣ m ∣ ! 1 4 π 2 l + 1 ( l − ∣ m ∣ )! ( l + ∣ m ∣ )! ( tan 2 ϑ ) ∣ m ∣ ( cos 2 ϑ ) 2 l F ( − l + ∣ m ∣ , − l ; ∣ m ∣ + 1 ; − tan 2 2 ϑ ) (935)
Y l m ( ϑ , φ ) = ( − 1 ) l − m ξ m 0 e i m φ 1 ∣ m ∣ ! 2 l + 1 4 π ( l + ∣ m ∣ ) ! ( l − ∣ m ∣ ) ! ( cot ϑ 2 ) ∣ m ∣ ( sin ϑ 2 ) 2 l × F ( − l + ∣ m ∣ , − l ; ∣ m ∣ + 1 ; − cot 2 ϑ 2 ) (936) Y_{lm}(\vartheta,\varphi)=(-1)^{l-m}\xi_{m0}e^{im\varphi}\frac{1}{|m|!}\sqrt{\frac{2l+1}{4\pi}\frac{(l+|m|)!}{(l-|m|)!}}\left(\cot\frac{\vartheta}{2}\right)^{|m|}\left(\sin\frac{\vartheta}{2}\right)^{2l}
\times F\left(-l+|m|,-l;|m|+1;-\cot^2\frac{\vartheta}{2}\right) \tag*{(936)} Y l m ( ϑ , φ ) = ( − 1 ) l − m ξ m 0 e im φ ∣ m ∣ ! 1 4 π 2 l + 1 ( l − ∣ m ∣ )! ( l + ∣ m ∣ )! ( cot 2 ϑ ) ∣ m ∣ ( sin 2 ϑ ) 2 l × F ( − l + ∣ m ∣ , − l ; ∣ m ∣ + 1 ; − cot 2 2 ϑ ) (936)
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and the Hypergeometric Functions with Arguments Expressed in Terms of Trigonometric Functions of ϑ \vartheta ϑ
Y l m ( ϑ , φ ) = ξ m 0 e i m φ 2 l + 1 4 π ( l + ∣ m ∣ ) ! ( l − ∣ m ∣ ) ! ( sin ϑ ) ∣ m ∣ 2 ∣ m ∣ ∣ m ∣ ! × { F ( − l − ∣ m ∣ 2 , l + ∣ m ∣ + 1 2 ; ∣ m ∣ + 1 ; sin 2 ϑ ) , if l + m is even , cos ϑ F ( − l − ∣ m ∣ − 1 2 , l + ∣ m ∣ + 2 2 ; ∣ m ∣ + 1 ; sin 2 ϑ ) , if l + m is odd , (937) Y_{lm}(\vartheta,\varphi)=\xi_{m0}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+|m|)!}{(l-|m|)!}\frac{(\sin\vartheta)^{|m|}}{2^{|m|}|m|!}}
\times
\begin{cases}
F\left(-\frac{l-|m|}{2},\frac{l+|m|+1}{2};|m|+1;\sin^2\vartheta\right), & \text{if } l+m\text{ is even},\\
\cos\vartheta F\left(-\frac{l-|m|-1}{2},\frac{l+|m|+2}{2};|m|+1;\sin^2\vartheta\right), & \text{if } l+m\text{ is odd},
\end{cases} \tag*{(937)} Y l m ( ϑ , φ ) = ξ m 0 e im φ 4 π 2 l + 1 ( l − ∣ m ∣ )! ( l + ∣ m ∣ )! 2 ∣ m ∣ ∣ m ∣ ! ( sin ϑ ) ∣ m ∣ × ⎩ ⎨ ⎧ F ( − 2 l − ∣ m ∣ , 2 l + ∣ m ∣ + 1 ; ∣ m ∣ + 1 ; sin 2 ϑ ) , cos ϑF ( − 2 l − ∣ m ∣ − 1 , 2 l + ∣ m ∣ + 2 ; ∣ m ∣ + 1 ; sin 2 ϑ ) , if l + m is even , if l + m is odd , (937)
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( 2 l − 1 ) ! ! ( sin ϑ ) l × { ( − 1 ) l + m 2 F ( − l + m 2 , − l − m 2 ; − 2 l − 1 2 ; 1 sin 2 ϑ ) , if l + m is even , ( − 1 ) l + m − 1 2 cot ϑ F ( − l + m − 1 2 , − l − m − 1 2 ; − 2 l − 1 2 ; 1 sin 2 ϑ ) , if l + m is odd , (938) Y_{lm}(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}(2l-1)!!(\sin\vartheta)^l}
\times
\begin{cases}
(-1)^{\frac{l+m}{2}}F\left(-\frac{l+m}{2},-\frac{l-m}{2};-\frac{2l-1}{2};\frac{1}{\sin^2\vartheta}\right), & \text{if } l+m\text{ is even},\\
(-1)^{\frac{l+m-1}{2}}\cot\vartheta F\left(-\frac{l+m-1}{2},-\frac{l-m-1}{2};-\frac{2l-1}{2};\frac{1}{\sin^2\vartheta}\right), & \text{if } l+m\text{ is odd},
\end{cases} \tag*{(938)} Y l m ( ϑ , φ ) = e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 ( 2 l − 1 )!! ( sin ϑ ) l × { ( − 1 ) 2 l + m F ( − 2 l + m , − 2 l − m ; − 2 2 l − 1 ; s i n 2 ϑ 1 ) , ( − 1 ) 2 l + m − 1 cot ϑF ( − 2 l + m − 1 , − 2 l − m − 1 ; − 2 2 l − 1 ; s i n 2 ϑ 1 ) , if l + m is even , if l + m is odd , (938)
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( sin ϑ ) m { ( − 1 ) l + m 2 ( l + m − 1 ) ! ! ( l − m − 1 ) ! ! F ( − l − m 2 , l + m + 1 2 ; 1 2 ; cos 2 ϑ ) if l + m is even ( − 1 ) l + m − 1 2 ( l + m ) ! ! ( l − m ) ! ! cos ϑ F ( − l − m − 1 2 , l + m + 2 2 ; 3 2 ; cos 2 ϑ ) if l + m is odd , (939) Y_{lm}(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}(\sin\vartheta)^m
\begin{cases}
(-1)^{\frac{l+m}{2}}(l+m-1)!!(l-m-1)!!F\left(-\frac{l-m}{2},\frac{l+m+1}{2};\frac{1}{2};\cos^2\vartheta\right) & \text{if }l+m\text{ is even}\\
(-1)^{\frac{l+m-1}{2}}(l+m)!!(l-m)!!\cos\vartheta F\left(-\frac{l-m-1}{2},\frac{l+m+2}{2};\frac{3}{2};\cos^2\vartheta\right) & \text{if }l+m\text{ is odd},
\end{cases} \tag*{(939)} Y l m ( ϑ , φ ) = e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 ( sin ϑ ) m { ( − 1 ) 2 l + m ( l + m − 1 )!! ( l − m − 1 )!! F ( − 2 l − m , 2 l + m + 1 ; 2 1 ; cos 2 ϑ ) ( − 1 ) 2 l + m − 1 ( l + m )!! ( l − m )!! cos ϑF ( − 2 l − m − 1 , 2 l + m + 2 ; 2 3 ; cos 2 ϑ ) if l + m is even if l + m is odd , (939)
Y l m ( ϑ , φ ) = ( − 1 ) m e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( 2 l − 1 ) ! ! ( cos ϑ ) l ( tan ϑ ) m F ( l − m 2 , l − m − 1 2 ; 2 l − 1 2 ; 1 cos 2 ϑ ) , (940) Y_{lm}(\vartheta,\varphi)=(-1)^m e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}(2l-1)!!(\cos\vartheta)^l(\tan\vartheta)^m
F\left(\frac{l-m}{2},\frac{l-m-1}{2};\frac{2l-1}{2};\frac{1}{\cos^2\vartheta}\right), \tag*{(940)} Y l m ( ϑ , φ ) = ( − 1 ) m e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 ( 2 l − 1 )!! ( cos ϑ ) l ( tan ϑ ) m F ( 2 l − m , 2 l − m − 1 ; 2 2 l − 1 ; cos 2 ϑ 1 ) , (940)
Y l m ( ϑ , φ ) = ξ m 0 e i m φ 2 l + 1 4 π ( l + ∣ m ∣ ) ! ( l − ∣ m ∣ ) ! ( cos ϑ ) l ( tan ϑ ) ∣ m ∣ 2 ∣ m ∣ ∣ m ∣ ! F ( − l − ∣ m ∣ 2 , − l − ∣ m ∣ − 1 2 ; ∣ m ∣ + 1 ; − tan 2 ϑ ) , (941) Y_{lm}(\vartheta,\varphi)=\xi_{m0}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+|m|)!}{(l-|m|)!}}(\cos\vartheta)^l\frac{(\tan\vartheta)^{|m|}}{2^{|m|}|m|!}
F\left(-\frac{l-|m|}{2},-\frac{l-|m|-1}{2};|m|+1;-\tan^2\vartheta\right), \tag*{(941)} Y l m ( ϑ , φ ) = ξ m 0 e im φ 4 π 2 l + 1 ( l − ∣ m ∣ )! ( l + ∣ m ∣ )! ( cos ϑ ) l 2 ∣ m ∣ ∣ m ∣ ! ( tan ϑ ) ∣ m ∣ F ( − 2 l − ∣ m ∣ , − 2 l − ∣ m ∣ − 1 ; ∣ m ∣ + 1 ; − tan 2 ϑ ) , (941)
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( sin ϑ ) l { ( − 1 ) l + m 2 ( l + m − 1 ) ! ! ( l − m − 1 ) ! ! F ( − l − m 2 , − l + m 2 ; 1 2 ; − cot 2 ϑ ) if l + m is even ( − 1 ) l + m − 1 2 ( l + m ) ! ! ( l − m ) ! ! cot ϑ F ( − l − m − 1 2 , − l + m − 1 2 ; 3 2 ; − cot 2 ϑ ) if l + m is odd , (942) Y_{lm}(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}(\sin\vartheta)^l
\begin{cases}
(-1)^{\frac{l+m}{2}}(l+m-1)!!(l-m-1)!!F\left(-\frac{l-m}{2},-\frac{l+m}{2};\frac{1}{2};-\cot^2\vartheta\right) & \text{if }l+m\text{ is even}\\
(-1)^{\frac{l+m-1}{2}}(l+m)!!(l-m)!!\cot\vartheta F\left(-\frac{l-m-1}{2},-\frac{l+m-1}{2};\frac{3}{2};-\cot^2\vartheta\right) & \text{if }l+m\text{ is odd},
\end{cases} \tag*{(942)} Y l m ( ϑ , φ ) = e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 ( sin ϑ ) l { ( − 1 ) 2 l + m ( l + m − 1 )!! ( l − m − 1 )!! F ( − 2 l − m , − 2 l + m ; 2 1 ; − cot 2 ϑ ) ( − 1 ) 2 l + m − 1 ( l + m )!! ( l − m )!! cot ϑF ( − 2 l − m − 1 , − 2 l + m − 1 ; 2 3 ; − cot 2 ϑ ) if l + m is even if l + m is odd , (942)
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and the Hypergeometric Functions with Arguments Expressed in Terms of Exponential Functions
Y l m ( ϑ , φ ) = i e i m φ π 2 l + 1 4 π ( l + m ) ! ( l − m ) ! 2 l + m + 1 ( sin ϑ ) m ( 2 l + 1 ) ! ! { e − i ( l + m + 1 ) ϑ F ( m + 1 2 , l + m + 1 ; l + 3 2 ; e − 2 i ϑ ) − e i ( l + m + 1 ) ϑ F ( m + 1 2 , l + m + 1 ; l + 3 2 ; e 2 i ϑ ) } , (943) Y_{lm}(\vartheta,\varphi)=\frac{ie^{im\varphi}}{\pi}\sqrt{\frac{2l+1}{4\pi}(l+m)!(l-m)!}\frac{2^{l+m+1}(\sin\vartheta)^m}{(2l+1)!!}
\left\{e^{-i(l+m+1)\vartheta}F\left(m+\frac{1}{2},l+m+1;l+\frac{3}{2};e^{-2i\vartheta}\right)-e^{i(l+m+1)\vartheta}F\left(m+\frac{1}{2},l+m+1;l+\frac{3}{2};e^{2i\vartheta}\right)\right\}, \tag*{(943)} Y l m ( ϑ , φ ) = π i e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( 2 l + 1 )!! 2 l + m + 1 ( sin ϑ ) m { e − i ( l + m + 1 ) ϑ F ( m + 2 1 , l + m + 1 ; l + 2 3 ; e − 2 i ϑ ) − e i ( l + m + 1 ) ϑ F ( m + 2 1 , l + m + 1 ; l + 2 3 ; e 2 i ϑ ) } , (943)
Y l m ( ϑ , φ ) = i π e i m φ 2 l + 1 8 π ( l + m ) ! ( l − m ) ! 2 l + 1 ( 2 l + 1 ) ! ! sin ϑ { e − i [ ( 2 l + 1 ) ϑ 2 + ( 2 m + 1 ) π 4 ] F ( m + 1 2 , 1 2 − m ; l + 3 2 ; − e − i ( ϑ + π 2 ) 2 sin ϑ ) − e i [ ( 2 l + 1 ) ϑ 2 + ( 2 m + 1 ) π 4 ] F ( m + 1 2 , 1 2 − m ; l + 3 2 ; − e i ( ϑ + π 2 ) 2 sin ϑ ) } . (944) Y_{lm}(\vartheta,\varphi)=\frac{i}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{8\pi}(l+m)!(l-m)!}\frac{2^{l+1}}{(2l+1)!!\sqrt{\sin\vartheta}}
\left\{e^{-i\left[\frac{(2l+1)\vartheta}{2}+\frac{(2m+1)\pi}{4}\right]}F\left(m+\frac{1}{2},\frac{1}{2}-m;l+\frac{3}{2};-\frac{e^{-i(\vartheta+\frac{\pi}{2})}}{2\sin\vartheta}\right)-e^{i\left[\frac{(2l+1)\vartheta}{2}+\frac{(2m+1)\pi}{4}\right]}F\left(m+\frac{1}{2},\frac{1}{2}-m;l+\frac{3}{2};-\frac{e^{i(\vartheta+\frac{\pi}{2})}}{2\sin\vartheta}\right)\right\}. \tag*{(944)} Y l m ( ϑ , φ ) = π i e im φ 8 π 2 l + 1 ( l + m )! ( l − m )! ( 2 l + 1 )!! sin ϑ 2 l + 1 { e − i [ 2 ( 2 l + 1 ) ϑ + 4 ( 2 m + 1 ) π ] F ( m + 2 1 , 2 1 − m ; l + 2 3 ; − 2 sin ϑ e − i ( ϑ + 2 π ) ) − e i [ 2 ( 2 l + 1 ) ϑ + 4 ( 2 m + 1 ) π ] F ( m + 2 1 , 2 1 − m ; l + 2 3 ; − 2 sin ϑ e i ( ϑ + 2 π ) ) } . (944)
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and Other Special Functions
(a) The relation between Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and the Wigner D-function (Chap. 4)
Y l m ( ϑ , φ ) = 2 l + 1 4 π D 0 − m l ( χ , ϑ , φ ) = ( − 1 ) m 2 l + 1 4 π D 0 m l ∗ ( χ , ϑ , φ ) = ( − 1 ) m 2 l + 1 4 π D − m 0 l ( φ , ϑ , χ ) = 2 l + 1 4 π D m 0 l ∗ ( φ , ϑ , χ ) , (945) Y_{lm}(\vartheta,\varphi)=\sqrt{\frac{2l+1}{4\pi}}D^l_{0-m}(\chi,\vartheta,\varphi)=(-1)^m\sqrt{\frac{2l+1}{4\pi}}D^{l*}_{0m}(\chi,\vartheta,\varphi)
=\left(-1\right)^m\sqrt{\frac{2l+1}{4\pi}}D^l_{-m0}(\varphi,\vartheta,\chi)=\sqrt{\frac{2l+1}{4\pi}}D^{l*}_{m0}(\varphi,\vartheta,\chi), \tag*{(945)} Y l m ( ϑ , φ ) = 4 π 2 l + 1 D 0 − m l ( χ , ϑ , φ ) = ( − 1 ) m 4 π 2 l + 1 D 0 m l ∗ ( χ , ϑ , φ ) = ( − 1 ) m 4 π 2 l + 1 D − m 0 l ( φ , ϑ , χ ) = 4 π 2 l + 1 D m 0 l ∗ ( φ , ϑ , χ ) , (945)
where χ \chi χ is an arbitrary angle.
(b) Y l m ( ϑ , φ ) Y_l^m(\vartheta,\varphi) Y l m ( ϑ , φ ) can be related to the Jacobi polynomials P l − ∣ m ∣ ( ∣ m ∣ , ∣ m ∣ ) ( x ) P_{l-|m|}^{(|m|,|m|)}(x) P l − ∣ m ∣ ( ∣ m ∣ , ∣ m ∣ ) ( x )
Y l m ( ϑ , φ ) = ξ m 0 e i m φ 2 ∣ m ∣ ! 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( sin ϑ ) ∣ m ∣ P l − ∣ m ∣ ( ∣ m ∣ , ∣ m ∣ ) ( cos ϑ ) . (946) Y_l^m(\vartheta,\varphi)=\frac{\xi_{m0}e^{im\varphi}}{2|m|!}\sqrt{\frac{2l+1}{4\pi}(l+m)!(l-m)!}(\sin\vartheta)^{|m|}P_{l-|m|}^{(|m|,|m|)}(\cos\vartheta). \tag*{(946)} Y l m ( ϑ , φ ) = 2∣ m ∣ ! ξ m 0 e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( sin ϑ ) ∣ m ∣ P l − ∣ m ∣ ( ∣ m ∣ , ∣ m ∣ ) ( cos ϑ ) . (946)
(c) Y l m ( ϑ , φ ) Y_l^m(\vartheta,\varphi) Y l m ( ϑ , φ ) can be expressed in terms of Gegenbauer polynomials C ν λ ( x ) C_\nu^\lambda(x) C ν λ ( x )
Y l m ( ϑ , φ ) = ξ m 0 e i m φ 2 l + 1 4 π ( l − ∣ m ∣ ) ! ( l + ∣ m ∣ ) ! ( 2 ∣ m ∣ − 1 ) ! ! ( sin ϑ ) ∣ m ∣ C l − ∣ m ∣ 1 2 + ∣ m ∣ ( cos ϑ ) . (947) Y_l^m(\vartheta,\varphi)=\xi_{m0}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-|m|)!}{(l+|m|)!}}(2|m|-1)!!(\sin\vartheta)^{|m|}C_{l-|m|}^{\frac{1}{2}+|m|}(\cos\vartheta). \tag*{(947)} Y l m ( ϑ , φ ) = ξ m 0 e im φ 4 π 2 l + 1 ( l + ∣ m ∣ )! ( l − ∣ m ∣ )! ( 2∣ m ∣ − 1 )!! ( sin ϑ ) ∣ m ∣ C l − ∣ m ∣ 2 1 + ∣ m ∣ ( cos ϑ ) . (947)
Y l m ( ϑ , φ ) Y_l^m(\vartheta,\varphi) Y l m ( ϑ , φ ) as an Irreducible Tensor Product
For any position vector r \boldsymbol{r} r specified by polar coordinates r , ϑ , φ r,\vartheta,\varphi r , ϑ , φ one has
Y l m ( ϑ , φ ) = 1 r l ( 2 l + 1 ) ! ! 4 π l ! { … { { r ⊗ r } 2 ⊗ r } 3 … ⊗ r } l m . (948) Y_l^m(\vartheta,\varphi)=\frac{1}{r^l}\sqrt{\frac{(2l+1)!!}{4\pi l!}}\left\{\ldots\left\{\left\{\boldsymbol{r}\otimes\boldsymbol{r}\right\}_2\otimes\boldsymbol{r}\right\}_3\ldots\otimes\boldsymbol{r}\right\}_{lm}. \tag*{(948)} Y l m ( ϑ , φ ) = r l 1 4 π l ! ( 2 l + 1 )!! { … { { r ⊗ r } 2 ⊗ r } 3 … ⊗ r } l m . (948)
Integral representations of the spherical harmonics [4,22,27]
Y l m ( ϑ , φ ) Y_l^m(\vartheta,\varphi) Y l m ( ϑ , φ ) in the Form of Indefinite Integrals
(a)
Y l m ( ϑ , φ ) = e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! 1 ( sin ϑ ) m ∫ 1 cos ϑ ∫ 1 cos ϑ ⋯ ∫ 1 cos ϑ P l ( cos ϑ ) ( d cos ϑ ) m , ( m ≥ 0 ) . (949) Y_l^m(\vartheta,\varphi)=e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!}}\frac{1}{(\sin\vartheta)^m}\int_1^{\cos\vartheta}\int_1^{\cos\vartheta}\cdots\int_1^{\cos\vartheta}P_l(\cos\vartheta)(d\cos\vartheta)^m,\qquad(m\ge0). \tag*{(949)} Y l m ( ϑ , φ ) = e im φ 4 π 2 l + 1 ( l − m )! ( l + m )! ( sin ϑ ) m 1 ∫ 1 c o s ϑ ∫ 1 c o s ϑ ⋯ ∫ 1 c o s ϑ P l ( cos ϑ ) ( d cos ϑ ) m , ( m ≥ 0 ) . (949)
This formula represents the analytic continuation of the differential expression (5.2.6) to the case m < 0 m<0 m < 0 .
(b) The Mehler-Dirichlet formulas
Y l m ( ϑ , φ ) = ( − 1 ) m 2 π e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( 2 m − 1 ) ! ! ( sin ϑ 2 ) m ∫ 0 ϑ cos [ ( 2 l + 1 ) ψ 2 ] ( cos ψ − cos ϑ ) m + 1 2 d ψ , ( m ≥ 0 ) (950) Y_l^m(\vartheta,\varphi)=(-1)^m\frac{\sqrt{2}}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}(2m-1)!!\left(\frac{\sin\vartheta}{2}\right)^m\int_0^\vartheta\frac{\cos\left[\frac{(2l+1)\psi}{2}\right]}{(\cos\psi-\cos\vartheta)^{m+\frac{1}{2}}}\,d\psi,\qquad(m\ge0) \tag*{(950)} Y l m ( ϑ , φ ) = ( − 1 ) m π 2 e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( 2 m − 1 )!! ( 2 sin ϑ ) m ∫ 0 ϑ ( cos ψ − cos ϑ ) m + 2 1 cos [ 2 ( 2 l + 1 ) ψ ] d ψ , ( m ≥ 0 ) (950)
Y l m ( ϑ , φ ) = 2 π e i m φ 2 l + 1 4 π ( l − m ) ! ( l + m ) ! ( 2 m − 1 ) ! ! ( sin ϑ 2 ) m ∫ ϑ π sin [ ( 2 l + 1 ) ψ 2 ] ( cos ϑ − cos ψ ) m + 1 2 d ψ , ( m ≥ 0 ) . (951) Y_l^m(\vartheta,\varphi)=\frac{\sqrt{2}}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!}{(l+m)!}}(2m-1)!!\left(\frac{\sin\vartheta}{2}\right)^m\int_\vartheta^\pi\frac{\sin\left[\frac{(2l+1)\psi}{2}\right]}{(\cos\vartheta-\cos\psi)^{m+\frac{1}{2}}}\,d\psi,\qquad(m\ge0). \tag*{(951)} Y l m ( ϑ , φ ) = π 2 e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! ( 2 m − 1 )!! ( 2 sin ϑ ) m ∫ ϑ π ( cos ϑ − cos ψ ) m + 2 1 sin [ 2 ( 2 l + 1 ) ψ ] d ψ , ( m ≥ 0 ) . (951)
(c)
Y l m ( ϑ , φ ) = ( − 1 ) m e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! 1 ( m − 1 ) ! ( sin ϑ ) m ∫ cos ϑ 1 P l ( cos ψ ) [ cos ψ − cos ϑ ] m − 1 d cos ψ , ( m > 0 ) . (952) Y_l^m(\vartheta,\varphi)=(-1)^m e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!}}\frac{1}{(m-1)!(\sin\vartheta)^m}\int_{\cos\vartheta}^{1}P_l(\cos\psi)[\cos\psi-\cos\vartheta]^{m-1}\,d\cos\psi,\qquad(m>0). \tag*{(952)} Y l m ( ϑ , φ ) = ( − 1 ) m e im φ 4 π 2 l + 1 ( l − m )! ( l + m )! ( m − 1 )! ( sin ϑ ) m 1 ∫ c o s ϑ 1 P l ( cos ψ ) [ cos ψ − cos ϑ ] m − 1 d cos ψ , ( m > 0 ) . (952)
Y l m ( ϑ , φ ) Y_l^m(\vartheta,\varphi) Y l m ( ϑ , φ ) in the Form of Definite Integrals
(a)
Y l m ( ϑ , φ ) = ( ± i ) m π e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! 1 l ! ∫ 0 π [ cos ϑ ± i sin ϑ cos ψ ] l cos ( m ψ ) d ψ (953) Y_l^m(\vartheta,\varphi)=\frac{(\pm i)^m}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}(l+m)!(l-m)!}\frac{1}{l!}\int_0^\pi[\cos\vartheta\pm i\sin\vartheta\cos\psi]^l\cos(m\psi)\,d\psi \tag*{(953)} Y l m ( ϑ , φ ) = π ( ± i ) m e im φ 4 π 2 l + 1 ( l + m )! ( l − m )! l ! 1 ∫ 0 π [ cos ϑ ± i sin ϑ cos ψ ] l cos ( m ψ ) d ψ (953)
Y l m ( ϑ , φ ) = ( ± i ) m π e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! 1 l ! ∫ 0 π cos ( m ψ ) d ψ ( cos ϑ ∓ i sin ϑ cos ψ ) l + 1 (954) Y_l^m(\vartheta,\varphi)=\frac{(\pm i)^m}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}\frac{1}{l!}\int_0^\pi\frac{\cos(m\psi)\,d\psi}{(\cos\vartheta\mp i\sin\vartheta\cos\psi)^{l+1}} \tag*{(954)} Y l m ( ϑ , φ ) = π ( ± i ) m e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 l ! 1 ∫ 0 π ( cos ϑ ∓ i sin ϑ cos ψ ) l + 1 cos ( m ψ ) d ψ (954)
For φ = 0 \varphi=0 φ = 0 , complex conjugation does not affect the right-hand side of Eq. (954) , because Y l m ( ϑ , 0 ) Y_l^m(\vartheta,0) Y l m ( ϑ , 0 ) is real.
(b)
Y l m ( ϑ , φ ) = ( ± i ) m 2 π 2 l + 1 4 π ( l + m ) ! ( l − m ) ! l ! ∫ 0 2 π [ cos ϑ ± i sin ϑ cos ( ψ − φ ) ] l e i m ψ d ψ (955) Y_{lm}(\vartheta,\varphi)=\frac{(\pm i)^m}{2\pi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!(l-m)!}{l!}}\int_{0}^{2\pi}[\cos\vartheta\pm i\sin\vartheta\cos(\psi-\varphi)]^l e^{im\psi}\,d\psi \tag*{(955)} Y l m ( ϑ , φ ) = 2 π ( ± i ) m 4 π 2 l + 1 l ! ( l + m )! ( l − m )! ∫ 0 2 π [ cos ϑ ± i sin ϑ cos ( ψ − φ ) ] l e im ψ d ψ (955)
Y l m ( ϑ , φ ) = ( ± i ) m 2 π 2 l + 1 4 π ( l + m ) ! ( l − m ) ! l ! ∫ 0 2 π [ cos ϑ ∓ i sin ϑ cos ( ψ − φ ) ] − l − 1 e i m ψ d ψ (956) Y_{lm}(\vartheta,\varphi)=\frac{(\pm i)^m}{2\pi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}l!\int_{0}^{2\pi}[\cos\vartheta\mp i\sin\vartheta\cos(\psi-\varphi)]^{-l-1}e^{im\psi}\,d\psi \tag*{(956)} Y l m ( ϑ , φ ) = 2 π ( ± i ) m 4 π ( l + m )! ( l − m )! 2 l + 1 l ! ∫ 0 2 π [ cos ϑ ∓ i sin ϑ cos ( ψ − φ ) ] − l − 1 e im ψ d ψ (956)
Equations (5.3.7) and (5.3.8) represent modifications of Eqs. (5.3.5) and (5.3.6).
(c)
Y l m ( ϑ , φ ) = ( − 1 ) m π e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( 2 m − 1 ) ! ! ( sin ϑ ) m ∫ 0 π ( cos ϑ ± i sin ϑ cos ψ ) l − m ( sin ψ ) 2 m d ψ , ( m ≥ 0 ) (957) Y_{lm}(\vartheta,\varphi)=\frac{(-1)^m}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!(2m-1)!!}}(\sin\vartheta)^m\int_{0}^{\pi}(\cos\vartheta\pm i\sin\vartheta\cos\psi)^{l-m}(\sin\psi)^{2m}\,d\psi,\quad(m\geq0) \tag*{(957)} Y l m ( ϑ , φ ) = π ( − 1 ) m e im φ 4 π 2 l + 1 ( l − m )! ( 2 m − 1 )!! ( l + m )! ( sin ϑ ) m ∫ 0 π ( cos ϑ ± i sin ϑ cos ψ ) l − m ( sin ψ ) 2 m d ψ , ( m ≥ 0 ) (957)
Y l m ( ϑ , φ ) = ( − 1 ) m π e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( 2 m − 1 ) ! ! ( sin ϑ ) m ∫ 0 π ( sin χ ) 2 m d χ ( cos ϑ ∓ i sin ϑ cos χ ) l + m + 1 , ( m ≥ 0 ) (958) Y_{lm}(\vartheta,\varphi)=\frac{(-1)^m}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!(2m-1)!!}}(\sin\vartheta)^m\int_{0}^{\pi}\frac{(\sin\chi)^{2m}\,d\chi}{(\cos\vartheta\mp i\sin\vartheta\cos\chi)^{l+m+1}},\quad(m\geq0) \tag*{(958)} Y l m ( ϑ , φ ) = π ( − 1 ) m e im φ 4 π 2 l + 1 ( l − m )! ( 2 m − 1 )!! ( l + m )! ( sin ϑ ) m ∫ 0 π ( cos ϑ ∓ i sin ϑ cos χ ) l + m + 1 ( sin χ ) 2 m d χ , ( m ≥ 0 ) (958)
Equations (5.3.9) and (5.3.10) transform to each other at replacing the integration variables according to
cos ψ = cos ϑ cos χ + i sin ϑ cos ϑ + i sin ϑ cos χ , cos χ = cos ϑ cos ψ − i sin ϑ cos ϑ − i sin ϑ cos ψ . (959) \cos\psi=\frac{\cos\vartheta\cos\chi+i\sin\vartheta}{\cos\vartheta+i\sin\vartheta\cos\chi},\qquad\cos\chi=\frac{\cos\vartheta\cos\psi-i\sin\vartheta}{\cos\vartheta-i\sin\vartheta\cos\psi}. \tag*{(959)} cos ψ = cos ϑ + i sin ϑ cos χ cos ϑ cos χ + i sin ϑ , cos χ = cos ϑ − i sin ϑ cos ψ cos ϑ cos ψ − i sin ϑ . (959)
This replacing is equivalent to the transformation of ”mirror” symmetry, i.e., to the replacement l l l by l = − l − 1 l=-l-1 l = − l − 1 .
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) in the Form of Improper Integrals
(a)
Y l m ( ϑ , φ ) = i m + 1 π e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! l ! { ( − 1 ) m ∫ 0 ∞ cosh ( m t ) d t ( cos ϑ + i sin ϑ cosh t ) l + 1 − ∫ 0 ∞ cosh ( m t ) d t ( cos ϑ − i sin ϑ cosh t ) l + 1 } , (960) Y_{lm}(\vartheta,\varphi)=\frac{i^{m+1}}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}l!
\left\{(-1)^m\int_{0}^{\infty}\frac{\cosh(mt)\,dt}{(\cos\vartheta+i\sin\vartheta\cosh t)^{l+1}}-\int_{0}^{\infty}\frac{\cosh(mt)\,dt}{(\cos\vartheta-i\sin\vartheta\cosh t)^{l+1}}\right\}, \tag*{(960)} Y l m ( ϑ , φ ) = π i m + 1 e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 l ! { ( − 1 ) m ∫ 0 ∞ ( cos ϑ + i sin ϑ cosh t ) l + 1 cosh ( m t ) d t − ∫ 0 ∞ ( cos ϑ − i sin ϑ cosh t ) l + 1 cosh ( m t ) d t } , (960)
Y l m ( ϑ , φ ) = i π e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( 2 m − 1 ) ! ! ( sin ϑ ) m { ∫ 0 ∞ ( sinh t ) 2 m d t ( cos ϑ + i sin ϑ cosh t ) l + m + 1 − ∫ 0 ∞ ( sinh t ) 2 m d t ( cos ϑ − i sin ϑ cosh t ) l + m + 1 } , ( m ≥ 0 ) . (961) Y_{lm}(\vartheta,\varphi)=\frac{i}{\pi}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!(2m-1)!!}}\frac{(\sin\vartheta)^m}{}
\left\{\int_{0}^{\infty}\frac{(\sinh t)^{2m}\,dt}{(\cos\vartheta+i\sin\vartheta\cosh t)^{l+m+1}}-\int_{0}^{\infty}\frac{(\sinh t)^{2m}\,dt}{(\cos\vartheta-i\sin\vartheta\cosh t)^{l+m+1}}\right\},\quad(m\geq0). \tag*{(961)} Y l m ( ϑ , φ ) = π i e im φ 4 π 2 l + 1 ( l − m )! ( 2 m − 1 )!! ( l + m )! ( sin ϑ ) m { ∫ 0 ∞ ( cos ϑ + i sin ϑ cosh t ) l + m + 1 ( sinh t ) 2 m d t − ∫ 0 ∞ ( cos ϑ − i sin ϑ cosh t ) l + m + 1 ( sinh t ) 2 m d t } , ( m ≥ 0 ) . (961)
(b)
Y l m ( ϑ , φ ) = ( − 1 ) m e i m φ 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ∫ 0 ∞ e − k cos ϑ J m ( k sin ϑ ) k l d k . (962) Y_{lm}(\vartheta,\varphi)=(-1)^m e^{im\varphi}\sqrt{\frac{2l+1}{4\pi(l+m)!(l-m)!}}\int_{0}^{\infty}e^{-k\cos\vartheta}J_m(k\sin\vartheta)k^l\,dk. \tag*{(962)} Y l m ( ϑ , φ ) = ( − 1 ) m e im φ 4 π ( l + m )! ( l − m )! 2 l + 1 ∫ 0 ∞ e − k c o s ϑ J m ( k sin ϑ ) k l d k . (962)
where J m ( x ) J_m(x) J m ( x ) is a Bessel function.
(c) ∣ Y l m ( ϑ , φ ) ∣ 2 |Y_{lm}(\vartheta,\varphi)|^2 ∣ Y l m ( ϑ , φ ) ∣ 2 may be represented as
∣ Y l m ( ϑ , φ ) ∣ 2 = 2 l + 1 4 π ∫ 0 ∞ [ J m ( t sin ϑ 2 ) ] 2 J 2 l + 1 ( t ) d t . (963) |Y_{lm}(\vartheta,\varphi)|^2=\frac{2l+1}{4\pi}\int_{0}^{\infty}\left[J_m\left(\frac{t\sin\vartheta}{2}\right)\right]^2J_{2l+1}(t)\,dt. \tag*{(963)} ∣ Y l m ( ϑ , φ ) ∣ 2 = 4 π 2 l + 1 ∫ 0 ∞ [ J m ( 2 t sin ϑ ) ] 2 J 2 l + 1 ( t ) d t . (963)
Symmetry properties
The symmetry relations given below couple the harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) with different values of ϑ \vartheta ϑ , φ \varphi φ and l , m l,m l , m . These relations permit us to extend the domain of allowed ϑ , φ \vartheta,\varphi ϑ , φ and generalise Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) to the case of negative l l l . (a) Complex conjugation
Y l m ∗ ( ϑ , φ ) = Y l m ( ϑ , − φ ) = ( − 1 ) m Y l , − m ( ϑ , φ ) . (964) Y_{lm}^{*}(\vartheta,\varphi)=Y_{lm}(\vartheta,-\varphi)=(-1)^mY_{l,-m}(\vartheta,\varphi). \tag*{(964)} Y l m ∗ ( ϑ , φ ) = Y l m ( ϑ , − φ ) = ( − 1 ) m Y l , − m ( ϑ , φ ) . (964)
(b) Sign reversal of m m m
Y l , − m ( ϑ , φ ) = ( − 1 ) m Y l m ( ϑ , − φ ) = ( − 1 ) m e − i 2 m φ Y l m ( ϑ , φ ) . (965) Y_{l,-m}(\vartheta,\varphi)=(-1)^mY_{lm}(\vartheta,-\varphi)=(-1)^m e^{-i2m\varphi}Y_{lm}(\vartheta,\varphi). \tag*{(965)} Y l , − m ( ϑ , φ ) = ( − 1 ) m Y l m ( ϑ , − φ ) = ( − 1 ) m e − i 2 m φ Y l m ( ϑ , φ ) . (965)
(c) ”Mirror” symmetry (replacement l l l by l ˉ = − l − 1 \bar{l}=-l-1 l ˉ = − l − 1 )
Y l ˉ m ( ϑ , φ ) = ( − 1 ) m Y l m ( ϑ , φ ) (966) Y_{\bar{l}m}(\vartheta,\varphi)=(-1)^mY_{lm}(\vartheta,\varphi) \tag*{(966)} Y l ˉ m ( ϑ , φ ) = ( − 1 ) m Y l m ( ϑ , φ ) (966)
(d) Replacement ϑ → π − ϑ \vartheta\to\pi-\vartheta ϑ → π − ϑ and φ = π + φ \varphi=\pi+\varphi φ = π + φ
Y l m ( π − ϑ , φ ) = ( − 1 ) l + m Y l m ( ϑ , φ ) (967) Y_{lm}(\pi-\vartheta,\varphi)=(-1)^{l+m}Y_{lm}(\vartheta,\varphi) \tag*{(967)} Y l m ( π − ϑ , φ ) = ( − 1 ) l + m Y l m ( ϑ , φ ) (967)
Y l m ( ϑ , π + φ ) = ( − 1 ) m Y l m ( ϑ , φ ) (968) Y_{lm}(\vartheta,\pi+\varphi)=(-1)^mY_{lm}(\vartheta,\varphi) \tag*{(968)} Y l m ( ϑ , π + φ ) = ( − 1 ) m Y l m ( ϑ , φ ) (968)
Y l m ( π − ϑ , π + φ ) = ( − 1 ) l Y l m ( ϑ , φ ) (969) Y_{lm}(\pi-\vartheta,\pi+\varphi)=(-1)^lY_{lm}(\vartheta,\varphi) \tag*{(969)} Y l m ( π − ϑ , π + φ ) = ( − 1 ) l Y l m ( ϑ , φ ) (969)
(e) Change of argument signs
Y l m ( − ϑ , φ ) = ( − 1 ) m Y l m ( ϑ , φ ) (970) Y_{lm}(-\vartheta,\varphi)=(-1)^mY_{lm}(\vartheta,\varphi) \tag*{(970)} Y l m ( − ϑ , φ ) = ( − 1 ) m Y l m ( ϑ , φ ) (970)
Y l m ( ϑ , − φ ) = ( − 1 ) m Y l , − m ( ϑ , φ ) (971) Y_{lm}(\vartheta,-\varphi)=(-1)^mY_{l,-m}(\vartheta,\varphi) \tag*{(971)} Y l m ( ϑ , − φ ) = ( − 1 ) m Y l , − m ( ϑ , φ ) (971)
Y l m ( − ϑ , − φ ) = Y l , − m ( ϑ , φ ) (972) Y_{lm}(-\vartheta,-\varphi)=Y_{l,-m}(\vartheta,\varphi) \tag*{(972)} Y l m ( − ϑ , − φ ) = Y l , − m ( ϑ , φ ) (972)
(f) The periodicity in ϑ \vartheta ϑ and φ \varphi φ
Y l m ( ϑ ± π n , φ ) = { ( − 1 ) l Y l m ( ϑ , φ ) , if n is odd Y l m ( ϑ , φ ) , if n is even (973) Y_{lm}(\vartheta\pm\pi n,\varphi)=
\begin{cases}
(-1)^lY_{lm}(\vartheta,\varphi), & \text{if } n\text{ is odd}\\
Y_{lm}(\vartheta,\varphi), & \text{if } n\text{ is even}
\end{cases} \tag*{(973)} Y l m ( ϑ ± π n , φ ) = { ( − 1 ) l Y l m ( ϑ , φ ) , Y l m ( ϑ , φ ) , if n is odd if n is even (973)
Y l m ( ϑ , φ ± n π ) = { ( − 1 ) m Y l m ( ϑ , φ ) , if n is odd Y l m ( ϑ , φ ) if n is even (974) Y_{lm}(\vartheta,\varphi\pm n\pi)=
\begin{cases}
(-1)^mY_{lm}(\vartheta,\varphi), & \text{if } n\text{ is odd}\\
Y_{lm}(\vartheta,\varphi) & \text{if } n\text{ is even}
\end{cases} \tag*{(974)} Y l m ( ϑ , φ ± nπ ) = { ( − 1 ) m Y l m ( ϑ , φ ) , Y l m ( ϑ , φ ) if n is odd if n is even (974)
Making use of the above symmetry properties, one gets
Y l m ( ϑ , φ ) = ( − 1 ) m e 2 i m φ Y l , − m ( ϑ , φ ) = e 2 i m φ Y l m ∗ ( ϑ , φ ) = ( − 1 ) m Y l m ( − ϑ , φ ) = e 2 i m φ Y l , − m ( − ϑ , φ ) = ( − 1 ) m Y l , − m ∗ ( ϑ , φ ) = ( − 1 ) m Y l , − m ( ϑ , − φ ) = e 2 i m φ Y l m ( ϑ , − φ ) = ( − 1 ) m e 2 i m φ Y l m ∗ ( − ϑ , φ ) = Y l , − m ∗ ( − ϑ , φ ) = Y l , − m ( − ϑ , − φ ) = ( − 1 ) m e 2 i m φ Y l m ( − ϑ , − φ ) = e 2 i m φ Y l , − m ∗ ( − ϑ , − φ ) = ( − 1 ) m Y l m ∗ ( − ϑ , − φ ) . (975) \begin{aligned}
Y_{lm}(\vartheta,\varphi)
&=(-1)^m e^{2im\varphi}Y_{l,-m}(\vartheta,\varphi)
=e^{2im\varphi}Y_{lm}^{*}(\vartheta,\varphi)\\
&=(-1)^mY_{lm}(-\vartheta,\varphi)
=e^{2im\varphi}Y_{l,-m}(-\vartheta,\varphi)
=(-1)^mY_{l,-m}^{*}(\vartheta,\varphi)\\
&=(-1)^mY_{l,-m}(\vartheta,-\varphi)
=e^{2im\varphi}Y_{lm}(\vartheta,-\varphi)
=(-1)^m e^{2im\varphi}Y_{lm}^{*}(-\vartheta,\varphi)
=Y_{l,-m}^{*}(-\vartheta,\varphi)\\
&=Y_{l,-m}(-\vartheta,-\varphi)
=(-1)^m e^{2im\varphi}Y_{lm}(-\vartheta,-\varphi)
=e^{2im\varphi}Y_{l,-m}^{*}(-\vartheta,-\varphi)
=(-1)^mY_{lm}^{*}(-\vartheta,-\varphi). \tag*{(975)}
\end{aligned} Y l m ( ϑ , φ ) = ( − 1 ) m e 2 im φ Y l , − m ( ϑ , φ ) = e 2 im φ Y l m ∗ ( ϑ , φ ) = ( − 1 ) m Y l m ( − ϑ , φ ) = e 2 im φ Y l , − m ( − ϑ , φ ) = ( − 1 ) m Y l , − m ∗ ( ϑ , φ ) = ( − 1 ) m Y l , − m ( ϑ , − φ ) = e 2 im φ Y l m ( ϑ , − φ ) = ( − 1 ) m e 2 im φ Y l m ∗ ( − ϑ , φ ) = Y l , − m ∗ ( − ϑ , φ ) = Y l , − m ( − ϑ , − φ ) = ( − 1 ) m e 2 im φ Y l m ( − ϑ , − φ ) = e 2 im φ Y l , − m ∗ ( − ϑ , − φ ) = ( − 1 ) m Y l m ∗ ( − ϑ , − φ ) . (975)
Behaviour of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) under transformations of coordinate systems
Any transformation of coordinate system in 3-dimensional space, which does not affect the orthogonality of coordinate axes, may be represented as a result of three operations: (1) rotation, (2) inversion and (3) parallel translation.
Rotation
The spherical harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) are covariant components of some irreducible tensor of rank l l l . Hence, under arbitrary rotation S { x , y , z } → S ′ { x ′ , y ′ , z ′ } \mathcal{S}\{x,y,z\}\to\mathcal{S}'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } of the coordinate system described by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ (§ 9.9 ) the spherical harmonics transform in accordance with
D ^ ( α , β , γ ) Y l m ′ ( ϑ , φ ) = Y l m ′ ( ϑ ′ , φ ′ ) = ∑ m Y l m ( ϑ , φ ) D m m ′ l ( α , β , γ ) (976) \hat{D}(\alpha,\beta,\gamma)Y_{lm'}(\vartheta,\varphi)=Y_{lm'}(\vartheta',\varphi')=\sum_m Y_{lm}(\vartheta,\varphi)D^l_{mm'}(\alpha,\beta,\gamma) \tag*{(976)} D ^ ( α , β , γ ) Y l m ′ ( ϑ , φ ) = Y l m ′ ( ϑ ′ , φ ′ ) = m ∑ Y l m ( ϑ , φ ) D m m ′ l ( α , β , γ ) (976)
Here D m m ′ l ( α , β , γ ) D^l_{mm'}(\alpha,\beta,\gamma) D m m ′ l ( α , β , γ ) is a Wigner D D D -function (Chap. 4), ϑ , φ \vartheta,\varphi ϑ , φ and ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ are polar angles of the position vector in the original and final coordinate systems, S \mathcal{S} S and S ′ \mathcal{S}' S ′ , respectively. The angles ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ may be expressed through ϑ , φ \vartheta,\varphi ϑ , φ and the Euler angles according to Eqs. 1.(648) .
Inversion
Under inversion S { x , y , z } → S ′ { − x , − y , − z } S\{x,y,z\} \to S'\{-x,-y,-z\} S { x , y , z } → S ′ { − x , − y , − z } the spherical harmonics transform as
P ^ r Y l m ( ϑ , φ ) = Y l m ( π − ϑ , φ + π ) = ( − 1 ) l Y l m ( ϑ , φ ) . (977) \hat{P}_{r}Y_{lm}(\vartheta,\varphi)=Y_{lm}(\pi-\vartheta,\varphi+\pi)=(-1)^{l}Y_{lm}(\vartheta,\varphi). \tag*{(977)} P ^ r Y l m ( ϑ , φ ) = Y l m ( π − ϑ , φ + π ) = ( − 1 ) l Y l m ( ϑ , φ ) . (977)
Parallel Translation
At parallel displacement S { x , y , z } → S ′ { x ′ , y ′ , z ′ } S\{x,y,z\} \to S'\{x',y',z'\} S { x , y , z } → S ′ { x ′ , y ′ , z ′ } by a vector a ( a , θ , ϕ ) \mathbf{a}(a,\theta,\phi) a ( a , θ , ϕ ) one has r ′ = r − a \mathbf{r}'=\mathbf{r}-\mathbf{a} r ′ = r − a . In this case the spherical harmonics transform as
T ^ ( a ) Y l ′ m ′ ( ϑ , φ ) = Y l ′ m ′ ( ϑ ′ , φ ′ ) = ∑ l = 0 l ′ ( − 1 ) l ′ + l [ 4 π ( 2 l ′ + 1 ) ! ( 2 l + 1 ) ! ( 2 l ′ − 2 l + 1 ) ! ] 1 2 ( a r ′ ) l ′ ( r a ) l { Y l ( ϑ , φ ) ⊗ Y l ′ − l ( Θ , Φ ) } l ′ m ′ . (978) \hat{T}(\mathbf{a})Y_{l'm'}(\vartheta,\varphi)=Y_{l'm'}(\vartheta',\varphi')=\sum_{l=0}^{l'}(-1)^{l'+l}\left[\frac{4\pi(2l'+1)!}{(2l+1)!(2l'-2l+1)!}\right]^{\frac{1}{2}}\left(\frac{a}{r'}\right)^{l'}\left(\frac{r}{a}\right)^{l}\{Y_l(\vartheta,\varphi)\otimes Y_{l'-l}(\Theta,\Phi)\}_{l'm'}. \tag*{(978)} T ^ ( a ) Y l ′ m ′ ( ϑ , φ ) = Y l ′ m ′ ( ϑ ′ , φ ′ ) = l = 0 ∑ l ′ ( − 1 ) l ′ + l [ ( 2 l + 1 )! ( 2 l ′ − 2 l + 1 )! 4 π ( 2 l ′ + 1 )! ] 2 1 ( r ′ a ) l ′ ( a r ) l { Y l ( ϑ , φ ) ⊗ Y l ′ − l ( Θ , Φ ) } l ′ m ′ . (978)
where T ^ ( a ) = e − a ⋅ ∇ \hat{T}(\mathbf{a})=e^{-\mathbf{a}\cdot\nabla} T ^ ( a ) = e − a ⋅ ∇ is the displacement operator, { Y l 1 ⊗ Y l 2 } t m \{Y_{l_1}\otimes Y_{l_2}\}_{tm} { Y l 1 ⊗ Y l 2 } t m is the irreducible tensor product (Sec. 3.1). The polar coordinates of the vectors r ( r , ϑ , φ ) \mathbf{r}(r,\vartheta,\varphi) r ( r , ϑ , φ ) and r ′ ( r ′ , ϑ ′ , φ ′ ) \mathbf{r}'(r',\vartheta',\varphi') r ′ ( r ′ , ϑ ′ , φ ′ ) are related by
r ′ 2 = r 2 + a 2 − 2 r a cos ω 12 r'^{2}=r^{2}+a^{2}-2ra\cos\omega_{12} r ′ 2 = r 2 + a 2 − 2 r a cos ω 12
cos ϑ ′ = r cos ϑ − a cos θ r 2 + a 2 − 2 r a cos ω 12 (979) \cos\vartheta'=\frac{r\cos\vartheta-a\cos\theta}{\sqrt{r^{2}+a^{2}-2ra\cos\omega_{12}}} \tag*{(979)} cos ϑ ′ = r 2 + a 2 − 2 r a cos ω 12 r cos ϑ − a cos θ (979)
tan φ ′ = r sin ϑ sin φ − a sin θ sin Φ r sin ϑ cos φ − a sin θ cos Φ \tan\varphi'=\frac{r\sin\vartheta\sin\varphi-a\sin\theta\sin\Phi}{r\sin\vartheta\cos\varphi-a\sin\theta\cos\Phi} tan φ ′ = r sin ϑ cos φ − a sin θ cos Φ r sin ϑ sin φ − a sin θ sin Φ
where
cos ω 12 = cos ϑ cos θ + sin ϑ sin θ cos ( φ − Φ ) . (980) \cos\omega_{12}=\cos\vartheta\cos\theta+\sin\vartheta\sin\theta\cos(\varphi-\Phi). \tag*{(980)} cos ω 12 = cos ϑ cos θ + sin ϑ sin θ cos ( φ − Φ ) . (980)
Special Cases of Coordinate-System Transformations
(a) Rotation about the coordinate axes through the angle π \pi π
about the x x x axis
\begin{aligned}
r'&=r,\\
\vartheta'&=\pi-\vartheta,\\
\varphi'&=2\pi-\varphi,
\end{aligned}
\qquad
Y_{lm}(\pi-\vartheta,2\pi-\varphi)=(-1)^{l}Y_{l,-m}(\vartheta,\varphi) \tag*{(981)}\end{aligned}
about the y y y axis
\begin{aligned}
r'&=r,\\
\vartheta'&=\pi-\vartheta,\\
\varphi'&=\pi-\varphi,
\end{aligned}
\qquad
Y_{lm}(\pi-\vartheta,\pi-\varphi)=(-1)^{l-m}Y_{l,-m}(\vartheta,\varphi) \tag*{(982)}\end{aligned}
about the z z z axis
\begin{aligned}
r'&=r,\\
\vartheta'&=\vartheta,\\
\varphi'&=\pi+\varphi,
\end{aligned}
\qquad
Y_{lm}(\vartheta,\pi+\varphi)=(-1)^{m}Y_{lm}(\vartheta,\varphi). \tag*{(983)}\end{aligned}
(b) Rotation about the z z z axis through an arbitrary angle χ \chi χ
\begin{aligned}
r'&=r,\\
\vartheta'&=\vartheta,\\
\varphi'&=\varphi-\chi,
\end{aligned}
\qquad
Y_{lm}(\vartheta,\varphi-\chi)=e^{-im\chi}Y_{lm}(\vartheta,\varphi). \tag*{(984)}\end{aligned}
(c) Rotation about an arbitrary direction n ( Θ , Φ ) \mathbf{n}(\Theta,\Phi) n ( Θ , Φ ) through a small angle ω ( ω ≪ π / 2 ) \omega(\omega\ll\pi/2) ω ( ω ≪ π /2 )
D ^ Y l m ( ϑ , φ ) ≈ Y l m ( ϑ , φ ) − i ω { m cos Θ Y l m ( ϑ , φ ) + sin Θ 2 [ e − i Φ l ( l + 1 ) − m ( m + 1 ) Y l , m + 1 ( ϑ , φ ) + e i Φ l ( l + 1 ) − m ( m − 1 ) Y l , m − 1 ( ϑ , φ ) ] } . (985) \hat{D}Y_{lm}(\vartheta,\varphi)\approx Y_{lm}(\vartheta,\varphi)-i\omega\left\{m\cos\Theta Y_{lm}(\vartheta,\varphi)+\frac{\sin\Theta}{2}\left[e^{-i\Phi}\sqrt{l(l+1)-m(m+1)}Y_{l,m+1}(\vartheta,\varphi)+e^{i\Phi}\sqrt{l(l+1)-m(m-1)}Y_{l,m-1}(\vartheta,\varphi)\right]\right\}. \tag*{(985)} D ^ Y l m ( ϑ , φ ) ≈ Y l m ( ϑ , φ ) − iω { m cos Θ Y l m ( ϑ , φ ) + 2 sin Θ [ e − i Φ l ( l + 1 ) − m ( m + 1 ) Y l , m + 1 ( ϑ , φ ) + e i Φ l ( l + 1 ) − m ( m − 1 ) Y l , m − 1 ( ϑ , φ ) ] } . (985)
(d) Reflection of the coordinate system with respect to the equatorial plane, ϑ = π / 2 \vartheta=\pi/2 ϑ = π /2
\begin{aligned}
r'&=r,\\
\vartheta'&=\pi-\vartheta,\\
\varphi'&=\varphi,
\end{aligned}
\qquad
Y_{lm}(\pi-\vartheta,\varphi)=(-1)^{l+m}Y_{lm}(\vartheta,\varphi) \tag*{(986)}\end{aligned}
(e) Reflection of the coordinate system with respect to a meridian plane, φ = φ 0 \varphi=\varphi_0 φ = φ 0 and φ = π + φ 0 \varphi=\pi+\varphi_0 φ = π + φ 0
\begin{aligned}
r'&=r,\\
\vartheta'&=\vartheta,\\
\varphi'&=2\varphi_0-\varphi,
\end{aligned}
\qquad
Y_{lm}(\vartheta,2\varphi_0-\varphi)=e^{i2m\varphi_0}(-1)^mY_{l-m}(\vartheta,\varphi) \tag*{(987)}\end{aligned}
Expansions in series of the spherical harmonics
General Relations
A collection of the spherical harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) with all integer non-negative l l l and integer m m m (∣ m ∣ ≤ l |m|\leq l ∣ m ∣ ≤ l ) constitutes a complete orthonormal set of functions of two real variables ϑ \vartheta ϑ and φ \varphi φ defined within 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π . The completeness relation for the spherical harmonics is given by
∑ l = 0 ∞ ∑ m = − l l Y l m ∗ ( ϑ , φ ) Y l m ( ϑ ′ , φ ′ ) = δ ( φ − φ ′ ) δ ( cos ϑ − cos ϑ ′ ) (988) \sum_{l=0}^{\infty}\sum_{m=-l}^{l}Y_{lm}^{*}(\vartheta,\varphi)Y_{lm}(\vartheta',\varphi')=\delta(\varphi-\varphi')\delta(\cos\vartheta-\cos\vartheta') \tag*{(988)} l = 0 ∑ ∞ m = − l ∑ l Y l m ∗ ( ϑ , φ ) Y l m ( ϑ ′ , φ ′ ) = δ ( φ − φ ′ ) δ ( cos ϑ − cos ϑ ′ ) (988)
The orthogonality and normalization relation is as follows: (§ 21.3.1 )
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l m ∗ ( ϑ , φ ) Y l ′ m ′ ( ϑ , φ ) = δ l l ′ δ m m ′ (989) \int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Y_{lm}^{*}(\vartheta,\varphi)Y_{l'm'}(\vartheta,\varphi)=\delta_{ll'}\delta_{mm'} \tag*{(989)} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l m ∗ ( ϑ , φ ) Y l ′ m ′ ( ϑ , φ ) = δ l l ′ δ m m ′ (989)
An arbitrary function f ( ϑ , φ ) f(\vartheta,\varphi) f ( ϑ , φ ) which is defined in the interval 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π and satisfies the condition
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ ∣ f ( ϑ , φ ) ∣ 2 < ∞ (990) \int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta|f(\vartheta,\varphi)|^2<\infty \tag*{(990)} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ ∣ f ( ϑ , φ ) ∣ 2 < ∞ (990)
can be expanded into a series of the spherical harmonics as
f ( ϑ , φ ) = ∑ l = 0 ∞ ∑ m = − l l a l m Y l m ( ϑ , φ ) (991) f(\vartheta,\varphi)=\sum_{l=0}^{\infty}\sum_{m=-l}^{l}a_{lm}Y_{lm}(\vartheta,\varphi) \tag*{(991)} f ( ϑ , φ ) = l = 0 ∑ ∞ m = − l ∑ l a l m Y l m ( ϑ , φ ) (991)
The expansion coefficients a l m a_{lm} a l m are given by the relation
a l m = ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l m ∗ ( ϑ , φ ) f ( ϑ , φ ) (992) a_{lm}=\int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Y_{lm}^{*}(\vartheta,\varphi)f(\vartheta,\varphi) \tag*{(992)} a l m = ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l m ∗ ( ϑ , φ ) f ( ϑ , φ ) (992)
This relation may be treated as an integral transformation of f ( ϑ , φ ) f(\vartheta,\varphi) f ( ϑ , φ ) from the continuous variables ϑ , φ \vartheta,\varphi ϑ , φ to the discrete variables l , m l,m l , m . In this case Y l m ( ϑ , φ ) ≡ ⟨ ϑ , φ ∣ l m ⟩ Y_{lm}(\vartheta,\varphi)\equiv\langle\vartheta,\varphi|lm\rangle Y l m ( ϑ , φ ) ≡ ⟨ ϑ , φ ∣ l m ⟩ plays the role of the transformation matrix
⟨ l m ∣ f ⟩ = ⟨ l m ∣ ϑ , φ ⟩ ⟨ ϑ , φ ∣ f ⟩ (993) \langle lm|f\rangle=\langle lm|\vartheta,\varphi\rangle\langle\vartheta,\varphi|f\rangle \tag*{(993)} ⟨ l m ∣ f ⟩ = ⟨ l m ∣ ϑ , φ ⟩ ⟨ ϑ , φ ∣ f ⟩ (993)
where
⟨ l m ∣ f ⟩ ≡ a l m , ⟨ l m ∣ ϑ , φ ⟩ = Y l m ∗ ( ϑ , φ ) , ⟨ ϑ , φ ∣ f ⟩ = f ( ϑ , φ ) \langle lm|f\rangle\equiv a_{lm},\qquad\langle lm|\vartheta,\varphi\rangle=Y_{lm}^{*}(\vartheta,\varphi),\qquad\langle\vartheta,\varphi|f\rangle=f(\vartheta,\varphi) ⟨ l m ∣ f ⟩ ≡ a l m , ⟨ l m ∣ ϑ , φ ⟩ = Y l m ∗ ( ϑ , φ ) , ⟨ ϑ , φ ∣ f ⟩ = f ( ϑ , φ )
As usual, in relations of this type, summation or integration is assumed over all variables which are repeated twice.
The transformation (5.6.5) is unitary, i.e.,
⟨ f ∣ l m ⟩ ⟨ l m ∣ f ⟩ = ⟨ f ∣ ϑ , φ ⟩ ⟨ ϑ , φ ∣ f ⟩ (994) \langle f|lm\rangle\langle lm|f\rangle=\langle f|\vartheta,\varphi\rangle\langle\vartheta,\varphi|f\rangle \tag*{(994)} ⟨ f ∣ l m ⟩ ⟨ l m ∣ f ⟩ = ⟨ f ∣ ϑ , φ ⟩ ⟨ ϑ , φ ∣ f ⟩ (994)
The expansion coefficients a l m a_{lm} a l m satisfy the Parseval condition
∑ l = 0 ∞ ∑ m = − l l ∣ a l m ∣ 2 = ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ ∣ f ( ϑ , φ ) ∣ 2 (995) \sum_{l=0}^{\infty}\sum_{m=-l}^{l}|a_{lm}|^2=\int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta|f(\vartheta,\varphi)|^2 \tag*{(995)} l = 0 ∑ ∞ m = − l ∑ l ∣ a l m ∣ 2 = ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ ∣ f ( ϑ , φ ) ∣ 2 (995)
The expansion (5.6.4) in terms of the spherical harmonics is widely used in different branches of physics. It is called the multipole expansion, and the a l m a_{lm} a l m are called multipole moments.
Expansion of Products of the Spherical Harmonics
Expansion of Products of the Spherical Harmonics
(a) A direct product of two spherical harmonics of the same arguments may be expanded in series as (the so-called Clebsch-Gordan series)
Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) = ∑ L M ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) 4 π ( 2 L + 1 ) C l 1 0 l 2 0 L 0 C l 1 m 1 l 2 m 2 L M Y L M ( ϑ , φ ) . (996) Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)=\sum_{LM}\sqrt{\frac{(2l_1+1)(2l_2+1)}{4\pi(2L+1)}}C_{l_1 0 l_2 0}^{L0}C_{l_1m_1l_2m_2}^{LM}Y_{LM}(\vartheta,\varphi). \tag*{(996)} Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) = L M ∑ 4 π ( 2 L + 1 ) ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) C l 1 0 l 2 0 L 0 C l 1 m 1 l 2 m 2 L M Y L M ( ϑ , φ ) . (996)
The inverse relation may be written as
4 π ( 2 L + 1 ) ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ∑ m 1 m 2 C l 1 m 1 l 2 m 2 L M Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) = C l 1 0 l 2 0 L 0 Y L M ( ϑ , φ ) . (997) \sqrt{\frac{4\pi(2L+1)}{(2l_1+1)(2l_2+1)}}\sum_{m_1m_2}C_{l_1m_1l_2m_2}^{LM}Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)=C_{l_1 0 l_2 0}^{L0}Y_{LM}(\vartheta,\varphi). \tag*{(997)} ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) 4 π ( 2 L + 1 ) m 1 m 2 ∑ C l 1 m 1 l 2 m 2 L M Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) = C l 1 0 l 2 0 L 0 Y L M ( ϑ , φ ) . (997)
Products of three and more spherical harmonics are decomposed according to
Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) Y l 3 m 3 ( ϑ , φ ) = ∑ L M L ′ M ′ ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 2 l 3 + 1 ) ( 4 π ) 2 ( 2 L + 1 ) C l 1 0 l 2 0 L ′ 0 C L ′ 0 l 3 0 L 0 C l 1 m 1 l 2 m 2 L ′ M ′ C L ′ M ′ l 3 m 3 L M Y L M ( ϑ , φ ) (998) Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)Y_{l_3m_3}(\vartheta,\varphi)=\sum_{LML'M'}\sqrt{\frac{(2l_1+1)(2l_2+1)(2l_3+1)}{(4\pi)^2(2L+1)}}C_{l_1 0 l_2 0}^{L'0}C_{L'0l_3 0}^{L0}C_{l_1m_1l_2m_2}^{L'M'}C_{L'M'l_3m_3}^{LM}Y_{LM}(\vartheta,\varphi) \tag*{(998)} Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) Y l 3 m 3 ( ϑ , φ ) = L M L ′ M ′ ∑ ( 4 π ) 2 ( 2 L + 1 ) ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 2 l 3 + 1 ) C l 1 0 l 2 0 L ′ 0 C L ′ 0 l 3 0 L 0 C l 1 m 1 l 2 m 2 L ′ M ′ C L ′ M ′ l 3 m 3 L M Y L M ( ϑ , φ ) (998)
Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) × … × Y l n m n ( ϑ , φ ) = ∑ L n B L n Y L n M n ( ϑ , φ ) (999) Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)\times\ldots\times Y_{l_nm_n}(\vartheta,\varphi)=\sum_{L_n}B_{L_n}Y_{L_nM_n}(\vartheta,\varphi) \tag*{(999)} Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) × … × Y l n m n ( ϑ , φ ) = L n ∑ B L n Y L n M n ( ϑ , φ ) (999)
where
B L n = 4 π 2 L n + 1 ∑ L 1 L 2 … L n − 1 M 1 M 2 , … M n − 1 ∏ i = 1 n ( 2 l i + 1 4 π C L i − 1 0 l i 0 L i 0 C L i − 1 M i − 1 l i m i L i M i ) (1000) B_{L_n}=\sqrt{\frac{4\pi}{2L_n+1}}\sum_{L_1L_2\ldots L_{n-1}\atop M_1M_2,\ldots M_{n-1}}\prod_{i=1}^{n}\left(\sqrt{\frac{2l_i+1}{4\pi}}C_{L_{i-1}0l_i0}^{L_i0}C_{L_{i-1}M_{i-1}l_im_i}^{L_iM_i}\right) \tag*{(1000)} B L n = 2 L n + 1 4 π M 1 M 2 , … M n − 1 L 1 L 2 … L n − 1 ∑ i = 1 ∏ n ( 4 π 2 l i + 1 C L i − 1 0 l i 0 L i 0 C L i − 1 M i − 1 l i m i L i M i ) (1000)
with L 0 ≡ M 0 ≡ 0 L_0\equiv M_0\equiv0 L 0 ≡ M 0 ≡ 0 .
(b) Irreducible tensor products (defined in Sec. 3.1) of the spherical harmonics may be expanded as
{ Y l 1 ( ϑ , φ ) ⊗ Y l 2 ( ϑ , φ ) } L M = ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) 4 π ( 2 L + 1 ) C l 1 0 l 2 0 L 0 Y L M ( ϑ , φ ) (1001) \{Y_{l_1}(\vartheta,\varphi)\otimes Y_{l_2}(\vartheta,\varphi)\}_{LM}=\sqrt{\frac{(2l_1+1)(2l_2+1)}{4\pi(2L+1)}}C_{l_1 0 l_2 0}^{L0}Y_{LM}(\vartheta,\varphi) \tag*{(1001)} { Y l 1 ( ϑ , φ ) ⊗ Y l 2 ( ϑ , φ ) } L M = 4 π ( 2 L + 1 ) ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) C l 1 0 l 2 0 L 0 Y L M ( ϑ , φ ) (1001)
{ { Y l 1 ( ϑ , φ ) ⊗ Y l 2 ( ϑ , φ ) } L ′ ⊗ Y l 3 ( ϑ , φ ) } L M = ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 2 l 3 + 1 ) ( 4 π ) 2 ( 2 L + 1 ) C l 1 0 l 2 0 L ′ 0 C L ′ 0 l 3 0 L 0 Y L M ( ϑ , φ ) (1002) \{\{Y_{l_1}(\vartheta,\varphi)\otimes Y_{l_2}(\vartheta,\varphi)\}_{L'}\otimes Y_{l_3}(\vartheta,\varphi)\}_{LM}=\sqrt{\frac{(2l_1+1)(2l_2+1)(2l_3+1)}{(4\pi)^2(2L+1)}}C_{l_1 0 l_2 0}^{L'0}C_{L'0l_3 0}^{L0}Y_{LM}(\vartheta,\varphi) \tag*{(1002)} {{ Y l 1 ( ϑ , φ ) ⊗ Y l 2 ( ϑ , φ ) } L ′ ⊗ Y l 3 ( ϑ , φ ) } L M = ( 4 π ) 2 ( 2 L + 1 ) ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 2 l 3 + 1 ) C l 1 0 l 2 0 L ′ 0 C L ′ 0 l 3 0 L 0 Y L M ( ϑ , φ ) (1002)
{ … { { Y l 1 ( ϑ , φ ) ⊗ Y l 2 ( ϑ , φ ) } L 2 ⊗ Y l 3 ( ϑ , φ ) } L 3 … ⊗ Y l n ( ϑ , φ ) } L n M n = 4 π 2 L n + 1 ∏ i = 1 n ( 2 l i + 1 4 π C L i − 1 0 l i 0 L i 0 ) Y L n M n ( ϑ , φ ) (1003) \{\ldots\{\{Y_{l_1}(\vartheta,\varphi)\otimes Y_{l_2}(\vartheta,\varphi)\}_{L_2}\otimes Y_{l_3}(\vartheta,\varphi)\}_{L_3}\ldots\otimes Y_{l_n}(\vartheta,\varphi)\}_{L_nM_n}=\sqrt{\frac{4\pi}{2L_n+1}}\prod_{i=1}^{n}\left(\sqrt{\frac{2l_i+1}{4\pi}}C_{L_{i-1}0l_i0}^{L_i0}\right)Y_{L_nM_n}(\vartheta,\varphi) \tag*{(1003)} { … {{ Y l 1 ( ϑ , φ ) ⊗ Y l 2 ( ϑ , φ ) } L 2 ⊗ Y l 3 ( ϑ , φ ) } L 3 … ⊗ Y l n ( ϑ , φ ) } L n M n = 2 L n + 1 4 π i = 1 ∏ n ( 4 π 2 l i + 1 C L i − 1 0 l i 0 L i 0 ) Y L n M n ( ϑ , φ ) (1003)
In particular, when l 1 = l 2 = … = l n = 1 l_1=l_2=\ldots=l_n=1 l 1 = l 2 = … = l n = 1 and L k = k L_k=k L k = k (k = 1 , 2 , 3 , … , n k=1,2,3,\ldots,n k = 1 , 2 , 3 , … , n ) one has
{ … { { Y 1 ( ϑ , φ ) ⊗ Y 1 ( ϑ , φ ) } 2 ⊗ Y 1 ( ϑ , φ ) } 3 … ⊗ Y 1 ( ϑ , φ ) } n m = 3 n ( 4 π ) n − 1 n ! ( 2 n + 1 ) ! ! Y n m ( ϑ , φ ) (1004) \{\ldots\{\{Y_1(\vartheta,\varphi)\otimes Y_1(\vartheta,\varphi)\}_2\otimes Y_1(\vartheta,\varphi)\}_3\ldots\otimes Y_1(\vartheta,\varphi)\}_{nm}=\sqrt{\frac{3^n}{(4\pi)^{n-1}}\frac{n!}{(2n+1)!!}}Y_{nm}(\vartheta,\varphi) \tag*{(1004)} { … {{ Y 1 ( ϑ , φ ) ⊗ Y 1 ( ϑ , φ ) } 2 ⊗ Y 1 ( ϑ , φ ) } 3 … ⊗ Y 1 ( ϑ , φ ) } nm = ( 4 π ) n − 1 3 n ( 2 n + 1 )!! n ! Y nm ( ϑ , φ ) (1004)
Note that when expanding the products of the spherical harmonics, the functions C l m ( ϑ , φ ) C_{lm}(\vartheta,\varphi) C l m ( ϑ , φ ) which contain an additional factor 4 π / ( 2 l + 1 ) \sqrt{4\pi/(2l+1)} 4 π / ( 2 l + 1 ) (Sec. 5.1.4), are more convenient.
Recursion relations
Recursion relations for the spherical harmonics may be obtained, for example, from the Clebsch-Gordan series (see Eq. (996) ).
− 2 m cot ϑ Y l m ( ϑ , φ ) = l ( l + 1 ) − m ( m + 1 ) e − i φ Y l , m + 1 ( ϑ , φ ) + l ( l + 1 ) − m ( m − 1 ) e i φ Y l , m − 1 ( ϑ , φ ) (1005) -2m\cot\vartheta Y_{lm}(\vartheta,\varphi)=\sqrt{l(l+1)-m(m+1)}e^{-i\varphi}Y_{l,m+1}(\vartheta,\varphi)+\sqrt{l(l+1)-m(m-1)}e^{i\varphi}Y_{l,m-1}(\vartheta,\varphi) \tag*{(1005)} − 2 m cot ϑ Y l m ( ϑ , φ ) = l ( l + 1 ) − m ( m + 1 ) e − i φ Y l , m + 1 ( ϑ , φ ) + l ( l + 1 ) − m ( m − 1 ) e i φ Y l , m − 1 ( ϑ , φ ) (1005)
cos ϑ Y l m ( ϑ , φ ) = ( l − m + 1 ) ( l + m + 1 ) ( 2 l + 1 ) ( 2 l + 3 ) Y l + 1 , m ( ϑ , φ ) + ( l − m ) ( l + m ) ( 2 l − 1 ) ( 2 l + 1 ) Y l − 1 , m ( ϑ , φ ) (1006) \cos\vartheta Y_{lm}(\vartheta,\varphi)=\sqrt{\frac{(l-m+1)(l+m+1)}{(2l+1)(2l+3)}}Y_{l+1,m}(\vartheta,\varphi)+\sqrt{\frac{(l-m)(l+m)}{(2l-1)(2l+1)}}Y_{l-1,m}(\vartheta,\varphi) \tag*{(1006)} cos ϑ Y l m ( ϑ , φ ) = ( 2 l + 1 ) ( 2 l + 3 ) ( l − m + 1 ) ( l + m + 1 ) Y l + 1 , m ( ϑ , φ ) + ( 2 l − 1 ) ( 2 l + 1 ) ( l − m ) ( l + m ) Y l − 1 , m ( ϑ , φ ) (1006)
sin ϑ Y l m ( ϑ , φ ) e − i φ = ( l − m + 1 ) ( l − m + 2 ) ( 2 l + 1 ) ( 2 l + 3 ) Y l + 1 , m − 1 ( ϑ , φ ) − ( l + m − 1 ) ( l + m ) ( 2 l − 1 ) ( 2 l + 1 ) Y l − 1 , m − 1 ( ϑ , φ ) = cos ϑ l − m + 1 l + m Y l , m − 1 ( ϑ , φ ) − 2 l + 1 2 l − 1 l + m − 1 l + m Y l − 1 , m − 1 ( ϑ , φ ) = − cos ϑ l + m l − m + 1 Y l , m − 1 ( ϑ , φ ) + 2 l + 1 2 l + 3 l − m + 2 l − m + 1 Y l + 1 , m − 1 ( ϑ , φ ) (1007) \begin{aligned}
\sin\vartheta Y_{lm}(\vartheta,\varphi)e^{-i\varphi}
&=\sqrt{\frac{(l-m+1)(l-m+2)}{(2l+1)(2l+3)}}Y_{l+1,m-1}(\vartheta,\varphi)-\sqrt{\frac{(l+m-1)(l+m)}{(2l-1)(2l+1)}}Y_{l-1,m-1}(\vartheta,\varphi)\\
&=\cos\vartheta\sqrt{\frac{l-m+1}{l+m}}Y_{l,m-1}(\vartheta,\varphi)-\sqrt{\frac{2l+1}{2l-1}\frac{l+m-1}{l+m}}Y_{l-1,m-1}(\vartheta,\varphi)\\
&=-\cos\vartheta\sqrt{\frac{l+m}{l-m+1}}Y_{l,m-1}(\vartheta,\varphi)+\sqrt{\frac{2l+1}{2l+3}\frac{l-m+2}{l-m+1}}Y_{l+1,m-1}(\vartheta,\varphi) \tag*{(1007)}
\end{aligned} sin ϑ Y l m ( ϑ , φ ) e − i φ = ( 2 l + 1 ) ( 2 l + 3 ) ( l − m + 1 ) ( l − m + 2 ) Y l + 1 , m − 1 ( ϑ , φ ) − ( 2 l − 1 ) ( 2 l + 1 ) ( l + m − 1 ) ( l + m ) Y l − 1 , m − 1 ( ϑ , φ ) = cos ϑ l + m l − m + 1 Y l , m − 1 ( ϑ , φ ) − 2 l − 1 2 l + 1 l + m l + m − 1 Y l − 1 , m − 1 ( ϑ , φ ) = − cos ϑ l − m + 1 l + m Y l , m − 1 ( ϑ , φ ) + 2 l + 3 2 l + 1 l − m + 1 l − m + 2 Y l + 1 , m − 1 ( ϑ , φ ) (1007)
sin ϑ Y l m ( ϑ , φ ) e i φ = − ( l + m + 1 ) ( l + m + 2 ) ( 2 l + 1 ) ( 2 l + 3 ) Y l + 1 , m + 1 ( ϑ , φ ) + ( l − m − 1 ) ( l − m ) ( 2 l − 1 ) ( 2 l + 1 ) Y l − 1 , m + 1 ( ϑ , φ ) = − cos ϑ l + m + 1 l − m Y l , m + 1 ( ϑ , φ ) + 2 l + 1 2 l − 1 l − m − 1 l − m Y l − 1 , m + 1 ( ϑ , φ ) = cos ϑ l − m l + m + 1 Y l , m + 1 ( ϑ , φ ) − 2 l + 1 2 l + 3 l + m + 2 l + m + 1 Y l + 1 , m + 1 ( ϑ , φ ) (1008) \begin{aligned}
\sin\vartheta Y_{lm}(\vartheta,\varphi)e^{i\varphi}
&=-\sqrt{\frac{(l+m+1)(l+m+2)}{(2l+1)(2l+3)}}Y_{l+1,m+1}(\vartheta,\varphi)+\sqrt{\frac{(l-m-1)(l-m)}{(2l-1)(2l+1)}}Y_{l-1,m+1}(\vartheta,\varphi)\\
&=-\cos\vartheta\sqrt{\frac{l+m+1}{l-m}}Y_{l,m+1}(\vartheta,\varphi)+\sqrt{\frac{2l+1}{2l-1}\frac{l-m-1}{l-m}}Y_{l-1,m+1}(\vartheta,\varphi)\\
&=\cos\vartheta\sqrt{\frac{l-m}{l+m+1}}Y_{l,m+1}(\vartheta,\varphi)-\sqrt{\frac{2l+1}{2l+3}\frac{l+m+2}{l+m+1}}Y_{l+1,m+1}(\vartheta,\varphi) \tag*{(1008)}
\end{aligned} sin ϑ Y l m ( ϑ , φ ) e i φ = − ( 2 l + 1 ) ( 2 l + 3 ) ( l + m + 1 ) ( l + m + 2 ) Y l + 1 , m + 1 ( ϑ , φ ) + ( 2 l − 1 ) ( 2 l + 1 ) ( l − m − 1 ) ( l − m ) Y l − 1 , m + 1 ( ϑ , φ ) = − cos ϑ l − m l + m + 1 Y l , m + 1 ( ϑ , φ ) + 2 l − 1 2 l + 1 l − m l − m − 1 Y l − 1 , m + 1 ( ϑ , φ ) = cos ϑ l + m + 1 l − m Y l , m + 1 ( ϑ , φ ) − 2 l + 3 2 l + 1 l + m + 1 l + m + 2 Y l + 1 , m + 1 ( ϑ , φ ) (1008)
( 2 l − 1 ) ( 2 l + 3 ) cos 2 ϑ Y l m ( ϑ , φ ) = ( 2 l − 1 ) ( l + 1 ) 2 − m 2 2 l + 1 ( l + 2 ) 2 − m 2 2 l + 5 Y l + 2 , m ( ϑ , φ ) + [ 2 l ( l + 1 ) − 2 m 2 − 1 ] Y l m ( ϑ , φ ) + ( 2 l + 3 ) l 2 − m 2 2 l + 1 ( l − 1 ) 2 − m 2 2 l − 3 Y l − 2 , m ( ϑ , φ ) (1009) \begin{aligned}
(2l-1)(2l+3)\cos^2\vartheta Y_{lm}(\vartheta,\varphi)
&=(2l-1)\sqrt{\frac{(l+1)^2-m^2}{2l+1}\frac{(l+2)^2-m^2}{2l+5}}Y_{l+2,m}(\vartheta,\varphi)\\
&+\left[2l(l+1)-2m^2-1\right]Y_{lm}(\vartheta,\varphi)\\
&+(2l+3)\sqrt{\frac{l^2-m^2}{2l+1}\frac{(l-1)^2-m^2}{2l-3}}Y_{l-2,m}(\vartheta,\varphi) \tag*{(1009)}
\end{aligned} ( 2 l − 1 ) ( 2 l + 3 ) cos 2 ϑ Y l m ( ϑ , φ ) = ( 2 l − 1 ) 2 l + 1 ( l + 1 ) 2 − m 2 2 l + 5 ( l + 2 ) 2 − m 2 Y l + 2 , m ( ϑ , φ ) + [ 2 l ( l + 1 ) − 2 m 2 − 1 ] Y l m ( ϑ , φ ) + ( 2 l + 3 ) 2 l + 1 l 2 − m 2 2 l − 3 ( l − 1 ) 2 − m 2 Y l − 2 , m ( ϑ , φ ) (1009)
( 2 l − 1 ) ( 2 l + 3 ) sin ϑ cos ϑ e i φ Y l m ( ϑ , φ ) = − ( 2 l − 1 ) ( l + 1 ) 2 − m 2 2 l + 1 ( l + m + 2 ) ( l + m + 3 ) 2 l + 5 Y l + 2 , m + 1 ( ϑ , φ ) − ( 2 m + 1 ) l ( l + 1 ) − m ( m + 1 ) Y l , m + 1 ( ϑ , φ ) + ( 2 l + 3 ) l 2 − m 2 2 l + 1 ( l − m − 1 ) ( l − m − 2 ) 2 l − 3 Y l − 2 , m + 1 ( ϑ , φ ) (1010) \begin{aligned}
(2l-1)(2l+3)\sin\vartheta\cos\vartheta e^{i\varphi}Y_{lm}(\vartheta,\varphi)
&=-(2l-1)\sqrt{\frac{(l+1)^2-m^2}{2l+1}\frac{(l+m+2)(l+m+3)}{2l+5}}Y_{l+2,m+1}(\vartheta,\varphi)\\
&-(2m+1)\sqrt{l(l+1)-m(m+1)}Y_{l,m+1}(\vartheta,\varphi)\\
&+(2l+3)\sqrt{\frac{l^2-m^2}{2l+1}\frac{(l-m-1)(l-m-2)}{2l-3}}Y_{l-2,m+1}(\vartheta,\varphi) \tag*{(1010)}
\end{aligned} ( 2 l − 1 ) ( 2 l + 3 ) sin ϑ cos ϑ e i φ Y l m ( ϑ , φ ) = − ( 2 l − 1 ) 2 l + 1 ( l + 1 ) 2 − m 2 2 l + 5 ( l + m + 2 ) ( l + m + 3 ) Y l + 2 , m + 1 ( ϑ , φ ) − ( 2 m + 1 ) l ( l + 1 ) − m ( m + 1 ) Y l , m + 1 ( ϑ , φ ) + ( 2 l + 3 ) 2 l + 1 l 2 − m 2 2 l − 3 ( l − m − 1 ) ( l − m − 2 ) Y l − 2 , m + 1 ( ϑ , φ ) (1010)
( 2 l − 1 ) ( 2 l + 3 ) sin ϑ cos ϑ e − i φ Y l m ( ϑ , φ ) = ( 2 l − 1 ) ( l + 1 ) 2 − m 2 2 l + 1 ( l − m + 2 ) ( l − m + 3 ) 2 l + 5 Y l + 2 , m − 1 ( ϑ , φ ) − ( 2 m − 1 ) l ( l + 1 ) − m ( m − 1 ) Y l , m − 1 ( ϑ , φ ) − ( 2 l + 3 ) l 2 − m 2 2 l + 1 ( l + m − 1 ) ( l + m − 2 ) 2 l − 3 Y l − 2 , m − 1 ( ϑ , φ ) (1011) \begin{aligned}
(2l-1)(2l+3)\sin\vartheta\cos\vartheta e^{-i\varphi}Y_{lm}(\vartheta,\varphi)
&=(2l-1)\sqrt{\frac{(l+1)^2-m^2}{2l+1}\frac{(l-m+2)(l-m+3)}{2l+5}}Y_{l+2,m-1}(\vartheta,\varphi)\\
&-(2m-1)\sqrt{l(l+1)-m(m-1)}Y_{l,m-1}(\vartheta,\varphi)\\
&-(2l+3)\sqrt{\frac{l^2-m^2}{2l+1}\frac{(l+m-1)(l+m-2)}{2l-3}}Y_{l-2,m-1}(\vartheta,\varphi) \tag*{(1011)}
\end{aligned} ( 2 l − 1 ) ( 2 l + 3 ) sin ϑ cos ϑ e − i φ Y l m ( ϑ , φ ) = ( 2 l − 1 ) 2 l + 1 ( l + 1 ) 2 − m 2 2 l + 5 ( l − m + 2 ) ( l − m + 3 ) Y l + 2 , m − 1 ( ϑ , φ ) − ( 2 m − 1 ) l ( l + 1 ) − m ( m − 1 ) Y l , m − 1 ( ϑ , φ ) − ( 2 l + 3 ) 2 l + 1 l 2 − m 2 2 l − 3 ( l + m − 1 ) ( l + m − 2 ) Y l − 2 , m − 1 ( ϑ , φ ) (1011)
( 2 l − 1 ) ( 2 l + 3 ) sin 2 ϑ e i 2 φ Y l m ( ϑ , φ ) = 2 l − 1 ( 2 l + 1 ) ( 2 l + 5 ) ( l + m + 4 ) ! ( l + m ) ! Y l + 2 , m + 2 ( ϑ , φ ) + 2 l + 3 ( 2 l + 1 ) ( 2 l − 3 ) ( l − m ) ! ( l − m − 4 ) ! Y l − 2 , m + 2 ( ϑ , φ ) − 2 ( l + m + 2 ) ! ( l − m ) ! ( l − m − 2 ) ! ( l + m ) ! Y l , m + 2 ( ϑ , φ ) (1012) \begin{aligned}
(2l-1)(2l+3)\sin^2\vartheta e^{i2\varphi}Y_{lm}(\vartheta,\varphi)
&=\frac{2l-1}{\sqrt{(2l+1)(2l+5)}}\sqrt{\frac{(l+m+4)!}{(l+m)!}}Y_{l+2,m+2}(\vartheta,\varphi)\\
&+\frac{2l+3}{\sqrt{(2l+1)(2l-3)}}\sqrt{\frac{(l-m)!}{(l-m-4)!}}Y_{l-2,m+2}(\vartheta,\varphi)\\
&-2\sqrt{\frac{(l+m+2)!(l-m)!}{(l-m-2)!(l+m)!}}Y_{l,m+2}(\vartheta,\varphi) \tag*{(1012)}
\end{aligned} ( 2 l − 1 ) ( 2 l + 3 ) sin 2 ϑ e i 2 φ Y l m ( ϑ , φ ) = ( 2 l + 1 ) ( 2 l + 5 ) 2 l − 1 ( l + m )! ( l + m + 4 )! Y l + 2 , m + 2 ( ϑ , φ ) + ( 2 l + 1 ) ( 2 l − 3 ) 2 l + 3 ( l − m − 4 )! ( l − m )! Y l − 2 , m + 2 ( ϑ , φ ) − 2 ( l − m − 2 )! ( l + m )! ( l + m + 2 )! ( l − m )! Y l , m + 2 ( ϑ , φ ) (1012)
( 2 l − 1 ) ( 2 l + 3 ) sin 2 ϑ e − i 2 φ Y l m ( ϑ , φ ) = 2 l − 1 ( 2 l + 1 ) ( 2 l + 5 ) ( l − m + 4 ) ! ( l − m ) ! Y l + 2 , m − 2 ( ϑ , φ ) + 2 l + 3 ( 2 l + 1 ) ( 2 l − 3 ) ( l + m ) ! ( l + m − 4 ) ! Y l − 2 , m − 2 ( ϑ , φ ) − 2 ( l − m + 2 ) ! ( l + m ) ! ( l + m − 2 ) ! ( l − m ) ! Y l , m − 2 ( ϑ , φ ) (1013) \begin{aligned}
(2l-1)(2l+3)\sin^2\vartheta e^{-i2\varphi}Y_{lm}(\vartheta,\varphi)
&=\frac{2l-1}{\sqrt{(2l+1)(2l+5)}}\sqrt{\frac{(l-m+4)!}{(l-m)!}}Y_{l+2,m-2}(\vartheta,\varphi)\\
&+\frac{2l+3}{\sqrt{(2l+1)(2l-3)}}\sqrt{\frac{(l+m)!}{(l+m-4)!}}Y_{l-2,m-2}(\vartheta,\varphi)\\
&-2\sqrt{\frac{(l-m+2)!(l+m)!}{(l+m-2)!(l-m)!}}Y_{l,m-2}(\vartheta,\varphi) \tag*{(1013)}
\end{aligned} ( 2 l − 1 ) ( 2 l + 3 ) sin 2 ϑ e − i 2 φ Y l m ( ϑ , φ ) = ( 2 l + 1 ) ( 2 l + 5 ) 2 l − 1 ( l − m )! ( l − m + 4 )! Y l + 2 , m − 2 ( ϑ , φ ) + ( 2 l + 1 ) ( 2 l − 3 ) 2 l + 3 ( l + m − 4 )! ( l + m )! Y l − 2 , m − 2 ( ϑ , φ ) − 2 ( l + m − 2 )! ( l − m )! ( l − m + 2 )! ( l + m )! Y l , m − 2 ( ϑ , φ ) (1013)
Differential relations
Action of the Operator of Orbital Angular Momentum on Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ )
Spherical components of the operator L \mathbf{L} L (Eq. (2.2.18)) act on Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) according to
L ^ ± 1 Y l m ( ϑ , φ ) = ∓ l ( l + 1 ) − m ( m ± 1 ) 2 Y l , m ± 1 ( ϑ , φ ) (1014) \hat{L}_{\pm1}Y_{lm}(\vartheta,\varphi)=\mp\sqrt{\frac{l(l+1)-m(m\pm1)}{2}}Y_{l,m\pm1}(\vartheta,\varphi) \tag*{(1014)} L ^ ± 1 Y l m ( ϑ , φ ) = ∓ 2 l ( l + 1 ) − m ( m ± 1 ) Y l , m ± 1 ( ϑ , φ ) (1014)
L ^ 0 Y l m ( ϑ , φ ) = m Y l m ( ϑ , φ ) \hat{L}_{0}Y_{lm}(\vartheta,\varphi)=mY_{lm}(\vartheta,\varphi) L ^ 0 Y l m ( ϑ , φ ) = m Y l m ( ϑ , φ )
or in a more compact form
L ^ μ Y l m ( ϑ , φ ) = l ( l + 1 ) C l m 1 μ l , m + μ Y l , m + μ ( ϑ , φ ) (1015) \hat{L}_{\mu}Y_{lm}(\vartheta,\varphi)=\sqrt{l(l+1)}C_{lm\,1\mu}^{l,m+\mu}Y_{l,m+\mu}(\vartheta,\varphi) \tag*{(1015)} L ^ μ Y l m ( ϑ , φ ) = l ( l + 1 ) C l m 1 μ l , m + μ Y l , m + μ ( ϑ , φ ) (1015)
The action of the operator L ^ 2 \hat{L}^{2} L ^ 2 on Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) yields
L ^ 2 Y l m ( ϑ , φ ) = l ( l + 1 ) Y l m ( ϑ , φ ) (1016) \hat{L}^{2}Y_{lm}(\vartheta,\varphi)=l(l+1)Y_{lm}(\vartheta,\varphi) \tag*{(1016)} L ^ 2 Y l m ( ϑ , φ ) = l ( l + 1 ) Y l m ( ϑ , φ ) (1016)
First and Second Order Derivatives of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ )
∂ ∂ φ Y l m ( ϑ , φ ) = i m Y l m ( ϑ , φ ) (1017) \frac{\partial}{\partial\varphi}Y_{lm}(\vartheta,\varphi)=imY_{lm}(\vartheta,\varphi) \tag*{(1017)} ∂ φ ∂ Y l m ( ϑ , φ ) = im Y l m ( ϑ , φ ) (1017)
∂ ∂ ϑ Y l m ( ϑ , φ ) = m cot ϑ Y l m ( ϑ , φ ) + l ( l + 1 ) − m ( m + 1 ) Y l , m + 1 ( ϑ , φ ) e − i φ = − m cot ϑ Y l m ( ϑ , φ ) − l ( l + 1 ) − m ( m − 1 ) Y l , m − 1 ( ϑ , φ ) e i φ = 1 2 l ( l + 1 ) − m ( m + 1 ) Y l , m + 1 ( ϑ , φ ) e − i φ − 1 2 l ( l + 1 ) − m ( m − 1 ) Y l , m − 1 ( ϑ , φ ) e i φ (1018) \begin{aligned}
\frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi)
&=m\cot\vartheta Y_{lm}(\vartheta,\varphi)+\sqrt{l(l+1)-m(m+1)}Y_{l,m+1}(\vartheta,\varphi)e^{-i\varphi}\\
&=-m\cot\vartheta Y_{lm}(\vartheta,\varphi)-\sqrt{l(l+1)-m(m-1)}Y_{l,m-1}(\vartheta,\varphi)e^{i\varphi}\\
&=\frac{1}{2}\sqrt{l(l+1)-m(m+1)}Y_{l,m+1}(\vartheta,\varphi)e^{-i\varphi}
-\frac{1}{2}\sqrt{l(l+1)-m(m-1)}Y_{l,m-1}(\vartheta,\varphi)e^{i\varphi} \tag*{(1018)}
\end{aligned} ∂ ϑ ∂ Y l m ( ϑ , φ ) = m cot ϑ Y l m ( ϑ , φ ) + l ( l + 1 ) − m ( m + 1 ) Y l , m + 1 ( ϑ , φ ) e − i φ = − m cot ϑ Y l m ( ϑ , φ ) − l ( l + 1 ) − m ( m − 1 ) Y l , m − 1 ( ϑ , φ ) e i φ = 2 1 l ( l + 1 ) − m ( m + 1 ) Y l , m + 1 ( ϑ , φ ) e − i φ − 2 1 l ( l + 1 ) − m ( m − 1 ) Y l , m − 1 ( ϑ , φ ) e i φ (1018)
sin ϑ ∂ ∂ ϑ Y l m ( ϑ , φ ) = − sin 2 ϑ ∂ ∂ cos ϑ Y l m ( ϑ , φ ) = l cos ϑ Y l m ( ϑ , φ ) − 2 l + 1 2 l − 1 ( l 2 − m 2 ) Y l − 1 , m ( ϑ , φ ) = − ( l + 1 ) cos ϑ Y l m ( ϑ , φ ) + 2 l + 1 2 l + 3 [ ( l + 1 ) 2 − m 2 ] Y l + 1 , m ( ϑ , φ ) = l ( l + 1 ) 2 − m 2 ( 2 l + 1 ) ( 2 l + 3 ) Y l + 1 , m ( ϑ , φ ) − ( l + 1 ) l 2 − m 2 ( 2 l + 1 ) ( 2 l − 1 ) Y l − 1 , m ( ϑ , φ ) (1019) \begin{aligned}
\sin\vartheta\frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi)
&=-\sin^{2}\vartheta\frac{\partial}{\partial\cos\vartheta}Y_{lm}(\vartheta,\varphi)\\
&=l\cos\vartheta Y_{lm}(\vartheta,\varphi)-\sqrt{\frac{2l+1}{2l-1}(l^{2}-m^{2})}Y_{l-1,m}(\vartheta,\varphi)\\
&=-(l+1)\cos\vartheta Y_{lm}(\vartheta,\varphi)+\sqrt{\frac{2l+1}{2l+3}[(l+1)^{2}-m^{2}]}Y_{l+1,m}(\vartheta,\varphi)\\
&=l\sqrt{\frac{(l+1)^{2}-m^{2}}{(2l+1)(2l+3)}}Y_{l+1,m}(\vartheta,\varphi)
-(l+1)\sqrt{\frac{l^{2}-m^{2}}{(2l+1)(2l-1)}}Y_{l-1,m}(\vartheta,\varphi) \tag*{(1019)}
\end{aligned} sin ϑ ∂ ϑ ∂ Y l m ( ϑ , φ ) = − sin 2 ϑ ∂ cos ϑ ∂ Y l m ( ϑ , φ ) = l cos ϑ Y l m ( ϑ , φ ) − 2 l − 1 2 l + 1 ( l 2 − m 2 ) Y l − 1 , m ( ϑ , φ ) = − ( l + 1 ) cos ϑ Y l m ( ϑ , φ ) + 2 l + 3 2 l + 1 [( l + 1 ) 2 − m 2 ] Y l + 1 , m ( ϑ , φ ) = l ( 2 l + 1 ) ( 2 l + 3 ) ( l + 1 ) 2 − m 2 Y l + 1 , m ( ϑ , φ ) − ( l + 1 ) ( 2 l + 1 ) ( 2 l − 1 ) l 2 − m 2 Y l − 1 , m ( ϑ , φ ) (1019)
∂ 2 ∂ ϑ 2 Y l m ( ϑ , φ ) = − [ l ( l + 1 ) − m 2 sin 2 ϑ ] Y l m ( ϑ , φ ) − cot ϑ ∂ ∂ ϑ Y l m ( ϑ , φ ) . (1020) \frac{\partial^{2}}{\partial\vartheta^{2}}Y_{lm}(\vartheta,\varphi)
=-\left[l(l+1)-\frac{m^{2}}{\sin^{2}\vartheta}\right]Y_{lm}(\vartheta,\varphi)-\cot\vartheta\frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi). \tag*{(1020)} ∂ ϑ 2 ∂ 2 Y l m ( ϑ , φ ) = − [ l ( l + 1 ) − sin 2 ϑ m 2 ] Y l m ( ϑ , φ ) − cot ϑ ∂ ϑ ∂ Y l m ( ϑ , φ ) . (1020)
sin 2 ϑ ∂ 2 ∂ ( cos ϑ ) 2 Y l m ( ϑ , φ ) = 2 cos ϑ ∂ ∂ cos ϑ Y l m ( ϑ , φ ) − [ l ( l + 1 ) − m 2 sin 2 ϑ ] Y l m ( ϑ , φ ) . (1021) \sin^{2}\vartheta\frac{\partial^{2}}{\partial(\cos\vartheta)^{2}}Y_{lm}(\vartheta,\varphi)
=2\cos\vartheta\frac{\partial}{\partial\cos\vartheta}Y_{lm}(\vartheta,\varphi)-\left[l(l+1)-\frac{m^{2}}{\sin^{2}\vartheta}\right]Y_{lm}(\vartheta,\varphi). \tag*{(1021)} sin 2 ϑ ∂ ( cos ϑ ) 2 ∂ 2 Y l m ( ϑ , φ ) = 2 cos ϑ ∂ cos ϑ ∂ Y l m ( ϑ , φ ) − [ l ( l + 1 ) − sin 2 ϑ m 2 ] Y l m ( ϑ , φ ) . (1021)
Vector Differentiation Operations
(a) The gradient of a function f ( r ) Y l m ( ϑ , φ ) f(r)Y_{lm}(\vartheta,\varphi) f ( r ) Y l m ( ϑ , φ ) , where f ( r ) f(r) f ( r ) is an arbitrary function of r ≡ ∣ r ∣ r\equiv|\mathbf{r}| r ≡ ∣ r ∣ can be expanded in terms of vector spherical harmonics (Sec. 7.3) as follows:
∇ [ f ( r ) Y l m ( ϑ , φ ) ] = − l + 1 2 l + 1 ( d f d r − l r f ) Y l m l + 1 ( ϑ , φ ) + l 2 l + 1 ( d f d r + l + 1 r f ) Y l m l − 1 ( ϑ , φ ) (1022) \nabla[f(r)Y_{lm}(\vartheta,\varphi)]=-\sqrt{\frac{l+1}{2l+1}}\left(\frac{df}{dr}-\frac{l}{r}f\right)\mathbf{Y}_{lm}^{l+1}(\vartheta,\varphi)+\sqrt{\frac{l}{2l+1}}\left(\frac{df}{dr}+\frac{l+1}{r}f\right)\mathbf{Y}_{lm}^{l-1}(\vartheta,\varphi) \tag*{(1022)} ∇ [ f ( r ) Y l m ( ϑ , φ )] = − 2 l + 1 l + 1 ( d r df − r l f ) Y l m l + 1 ( ϑ , φ ) + 2 l + 1 l ( d r df + r l + 1 f ) Y l m l − 1 ( ϑ , φ ) (1022)
or in component form
∇ 0 [ f ( r ) Y l m ( ϑ , φ ) ] = ( l + 1 ) 2 − m 2 ( 2 l + 1 ) ( 2 l + 3 ) ( d f d r − l r f ) Y l + 1 , m ( ϑ , φ ) + l 2 − m 2 ( 2 l − 1 ) ( 2 l + 1 ) ( d f d r + l + 1 r f ) Y l − 1 , m ( ϑ , φ ) (1023) \begin{aligned}
\nabla_{0}[f(r)Y_{lm}(\vartheta,\varphi)]
&=\sqrt{\frac{(l+1)^{2}-m^{2}}{(2l+1)(2l+3)}}\left(\frac{df}{dr}-\frac{l}{r}f\right)Y_{l+1,m}(\vartheta,\varphi)\\
&\quad+\sqrt{\frac{l^{2}-m^{2}}{(2l-1)(2l+1)}}\left(\frac{df}{dr}+\frac{l+1}{r}f\right)Y_{l-1,m}(\vartheta,\varphi) \tag*{(1023)}
\end{aligned} ∇ 0 [ f ( r ) Y l m ( ϑ , φ )] = ( 2 l + 1 ) ( 2 l + 3 ) ( l + 1 ) 2 − m 2 ( d r df − r l f ) Y l + 1 , m ( ϑ , φ ) + ( 2 l − 1 ) ( 2 l + 1 ) l 2 − m 2 ( d r df + r l + 1 f ) Y l − 1 , m ( ϑ , φ ) (1023)
∇ ± 1 [ f ( r ) Y l m ( ϑ , φ ) ] = ( l ± m + 1 ) ( l ± m + 2 ) 2 ( 2 l + 1 ) ( 2 l + 3 ) ( d f d r − l r f ) Y l + 1 , m ± 1 ( ϑ , φ ) − ( l ∓ m − 1 ) ( l ∓ m ) 2 ( 2 l − 1 ) ( 2 l + 1 ) ( d f d r + l + 1 r f ) Y l − 1 , m ± 1 ( ϑ , φ ) (1024) \begin{aligned}
\nabla_{\pm1}[f(r)Y_{lm}(\vartheta,\varphi)]
&=\sqrt{\frac{(l\pm m+1)(l\pm m+2)}{2(2l+1)(2l+3)}}\left(\frac{df}{dr}-\frac{l}{r}f\right)Y_{l+1,m\pm1}(\vartheta,\varphi)\\
&\quad-\sqrt{\frac{(l\mp m-1)(l\mp m)}{2(2l-1)(2l+1)}}\left(\frac{df}{dr}+\frac{l+1}{r}f\right)Y_{l-1,m\pm1}(\vartheta,\varphi) \tag*{(1024)}
\end{aligned} ∇ ± 1 [ f ( r ) Y l m ( ϑ , φ )] = 2 ( 2 l + 1 ) ( 2 l + 3 ) ( l ± m + 1 ) ( l ± m + 2 ) ( d r df − r l f ) Y l + 1 , m ± 1 ( ϑ , φ ) − 2 ( 2 l − 1 ) ( 2 l + 1 ) ( l ∓ m − 1 ) ( l ∓ m ) ( d r df + r l + 1 f ) Y l − 1 , m ± 1 ( ϑ , φ ) (1024)
Here ∇ 0 , ± 1 \nabla_{0,\pm1} ∇ 0 , ± 1 are spherical components of the operator ∇ \nabla ∇ (Sec. 1.3).
Let us consider some special cases of Eq. (1022) . If f ( r ) = j l ( k r ) f(r)=j_l(kr) f ( r ) = j l ( k r ) is a spherical Bessel function, one has,
1 k ∇ [ j l ( k r ) Y l m ( ϑ , φ ) ] = l + 1 2 l + 1 j l + 1 ( k r ) Y l m l + 1 ( ϑ , φ ) + l 2 l + 1 j l − 1 ( k r ) Y l m l − 1 ( ϑ , φ ) (1025) \frac{1}{k}\nabla\left[j_l(kr)Y_{lm}(\vartheta,\varphi)\right]=\sqrt{\frac{l+1}{2l+1}}j_{l+1}(kr)Y_{lm}^{l+1}(\vartheta,\varphi)+\sqrt{\frac{l}{2l+1}}j_{l-1}(kr)Y_{lm}^{l-1}(\vartheta,\varphi) \tag*{(1025)} k 1 ∇ [ j l ( k r ) Y l m ( ϑ , φ ) ] = 2 l + 1 l + 1 j l + 1 ( k r ) Y l m l + 1 ( ϑ , φ ) + 2 l + 1 l j l − 1 ( k r ) Y l m l − 1 ( ϑ , φ ) (1025)
Putting f ( r ) = r l f(r)=r^l f ( r ) = r l , one gets
∇ [ r l Y l m ( ϑ , φ ) ] = l ( 2 l + 1 ) r l − 1 Y l m l − 1 ( ϑ , φ ) (1026) \nabla\left[r^lY_{lm}(\vartheta,\varphi)\right]=\sqrt{l(2l+1)}r^{l-1}Y_{lm}^{l-1}(\vartheta,\varphi) \tag*{(1026)} ∇ [ r l Y l m ( ϑ , φ ) ] = l ( 2 l + 1 ) r l − 1 Y l m l − 1 ( ϑ , φ ) (1026)
If f ( r ) = r − l − 1 f(r)=r^{-l-1} f ( r ) = r − l − 1 , Eq. (1022) is reduced to
∇ [ r − l − 1 Y l m ( ϑ , φ ) ] = ( l + 1 ) ( 2 l + 1 ) r − l − 2 Y l m l + 1 ( ϑ , φ ) (1027) \nabla\left[r^{-l-1}Y_{lm}(\vartheta,\varphi)\right]=\sqrt{(l+1)(2l+1)}r^{-l-2}Y_{lm}^{l+1}(\vartheta,\varphi) \tag*{(1027)} ∇ [ r − l − 1 Y l m ( ϑ , φ ) ] = ( l + 1 ) ( 2 l + 1 ) r − l − 2 Y l m l + 1 ( ϑ , φ ) (1027)
(b) The divergence of a vector function f ( r ) L ^ Y l m ( ϑ , φ ) f(r)\hat{\mathbf{L}}Y_{lm}(\vartheta,\varphi) f ( r ) L ^ Y l m ( ϑ , φ ) vanishes for any function f ( r ) f(r) f ( r )
∇ ⋅ [ f ( r ) L ^ Y l m ( ϑ , φ ) ] = 0. (1028) \nabla\cdot\left[f(r)\hat{\mathbf{L}}Y_{lm}(\vartheta,\varphi)\right]=0. \tag*{(1028)} ∇ ⋅ [ f ( r ) L ^ Y l m ( ϑ , φ ) ] = 0. (1028)
(c) The curl of a vector function f ( r ) L ^ Y l m ( ϑ , φ ) f(r)\hat{\mathbf{L}}Y_{lm}(\vartheta,\varphi) f ( r ) L ^ Y l m ( ϑ , φ ) , where f ( r ) f(r) f ( r ) is an arbitrary function of r r r , is expressible in terms of a gradient of the corresponding scalar function
∇ × [ f ( r ) L ^ Y l m ( ϑ , φ ) ] = i ∇ [ ( r d f d r + f ) Y l m ( ϑ , φ ) ] − i r [ 1 r 2 d d r ( r 2 d f d r ) − l ( l + 1 ) r 2 f ] Y l m ( ϑ , φ ) . (1029) \nabla\times\left[f(r)\hat{\mathbf{L}}Y_{lm}(\vartheta,\varphi)\right]=i\nabla\left[\left(r\frac{df}{dr}+f\right)Y_{lm}(\vartheta,\varphi)\right]-ir\left[\frac{1}{r^2}\frac{d}{dr}\left(r^2\frac{df}{dr}\right)-\frac{l(l+1)}{r^2}f\right]Y_{lm}(\vartheta,\varphi). \tag*{(1029)} ∇ × [ f ( r ) L ^ Y l m ( ϑ , φ ) ] = i ∇ [ ( r d r df + f ) Y l m ( ϑ , φ ) ] − i r [ r 2 1 d r d ( r 2 d r df ) − r 2 l ( l + 1 ) f ] Y l m ( ϑ , φ ) . (1029)
In particular, if f ( r ) = j l ( k r ) f(r)=j_l(kr) f ( r ) = j l ( k r ) , one gets
∇ × [ j l ( k r ) L ^ Y l m ( ϑ , φ ) ] = i ∇ [ ( k r j l − 1 ( k r ) − l j l ( k r ) ) Y l m ( ϑ , φ ) ] + i r k 2 j l ( k r ) Y l m ( ϑ , φ ) . (1030) \nabla\times\left[j_l(kr)\hat{\mathbf{L}}Y_{lm}(\vartheta,\varphi)\right]=i\nabla\left[(krj_{l-1}(kr)-lj_l(kr))Y_{lm}(\vartheta,\varphi)\right]+irk^2j_l(kr)Y_{lm}(\vartheta,\varphi). \tag*{(1030)} ∇ × [ j l ( k r ) L ^ Y l m ( ϑ , φ ) ] = i ∇ [ ( k r j l − 1 ( k r ) − l j l ( k r )) Y l m ( ϑ , φ ) ] + i r k 2 j l ( k r ) Y l m ( ϑ , φ ) . (1030)
If f ( r ) = r f(r)=r f ( r ) = r ,
∇ × [ r L ^ Y l m ( ϑ , φ ) ] = i ( l + 1 ) l ( 2 l + 1 ) r l − 1 Y l m l − 1 ( ϑ , φ ) . (1031) \nabla\times\left[r\hat{\mathbf{L}}Y_{lm}(\vartheta,\varphi)\right]=i(l+1)\sqrt{l(2l+1)}r^{l-1}Y_{lm}^{l-1}(\vartheta,\varphi). \tag*{(1031)} ∇ × [ r L ^ Y l m ( ϑ , φ ) ] = i ( l + 1 ) l ( 2 l + 1 ) r l − 1 Y l m l − 1 ( ϑ , φ ) . (1031)
and if f ( r ) = r − l − 1 f(r)=r^{-l-1} f ( r ) = r − l − 1 ,
∇ × [ r − l − 1 L ^ Y l m ( ϑ , φ ) ] = − i l ( l + 1 ) ( 2 l + 1 ) r − l − 2 Y l m l + 1 ( ϑ , φ ) . (1032) \nabla\times\left[r^{-l-1}\hat{\mathbf{L}}Y_{lm}(\vartheta,\varphi)\right]=-il\sqrt{(l+1)(2l+1)}r^{-l-2}Y_{lm}^{l+1}(\vartheta,\varphi). \tag*{(1032)} ∇ × [ r − l − 1 L ^ Y l m ( ϑ , φ ) ] = − i l ( l + 1 ) ( 2 l + 1 ) r − l − 2 Y l m l + 1 ( ϑ , φ ) . (1032)
Some additional formulas for the vector differentiation of functions which contain Y l m ( θ , φ ) Y_{lm}(\theta,\varphi) Y l m ( θ , φ ) will be given in § 27.4.6 .
Some integrals involving spherical harmonics
Integrals over Total Solid Angle
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l m ( ϑ , φ ) = 4 π δ l 0 δ m 0 (1033) \int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Y_{lm}(\vartheta,\varphi)=\sqrt{4\pi}\delta_{l0}\delta_{m0} \tag*{(1033)} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l m ( ϑ , φ ) = 4 π δ l 0 δ m 0 (1033)
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ∗ ( ϑ , φ ) = δ l 1 l 2 δ m 1 m 2 (1034) \int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}^{*}(\vartheta,\varphi)=\delta_{l_1l_2}\delta_{m_1m_2} \tag*{(1034)} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ∗ ( ϑ , φ ) = δ l 1 l 2 δ m 1 m 2 (1034)
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) = ( − 1 ) m 2 δ l 1 l 2 δ − m 1 m 2 (1035) \int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)=(-1)^{m_2}\delta_{l_1l_2}\delta_{-m_1m_2} \tag*{(1035)} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) = ( − 1 ) m 2 δ l 1 l 2 δ − m 1 m 2 (1035)
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) Y l 3 m 3 ∗ ( ϑ , φ ) = ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) 4 π ( 2 l 3 + 1 ) C l 1 0 l 2 0 l 3 0 C l 1 m 1 l 2 m 2 l 3 m 3 (1036) \int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)Y_{l_3m_3}^{*}(\vartheta,\varphi)=\sqrt{\frac{(2l_1+1)(2l_2+1)}{4\pi(2l_3+1)}}C_{l_1 0\,l_2 0}^{l_3 0}C_{l_1m_1\,l_2m_2}^{l_3m_3} \tag*{(1036)} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) Y l 3 m 3 ∗ ( ϑ , φ ) = 4 π ( 2 l 3 + 1 ) ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) C l 1 0 l 2 0 l 3 0 C l 1 m 1 l 2 m 2 l 3 m 3 (1036)
∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) Y l 3 m 3 ( ϑ , φ ) = ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 2 l 3 + 1 ) 4 π ( l 1 l 2 l 3 0 0 0 ) ( l 1 l 2 l 3 m 1 m 2 m 3 ) (1037) \begin{aligned}
\int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)Y_{l_3m_3}(\vartheta,\varphi)
&=\sqrt{\frac{(2l_1+1)(2l_2+1)(2l_3+1)}{4\pi}}
\begin{pmatrix} l_1 & l_2 & l_3 \\ 0 & 0 & 0 \end{pmatrix}
\begin{pmatrix} l_1 & l_2 & l_3 \\ m_1 & m_2 & m_3 \end{pmatrix} \tag*{(1037)}
\end{aligned} ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) Y l 3 m 3 ( ϑ , φ ) = 4 π ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 2 l 3 + 1 ) ( l 1 0 l 2 0 l 3 0 ) ( l 1 m 1 l 2 m 2 l 3 m 3 ) (1037)
Integrals involving products of three and more spherical harmonics can be evaluated by reducing them to integrals which involve products of a smaller number of harmonics. For this purpose it is convenient to use the Clebsch-Gordan expansion (Eq. (996) ).
Fourier Transformations for Some Functions Which Contain Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ )
∫ 0 ∞ d r r 2 ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ e i k r j l ( q r ) Y l m ( ϑ , φ ) = 2 π 2 i l δ ( q − k ) q 2 Y l m ( ϑ k , φ k ) (1038) \int_{0}^{\infty} dr r^{2} \int_{0}^{2\pi} d\varphi\int_{0}^{\pi} d\vartheta\sin\vartheta e^{i\mathbf{k}\mathbf{r}} j_l(qr)Y_{lm}(\vartheta,\varphi)=2\pi^{2}i^l\frac{\delta(q-k)}{q^{2}}Y_{lm}(\vartheta_k,\varphi_k) \tag*{(1038)} ∫ 0 ∞ d r r 2 ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ e i kr j l ( q r ) Y l m ( ϑ , φ ) = 2 π 2 i l q 2 δ ( q − k ) Y l m ( ϑ k , φ k ) (1038)
∫ 0 ∞ d r r 2 ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ e i k r L ^ [ j l ( q r ) Y l m ( ϑ , φ ) ] = 2 π 2 i l δ ( q − k ) q 2 L ^ k Y l m ( ϑ k , φ k ) (1039) \int_{0}^{\infty} dr r^{2} \int_{0}^{2\pi} d\varphi\int_{0}^{\pi} d\vartheta\sin\vartheta e^{i\mathbf{k}\mathbf{r}}\hat{\mathbf{L}}\left[j_l(qr)Y_{lm}(\vartheta,\varphi)\right]=2\pi^{2}i^l\frac{\delta(q-k)}{q^{2}}\hat{\mathbf{L}}_{\mathbf{k}}Y_{lm}(\vartheta_k,\varphi_k) \tag*{(1039)} ∫ 0 ∞ d r r 2 ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ e i kr L ^ [ j l ( q r ) Y l m ( ϑ , φ ) ] = 2 π 2 i l q 2 δ ( q − k ) L ^ k Y l m ( ϑ k , φ k ) (1039)
∫ 0 ∞ d r r 2 ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ e i k r [ ∇ × L ^ ] ( j l ( q r ) Y l m ( ϑ , φ ) ) = 2 π 2 i l − 1 δ ( q − k ) q 2 [ k × L ^ k ] Y l m ( ϑ k , φ k ) (1040) \int_{0}^{\infty} dr r^{2} \int_{0}^{2\pi} d\varphi\int_{0}^{\pi} d\vartheta\sin\vartheta e^{i\mathbf{k}\mathbf{r}}[\nabla\times\hat{\mathbf{L}}]\left(j_l(qr)Y_{lm}(\vartheta,\varphi)\right)=2\pi^{2}i^{l-1}\frac{\delta(q-k)}{q^{2}}\left[\mathbf{k}\times\hat{\mathbf{L}}_{\mathbf{k}}\right]Y_{lm}(\vartheta_k,\varphi_k) \tag*{(1040)} ∫ 0 ∞ d r r 2 ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ e i kr [ ∇ × L ^ ] ( j l ( q r ) Y l m ( ϑ , φ ) ) = 2 π 2 i l − 1 q 2 δ ( q − k ) [ k × L ^ k ] Y l m ( ϑ k , φ k ) (1040)
Here j l ( x ) j_l(x) j l ( x ) is a spherical Bessel function, ϑ k \vartheta_k ϑ k and φ k \varphi_k φ k are polar angles of the vector k \mathbf{k} k ; L ^ \hat{\mathbf{L}} L ^ and L ^ k \hat{\mathbf{L}}_{\mathbf{k}} L ^ k are the operators of orbital angular momentum in the coordinate and momentum representations, respectively,
L ^ = − i [ r × ∇ ] , L ^ k = − i [ k × ∇ k ] . (1041) \hat{\mathbf{L}}=-i[\mathbf{r}\times\nabla],\qquad\hat{\mathbf{L}}_{\mathbf{k}}=-i[\mathbf{k}\times\nabla_{\mathbf{k}}]. \tag*{(1041)} L ^ = − i [ r × ∇ ] , L ^ k = − i [ k × ∇ k ] . (1041)
Integrals with Respect to
∫ 0 π Y l m ∗ ( ϑ , φ ) Y l ′ m ( ϑ , φ ) sin ϑ d ϑ = δ l l ′ 2 π . (1042) \int_{0}^{\pi}Y^{*}_{lm}(\vartheta,\varphi)Y_{l'm}(\vartheta,\varphi)\sin\vartheta\,d\vartheta=\frac{\delta_{ll'}}{2\pi}. \tag*{(1042)} ∫ 0 π Y l m ∗ ( ϑ , φ ) Y l ′ m ( ϑ , φ ) sin ϑ d ϑ = 2 π δ l l ′ . (1042)
∫ 0 π Y l m ∗ ( ϑ , φ ) Y l m ′ ( ϑ , φ ) d ϑ sin ϑ = 2 l + 1 4 π m δ m m ′ ( m , m ′ > 0 ) . (1043) \int_{0}^{\pi}Y^{*}_{lm}(\vartheta,\varphi)Y_{lm'}(\vartheta,\varphi)\frac{d\vartheta}{\sin\vartheta}=\frac{2l+1}{4\pi m}\delta_{mm'}\quad(m,m'>0). \tag*{(1043)} ∫ 0 π Y l m ∗ ( ϑ , φ ) Y l m ′ ( ϑ , φ ) sin ϑ d ϑ = 4 π m 2 l + 1 δ m m ′ ( m , m ′ > 0 ) . (1043)
∫ 0 π / 2 ( sin ϑ ) m + 1 ( cos ϑ ) n Y l m ( ϑ , φ ) d ϑ = 2 l + 1 4 π ⋅ ( l + m ) ! ( l − m ) ! ( − 1 ) m n ! e i m φ ( n + l + m + 1 ) ! ! ( n − l + m ) ! ! , ( m , n ≥ 0 ) (1044) \int_{0}^{\pi/2}(\sin\vartheta)^{m+1}(\cos\vartheta)^nY_{lm}(\vartheta,\varphi)\,d\vartheta
=\sqrt{\frac{2l+1}{4\pi}\cdot\frac{(l+m)!}{(l-m)!}}(-1)^m\frac{n!e^{im\varphi}}{(n+l+m+1)!!(n-l+m)!!},\quad(m,n\geq0) \tag*{(1044)} ∫ 0 π /2 ( sin ϑ ) m + 1 ( cos ϑ ) n Y l m ( ϑ , φ ) d ϑ = 4 π 2 l + 1 ⋅ ( l − m )! ( l + m )! ( − 1 ) m ( n + l + m + 1 )!! ( n − l + m )!! n ! e im φ , ( m , n ≥ 0 ) (1044)
∫ a b [ ( l 1 − l 2 ) ( l 1 + l 2 + 1 ) − m 1 2 − m 2 2 sin 2 ϑ ] Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) d ( cos ϑ ) = sin 2 ϑ [ Y l 1 m 1 ( ϑ , φ ) d d cos ϑ Y l 2 m 2 ( ϑ , φ ) − Y l 2 m 2 ( ϑ , φ ) d d cos ϑ Y l 1 m 1 ( ϑ , φ ) ] ∣ a b = [ cos ϑ ( l 1 − l 2 ) Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) + ( 2 l 2 + 1 ) ( l 2 2 − m 2 2 ) 2 l 2 − 1 Y l 1 m 1 ( ϑ , φ ) Y l 2 − 1 m 2 ( ϑ , φ ) − ( 2 l 1 + 1 ) ( l 1 2 − m 1 2 ) 2 l 1 − 1 Y l 1 − 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) ] a b . (1045) \begin{aligned}
&\int_{a}^{b}\left[(l_1-l_2)(l_1+l_2+1)-\frac{m_1^2-m_2^2}{\sin^2\vartheta}\right]Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)\,d(\cos\vartheta)\\
&=\left.\sin^2\vartheta\left[Y_{l_1m_1}(\vartheta,\varphi)\frac{d}{d\cos\vartheta}Y_{l_2m_2}(\vartheta,\varphi)-Y_{l_2m_2}(\vartheta,\varphi)\frac{d}{d\cos\vartheta}Y_{l_1m_1}(\vartheta,\varphi)\right]\right|_{a}^{b}\\
&=\left[\cos\vartheta(l_1-l_2)Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)+\sqrt{\frac{(2l_2+1)(l_2^2-m_2^2)}{2l_2-1}}Y_{l_1m_1}(\vartheta,\varphi)Y_{l_2-1m_2}(\vartheta,\varphi)\right.\\
&\qquad\left.-\sqrt{\frac{(2l_1+1)(l_1^2-m_1^2)}{2l_1-1}}Y_{l_1-1m_1}(\vartheta,\varphi)Y_{l_2m_2}(\vartheta,\varphi)\right]_{a}^{b}. \tag*{(1045)}
\end{aligned} ∫ a b [ ( l 1 − l 2 ) ( l 1 + l 2 + 1 ) − sin 2 ϑ m 1 2 − m 2 2 ] Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) d ( cos ϑ ) = sin 2 ϑ [ Y l 1 m 1 ( ϑ , φ ) d cos ϑ d Y l 2 m 2 ( ϑ , φ ) − Y l 2 m 2 ( ϑ , φ ) d cos ϑ d Y l 1 m 1 ( ϑ , φ ) ] a b = cos ϑ ( l 1 − l 2 ) Y l 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) + 2 l 2 − 1 ( 2 l 2 + 1 ) ( l 2 2 − m 2 2 ) Y l 1 m 1 ( ϑ , φ ) Y l 2 − 1 m 2 ( ϑ , φ ) − 2 l 1 − 1 ( 2 l 1 + 1 ) ( l 1 2 − m 1 2 ) Y l 1 − 1 m 1 ( ϑ , φ ) Y l 2 m 2 ( ϑ , φ ) a b . (1045)
Sums involving spherical harmonics
Sums over m m m (with fixed l l l )
∑ m = − l l ∣ Y l m ( ϑ , φ ) ∣ 2 = 2 l + 1 4 π , (1046) \sum_{m=-l}^{l}|Y_{lm}(\vartheta,\varphi)|^{2}=\frac{2l+1}{4\pi}, \tag*{(1046)} m = − l ∑ l ∣ Y l m ( ϑ , φ ) ∣ 2 = 4 π 2 l + 1 , (1046)
∑ m = − l l m ∣ Y l m ( ϑ , φ ) ∣ 2 = 0 , (1047) \sum_{m=-l}^{l}m|Y_{lm}(\vartheta,\varphi)|^{2}=0, \tag*{(1047)} m = − l ∑ l m ∣ Y l m ( ϑ , φ ) ∣ 2 = 0 , (1047)
∑ m = − l l m 2 ∣ Y l m ( ϑ , φ ) ∣ 2 = l ( l + 1 ) ( 2 l + 1 ) 8 π sin 2 ϑ , (1048) \sum_{m=-l}^{l}m^{2}|Y_{lm}(\vartheta,\varphi)|^{2}=\frac{l(l+1)(2l+1)}{8\pi}\sin^{2}\vartheta, \tag*{(1048)} m = − l ∑ l m 2 ∣ Y l m ( ϑ , φ ) ∣ 2 = 8 π l ( l + 1 ) ( 2 l + 1 ) sin 2 ϑ , (1048)
∑ m = − l l ( l 2 − m 2 ) [ ( l + 1 ) 2 − m 2 ] ( 2 l − 1 ) ( 2 l + 3 ) Y l − 1 m ( ϑ , φ ) Y l + 1 m ∗ ( ϑ , φ ) = l ( l + 1 ) 8 π ( 3 cos 2 ϑ − 1 ) , (1049) \sum_{m=-l}^{l}\sqrt{\frac{(l^{2}-m^{2})[(l+1)^{2}-m^{2}]}{(2l-1)(2l+3)}}Y_{l-1m}(\vartheta,\varphi)Y^{*}_{l+1m}(\vartheta,\varphi)=\frac{l(l+1)}{8\pi}(3\cos^{2}\vartheta-1), \tag*{(1049)} m = − l ∑ l ( 2 l − 1 ) ( 2 l + 3 ) ( l 2 − m 2 ) [( l + 1 ) 2 − m 2 ] Y l − 1 m ( ϑ , φ ) Y l + 1 m ∗ ( ϑ , φ ) = 8 π l ( l + 1 ) ( 3 cos 2 ϑ − 1 ) , (1049)
∑ m = − l l ( ∓ i ) m ( l − m ) ! ( l + m ) ! Y l m ( ϑ , φ ) = 2 l + 1 4 π ( cos ϑ ± i sin ϑ cos φ ) l l ! . (1050) \sum_{m=-l}^{l}\frac{(\mp i)^{m}}{\sqrt{(l-m)!(l+m)!}}Y_{lm}(\vartheta,\varphi)=\sqrt{\frac{2l+1}{4\pi}}\frac{(\cos\vartheta\pm i\sin\vartheta\cos\varphi)^{l}}{l!}. \tag*{(1050)} m = − l ∑ l ( l − m )! ( l + m )! ( ∓ i ) m Y l m ( ϑ , φ ) = 4 π 2 l + 1 l ! ( cos ϑ ± i sin ϑ cos φ ) l . (1050)
Sums over l l l (with fixed m ≥ 0 m \ge0 m ≥ 0 )
In the equations given below [25] we will use the notation R = 1 − 2 t cos ϑ + t 2 R=\sqrt{1-2t\cos\vartheta+t^2} R = 1 − 2 t cos ϑ + t 2 .
∑ l = m n + m ( − 1 ) l t n − l + m ( n − l + m ) ! ( l − m ) ! ( l + m ) ! 4 π 2 l + 1 Y l m ( ϑ , φ ) = ( − 1 ) n ( 2 m − 1 ) ! ! ( 2 m + n ) ! ( sin ϑ e i φ ) m ( 1 − 2 t cos ϑ + t 2 ) n / 2 C m + 1 / 2 n ( cos ϑ − t 1 − 2 t cos ϑ + t 2 ) , ( ∣ t ∣ < 1 ) , (1051) \sum_{l=m}^{n+m}(-1)^l\frac{t^{n-l+m}}{(n-l+m)!\sqrt{(l-m)!(l+m)!}}\sqrt{\frac{4\pi}{2l+1}}Y_{lm}(\vartheta,\varphi)
=(-1)^n\frac{(2m-1)!!}{(2m+n)!}(\sin\vartheta e^{i\varphi})^m(1-2t\cos\vartheta+t^2)^{n/2}C_{m+1/2}^{n}\left(\frac{\cos\vartheta-t}{\sqrt{1-2t\cos\vartheta+t^2}}\right),\quad(|t|<1), \tag*{(1051)} l = m ∑ n + m ( − 1 ) l ( n − l + m )! ( l − m )! ( l + m )! t n − l + m 2 l + 1 4 π Y l m ( ϑ , φ ) = ( − 1 ) n ( 2 m + n )! ( 2 m − 1 )!! ( sin ϑ e i φ ) m ( 1 − 2 t cos ϑ + t 2 ) n /2 C m + 1/2 n ( 1 − 2 t cos ϑ + t 2 cos ϑ − t ) , ( ∣ t ∣ < 1 ) , (1051)
∑ l = m ∞ 4 π ( 2 l + 1 ) ( l − m ) ! ( l + m ) ! 1 l ! t l − m Y l m ( ϑ , φ ) = ( − sin ϑ e i φ ) m 2 m ( m ! ) 2 0 F 1 ( ; m + 1 ; t cos 2 ϑ 2 ) 0 F 1 ( ; m + 1 ; − t sin 2 ϑ 2 ) , ( ∣ t ∣ < 1 ) , (1052) \sum_{l=m}^{\infty}\sqrt{\frac{4\pi}{(2l+1)(l-m)!(l+m)!}}\frac{1}{l!}t^{l-m}Y_{lm}(\vartheta,\varphi)
=\frac{(-\sin\vartheta e^{i\varphi})^m}{2^m(m!)^2}{}_0F_1\left(;m+1;t\cos^2\frac{\vartheta}{2}\right){}_0F_1\left(;m+1;-t\sin^2\frac{\vartheta}{2}\right),\quad(|t|<1), \tag*{(1052)} l = m ∑ ∞ ( 2 l + 1 ) ( l − m )! ( l + m )! 4 π l ! 1 t l − m Y l m ( ϑ , φ ) = 2 m ( m ! ) 2 ( − sin ϑ e i φ ) m 0 F 1 ( ; m + 1 ; t cos 2 2 ϑ ) 0 F 1 ( ; m + 1 ; − t sin 2 2 ϑ ) , ( ∣ t ∣ < 1 ) , (1052)
∑ l = m ∞ ( n + l − m ) ! t l − m ( l + m ) ! ( l − m ) ! 4 π 2 l + 1 Y l m ( ϑ , φ ) = n ! 2 m m ! ( − sin ϑ e i φ ) m ( 1 − t cos ϑ ) n + 1 F ( n + 1 2 , n 2 + 1 ; m + 1 ; − ( t sin ϑ 1 − t cos ϑ ) 2 ) , ( ∣ t ∣ < 1 ) , (1053) \sum_{l=m}^{\infty}\frac{(n+l-m)!t^{l-m}}{\sqrt{(l+m)!(l-m)!}}\sqrt{\frac{4\pi}{2l+1}}Y_{lm}(\vartheta,\varphi)
=\frac{n!}{2^m m!}\frac{(-\sin\vartheta e^{i\varphi})^m}{(1-t\cos\vartheta)^{n+1}}F\left(\frac{n+1}{2},\frac{n}{2}+1;m+1;-\left(\frac{t\sin\vartheta}{1-t\cos\vartheta}\right)^2\right),\quad(|t|<1), \tag*{(1053)} l = m ∑ ∞ ( l + m )! ( l − m )! ( n + l − m )! t l − m 2 l + 1 4 π Y l m ( ϑ , φ ) = 2 m m ! n ! ( 1 − t cos ϑ ) n + 1 ( − sin ϑ e i φ ) m F ( 2 n + 1 , 2 n + 1 ; m + 1 ; − ( 1 − t cos ϑ t sin ϑ ) 2 ) , ( ∣ t ∣ < 1 ) , (1053)
∑ l = m ∞ 4 π ( 2 l + 1 ) ( l − m ) ! ( l + m ) ! ( l + m − n ) ! ( l − m + n − 1 ) ! l ! t l − m Y l m ( ϑ , φ ) = ( 2 m − n ) ! ( n − 1 ) ! 2 m ( m ! ) 2 ( − sin ϑ e i φ ) m F ( n , 2 m + 1 − n ; m + 1 ; 1 − t − R 2 ) F ( n , 2 m + 1 − n ; m + 1 ; 1 + t − R 2 ) , ( ∣ t ∣ < 1 ) , (1054) \sum_{l=m}^{\infty}\sqrt{\frac{4\pi}{(2l+1)(l-m)!(l+m)!}}\frac{(l+m-n)!(l-m+n-1)!}{l!}t^{l-m}Y_{lm}(\vartheta,\varphi)
=\frac{(2m-n)!(n-1)!}{2^m(m!)^2}(-\sin\vartheta e^{i\varphi})^mF\left(n,2m+1-n;m+1;\frac{1-t-R}{2}\right)F\left(n,2m+1-n;m+1;\frac{1+t-R}{2}\right),\quad(|t|<1), \tag*{(1054)} l = m ∑ ∞ ( 2 l + 1 ) ( l − m )! ( l + m )! 4 π l ! ( l + m − n )! ( l − m + n − 1 )! t l − m Y l m ( ϑ , φ ) = 2 m ( m ! ) 2 ( 2 m − n )! ( n − 1 )! ( − sin ϑ e i φ ) m F ( n , 2 m + 1 − n ; m + 1 ; 2 1 − t − R ) F ( n , 2 m + 1 − n ; m + 1 ; 2 1 + t − R ) , ( ∣ t ∣ < 1 ) , (1054)
∑ l = m ∞ 4 π ( l − m ) ! ( 2 l + 1 ) ( l + m ) ! L l − m 2 m ( y ) t l − m Y l m ( ϑ , φ ) = 1 2 m m ! ( − sin ϑ e i φ ) m R 2 m + 1 exp [ − y t ( cos ϑ − t ) R 2 ] 0 F 1 ( ; m + 1 ; − y 2 t 2 sin 2 ϑ 4 R 4 ) , ( ∣ t ∣ < 1 ) , (1055) \sum_{l=m}^{\infty}\sqrt{\frac{4\pi(l-m)!}{(2l+1)(l+m)!}}L_{l-m}^{2m}(y)t^{l-m}Y_{lm}(\vartheta,\varphi)
=\frac{1}{2^m m!}\frac{(-\sin\vartheta e^{i\varphi})^m}{R^{2m+1}}\exp\left[-\frac{yt(\cos\vartheta-t)}{R^2}\right]{}_0F_1\left(;m+1;-\frac{y^2t^2\sin^2\vartheta}{4R^4}\right),\quad(|t|<1), \tag*{(1055)} l = m ∑ ∞ ( 2 l + 1 ) ( l + m )! 4 π ( l − m )! L l − m 2 m ( y ) t l − m Y l m ( ϑ , φ ) = 2 m m ! 1 R 2 m + 1 ( − sin ϑ e i φ ) m exp [ − R 2 y t ( cos ϑ − t ) ] 0 F 1 ( ; m + 1 ; − 4 R 4 y 2 t 2 sin 2 ϑ ) , ( ∣ t ∣ < 1 ) , (1055)
∑ l = m ∞ i l − m 4 π ( 2 l + 1 ) ( l + m ) ! ( l − m ) ! j l ( t ) Y l m ( ϑ , φ ) = ( − t sin ϑ e i φ ) m e i t cos ϑ , (1056) \sum_{l=m}^{\infty}i^{l-m}\sqrt{\frac{4\pi(2l+1)(l+m)!}{(l-m)!}}j_l(t)Y_{lm}(\vartheta,\varphi)=(-t\sin\vartheta e^{i\varphi})^m e^{it\cos\vartheta}, \tag*{(1056)} l = m ∑ ∞ i l − m ( l − m )! 4 π ( 2 l + 1 ) ( l + m )! j l ( t ) Y l m ( ϑ , φ ) = ( − t sin ϑ e i φ ) m e i t c o s ϑ , (1056)
∑ l = m , m + 2 , … ∞ i l − m 4 π ( 2 l + 1 ) ( l + m ) ! ( l − m ) ! j l ( t ) Y l m ( ϑ , φ ) = ( − t sin ϑ e i φ ) m cos ( t cos ϑ ) , (1057) \sum_{l=m,m+2,\ldots}^{\infty}i^{l-m}\sqrt{\frac{4\pi(2l+1)(l+m)!}{(l-m)!}}j_l(t)Y_{lm}(\vartheta,\varphi)=(-t\sin\vartheta e^{i\varphi})^m\cos(t\cos\vartheta), \tag*{(1057)} l = m , m + 2 , … ∑ ∞ i l − m ( l − m )! 4 π ( 2 l + 1 ) ( l + m )! j l ( t ) Y l m ( ϑ , φ ) = ( − t sin ϑ e i φ ) m cos ( t cos ϑ ) , (1057)
∑ l = m + 1 , m + 3 , … ∞ i l − m − 1 4 π ( 2 l + 1 ) ( l + m ) ! ( l − m ) ! j l ( t ) Y l m ( ϑ , φ ) = ( − t sin ϑ e i φ ) m sin ( t cos ϑ ) , (1058) \sum_{l=m+1,m+3,\ldots}^{\infty}i^{l-m-1}\sqrt{\frac{4\pi(2l+1)(l+m)!}{(l-m)!}}j_l(t)Y_{lm}(\vartheta,\varphi)=(-t\sin\vartheta e^{i\varphi})^m\sin(t\cos\vartheta), \tag*{(1058)} l = m + 1 , m + 3 , … ∑ ∞ i l − m − 1 ( l − m )! 4 π ( 2 l + 1 ) ( l + m )! j l ( t ) Y l m ( ϑ , φ ) = ( − t sin ϑ e i φ ) m sin ( t cos ϑ ) , (1058)
∑ l = m ∞ 4 π ( 2 l + 1 ) ( l + m ) ! ( l − m ) ! j l ( t ) z l ( t ) Y l m ( ϑ , φ ) = ( − t cos ϑ 2 e i φ ) m z m ( 2 t sin ϑ 2 ) , (1059) \sum_{l=m}^{\infty}\sqrt{\frac{4\pi(2l+1)(l+m)!}{(l-m)!}}j_l(t)z_l(t)Y_{lm}(\vartheta,\varphi)=\left(-t\cos\frac{\vartheta}{2}e^{i\varphi}\right)^m z_m\left(2t\sin\frac{\vartheta}{2}\right), \tag*{(1059)} l = m ∑ ∞ ( l − m )! 4 π ( 2 l + 1 ) ( l + m )! j l ( t ) z l ( t ) Y l m ( ϑ , φ ) = ( − t cos 2 ϑ e i φ ) m z m ( 2 t sin 2 ϑ ) , (1059)
∑ l = m ∞ 4 π ( 2 l + 1 ) ( l + m ) ! ( l − m ) ! j l ( x ) z l ( y ) Y l m ( ϑ , φ ) = ( − x y sin ϑ e i φ x 2 + y 2 − 2 x y cos ϑ ) m z m ( x 2 + y 2 − 2 x y cos ϑ ) , (1060) \sum_{l=m}^{\infty}\sqrt{\frac{4\pi(2l+1)(l+m)!}{(l-m)!}}j_l(x)z_l(y)Y_{lm}(\vartheta,\varphi)=\left(-\frac{xy\sin\vartheta e^{i\varphi}}{\sqrt{x^2+y^2-2xy\cos\vartheta}}\right)^m z_m\left(\sqrt{x^2+y^2-2xy\cos\vartheta}\right), \tag*{(1060)} l = m ∑ ∞ ( l − m )! 4 π ( 2 l + 1 ) ( l + m )! j l ( x ) z l ( y ) Y l m ( ϑ , φ ) = ( − x 2 + y 2 − 2 x y cos ϑ x y sin ϑ e i φ ) m z m ( x 2 + y 2 − 2 x y cos ϑ ) , (1060)
∑ l = m ∞ i l − m j l ( t ) Y l m ( ϑ 1 , φ 1 ) Y l m ∗ ( ϑ 2 , φ 2 ) = 1 4 π J m ( t sin ϑ 1 sin ϑ 2 ) e i t cos ϑ 1 cos ϑ 2 e i m ( φ 1 − φ 2 ) (1061) \sum_{l=m}^{\infty}i^{l-m}j_l(t)Y_{lm}(\vartheta_1,\varphi_1)Y_{lm}^{*}(\vartheta_2,\varphi_2)=\frac{1}{4\pi}J_m(t\sin\vartheta_1\sin\vartheta_2)e^{it\cos\vartheta_1\cos\vartheta_2}e^{im(\varphi_1-\varphi_2)} \tag*{(1061)} l = m ∑ ∞ i l − m j l ( t ) Y l m ( ϑ 1 , φ 1 ) Y l m ∗ ( ϑ 2 , φ 2 ) = 4 π 1 J m ( t sin ϑ 1 sin ϑ 2 ) e i t c o s ϑ 1 c o s ϑ 2 e im ( φ 1 − φ 2 ) (1061)
Generating functions for Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ )
For fixed m m m one has (see Refs. [23, 24])
1 R 2 m + 1 = { ( − 1 ) m ( 2 m − 1 ) ! ! ( sin ϑ ) m ∑ l = m ∞ t l − m 4 π 2 l + 1 ( l + m ) ! ( l − m ) ! Y l m ( ϑ , 0 ) , ∣ t ∣ < 1 , ( − 1 ) m ( 2 m − 1 ) ! ! ( sin ϑ ) m ∑ l = m ∞ 1 t l + m + 1 4 π 2 l + 1 ( l + m ) ! ( l − m ) ! Y l m ( ϑ , 0 ) , ∣ t ∣ > 1 (1062) \frac{1}{R^{2m+1}}=
\begin{cases}
\frac{(-1)^m}{(2m-1)!!(\sin\vartheta)^m}\sum_{l=m}^{\infty}t^{l-m}\sqrt{\frac{4\pi}{2l+1}\frac{(l+m)!}{(l-m)!}}Y_{lm}(\vartheta,0), & |t|<1,\\[1.2em]
\frac{(-1)^m}{(2m-1)!!(\sin\vartheta)^m}\sum_{l=m}^{\infty}\frac{1}{t^{l+m+1}}\sqrt{\frac{4\pi}{2l+1}\frac{(l+m)!}{(l-m)!}}Y_{lm}(\vartheta,0), & |t|>1
\end{cases} \tag*{(1062)} R 2 m + 1 1 = ⎩ ⎨ ⎧ ( 2 m − 1 )!! ( s i n ϑ ) m ( − 1 ) m ∑ l = m ∞ t l − m 2 l + 1 4 π ( l − m )! ( l + m )! Y l m ( ϑ , 0 ) , ( 2 m − 1 )!! ( s i n ϑ ) m ( − 1 ) m ∑ l = m ∞ t l + m + 1 1 2 l + 1 4 π ( l − m )! ( l + m )! Y l m ( ϑ , 0 ) , ∣ t ∣ < 1 , ∣ t ∣ > 1 (1062)
where
R = 1 − 2 t cos ϑ + t 2 R=\sqrt{1-2t\cos\vartheta+t^2} R = 1 − 2 t cos ϑ + t 2
In particular, for m = 0 m=0 m = 0 one gets
1 R = { ∑ l = 0 ∞ t l P l ( cos ϑ ) , ∣ t ∣ < 1 , ∑ l = 0 ∞ 1 t l + 1 P l ( cos ϑ ) , ∣ t ∣ > 1. (1063) \frac{1}{R}=
\begin{cases}
\sum_{l=0}^{\infty}t^lP_l(\cos\vartheta), & |t|<1,\\[6pt]
\sum_{l=0}^{\infty}\frac{1}{t^{l+1}}P_l(\cos\vartheta), & |t|>1.
\end{cases} \tag*{(1063)} R 1 = ⎩ ⎨ ⎧ ∑ l = 0 ∞ t l P l ( cos ϑ ) , ∑ l = 0 ∞ t l + 1 1 P l ( cos ϑ ) , ∣ t ∣ < 1 , ∣ t ∣ > 1. (1063)
[ ( 1 + R ) 2 − t 2 ] − m R = ( − 1 ) m 2 m ( sin ϑ ) m ∑ l = m ∞ t l − m l ! 4 π ( 2 l + 1 ) ( l + m ) ! ( l − m ) ! Y l m ( ϑ , 0 ) , ∣ t ∣ < 1. (1064) \frac{\left[(1+R)^2-t^2\right]^{-m}}{R}
=\frac{(-1)^m}{2^m(\sin\vartheta)^m}\sum_{l=m}^{\infty}t^{l-m}l!\sqrt{\frac{4\pi}{(2l+1)(l+m)!(l-m)!}}Y_{lm}(\vartheta,0),\quad|t|<1. \tag*{(1064)} R [ ( 1 + R ) 2 − t 2 ] − m = 2 m ( sin ϑ ) m ( − 1 ) m l = m ∑ ∞ t l − m l ! ( 2 l + 1 ) ( l + m )! ( l − m )! 4 π Y l m ( ϑ , 0 ) , ∣ t ∣ < 1. (1064)
In addition, we present the following expansions in terms of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) [23]
( sin ϑ ) m ( cos ψ − cos ϑ ) m + 1 2 Θ ( cos ψ − cos ϑ ) = ( − 1 ) m 2 m ( 2 m − 1 ) ! ! ∑ l = m ∞ 8 π 2 l + 1 ( l + m ) ! ( l − m ) ! cos [ ( 2 l + 1 ) ψ 2 ] Y l m ( ϑ , 0 ) (1065) \frac{(\sin\vartheta)^m}{(\cos\psi-\cos\vartheta)^{m+\frac{1}{2}}}\Theta(\cos\psi-\cos\vartheta)
=\frac{(-1)^m2^m}{(2m-1)!!}\sum_{l=m}^{\infty}\sqrt{\frac{8\pi}{2l+1}\frac{(l+m)!}{(l-m)!}}\cos\left[(2l+1)\frac{\psi}{2}\right]Y_{lm}(\vartheta,0) \tag*{(1065)} ( cos ψ − cos ϑ ) m + 2 1 ( sin ϑ ) m Θ ( cos ψ − cos ϑ ) = ( 2 m − 1 )!! ( − 1 ) m 2 m l = m ∑ ∞ 2 l + 1 8 π ( l − m )! ( l + m )! cos [ ( 2 l + 1 ) 2 ψ ] Y l m ( ϑ , 0 ) (1065)
( sin ϑ ) m ( cos ϑ − cos ψ ) m + 1 2 Θ ( cos ϑ − cos ψ ) = ( − 1 ) m 2 m ( 2 m − 1 ) ! ! ∑ l = m ∞ 8 π 2 l + 1 ( l + m ) ! ( l − m ) ! sin [ ( 2 l + 1 ) ψ 2 ] Y l m ( ϑ , 0 ) (1066) \frac{(\sin\vartheta)^m}{(\cos\vartheta-\cos\psi)^{m+\frac{1}{2}}}\Theta(\cos\vartheta-\cos\psi)
=\frac{(-1)^m2^m}{(2m-1)!!}\sum_{l=m}^{\infty}\sqrt{\frac{8\pi}{2l+1}\frac{(l+m)!}{(l-m)!}}\sin\left[(2l+1)\frac{\psi}{2}\right]Y_{lm}(\vartheta,0) \tag*{(1066)} ( cos ϑ − cos ψ ) m + 2 1 ( sin ϑ ) m Θ ( cos ϑ − cos ψ ) = ( 2 m − 1 )!! ( − 1 ) m 2 m l = m ∑ ∞ 2 l + 1 8 π ( l − m )! ( l + m )! sin [ ( 2 l + 1 ) 2 ψ ] Y l m ( ϑ , 0 ) (1066)
where
Θ ( x ) = { 1 if x ≥ 0 , 0 if x < 0. (1067) \Theta(x)=
\begin{cases}
1 & \text{if }x\geq0,\\
0 & \text{if }x<0.
\end{cases} \tag*{(1067)} Θ ( x ) = { 1 0 if x ≥ 0 , if x < 0. (1067)
In particular, when m = 0 m=0 m = 0 one has
Θ ( cos ψ − cos ϑ ) cos ψ − cos ϑ ± i Θ ( cos ϑ − cos ψ ) cos ϑ − cos ψ = 2 ∑ l = 0 ∞ e ± i ( 2 l + 1 ) ψ 2 P l ( cos ϑ ) . (1068) \frac{\Theta(\cos\psi-\cos\vartheta)}{\sqrt{\cos\psi-\cos\vartheta}}
\pm i\frac{\Theta(\cos\vartheta-\cos\psi)}{\sqrt{\cos\vartheta-\cos\psi}}
=\sqrt{2}\sum_{l=0}^{\infty}e^{\pm i(2l+1)\frac{\psi}{2}}P_l(\cos\vartheta). \tag*{(1068)} cos ψ − cos ϑ Θ ( cos ψ − cos ϑ ) ± i cos ϑ − cos ψ Θ ( cos ϑ − cos ψ ) = 2 l = 0 ∑ ∞ e ± i ( 2 l + 1 ) 2 ψ P l ( cos ϑ ) . (1068)
Asymptotic expressions for Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ )
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) for Large l l l
If l ≫ 1 l\gg1 l ≫ 1 and l ≫ m ≥ 0 l\gg m\geq0 l ≫ m ≥ 0 , a spherical harmonic Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) for ϵ ≤ ϑ ≤ π − ϵ \epsilon\leq\vartheta\leq\pi-\epsilon ϵ ≤ ϑ ≤ π − ϵ (0 < ϵ ≪ 1 / l 0<\epsilon\ll1/l 0 < ϵ ≪ 1/ l ) and 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π can be approximated by (see Ref. [23])
Y l m ( ϑ , φ ) ≈ e i m φ π cos [ ( 2 l + 1 ) ϑ 2 + ( 2 m − 1 ) π 4 ] sin ϑ + O ( 1 l ) . (1069) Y_{lm}(\vartheta,\varphi)\approx\frac{e^{im\varphi}}{\pi}\frac{\cos\left[(2l+1)\frac{\vartheta}{2}+(2m-1)\frac{\pi}{4}\right]}{\sqrt{\sin\vartheta}}+O\left(\frac{1}{l}\right). \tag*{(1069)} Y l m ( ϑ , φ ) ≈ π e im φ sin ϑ cos [ ( 2 l + 1 ) 2 ϑ + ( 2 m − 1 ) 4 π ] + O ( l 1 ) . (1069)
A more exact formula [24] is
Y l m ( ϑ , φ ) ≈ e i m φ π sin ϑ { ( 1 + 4 m 2 − 3 8 l ) cos [ ( 2 l + 1 ) ϑ 2 + ( 2 m − 1 ) π 4 ] − 4 m 2 − 1 8 l sin ϑ cos [ ( 2 l + 3 ) ϑ 2 + ( 2 m − 3 ) π 4 ] } + O ( 1 l 2 ) (1070) Y_{lm}(\vartheta,\varphi)\approx\frac{e^{im\varphi}}{\pi\sqrt{\sin\vartheta}}\left\{\left(1+\frac{4m^2-3}{8l}\right)\cos\left[(2l+1)\frac{\vartheta}{2}+(2m-1)\frac{\pi}{4}\right]-\frac{4m^2-1}{8l\sin\vartheta}\cos\left[(2l+3)\frac{\vartheta}{2}+(2m-3)\frac{\pi}{4}\right]\right\}+O\left(\frac{1}{l^2}\right) \tag*{(1070)} Y l m ( ϑ , φ ) ≈ π sin ϑ e im φ { ( 1 + 8 l 4 m 2 − 3 ) cos [ ( 2 l + 1 ) 2 ϑ + ( 2 m − 1 ) 4 π ] − 8 l sin ϑ 4 m 2 − 1 cos [ ( 2 l + 3 ) 2 ϑ + ( 2 m − 3 ) 4 π ] } + O ( l 2 1 ) (1070)
The terms O ( 1 / l ) O(1/l) O ( 1/ l ) and O ( 1 / l 2 ) O(1/l^2) O ( 1/ l 2 ) depend on m m m and ϑ \vartheta ϑ .
If l ≫ 1 l\gg1 l ≫ 1 and l ≫ m ≥ 0 l\gg m\geq0 l ≫ m ≥ 0 , one also obtains
∣ Y l ± m ( ϑ , φ ) ∣ < 2 π ( sin ϑ ) − ( m + 1 2 ) . (1071) \left|Y_{l\pm m}(\vartheta,\varphi)\right|<\frac{2}{\pi}(\sin\vartheta)^{-\left(m+\frac{1}{2}\right)}. \tag*{(1071)} ∣ Y l ± m ( ϑ , φ ) ∣ < π 2 ( sin ϑ ) − ( m + 2 1 ) . (1071)
Behaviour of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) in Neighbourhoods of ϑ = 0 \vartheta=0 ϑ = 0 , π \pi π and π / 2 \pi/2 π /2
For 0 ≤ ϑ ≤ ϵ 0\leq\vartheta\leq\epsilon 0 ≤ ϑ ≤ ϵ (ϵ ≪ 1 \epsilon\ll1 ϵ ≪ 1 ) one has
Y l ± m ( ϑ , φ ) ≈ ( ∓ 1 ) m e ± i m φ m ! 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( ϑ 2 ) m [ 1 − 3 l ( l + 1 ) − m ( m + 1 ) 3 ( m + 1 ) ( ϑ 2 ) 2 ] . (1072) Y_{l\pm m}(\vartheta,\varphi)\approx(\mp1)^m\frac{e^{\pm im\varphi}}{m!}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!}}\left(\frac{\vartheta}{2}\right)^m\left[1-\frac{3l(l+1)-m(m+1)}{3(m+1)}\left(\frac{\vartheta}{2}\right)^2\right]. \tag*{(1072)} Y l ± m ( ϑ , φ ) ≈ ( ∓ 1 ) m m ! e ± im φ 4 π 2 l + 1 ( l − m )! ( l + m )! ( 2 ϑ ) m [ 1 − 3 ( m + 1 ) 3 l ( l + 1 ) − m ( m + 1 ) ( 2 ϑ ) 2 ] . (1072)
For π − ϵ ≤ ϑ ≤ π \pi-\epsilon\leq\vartheta\leq\pi π − ϵ ≤ ϑ ≤ π , one obtains
Y l ± m ( ϑ , φ ) ≈ ( − 1 ) l ( ± 1 ) m e ± i m φ m ! 2 l + 1 4 π ( l + m ) ! ( l − m ) ! ( π − ϑ 2 ) m [ 1 − 3 l ( l + 1 ) − m ( m + 1 ) 3 ( m + 1 ) ( π − ϑ 2 ) 2 ] . (1073) Y_{l\pm m}(\vartheta,\varphi)\approx(-1)^l(\pm1)^m\frac{e^{\pm im\varphi}}{m!}\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!}}\left(\frac{\pi-\vartheta}{2}\right)^m\left[1-\frac{3l(l+1)-m(m+1)}{3(m+1)}\left(\frac{\pi-\vartheta}{2}\right)^2\right]. \tag*{(1073)} Y l ± m ( ϑ , φ ) ≈ ( − 1 ) l ( ± 1 ) m m ! e ± im φ 4 π 2 l + 1 ( l − m )! ( l + m )! ( 2 π − ϑ ) m [ 1 − 3 ( m + 1 ) 3 l ( l + 1 ) − m ( m + 1 ) ( 2 π − ϑ ) 2 ] . (1073)
In the case when π / 2 − ε ≤ ϑ ≤ π / 2 + ε \pi/2-\varepsilon\leq\vartheta\leq\pi/2+\varepsilon π /2 − ε ≤ ϑ ≤ π /2 + ε
Y l ± m ( ϑ , φ ) ≈ ( − 1 ) l ± m 2 e ± i m φ 2 l + 1 4 π ( l − m − 1 ) ! ! ( l − m ) ! ! ( l + m − 1 ) ! ! ( l + m ) ! ! [ 1 − l ( l + 1 ) − m 2 2 ( π 2 − ϑ ) 2 ] (1074) Y_{l\pm m}(\vartheta,\varphi)\approx(-1)^{\frac{l\pm m}{2}}e^{\pm im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m-1)!!}{(l-m)!!}\frac{(l+m-1)!!}{(l+m)!!}}\left[1-\frac{l(l+1)-m^2}{2}\left(\frac{\pi}{2}-\vartheta\right)^2\right] \tag*{(1074)} Y l ± m ( ϑ , φ ) ≈ ( − 1 ) 2 l ± m e ± im φ 4 π 2 l + 1 ( l − m )!! ( l − m − 1 )!! ( l + m )!! ( l + m − 1 )!! [ 1 − 2 l ( l + 1 ) − m 2 ( 2 π − ϑ ) 2 ] (1074)
if l + m l+m l + m is even and
Y l ± m ( ϑ , φ ) ≈ ( − 1 ) l ± m − 1 2 e ± i m φ 2 l + 1 4 π ( l − m ) ! ! ( l − m − 1 ) ! ! ( l + m ) ! ! ( l + m − 1 ) ! ! ( π 2 − ϑ ) Y_{l\pm m}(\vartheta,\varphi)\approx(-1)^{\frac{l\pm m-1}{2}}e^{\pm im\varphi}\sqrt{\frac{2l+1}{4\pi}\frac{(l-m)!!}{(l-m-1)!!}\frac{(l+m)!!}{(l+m-1)!!}}\left(\frac{\pi}{2}-\vartheta\right) Y l ± m ( ϑ , φ ) ≈ ( − 1 ) 2 l ± m − 1 e ± im φ 4 π 2 l + 1 ( l − m − 1 )!! ( l − m )!! ( l + m − 1 )!! ( l + m )!! ( 2 π − ϑ )
× [ 1 − l ( l + 1 ) − ( m 2 + 1 ) 6 ( π 2 − ϑ ) 2 ] (1075) {}\times\left[1-\frac{l(l+1)-(m^2+1)}{6}\left(\frac{\pi}{2}-\vartheta\right)^2\right] \tag*{(1075)} × [ 1 − 6 l ( l + 1 ) − ( m 2 + 1 ) ( 2 π − ϑ ) 2 ] (1075)
if l + m l+m l + m is odd.
If ϑ \vartheta ϑ is small, Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) may be approximated by the Bessel function (McDonald formula [23])
Y l − m ( ϑ , φ ) = 2 l + 1 4 π ( l + m ) ! ( l − m ) ! e − i m φ [ ( l + 1 2 ) cos ϑ 2 ] − m Y_{l-m}(\vartheta,\varphi)=\sqrt{\frac{2l+1}{4\pi}\frac{(l+m)!}{(l-m)!}}e^{-im\varphi}\left[\left(l+\frac{1}{2}\right)\cos\frac{\vartheta}{2}\right]^{-m} Y l − m ( ϑ , φ ) = 4 π 2 l + 1 ( l − m )! ( l + m )! e − im φ [ ( l + 2 1 ) cos 2 ϑ ] − m
× { J m ( α ) + ( sin ϑ 2 ) 2 [ α 6 J m + 3 ( α ) − J m + 2 ( α ) + 1 2 α J m + 1 ( α ) ] + O [ ( sin ϑ 2 ) 4 ] } (1076) {}\times\left\{J_m(\alpha)+\left(\sin\frac{\vartheta}{2}\right)^2\left[\frac{\alpha}{6}J_{m+3}(\alpha)-J_{m+2}(\alpha)+\frac{1}{2\alpha}J_{m+1}(\alpha)\right]+O\left[\left(\sin\frac{\vartheta}{2}\right)^4\right]\right\} \tag*{(1076)} × { J m ( α ) + ( sin 2 ϑ ) 2 [ 6 α J m + 3 ( α ) − J m + 2 ( α ) + 2 α 1 J m + 1 ( α ) ] + O [ ( sin 2 ϑ ) 4 ] } (1076)
Here α = ( 2 l + 1 ) sin ( ϑ / 2 ) \alpha=(2l+1)\sin(\vartheta/2) α = ( 2 l + 1 ) sin ( ϑ /2 ) , J λ ( α ) J_\lambda(\alpha) J λ ( α ) is the Bessel function.
Asymptotic Expression for Fixed m m m , l → ∞ l\to\infty l → ∞ , ϑ → 0 \vartheta\to0 ϑ → 0 and Finite l ϑ l\vartheta l ϑ
In this case
Y l − m ( ϑ , φ ) ≈ l 2 π e − i m φ J m ( l ϑ ) (1077) Y_{l-m}(\vartheta,\varphi)\approx\sqrt{\frac{l}{2\pi}}e^{-im\varphi}J_m(l\vartheta) \tag*{(1077)} Y l − m ( ϑ , φ ) ≈ 2 π l e − im φ J m ( l ϑ ) (1077)
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) for special values of l l l and m m m
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) for l ≤ 5 l\leq5 l ≤ 5
Please use sympy or Mathematica to generate the explicit forms of the spherical harmonics for l = 0 , 1 , 2 , 3 , 4 , 5 l=0,1,2,3,4,5 l = 0 , 1 , 2 , 3 , 4 , 5 in terms of trigonometric functions.
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) for ∣ m ∣ = l , l − 1 , l − 2 , l − 3 , l − 4 , l − 5 |m|=l,l-1,l-2,l-3,l-4,l-5 ∣ m ∣ = l , l − 1 , l − 2 , l − 3 , l − 4 , l − 5 and Any Integer l l l
In this case the spherical harmonics are conveniently expressed in terms of trigonometric functions.
Y l ± l ( ϑ , φ ) = ( ∓ 1 ) l e ± i l φ ( 2 l + 1 ) ! ! 4 π ( 2 l ) ! ! ( sin ϑ ) l (1078) Y_{l\pm l}(\vartheta,\varphi)=(\mp1)^l e^{\pm il\varphi}\sqrt{\frac{(2l+1)!!}{4\pi(2l)!!}}(\sin\vartheta)^l \tag*{(1078)} Y l ± l ( ϑ , φ ) = ( ∓ 1 ) l e ± i l φ 4 π ( 2 l )!! ( 2 l + 1 )!! ( sin ϑ ) l (1078)
Y l ± ( l − 1 ) ( ϑ , φ ) = ( ∓ 1 ) l − 1 e ± i ( l − 1 ) φ ( 2 l + 1 ) ! ! 4 π ( 2 l − 2 ) ! ! cos ϑ ( sin ϑ ) l − 1 (1079) Y_{l\pm(l-1)}(\vartheta,\varphi)=(\mp1)^{l-1}e^{\pm i(l-1)\varphi}\sqrt{\frac{(2l+1)!!}{4\pi(2l-2)!!}}\cos\vartheta(\sin\vartheta)^{l-1} \tag*{(1079)} Y l ± ( l − 1 ) ( ϑ , φ ) = ( ∓ 1 ) l − 1 e ± i ( l − 1 ) φ 4 π ( 2 l − 2 )!! ( 2 l + 1 )!! cos ϑ ( sin ϑ ) l − 1 (1079)
Y l ± ( l − 2 ) ( ϑ , φ ) = ( ∓ 1 ) l − 2 e ± i ( l − 2 ) φ 2 l + 1 8 π ( 2 l − 3 ) ! ! ( 2 l − 2 ) ! ! ( sin ϑ ) l − 2 [ ( 2 l − 1 ) cos 2 ϑ − 1 ] (1080) Y_{l\pm(l-2)}(\vartheta,\varphi)=(\mp1)^{l-2}e^{\pm i(l-2)\varphi}\sqrt{\frac{2l+1}{8\pi}\frac{(2l-3)!!}{(2l-2)!!}}(\sin\vartheta)^{l-2}\left[(2l-1)\cos^2\vartheta-1\right] \tag*{(1080)} Y l ± ( l − 2 ) ( ϑ , φ ) = ( ∓ 1 ) l − 2 e ± i ( l − 2 ) φ 8 π 2 l + 1 ( 2 l − 2 )!! ( 2 l − 3 )!! ( sin ϑ ) l − 2 [ ( 2 l − 1 ) cos 2 ϑ − 1 ] (1080)
Y l ± ( l − 3 ) ( ϑ , φ ) = ( ∓ 1 ) l − 3 e ± i ( l − 3 ) φ 2 l + 1 24 π ( 2 l − 3 ) ! ! ( 2 l − 4 ) ! ! ( sin ϑ ) l − 3 cos ϑ [ ( 2 l − 1 ) cos 2 ϑ − 3 ] (1081) Y_{l\pm(l-3)}(\vartheta,\varphi)=(\mp1)^{l-3}e^{\pm i(l-3)\varphi}\sqrt{\frac{2l+1}{24\pi}\frac{(2l-3)!!}{(2l-4)!!}}(\sin\vartheta)^{l-3}\cos\vartheta\left[(2l-1)\cos^2\vartheta-3\right] \tag*{(1081)} Y l ± ( l − 3 ) ( ϑ , φ ) = ( ∓ 1 ) l − 3 e ± i ( l − 3 ) φ 24 π 2 l + 1 ( 2 l − 4 )!! ( 2 l − 3 )!! ( sin ϑ ) l − 3 cos ϑ [ ( 2 l − 1 ) cos 2 ϑ − 3 ] (1081)
Y l ± ( l − 4 ) ( ϑ , φ ) = ( ∓ 1 ) l − 4 e ± i ( l − 4 ) φ 2 l + 1 96 π ( 2 l − 5 ) ! ! ( 2 l − 4 ) ! ! ( sin ϑ ) l − 4 [ ( 2 l − 1 ) ( 2 l − 3 ) cos 4 ϑ − 6 ( 2 l − 3 ) cos 2 ϑ + 3 ] (1082) Y_{l\pm(l-4)}(\vartheta,\varphi)=(\mp1)^{l-4}e^{\pm i(l-4)\varphi}\sqrt{\frac{2l+1}{96\pi}\frac{(2l-5)!!}{(2l-4)!!}}(\sin\vartheta)^{l-4}
\left[(2l-1)(2l-3)\cos^4\vartheta-6(2l-3)\cos^2\vartheta+3\right] \tag*{(1082)} Y l ± ( l − 4 ) ( ϑ , φ ) = ( ∓ 1 ) l − 4 e ± i ( l − 4 ) φ 96 π 2 l + 1 ( 2 l − 4 )!! ( 2 l − 5 )!! ( sin ϑ ) l − 4 [ ( 2 l − 1 ) ( 2 l − 3 ) cos 4 ϑ − 6 ( 2 l − 3 ) cos 2 ϑ + 3 ] (1082)
Y l ± ( l − 5 ) ( ϑ , φ ) = ( ∓ 1 ) l − 5 e ± i ( l − 5 ) φ 2 l + 1 480 π ( 2 l − 5 ) ! ! ( 2 l − 6 ) ! ! ( sin ϑ ) l − 5 cos ϑ [ ( 2 l − 1 ) ( 2 l − 3 ) cos 4 ϑ − 10 ( 2 l − 3 ) cos 2 ϑ + 15 ] (1083) Y_{l\pm(l-5)}(\vartheta,\varphi)=(\mp1)^{l-5}e^{\pm i(l-5)\varphi}\sqrt{\frac{2l+1}{480\pi}\frac{(2l-5)!!}{(2l-6)!!}}(\sin\vartheta)^{l-5}\cos\vartheta
\left[(2l-1)(2l-3)\cos^4\vartheta-10(2l-3)\cos^2\vartheta+15\right] \tag*{(1083)} Y l ± ( l − 5 ) ( ϑ , φ ) = ( ∓ 1 ) l − 5 e ± i ( l − 5 ) φ 480 π 2 l + 1 ( 2 l − 6 )!! ( 2 l − 5 )!! ( sin ϑ ) l − 5 cos ϑ [ ( 2 l − 1 ) ( 2 l − 3 ) cos 4 ϑ − 10 ( 2 l − 3 ) cos 2 ϑ + 15 ] (1083)
Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ ) for special ϑ \vartheta ϑ
Y l m ( 0 , φ ) = δ m 0 2 l + 1 4 π , (1084) Y_{lm}(0,\varphi)=\delta_{m0}\sqrt{\frac{2l+1}{4\pi}}, \tag*{(1084)} Y l m ( 0 , φ ) = δ m 0 4 π 2 l + 1 , (1084)
Y l m ( π 2 , φ ) = { ( − 1 ) l + m 2 e i m φ 2 l + 1 4 π ( l + m − 1 ) ! ! ( l + m ) ! ! ( l − m − 1 ) ! ! ( l − m ) ! ! if l + m is even , 0 if l + m is odd , (1085) Y_{lm}\left(\frac{\pi}{2},\varphi\right)=
\begin{cases}
(-1)^{\frac{l+m}{2}}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}}\frac{(l+m-1)!!}{(l+m)!!}\frac{(l-m-1)!!}{(l-m)!!} & \text{if }l+m\text{ is even},\\
0 & \text{if }l+m\text{ is odd},
\end{cases} \tag*{(1085)} Y l m ( 2 π , φ ) = { ( − 1 ) 2 l + m e im φ 4 π 2 l + 1 ( l + m )!! ( l + m − 1 )!! ( l − m )!! ( l − m − 1 )!! 0 if l + m is even , if l + m is odd , (1085)
Y l m ( π , φ ) = δ m 0 ( − 1 ) l 2 l + 1 4 π , (1086) Y_{lm}(\pi,\varphi)=\delta_{m0}(-1)^l\sqrt{\frac{2l+1}{4\pi}}, \tag*{(1086)} Y l m ( π , φ ) = δ m 0 ( − 1 ) l 4 π 2 l + 1 , (1086)
Y l m ( ± n π , φ ) = δ m 0 ( − 1 ) n l 2 l + 1 4 π . (1087) Y_{lm}(\pm n\pi,\varphi)=\delta_{m0}(-1)^{nl}\sqrt{\frac{2l+1}{4\pi}}. \tag*{(1087)} Y l m ( ± nπ , φ ) = δ m 0 ( − 1 ) n l 4 π 2 l + 1 . (1087)
∂ ∂ ϑ Y l m ( ϑ , φ ) ∣ ϑ = 0 = ( δ m − 1 e − i φ − δ m 1 e i φ ) l ( l + 1 ) ( 2 l + 1 ) 16 π (1088) \left.\frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi)\right|_{\vartheta=0}
=(\delta_{m-1}e^{-i\varphi}-\delta_{m1}e^{i\varphi})\sqrt{\frac{l(l+1)(2l+1)}{16\pi}} \tag*{(1088)} ∂ ϑ ∂ Y l m ( ϑ , φ ) ϑ = 0 = ( δ m − 1 e − i φ − δ m 1 e i φ ) 16 π l ( l + 1 ) ( 2 l + 1 ) (1088)
∂ ∂ ϑ Y l m ( ϑ , φ ) ∣ ϑ = π 2 = { ( − 1 ) l + m + 1 2 e i m φ 2 l + 1 4 π ( l + m ) ! ! ( l − m ) ! ! ( l + m − 1 ) ! ! ( l − m − 1 ) ! ! if l + m is odd , 0 if l + m is even (1089) \left.\frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi)\right|_{\vartheta=\frac{\pi}{2}}=
\begin{cases}
(-1)^{\frac{l+m+1}{2}}e^{im\varphi}\sqrt{\frac{2l+1}{4\pi}}\frac{(l+m)!!(l-m)!!}{(l+m-1)!!(l-m-1)!!} & \text{if }l+m\text{ is odd},\\
0 & \text{if }l+m\text{ is even}
\end{cases} \tag*{(1089)} ∂ ϑ ∂ Y l m ( ϑ , φ ) ϑ = 2 π = { ( − 1 ) 2 l + m + 1 e im φ 4 π 2 l + 1 ( l + m − 1 )!! ( l − m − 1 )!! ( l + m )!! ( l − m )!! 0 if l + m is odd , if l + m is even (1089)
∂ ∂ ϑ Y l m ( ϑ , φ ) ∣ ϑ = π = ( − 1 ) l ( δ m − 1 e − i φ − δ m 1 e i φ ) l ( l + 1 ) ( 2 l + 1 ) 16 π (1090) \left.\frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi)\right|_{\vartheta=\pi}
=(-1)^l(\delta_{m-1}e^{-i\varphi}-\delta_{m1}e^{i\varphi})\sqrt{\frac{l(l+1)(2l+1)}{16\pi}} \tag*{(1090)} ∂ ϑ ∂ Y l m ( ϑ , φ ) ϑ = π = ( − 1 ) l ( δ m − 1 e − i φ − δ m 1 e i φ ) 16 π l ( l + 1 ) ( 2 l + 1 ) (1090)
∂ ∂ ϑ Y l m ( ϑ , φ ) ∣ ϑ = ± n π = ( − 1 ) n l ( δ m − 1 e − i φ − δ m 1 e i φ ) l ( l + 1 ) ( 2 l + 1 ) 16 π (1091) \left.\frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi)\right|_{\vartheta=\pm n\pi}
=(-1)^{nl}(\delta_{m-1}e^{-i\varphi}-\delta_{m1}e^{i\varphi})\sqrt{\frac{l(l+1)(2l+1)}{16\pi}} \tag*{(1091)} ∂ ϑ ∂ Y l m ( ϑ , φ ) ϑ = ± nπ = ( − 1 ) n l ( δ m − 1 e − i φ − δ m 1 e i φ ) 16 π l ( l + 1 ) ( 2 l + 1 ) (1091)
Zeros of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ )
In the interval 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π the spherical harmonics Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and their derivatives ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ ) possess some zeros.
The number of zeros, i.e., the number of roots of the equations
Y l m ( ϑ , φ ) = 0 and ∂ ∂ ϑ Y l m ( ϑ , φ ) = 0 (1092) Y_{lm}(\vartheta,\varphi)=0\quad\text{and}\quad\frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi)=0 \tag*{(1092)} Y l m ( ϑ , φ ) = 0 and ∂ ϑ ∂ Y l m ( ϑ , φ ) = 0 (1092)
is finite in interval 0 < ϑ < π 0<\vartheta<\pi 0 < ϑ < π . All these roots are non-degenerate. Below we denote the zeros of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ ) by ϑ α \vartheta_\alpha ϑ α and ϑ β \vartheta_\beta ϑ β , respectively. Note that the functions in question have no roots with respect to φ \varphi φ in the domain 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π .
Zeros of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ )
If m = 0 m=0 m = 0 , Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) has l l l zeros in the interval 0 < ϑ < π 0<\vartheta<\pi 0 < ϑ < π . If m ≠ 0 m\neq0 m = 0 , Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) has ( l − ∣ m ∣ ) (l-|m|) ( l − ∣ m ∣ ) zeros inside the same interval of ϑ \vartheta ϑ and also two zeros at ϑ α = 0 \vartheta_\alpha=0 ϑ α = 0 and ϑ α = π \vartheta_\alpha=\pi ϑ α = π . All these zeros lie symmetrically with respect to ϑ = π / 2 \vartheta=\pi/2 ϑ = π /2 . Hence, Y l m ( π / 2 , φ ) = 0 Y_{lm}(\pi/2,\varphi)=0 Y l m ( π /2 , φ ) = 0 provided ( l − m ) (l-m) ( l − m ) is odd. If m ≠ 0 m\neq0 m = 0 , the zeros of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) are determined by the equations
m = ± l , cos 2 ϑ α = 1 ; m = ± ( l − 1 ) , cos 2 ϑ α = 1 , 0 ; m = ± ( l − 2 ) , cos 2 ϑ α = 1 , 1 2 l − 1 ; m = ± ( l − 3 ) , cos 2 ϑ α = 1 , 3 2 l − 1 , 0 ; m = ± ( l − 4 ) , cos 2 ϑ α = 1 , 3 ± 2 3 ( l − 2 ) / ( 2 l − 3 ) 2 l − 1 ; m = ± ( l − 5 ) , cos 2 ϑ α = 1 , 5 ± 2 5 ( l − 3 ) / ( 2 l − 3 ) 2 l − 1 , 0. (1093) \begin{aligned}
m&=\pm l,\qquad\cos^2\vartheta_\alpha=1;\\
m&=\pm(l-1),\qquad\cos^2\vartheta_\alpha=1,0;\\
m&=\pm(l-2),\qquad\cos^2\vartheta_\alpha=1,\frac{1}{2l-1};\\
m&=\pm(l-3),\qquad\cos^2\vartheta_\alpha=1,\frac{3}{2l-1},0;\\
m&=\pm(l-4),\qquad\cos^2\vartheta_\alpha=1,\frac{3\pm2\sqrt{3(l-2)/(2l-3)}}{2l-1};\\
m&=\pm(l-5),\qquad\cos^2\vartheta_\alpha=1,\frac{5\pm2\sqrt{5(l-3)/(2l-3)}}{2l-1},0. \tag*{(1093)}
\end{aligned} m m m m m m = ± l , cos 2 ϑ α = 1 ; = ± ( l − 1 ) , cos 2 ϑ α = 1 , 0 ; = ± ( l − 2 ) , cos 2 ϑ α = 1 , 2 l − 1 1 ; = ± ( l − 3 ) , cos 2 ϑ α = 1 , 2 l − 1 3 , 0 ; = ± ( l − 4 ) , cos 2 ϑ α = 1 , 2 l − 1 3 ± 2 3 ( l − 2 ) / ( 2 l − 3 ) ; = ± ( l − 5 ) , cos 2 ϑ α = 1 , 2 l − 1 5 ± 2 5 ( l − 3 ) / ( 2 l − 3 ) , 0. (1093)
Numerical values of ϑ α \vartheta_\alpha ϑ α for [missing text] can be found using computer programs. If l ≫ 1 l \gg1 l ≫ 1 , approximate values of ϑ α \vartheta_\alpha ϑ α may be found from the relation
ϑ α ≈ π 2 ( 4 k + 3 − 2 m 2 l + 1 ) , (1094) \vartheta_\alpha\approx\frac{\pi}{2}\left(\frac{4k+3-2m}{2l+1}\right), \tag*{(1094)} ϑ α ≈ 2 π ( 2 l + 1 4 k + 3 − 2 m ) , (1094)
where k k k is integer and 2 ∣ m ∣ − 3 4 ≤ k ≤ 4 l + 2 ∣ m ∣ − 1 4 \frac{2|m|-3}{4}\leq k\leq\frac{4l+2|m|-1}{4} 4 2∣ m ∣ − 3 ≤ k ≤ 4 4 l + 2∣ m ∣ − 1 .
Zeros of ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ )
If m = 0 m=0 m = 0 , ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ ) has l − 1 l-1 l − 1 zeros in the interval 0 < ϑ < π 0<\vartheta<\pi 0 < ϑ < π and also two zeros at ϑ β = 0 \vartheta_\beta=0 ϑ β = 0 and ϑ β = π \vartheta_\beta=\pi ϑ β = π . If m ≠ 0 m\ne0 m = 0 , ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ ) has ( l − ∣ m ∣ + 1 ) (l-|m|+1) ( l − ∣ m ∣ + 1 ) zeros in the same interval and two additional ones at ϑ β = 0 \vartheta_\beta=0 ϑ β = 0 and ϑ β = π \vartheta_\beta=\pi ϑ β = π provided m ≥ 2 m\geq2 m ≥ 2 . All zeros of ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ ) lie symmetrically with respect to ϑ = π / 2 \vartheta=\pi/2 ϑ = π /2 . Hence ∂ ∂ ϑ Y l m ( ϑ , φ ) = 0 \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi)=0 ∂ ϑ ∂ Y l m ( ϑ , φ ) = 0 for ϑ = π / 2 \vartheta=\pi/2 ϑ = π /2 provided ( l − m + 1 ) (l-m+1) ( l − m + 1 ) is odd. When m ≠ 0 m\ne0 m = 0 the values of ϑ β \vartheta_\beta ϑ β are determined by the equations
m = ± l cos 2 ϑ β = 0 ; 1 ( l ≥ 2 ) ; m = ± ( l − 1 ) cos 2 ϑ β = 1 l ; 1 ( l ≥ 3 ) ; m = ± ( l − 2 ) cos 2 ϑ β = 5 l − 4 l ( 2 l − 1 ) ; 0 ; 1 ( l ≥ 4 ) ; m = ± ( l − 3 ) cos 2 ϑ β = 9 ( l − 1 ) ± 3 ( 19 l 2 − 50 l + 27 ) 2 l ( 2 l − 1 ) ; 1 ( l ≥ 5 ) ; m = ± ( l − 4 ) cos 2 ϑ β = 7 l − 8 ± 2 ( 11 l 3 − 62 l 2 + 104 l − 48 ) / ( 2 l − 3 ) l ( 2 l − 1 ) ; 0 ; 1 ( l ≥ 6 ) . (1095) \begin{aligned}
m&=\pm l && \cos^2\vartheta_\beta=0;\quad1(l\geq2);\\
m&=\pm(l-1) && \cos^2\vartheta_\beta=\frac{1}{l};\quad1(l\geq3);\\
m&=\pm(l-2) && \cos^2\vartheta_\beta=\frac{5l-4}{l(2l-1)};\quad0;\ 1(l\geq4);\\
m&=\pm(l-3) && \cos^2\vartheta_\beta=\frac{9(l-1)\pm\sqrt{3(19l^2-50l+27)}}{2l(2l-1)};\quad1(l\geq5);\\
m&=\pm(l-4) && \cos^2\vartheta_\beta=\frac{7l-8\pm2\sqrt{(11l^3-62l^2+104l-48)/(2l-3)}}{l(2l-1)};\quad0;\ 1(l\geq6). \tag*{(1095)}
\end{aligned} m m m m m = ± l = ± ( l − 1 ) = ± ( l − 2 ) = ± ( l − 3 ) = ± ( l − 4 ) cos 2 ϑ β = 0 ; 1 ( l ≥ 2 ) ; cos 2 ϑ β = l 1 ; 1 ( l ≥ 3 ) ; cos 2 ϑ β = l ( 2 l − 1 ) 5 l − 4 ; 0 ; 1 ( l ≥ 4 ) ; cos 2 ϑ β = 2 l ( 2 l − 1 ) 9 ( l − 1 ) ± 3 ( 19 l 2 − 50 l + 27 ) ; 1 ( l ≥ 5 ) ; cos 2 ϑ β = l ( 2 l − 1 ) 7 l − 8 ± 2 ( 11 l 3 − 62 l 2 + 104 l − 48 ) / ( 2 l − 3 ) ; 0 ; 1 ( l ≥ 6 ) . (1095)
It is evident that Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) has an extremum at ϑ = ϑ β \vartheta=\vartheta_\beta ϑ = ϑ β . Zeros of Y l m ( ϑ , φ ) Y_{lm}(\vartheta,\varphi) Y l m ( ϑ , φ ) and ∂ ∂ ϑ Y l m ( ϑ , φ ) \frac{\partial}{\partial\vartheta}Y_{lm}(\vartheta,\varphi) ∂ ϑ ∂ Y l m ( ϑ , φ ) alternate in the domain 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π .
All harmonics Y l m ( ϑ , 0 ) Y_{lm}(\vartheta,0) Y l m ( ϑ , 0 ) and Y l + s , m + s ( ϑ , 0 ) Y_{l+s,m+s}(\vartheta,0) Y l + s , m + s ( ϑ , 0 ) with m > 0 m>0 m > 0 , and m + s > 0 m+s>0 m + s > 0 have the same number of maxima, minima and zeros, i.e., can be represented by topologically equivalent plots (Figure 5.1 ). With increasing l + s l+s l + s the positions of all maxima, minima and zeros of Y l + s , m + s ( ϑ , 0 ) Y_{l+s,m+s}(\vartheta,0) Y l + s , m + s ( ϑ , 0 ) shift to ϑ = π / 2 \vartheta=\pi/2 ϑ = π /2 .
Figure 5.1. Spherical harmonics Y l m ( ϑ , 0 ) Y_{lm}(\vartheta,0) Y l m ( ϑ , 0 ) are shown as functions of the polar angle ϑ \vartheta ϑ . Red and Blue colours indicate the sign of the function (positive and negative respectively).
Bipolar and tripolar spherical harmonics
In many applications one has to deal with functions which depend on two or three vector directions. Convenient bases for expansions of such functions are provided by the bipolar and tripolar harmonics.
Bipolar Spherical Harmonics
Bipolar spherical harmonics are given by irreducible tensor product of the spherical harmonics with different arguments
{ Y l 1 ( ϑ 1 , φ 1 ) ⊗ Y l 2 ( ϑ 2 , φ 2 ) } L M = ∑ m 1 m 2 C l 1 m 1 l 2 m 2 L M Y l 1 m 1 ( ϑ 1 , φ 1 ) Y l 2 m 2 ( ϑ 2 , φ 2 ) (1096) \left\{Y_{l_1}(\vartheta_1,\varphi_1)\otimes Y_{l_2}(\vartheta_2,\varphi_2)\right\}_{LM}
=\sum_{m_1m_2}C^{LM}_{l_1m_1\,l_2m_2}Y_{l_1m_1}(\vartheta_1,\varphi_1)Y_{l_2m_2}(\vartheta_2,\varphi_2) \tag*{(1096)} { Y l 1 ( ϑ 1 , φ 1 ) ⊗ Y l 2 ( ϑ 2 , φ 2 ) } L M = m 1 m 2 ∑ C l 1 m 1 l 2 m 2 L M Y l 1 m 1 ( ϑ 1 , φ 1 ) Y l 2 m 2 ( ϑ 2 , φ 2 ) (1096)
The variety of bipolar spherical harmonics with different l 1 , l 2 , L l_1,l_2,L l 1 , l 2 , L and M M M forms a complete orthonormal set of functions which depend on two unit vectors, n 1 \mathbf{n}_1 n 1 and n 2 \mathbf{n}_2 n 2 . These functions possess simple transformation properties under rotations and inversion of coordinate system.
The orthogonality and normalization relation for these harmonics is given by
∬ d Ω 1 d Ω 2 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M ∗ { Y l 1 ′ ( Ω 1 ) ⊗ Y l 2 ′ ( Ω 2 ) } L ′ M ′ = δ l 1 l 1 ′ δ l 2 l 2 ′ δ L L ′ δ M M ′ (1097) \iint d\Omega_1d\Omega_2\left\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\right\}_{LM}^{*}\left\{Y_{l'_1}(\Omega_1)\otimes Y_{l'_2}(\Omega_2)\right\}_{L'M'}=\delta_{l_1l'_1}\delta_{l_2l'_2}\delta_{LL'}\delta_{MM'} \tag*{(1097)} ∬ d Ω 1 d Ω 2 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M ∗ { Y l 1 ′ ( Ω 1 ) ⊗ Y l 2 ′ ( Ω 2 ) } L ′ M ′ = δ l 1 l 1 ′ δ l 2 l 2 ′ δ L L ′ δ M M ′ (1097)
where
Ω ≡ { ϑ , φ } , ∫ d Ω = ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ \Omega\equiv\{\vartheta,\varphi\},\qquad\int d\Omega=\int_{0}^{2\pi}d\varphi\int_{0}^{\pi}d\vartheta\sin\vartheta Ω ≡ { ϑ , φ } , ∫ d Ω = ∫ 0 2 π d φ ∫ 0 π d ϑ sin ϑ
The completeness condition has the form
∑ l 1 l 2 L M { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M { Y l 1 ( Ω 1 ′ ) ⊗ Y l 2 ( Ω 2 ′ ) } L M ∗ = δ ( Ω 1 − Ω 1 ′ ) δ ( Ω 2 − Ω 2 ′ ) (1098) \sum_{l_1l_2LM}\left\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\right\}_{LM}\left\{Y_{l_1}(\Omega'_1)\otimes Y_{l_2}(\Omega'_2)\right\}_{LM}^{*}=\delta(\Omega_1-\Omega'_1)\delta(\Omega_2-\Omega'_2) \tag*{(1098)} l 1 l 2 L M ∑ { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M { Y l 1 ( Ω 1 ′ ) ⊗ Y l 2 ( Ω 2 ′ ) } L M ∗ = δ ( Ω 1 − Ω 1 ′ ) δ ( Ω 2 − Ω 2 ′ ) (1098)
where
δ ( Ω − Ω ′ ) ≡ δ ( cos ϑ − cos ϑ ′ ) δ ( φ − φ ′ ) . \delta(\Omega-\Omega')\equiv\delta(\cos\vartheta-\cos\vartheta')\delta(\varphi-\varphi'). δ ( Ω − Ω ′ ) ≡ δ ( cos ϑ − cos ϑ ′ ) δ ( φ − φ ′ ) .
In addition, let us note the following equation
∑ L M ∣ { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M ∣ 2 = ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 4 π ) 2 (1099) \sum_{LM}\left|\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM}\right|^2=\frac{(2l_1+1)(2l_2+1)}{(4\pi)^2} \tag*{(1099)} L M ∑ ∣ { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M ∣ 2 = ( 4 π ) 2 ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) (1099)
Under a rotation of the coordinate system the bipolar harmonics transform according to
D ^ ( α , β , γ ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M ′ = { Y l 1 ( Ω 1 ′ ) ⊗ Y l 2 ( Ω 2 ′ ) } L M ′ = ∑ M D M M ′ L ( α , β , γ ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1100) \hat{D}(\alpha,\beta,\gamma)\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM'}=\{Y_{l_1}(\Omega'_1)\otimes Y_{l_2}(\Omega'_2)\}_{LM'}=\sum_M D^L_{MM'}(\alpha,\beta,\gamma)\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM} \tag*{(1100)} D ^ ( α , β , γ ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M ′ = { Y l 1 ( Ω 1 ′ ) ⊗ Y l 2 ( Ω 2 ′ ) } L M ′ = M ∑ D M M ′ L ( α , β , γ ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1100)
where the angles Ω i ′ ≡ { ϑ i ′ , φ i ′ } \Omega'_i\equiv\{\vartheta'_i,\varphi'_i\} Ω i ′ ≡ { ϑ i ′ , φ i ′ } and Ω i ≡ { ϑ i , φ i } \Omega_i\equiv\{\vartheta_i,\varphi_i\} Ω i ≡ { ϑ i , φ i } are related by Eqs. (131) , (132) . Under inversion of coordinate system the phases of the bipolar harmonics change according to
P ^ r { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M = ( − 1 ) l 1 + l 2 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1101) \hat{P}_r\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM}=(-1)^{l_1+l_2}\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM} \tag*{(1101)} P ^ r { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M = ( − 1 ) l 1 + l 2 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1101)
The Clebsch-Gordan series for product of the bipolar harmonics may be presented in the form
{ Y l 1 ′ ( Ω 1 ) ⊗ Y l 2 ′ ( Ω 2 ) } L ′ M ′ { Y l 1 ′ ′ ( Ω 1 ) ⊗ Y l 2 ′ ′ ( Ω 2 ) } L ′ ′ M ′ ′ = ∑ L M C L ′ M ′ L ′ ′ M ′ ′ L M ∑ l 1 l 2 B l 1 ′ l 2 ′ L ′ l 1 ′ ′ l 2 ′ ′ L ′ ′ l 1 l 2 L { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M . (1102) \begin{aligned}
\{Y_{l'_1}(\Omega_1)\otimes Y_{l'_2}(\Omega_2)\}_{L'M'}\{Y_{l''_1}(\Omega_1)\otimes Y_{l''_2}(\Omega_2)\}_{L''M''}
={}&\sum_{LM}C^{LM}_{L'M'L''M''}\sum_{l_1l_2}B^{l_1l_2L}_{l'_1l'_2L'l''_1l''_2L''}\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM}. \tag*{(1102)}
\end{aligned} { Y l 1 ′ ( Ω 1 ) ⊗ Y l 2 ′ ( Ω 2 ) } L ′ M ′ { Y l 1 ′′ ( Ω 1 ) ⊗ Y l 2 ′′ ( Ω 2 ) } L ′′ M ′′ = L M ∑ C L ′ M ′ L ′′ M ′′ L M l 1 l 2 ∑ B l 1 ′ l 2 ′ L ′ l 1 ′′ l 2 ′′ L ′′ l 1 l 2 L { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M . (1102)
where
B l 1 ′ l 2 ′ L ′ l 1 ′ ′ l 2 ′ ′ L ′ ′ l 1 l 2 L = ( 2 l 1 ′ + 1 ) ( 2 l 2 ′ + 1 ) ( 2 l 1 ′ ′ + 1 ) ( 2 l 2 ′ ′ + 1 ) ( 2 L ′ + 1 ) ( 2 L ′ ′ + 1 ) ( 4 π ) 2 C l 1 ′ 0 l 1 ′ ′ 0 l 1 0 C l 2 ′ 0 l 2 ′ ′ 0 l 2 0 { l 1 ′ l 1 ′ ′ l 1 l 2 ′ l 2 ′ ′ l 2 L ′ L ′ ′ L } . (1103) B^{l_1l_2L}_{l'_1l'_2L'l''_1l''_2L''}=\sqrt{\frac{(2l'_1+1)(2l'_2+1)(2l''_1+1)(2l''_2+1)(2L'+1)(2L''+1)}{(4\pi)^2}}C^{l_1 0}_{l'_1 0\,l''_1 0}C^{l_2 0}_{l'_2 0\,l''_2 0}\left\{\begin{matrix} l'_1 & l''_1 & l_1 \\ l'_2 & l''_2 & l_2 \\ L' & L'' & L \end{matrix}\right\}. \tag*{(1103)} B l 1 ′ l 2 ′ L ′ l 1 ′′ l 2 ′′ L ′′ l 1 l 2 L = ( 4 π ) 2 ( 2 l 1 ′ + 1 ) ( 2 l 2 ′ + 1 ) ( 2 l 1 ′′ + 1 ) ( 2 l 2 ′′ + 1 ) ( 2 L ′ + 1 ) ( 2 L ′′ + 1 ) C l 1 ′ 0 l 1 ′′ 0 l 1 0 C l 2 ′ 0 l 2 ′′ 0 l 2 0 ⎩ ⎨ ⎧ l 1 ′ l 2 ′ L ′ l 1 ′′ l 2 ′′ L ′′ l 1 l 2 L ⎭ ⎬ ⎫ . (1103)
For the important special case L = 0 L=0 L = 0 a bipolar harmonic is reduced to a scalar product of the spherical harmonics
{ Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } 00 = ( − 1 ) l 1 2 l 1 + 1 ( Y l 1 ( Ω 1 ) Y l 1 ( Ω 2 ) ) δ l 1 l 2 (1104) \{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{00}=\frac{(-1)^{l_1}}{\sqrt{2l_1+1}}(Y_{l_1}(\Omega_1)Y_{l_1}(\Omega_2))\delta_{l_1l_2} \tag*{(1104)} { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } 00 = 2 l 1 + 1 ( − 1 ) l 1 ( Y l 1 ( Ω 1 ) Y l 1 ( Ω 2 )) δ l 1 l 2 (1104)
In accordance with Eq. (464) this scalar product is given by
( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) = ∑ m = − l l Y l m ∗ ( Ω 1 ) Y l m ( Ω 2 ) (1105) (Y_l(\Omega_1)\cdot Y_l(\Omega_2))=\sum_{m=-l}^{l}Y^*_{lm}(\Omega_1)Y_{lm}(\Omega_2) \tag*{(1105)} ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 )) = m = − l ∑ l Y l m ∗ ( Ω 1 ) Y l m ( Ω 2 ) (1105)
Tripolar Spherical Harmonics
A tripolar spherical harmonic is defined as an irreducible tensor product of three spherical harmonics with different arguments,
{ Y l 1 ( ϑ 1 , φ 1 ) ⊗ { Y l 2 ( ϑ 2 , φ 2 ) ⊗ Y l 3 ( ϑ 3 , φ 3 ) } l 23 } L M = ∑ m 1 m 2 m 3 m 23 C l 1 m 1 l 23 m 23 L M C l 2 m 2 l 3 m 3 l 23 m 23 Y l 1 m 1 ( ϑ 1 , φ 1 ) Y l 2 m 2 ( ϑ 2 , φ 2 ) Y l 3 m 3 ( ϑ 3 , φ 3 ) (1106) \{Y_{l_1}(\vartheta_1,\varphi_1)\otimes\{Y_{l_2}(\vartheta_2,\varphi_2)\otimes Y_{l_3}(\vartheta_3,\varphi_3)\}_{l_{23}}\}_{LM}
=\sum_{\substack{m_1m_2m_3\\m_{23}}}C^{LM}_{l_1m_1l_{23}m_{23}}C^{l_{23}m_{23}}_{l_2m_2l_3m_3}Y_{l_1m_1}(\vartheta_1,\varphi_1)Y_{l_2m_2}(\vartheta_2,\varphi_2)Y_{l_3m_3}(\vartheta_3,\varphi_3) \tag*{(1106)} { Y l 1 ( ϑ 1 , φ 1 ) ⊗ { Y l 2 ( ϑ 2 , φ 2 ) ⊗ Y l 3 ( ϑ 3 , φ 3 ) } l 23 } L M = m 1 m 2 m 3 m 23 ∑ C l 1 m 1 l 23 m 23 L M C l 2 m 2 l 3 m 3 l 23 m 23 Y l 1 m 1 ( ϑ 1 , φ 1 ) Y l 2 m 2 ( ϑ 2 , φ 2 ) Y l 3 m 3 ( ϑ 3 , φ 3 ) (1106)
In contrast to bipolar harmonics, the tripolar harmonics allow different coupling schemes of angular momenta, namely, l 1 + ( l 2 + l 3 ) l 23 = L l_1+(l_2+l_3)_{l_{23}}=L l 1 + ( l 2 + l 3 ) l 23 = L , ( l 1 + l 2 ) l 12 + l 3 = L (l_1+l_2)_{l_{12}}+l_3=L ( l 1 + l 2 ) l 12 + l 3 = L or ( l 1 + l 3 ) l 13 + l 2 = L (l_1+l_3)_{l_{13}}+l_2=L ( l 1 + l 3 ) l 13 + l 2 = L . The relations between different coupling schemes are considered in § 13 .
The tripolar spherical harmonics form a complete orthonormal set of functions which depend on three unit vectors n 1 n_1 n 1 , n 2 n_2 n 2 and n 3 n_3 n 3 .
The orthogonality and normalization relations is as follows
∫ ∫ ∫ d Ω 1 d Ω 2 d Ω 3 { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M { Y l 1 ′ ( Ω 1 ) ⊗ { Y l 2 ′ ( Ω 2 ) ⊗ Y l 3 ′ ( Ω 3 ) } λ ′ } L ′ M ′ ∗ = δ l 1 l 1 ′ δ l 2 l 2 ′ δ l 3 l 3 ′ δ λ λ ′ δ L L ′ δ M M ′ . (1107) \begin{aligned}
\int\!\!\int\!\!\int d\Omega_1d\Omega_2d\Omega_3\ \{Y_{l_1}(\Omega_1)\otimes\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\}_{\lambda}\}_{LM}\{Y_{l'_1}(\Omega_1)\otimes\{Y_{l'_2}(\Omega_2)\otimes Y_{l'_3}(\Omega_3)\}_{\lambda'}\}_{L'M'}^*
=\delta_{l_1l'_1}\delta_{l_2l'_2}\delta_{l_3l'_3}\delta_{\lambda\lambda'}\delta_{LL'}\delta_{MM'}. \tag*{(1107)}
\end{aligned} ∫ ∫ ∫ d Ω 1 d Ω 2 d Ω 3 { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M { Y l 1 ′ ( Ω 1 ) ⊗ { Y l 2 ′ ( Ω 2 ) ⊗ Y l 3 ′ ( Ω 3 ) } λ ′ } L ′ M ′ ∗ = δ l 1 l 1 ′ δ l 2 l 2 ′ δ l 3 l 3 ′ δ λ λ ′ δ L L ′ δ M M ′ . (1107)
The completeness relation may be presented in the form
∑ l 1 l 2 l 3 λ L M { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M { Y l 1 ( Ω 1 ′ ) ⊗ { Y l 2 ( Ω 2 ′ ) ⊗ Y l 3 ( Ω 3 ′ ) } λ } L M ∗ = δ ( Ω 1 − Ω 1 ′ ) δ ( Ω 2 − Ω 2 ′ ) δ ( Ω 3 − Ω 3 ′ ) . (1108) \begin{aligned}
\sum_{\substack{l_1l_2l_3\\\lambda LM}}\{Y_{l_1}(\Omega_1)\otimes\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\}_{\lambda}\}_{LM}\{Y_{l_1}(\Omega'_1)\otimes\{Y_{l_2}(\Omega'_2)\otimes Y_{l_3}(\Omega'_3)\}_{\lambda}\}_{LM}^*
=\delta(\Omega_1-\Omega'_1)\delta(\Omega_2-\Omega'_2)\delta(\Omega_3-\Omega'_3). \tag*{(1108)}
\end{aligned} l 1 l 2 l 3 λ L M ∑ { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M { Y l 1 ( Ω 1 ′ ) ⊗ { Y l 2 ( Ω 2 ′ ) ⊗ Y l 3 ( Ω 3 ′ ) } λ } L M ∗ = δ ( Ω 1 − Ω 1 ′ ) δ ( Ω 2 − Ω 2 ′ ) δ ( Ω 3 − Ω 3 ′ ) . (1108)
In addition we also have the following equation
∑ λ L M ∣ { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M ∣ 2 = ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 2 l 3 + 1 ) ( 4 π ) 3 (1109) \sum_{\lambda L M}\left|\left\{Y_{l_1}(\Omega_1)\otimes\left\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\right\}_{\lambda}\right\}_{LM}\right|^2=\frac{(2l_1+1)(2l_2+1)(2l_3+1)}{(4\pi)^3} \tag*{(1109)} λ L M ∑ { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M 2 = ( 4 π ) 3 ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) ( 2 l 3 + 1 ) (1109)
Under a rotation of the coordinate system the tripolar harmonics transform according to
D ^ ( α , β , γ ) { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M ′ = { Y l 1 ( Ω 1 ′ ) ⊗ { Y l 2 ( Ω 2 ′ ) ⊗ Y l 3 ( Ω 3 ′ ) } λ } L M ′ = ∑ M D M M ′ L ( α , β , γ ) { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M (1110) \begin{aligned}
\hat{D}(\alpha,\beta,\gamma)\left\{Y_{l_1}(\Omega_1)\otimes\left\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\right\}_{\lambda}\right\}_{LM'}=\left\{Y_{l_1}(\Omega'_1)\otimes\left\{Y_{l_2}(\Omega'_2)\otimes Y_{l_3}(\Omega'_3)\right\}_{\lambda}\right\}_{LM'} \\
&=\sum_M D^L_{MM'}(\alpha,\beta,\gamma)\left\{Y_{l_1}(\Omega_1)\otimes\left\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\right\}_{\lambda}\right\}_{LM} \tag*{(1110)}
\end{aligned} D ^ ( α , β , γ ) { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M ′ = { Y l 1 ( Ω 1 ′ ) ⊗ { Y l 2 ( Ω 2 ′ ) ⊗ Y l 3 ( Ω 3 ′ ) } λ } L M ′ = M ∑ D M M ′ L ( α , β , γ ) { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M (1110)
where the angles Ω i ′ ≡ { ϑ i ′ , φ i ′ } \Omega'_i\equiv\{\vartheta'_i,\varphi'_i\} Ω i ′ ≡ { ϑ i ′ , φ i ′ } and Ω i ≡ { ϑ i , φ i } \Omega_i\equiv\{\vartheta_i,\varphi_i\} Ω i ≡ { ϑ i , φ i } are related through Eqs. (131) , (132) .
Under inversion of the coordinate system the tripolar harmonics acquire the phase factor
P ^ r { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M = ( − 1 ) l 1 + l 2 + l 3 { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M (1111) \hat{P}_{r}\left\{Y_{l_1}(\Omega_1)\otimes\left\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\right\}_{\lambda}\right\}_{LM}=(-1)^{l_1+l_2+l_3}\left\{Y_{l_1}(\Omega_1)\otimes\left\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\right\}_{\lambda}\right\}_{LM} \tag*{(1111)} P ^ r { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M = ( − 1 ) l 1 + l 2 + l 3 { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M (1111)
The Clebsch-Gordan series for the tripolar harmonics is given by
{ Y l 1 ′ ( Ω 1 ) ⊗ { Y l 2 ′ ( Ω 2 ) ⊗ Y l 3 ′ ( Ω 3 ) } λ ′ } L ′ M ′ , { Y l 1 ′ ′ ( Ω 1 ) ⊗ { Y l 2 ′ ′ ( Ω 2 ) ⊗ Y l 3 ′ ′ ( Ω 3 ) } λ ′ ′ } L ′ ′ M ′ ′ = ∑ L M C L ′ M ′ L ′ ′ M ′ ′ L M ∑ l 1 l 2 l 3 λ B l 1 ′ l 2 ′ l 3 ′ λ ′ L ′ l 1 ′ ′ l 2 ′ ′ l 3 ′ ′ λ ′ ′ L ′ ′ l 1 l 2 l 3 λ L { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M (1112) \begin{aligned}
\left\{Y_{l'_1}(\Omega_1)\otimes\left\{Y_{l'_2}(\Omega_2)\otimes Y_{l'_3}(\Omega_3)\right\}_{\lambda'}\right\}_{L'M'},\left\{Y_{l''_1}(\Omega_1)\otimes\left\{Y_{l''_2}(\Omega_2)\otimes Y_{l''_3}(\Omega_3)\right\}_{\lambda''}\right\}_{L''M''} \\
&=\sum_{LM}C^{LM}_{L'M'L''M''}\sum_{l_1l_2l_3\lambda}B^{l_1l_2l_3\lambda L}_{l'_1l'_2l'_3\lambda'L'l''_1l''_2l''_3\lambda''L''}\left\{Y_{l_1}(\Omega_1)\otimes\left\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\right\}_{\lambda}\right\}_{LM} \tag*{(1112)}
\end{aligned} { Y l 1 ′ ( Ω 1 ) ⊗ { Y l 2 ′ ( Ω 2 ) ⊗ Y l 3 ′ ( Ω 3 ) } λ ′ } L ′ M ′ , { Y l 1 ′′ ( Ω 1 ) ⊗ { Y l 2 ′′ ( Ω 2 ) ⊗ Y l 3 ′′ ( Ω 3 ) } λ ′′ } L ′′ M ′′ = L M ∑ C L ′ M ′ L ′′ M ′′ L M l 1 l 2 l 3 λ ∑ B l 1 ′ l 2 ′ l 3 ′ λ ′ L ′ l 1 ′′ l 2 ′′ l 3 ′′ λ ′′ L ′′ l 1 l 2 l 3 λ L { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } L M (1112)
where
B l 1 ′ l 2 ′ l 3 ′ λ ′ L ′ l 1 ′ ′ l 2 ′ ′ l 3 ′ ′ λ ′ ′ L ′ ′ l 1 l 2 l 3 λ L = ( 2 l 1 ′ + 1 ) ( 2 l 1 ′ ′ + 1 ) ( 2 l 2 ′ + 1 ) ( 2 l 2 ′ ′ + 1 ) ( 2 l 3 ′ ′ + 1 ) ( 2 l 3 ′ + 1 ) ( 2 L ′ + 1 ) ( 2 L ′ ′ + 1 ) ( 4 π ) 3 × ( 2 λ ′ + 1 ) ( 2 λ ′ ′ + 1 ) ( 2 λ + 1 ) C l 1 ′ 0 l 1 ′ ′ 0 l 1 0 C l 2 ′ 0 l 2 ′ ′ 0 l 2 0 C l 3 ′ 0 l 3 ′ ′ 0 l 3 0 { l 1 ′ l 1 ′ ′ l 1 λ ′ λ ′ ′ λ L ′ L ′ ′ L } { l 2 ′ l 2 ′ ′ l 2 l 3 ′ l 3 ′ ′ l 3 λ ′ λ ′ ′ λ } (1113) \begin{aligned}
B^{l_1l_2l_3\lambda L}_{l'_1l'_2l'_3\lambda'L'l''_1l''_2l''_3\lambda''L''}=
\sqrt{\frac{(2l'_1+1)(2l''_1+1)(2l'_2+1)(2l''_2+1)(2l''_3+1)(2l'_3+1)(2L'+1)(2L''+1)}{(4\pi)^3}} \\
&\times\sqrt{(2\lambda'+1)(2\lambda''+1)(2\lambda+1)}
C^{l_1 0}_{l'_1 0 l''_1 0}C^{l_2 0}_{l'_2 0 l''_2 0}C^{l_3 0}_{l'_3 0 l''_3 0}
\begin{Bmatrix}
l'_1 & l''_1 & l_1\\
\lambda' & \lambda'' & \lambda\\
L' & L'' & L
\end{Bmatrix}
\begin{Bmatrix}
l'_2 & l''_2 & l_2\\
l'_3 & l''_3 & l_3\\
\lambda' & \lambda'' & \lambda
\end{Bmatrix} \tag*{(1113)}
\end{aligned} B l 1 ′ l 2 ′ l 3 ′ λ ′ L ′ l 1 ′′ l 2 ′′ l 3 ′′ λ ′′ L ′′ l 1 l 2 l 3 λ L = ( 4 π ) 3 ( 2 l 1 ′ + 1 ) ( 2 l 1 ′′ + 1 ) ( 2 l 2 ′ + 1 ) ( 2 l 2 ′′ + 1 ) ( 2 l 3 ′′ + 1 ) ( 2 l 3 ′ + 1 ) ( 2 L ′ + 1 ) ( 2 L ′′ + 1 ) × ( 2 λ ′ + 1 ) ( 2 λ ′′ + 1 ) ( 2 λ + 1 ) C l 1 ′ 0 l 1 ′′ 0 l 1 0 C l 2 ′ 0 l 2 ′′ 0 l 2 0 C l 3 ′ 0 l 3 ′′ 0 l 3 0 ⎩ ⎨ ⎧ l 1 ′ λ ′ L ′ l 1 ′′ λ ′′ L ′′ l 1 λ L ⎭ ⎬ ⎫ ⎩ ⎨ ⎧ l 2 ′ l 3 ′ λ ′ l 2 ′′ l 3 ′′ λ ′′ l 2 l 3 λ ⎭ ⎬ ⎫ (1113)
Putting L = 0 L=0 L = 0 in Eq. (1106) , we get an important special case for the tripolar scalar harmonic:
{ Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } 00 = ( − 1 ) l 1 + l 2 + l 3 δ λ l 1 ∑ m 1 m 2 m 3 ( l 1 l 2 l 3 m 1 m 2 m 3 ) Y l 1 m 1 ( Ω 1 ) Y l 2 m 2 ( Ω 2 ) Y l 3 m 3 ( Ω 3 ) (1114) \left\{Y_{l_1}(\Omega_1)\otimes\left\{Y_{l_2}(\Omega_2)\otimes Y_{l_3}(\Omega_3)\right\}_{\lambda}\right\}_{00}
=(-1)^{l_1+l_2+l_3}\delta_{\lambda l_1}\sum_{m_1m_2m_3}
\begin{pmatrix}
l_1 & l_2 & l_3\\
m_1 & m_2 & m_3
\end{pmatrix}
Y_{l_1m_1}(\Omega_1)Y_{l_2m_2}(\Omega_2)Y_{l_3m_3}(\Omega_3) \tag*{(1114)} { Y l 1 ( Ω 1 ) ⊗ { Y l 2 ( Ω 2 ) ⊗ Y l 3 ( Ω 3 ) } λ } 00 = ( − 1 ) l 1 + l 2 + l 3 δ λ l 1 m 1 m 2 m 3 ∑ ( l 1 m 1 l 2 m 2 l 3 m 3 ) Y l 1 m 1 ( Ω 1 ) Y l 2 m 2 ( Ω 2 ) Y l 3 m 3 ( Ω 3 ) (1114)
Expansions of functions which depend on two vectors
Preliminary Remarks
Any function f ( r 1 , r 2 ) f(\mathbf{r}_1,\mathbf{r}_2) f ( r 1 , r 2 ) which depends on two arbitrary vectors r 1 ( r 1 , ϑ 1 , φ 1 ) \mathbf{r}_1(r_1,\vartheta_1,\varphi_1) r 1 ( r 1 , ϑ 1 , φ 1 ) and r 2 ( r 2 , ϑ 2 , φ 2 ) \mathbf{r}_2(r_2,\vartheta_2,\varphi_2) r 2 ( r 2 , ϑ 2 , φ 2 ) may be expanded into a series of the bipolar harmonics (§ 24.9 ):
f ( r 1 , r 2 ) = ∑ l 1 l 2 L M f l 1 l 2 L M ( r 1 , r 2 ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1115) f(\mathbf{r}_1,\mathbf{r}_2)=\sum_{l_1l_2LM}f^{LM}_{l_1l_2}(r_1,r_2)\left\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\right\}_{LM} \tag*{(1115)} f ( r 1 , r 2 ) = l 1 l 2 L M ∑ f l 1 l 2 L M ( r 1 , r 2 ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1115)
The function f ( r 1 , r 2 ) f(\mathbf{r}_1,\mathbf{r}_2) f ( r 1 , r 2 ) may be presented in terms of r 1 \mathbf{r}_1 r 1 and r 2 \mathbf{r}_2 r 2 as well as in terms of R \mathbf{R} R and r \mathbf{r} r , where
R = r 1 + r 2 , r = r 1 − r 2 , r 1 = 1 2 ( R + r ) , r 2 = 1 2 ( R − r ) (1116) \mathbf{R}=\mathbf{r}_1+\mathbf{r}_2,\quad\mathbf{r}=\mathbf{r}_1-\mathbf{r}_2,\quad\mathbf{r}_1=\frac{1}{2}(\mathbf{R}+\mathbf{r}),\quad\mathbf{r}_2=\frac{1}{2}(\mathbf{R}-\mathbf{r}) \tag*{(1116)} R = r 1 + r 2 , r = r 1 − r 2 , r 1 = 2 1 ( R + r ) , r 2 = 2 1 ( R − r ) (1116)
Polar coordinates of the vectors R ( R , Θ , Φ ) \mathbf{R}(R,\Theta,\Phi) R ( R , Θ , Φ ) and r ( r , ϑ , φ ) \mathbf{r}(r,\vartheta,\varphi) r ( r , ϑ , φ ) are related to polar coordinates of the vectors r 1 ( r 1 , ϑ 1 , φ 1 ) \mathbf{r}_1(r_1,\vartheta_1,\varphi_1) r 1 ( r 1 , ϑ 1 , φ 1 ) and r 2 ( r 2 , ϑ 2 , φ 2 ) \mathbf{r}_2(r_2,\vartheta_2,\varphi_2) r 2 ( r 2 , ϑ 2 , φ 2 ) by
R 2 = r 1 2 + r 2 2 + 2 r 1 r 2 cos ω 12 , R^2=r_1^2+r_2^2+2r_1r_2\cos\omega_{12}, R 2 = r 1 2 + r 2 2 + 2 r 1 r 2 cos ω 12 ,
cos Θ = r 1 cos ϑ 1 + r 2 cos ϑ 2 r 1 2 + r 2 2 + 2 r 1 r 2 cos ω 12 , \cos\Theta=\frac{r_1\cos\vartheta_1+r_2\cos\vartheta_2}{\sqrt{r_1^2+r_2^2+2r_1r_2\cos\omega_{12}}}, cos Θ = r 1 2 + r 2 2 + 2 r 1 r 2 cos ω 12 r 1 cos ϑ 1 + r 2 cos ϑ 2 ,
tan Φ = r 1 sin ϑ 1 sin φ 1 + r 2 sin ϑ 2 sin φ 2 r 1 sin ϑ 1 cos φ 1 + r 2 sin ϑ 2 cos φ 2 , \tan\Phi=\frac{r_1\sin\vartheta_1\sin\varphi_1+r_2\sin\vartheta_2\sin\varphi_2}{r_1\sin\vartheta_1\cos\varphi_1+r_2\sin\vartheta_2\cos\varphi_2}, tan Φ = r 1 sin ϑ 1 cos φ 1 + r 2 sin ϑ 2 cos φ 2 r 1 sin ϑ 1 sin φ 1 + r 2 sin ϑ 2 sin φ 2 ,
r 2 = r 1 2 + r 2 2 − 2 r 1 r 2 cos ω 12 , r^2=r_1^2+r_2^2-2r_1r_2\cos\omega_{12}, r 2 = r 1 2 + r 2 2 − 2 r 1 r 2 cos ω 12 ,
cos ϑ = r 1 cos ϑ 1 − r 2 cos ϑ 2 r 1 2 + r 2 2 − 2 r 1 r 2 cos ω 12 , \cos\vartheta=\frac{r_1\cos\vartheta_1-r_2\cos\vartheta_2}{\sqrt{r_1^2+r_2^2-2r_1r_2\cos\omega_{12}}}, cos ϑ = r 1 2 + r 2 2 − 2 r 1 r 2 cos ω 12 r 1 cos ϑ 1 − r 2 cos ϑ 2 ,
tan φ = r 1 sin ϑ 1 sin φ 1 − r 2 sin ϑ 2 sin φ 2 r 1 sin ϑ 1 cos φ 1 − r 2 sin ϑ 2 cos φ 2 (1117) \tan\varphi=\frac{r_1\sin\vartheta_1\sin\varphi_1-r_2\sin\vartheta_2\sin\varphi_2}{r_1\sin\vartheta_1\cos\varphi_1-r_2\sin\vartheta_2\cos\varphi_2} \tag*{(1117)} tan φ = r 1 sin ϑ 1 cos φ 1 − r 2 sin ϑ 2 cos φ 2 r 1 sin ϑ 1 sin φ 1 − r 2 sin ϑ 2 sin φ 2 (1117)
Here ω 12 \omega_{12} ω 12 is the angle between r 1 \mathbf{r}_1 r 1 and r 2 \mathbf{r}_2 r 2 (Figure 5.2 ),
Figure 5.2. Relations between the vectors R \mathbf{R} R , r \mathbf{r} r and r 1 , r 2 \mathbf{r}_1,\mathbf{r}_2 r 1 , r 2 .
cos ω 12 = cos ϑ 1 cos ϑ 2 + sin ϑ 1 sin ϑ 2 cos ( φ 1 − φ 2 ) . (1118) \cos\omega_{12}=\cos\vartheta_1\cos\vartheta_2+\sin\vartheta_1\sin\vartheta_2\cos(\varphi_1-\varphi_2). \tag*{(1118)} cos ω 12 = cos ϑ 1 cos ϑ 2 + sin ϑ 1 sin ϑ 2 cos ( φ 1 − φ 2 ) . (1118)
The inverse relations to (5.17.3) can be written in the form
4 r 1 2 = R 2 + r 2 + 2 R r cos ω , 4 r 2 2 = R 2 + r 2 − 2 R r cos ω , cos ϑ 1 = R cos Θ + r cos ϑ R 2 + r 2 + 2 R r cos ω , cos ϑ 2 = R cos Θ − r cos ϑ R 2 + r 2 − 2 R r cos ω , tan φ 1 = R sin Θ sin Φ + r sin ϑ sin φ R sin Θ cos Φ + r sin ϑ cos φ , tan φ 2 = R sin Θ sin Φ − r sin ϑ sin φ R sin Θ cos Φ − r sin ϑ cos φ . (1119) \begin{aligned}
4r_1^2&=R^2+r^2+2Rr\cos\omega, & 4r_2^2&=R^2+r^2-2Rr\cos\omega,\\
\cos\vartheta_1&=\frac{R\cos\Theta+r\cos\vartheta}{\sqrt{R^2+r^2+2Rr\cos\omega}}, &
\cos\vartheta_2&=\frac{R\cos\Theta-r\cos\vartheta}{\sqrt{R^2+r^2-2Rr\cos\omega}},\\
\tan\varphi_1&=\frac{R\sin\Theta\sin\Phi+r\sin\vartheta\sin\varphi}{R\sin\Theta\cos\Phi+r\sin\vartheta\cos\varphi}, &
\tan\varphi_2&=\frac{R\sin\Theta\sin\Phi-r\sin\vartheta\sin\varphi}{R\sin\Theta\cos\Phi-r\sin\vartheta\cos\varphi}. \tag*{(1119)}
\end{aligned} 4 r 1 2 cos ϑ 1 tan φ 1 = R 2 + r 2 + 2 R r cos ω , = R 2 + r 2 + 2 R r cos ω R cos Θ + r cos ϑ , = R sin Θ cos Φ + r sin ϑ cos φ R sin Θ sin Φ + r sin ϑ sin φ , 4 r 2 2 cos ϑ 2 tan φ 2 = R 2 + r 2 − 2 R r cos ω , = R 2 + r 2 − 2 R r cos ω R cos Θ − r cos ϑ , = R sin Θ cos Φ − r sin ϑ cos φ R sin Θ sin Φ − r sin ϑ sin φ . (1119)
ω \omega ω being the angle between R \mathbf{R} R and r \mathbf{r} r (Figure 5.2 )
cos ω = cos Θ cos ϑ + sin Θ sin ϑ cos ( Φ − φ ) . (1120) \cos\omega=\cos\Theta\cos\vartheta+\sin\Theta\sin\vartheta\cos(\Phi-\varphi). \tag*{(1120)} cos ω = cos Θ cos ϑ + sin Θ sin ϑ cos ( Φ − φ ) . (1120)
Note that the expressions for the polar coordinates of R \mathbf{R} R and r \mathbf{r} r are interchanged by the replacement r 2 → − r 2 \mathbf{r}_2\to-\mathbf{r}_2 r 2 → − r 2 , i.e., ϑ 2 → π − ϑ 2 \vartheta_2\to\pi-\vartheta_2 ϑ 2 → π − ϑ 2 , φ 2 → π + φ 2 \varphi_2\to\pi+\varphi_2 φ 2 → π + φ 2 . Hence, the expansions of functions F ( R , r ) F(\mathbf{R},\mathbf{r}) F ( R , r ) and F ( r , R ) F(\mathbf{r},\mathbf{R}) F ( r , R ) , which may be reduced to each other by the replacement R ⇄ r \mathbf{R}\rightleftarrows\mathbf{r} R ⇄ r are mutually related. If
F ( R , r ) = f ( r 1 , r 2 ) = ∑ l 1 l 2 L M C l 1 l 2 L M ( r 1 , r 2 ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1121) F(\mathbf{R},\mathbf{r})=f(\mathbf{r}_1,\mathbf{r}_2)=\sum_{l_1l_2LM}C^{LM}_{l_1l_2}(r_1,r_2)\{\mathbf{Y}_{l_1}(\Omega_1)\otimes\mathbf{Y}_{l_2}(\Omega_2)\}_{LM} \tag*{(1121)} F ( R , r ) = f ( r 1 , r 2 ) = l 1 l 2 L M ∑ C l 1 l 2 L M ( r 1 , r 2 ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1121)
then
F ( r , R ) = f ( r 1 , − r 2 ) = ∑ l 1 l 2 L M ( − 1 ) l 2 C l 1 l 2 L M ( r 1 , r 2 ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1122) F(\mathbf{r},\mathbf{R})=f(\mathbf{r}_1,-\mathbf{r}_2)=\sum_{l_1l_2LM}(-1)^{l_2}C^{LM}_{l_1l_2}(r_1,r_2)\{\mathbf{Y}_{l_1}(\Omega_1)\otimes\mathbf{Y}_{l_2}(\Omega_2)\}_{LM} \tag*{(1122)} F ( r , R ) = f ( r 1 , − r 2 ) = l 1 l 2 L M ∑ ( − 1 ) l 2 C l 1 l 2 L M ( r 1 , r 2 ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1122)
Addition Theorem for the Spherical Harmonics
The functions f ( r 1 , r 2 ) f(\mathbf{r}_1,\mathbf{r}_2) f ( r 1 , r 2 ) which are invariant under rotations of coordinate systems play an especially important role in applications. These functions depend only on three variables r 1 = ∣ r 1 ∣ r_1=|\mathbf{r}_1| r 1 = ∣ r 1 ∣ , r 2 = ∣ r 2 ∣ r_2=|\mathbf{r}_2| r 2 = ∣ r 2 ∣ , ( r 1 ⋅ r 2 ) = r 1 r 2 cos ω 12 (\mathbf{r}_1\cdot\mathbf{r}_2)=r_1r_2\cos\omega_{12} ( r 1 ⋅ r 2 ) = r 1 r 2 cos ω 12 or, equivalently, on R = ∣ r 1 + r 2 ∣ R=|\mathbf{r}_1+\mathbf{r}_2| R = ∣ r 1 + r 2 ∣ , r = ∣ r 1 − r 2 ∣ r=|\mathbf{r}_1-\mathbf{r}_2| r = ∣ r 1 − r 2 ∣ and ( R ⋅ r ) = R r cos ω (\mathbf{R}\cdot\mathbf{r})=Rr\cos\omega ( R ⋅ r ) = R r cos ω . The expansions of these functions will evidently contain only the bipolar harmonics of zero rank (L = 0 L=0 L = 0 ), i.e., only scalar products ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (\mathbf{Y}_{l}(\Omega_1)\cdot\mathbf{Y}_{l}(\Omega_2)) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 )) (Eqs. (1105) ). Such expansions are equivalent to the expansions in terms of the Legendre polynomials, which depend on cos ω 12 \cos\omega_{12} cos ω 12 because
( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) = 2 l + 1 4 π P l ( cos ω 12 ) . (1123) (\mathbf{Y}_{l}(\Omega_1)\cdot\mathbf{Y}_{l}(\Omega_2))=\frac{2l+1}{4\pi}P_l(\cos\omega_{12}). \tag*{(1123)} ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 )) = 4 π 2 l + 1 P l ( cos ω 12 ) . (1123)
This relation is called the addition theorem for the spherical harmonics.
Expansions of Some Functions Which Depend on ( r 1 ⋅ r 2 ) (\mathbf{r}_1 \cdot\mathbf{r}_2) ( r 1 ⋅ r 2 )
( r 1 ⋅ r 2 ) = 4 π 3 r 1 r 2 ( Y 1 ( ϑ 1 , φ 1 ) ⋅ Y 1 ( ϑ 2 , φ 2 ) ) (1124) (\mathbf{r}_1 \cdot\mathbf{r}_2)=\frac{4\pi}{3}r_1r_2\left(\mathbf{Y}_1(\vartheta_1,\varphi_1)\cdot\mathbf{Y}_1(\vartheta_2,\varphi_2)\right) \tag*{(1124)} ( r 1 ⋅ r 2 ) = 3 4 π r 1 r 2 ( Y 1 ( ϑ 1 , φ 1 ) ⋅ Y 1 ( ϑ 2 , φ 2 ) ) (1124)
( r 1 ⋅ r 2 ) 2 = 4 π 3 r 1 2 r 2 2 [ ( Y 0 ( ϑ 1 , φ 1 ) ⋅ Y 0 ( ϑ 2 , φ 2 ) ) + 2 5 ( Y 2 ( ϑ 1 , φ 1 ) ⋅ Y 2 ( ϑ 2 , φ 2 ) ) ] (1125) (\mathbf{r}_1 \cdot\mathbf{r}_2)^2=\frac{4\pi}{3}r_1^2r_2^2\left[\left(\mathbf{Y}_0(\vartheta_1,\varphi_1)\cdot\mathbf{Y}_0(\vartheta_2,\varphi_2)\right)+\frac{2}{5}\left(\mathbf{Y}_2(\vartheta_1,\varphi_1)\cdot\mathbf{Y}_2(\vartheta_2,\varphi_2)\right)\right] \tag*{(1125)} ( r 1 ⋅ r 2 ) 2 = 3 4 π r 1 2 r 2 2 [ ( Y 0 ( ϑ 1 , φ 1 ) ⋅ Y 0 ( ϑ 2 , φ 2 ) ) + 5 2 ( Y 2 ( ϑ 1 , φ 1 ) ⋅ Y 2 ( ϑ 2 , φ 2 ) ) ] (1125)
( r 1 ⋅ r 2 ) 3 = 4 π 5 r 1 3 r 2 3 [ ( Y 1 ( ϑ 1 , φ 1 ) ⋅ Y 1 ( ϑ 2 , φ 2 ) ) + 2 7 ( Y 3 ( ϑ 1 , φ 1 ) ⋅ Y 3 ( ϑ 2 , φ 2 ) ) ] (1126) (\mathbf{r}_1 \cdot\mathbf{r}_2)^3=\frac{4\pi}{5}r_1^3r_2^3\left[\left(\mathbf{Y}_1(\vartheta_1,\varphi_1)\cdot\mathbf{Y}_1(\vartheta_2,\varphi_2)\right)+\frac{2}{7}\left(\mathbf{Y}_3(\vartheta_1,\varphi_1)\cdot\mathbf{Y}_3(\vartheta_2,\varphi_2)\right)\right] \tag*{(1126)} ( r 1 ⋅ r 2 ) 3 = 5 4 π r 1 3 r 2 3 [ ( Y 1 ( ϑ 1 , φ 1 ) ⋅ Y 1 ( ϑ 2 , φ 2 ) ) + 7 2 ( Y 3 ( ϑ 1 , φ 1 ) ⋅ Y 3 ( ϑ 2 , φ 2 ) ) ] (1126)
In the general case [48], for integer n n n , one has
( r 1 ⋅ r 2 ) n = 4 π r 1 n r 2 n ∑ l n ! ( n − l ) ! ! ( n + l + 1 ) ! ! ( Y l ( ϑ 1 , φ 1 ) ⋅ Y l ( ϑ 2 , φ 2 ) ) . (1127) (\mathbf{r}_1 \cdot\mathbf{r}_2)^n=4\pi r_1^n r_2^n\sum_l\frac{n!}{(n-l)!!(n+l+1)!!}\left(\mathbf{Y}_l(\vartheta_1,\varphi_1)\cdot\mathbf{Y}_l(\vartheta_2,\varphi_2)\right). \tag*{(1127)} ( r 1 ⋅ r 2 ) n = 4 π r 1 n r 2 n l ∑ ( n − l )!! ( n + l + 1 )!! n ! ( Y l ( ϑ 1 , φ 1 ) ⋅ Y l ( ϑ 2 , φ 2 ) ) . (1127)
In this formula the summation index l l l assumes the values l = 0 , 2 , … , n − 2 , n l=0,2,\ldots,n-2,n l = 0 , 2 , … , n − 2 , n if n n n is even, and l = 1 , 3 , … , n − 2 , n l=1,3,\ldots,n-2,n l = 1 , 3 , … , n − 2 , n if n n n is odd.
The expansion of the exponential function is given by
e i ( r 1 ⋅ r 2 ) = 4 π ∑ l = 0 ∞ i l j l ( r 1 r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1128) e^{i(\mathbf{r}_1\cdot\mathbf{r}_2)}=4\pi\sum_{l=0}^{\infty}i^l j_l(r_1r_2)\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right) \tag*{(1128)} e i ( r 1 ⋅ r 2 ) = 4 π l = 0 ∑ ∞ i l j l ( r 1 r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1128)
where j l ( x ) j_l(x) j l ( x ) is a spherical Bessel function.
An arbitrary function f ( r 1 ⋅ r 2 ) f(\mathbf{r}_1\cdot\mathbf{r}_2) f ( r 1 ⋅ r 2 ) which can be written as a power series
f ( r 1 ⋅ r 2 ) = ∑ n = 0 ∞ c n ( r 1 ⋅ r 2 ) n (1129) f(\mathbf{r}_1\cdot\mathbf{r}_2)=\sum_{n=0}^{\infty}c_n(\mathbf{r}_1\cdot\mathbf{r}_2)^n \tag*{(1129)} f ( r 1 ⋅ r 2 ) = n = 0 ∑ ∞ c n ( r 1 ⋅ r 2 ) n (1129)
may be expanded in terms of the spherical harmonics:
f ( r 1 ⋅ r 2 ) = ∑ l = 0 ∞ f l ( r 1 r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1130) f(\mathbf{r}_1\cdot\mathbf{r}_2)=\sum_{l=0}^{\infty}f_l(r_1r_2)\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right) \tag*{(1130)} f ( r 1 ⋅ r 2 ) = l = 0 ∑ ∞ f l ( r 1 r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1130)
The expansion coefficients are determined by
f l ( r 1 r 2 ) = 4 π ∑ n = l , l + 2 ∞ c n n ! ( r 1 r 2 ) n ( n − 1 ) ! ! ( n + l + 1 ) ! ! (1131) f_l(r_1r_2)=4\pi\sum_{n=l,l+2}^{\infty}c_n\frac{n!(r_1r_2)^n}{(n-1)!!(n+l+1)!!} \tag*{(1131)} f l ( r 1 r 2 ) = 4 π n = l , l + 2 ∑ ∞ c n ( n − 1 )!! ( n + l + 1 )!! n ! ( r 1 r 2 ) n (1131)
Expansions of Some Functions Which Depend on r = ∣ r 1 − r 2 ∣ r=|\mathbf{r}_1-\mathbf{r}_2| r = ∣ r 1 − r 2 ∣
An expansion of the delta function δ ( r 1 − r 2 ) \delta(\mathbf{r}_1-\mathbf{r}_2) δ ( r 1 − r 2 ) is
δ ( r 1 − r 2 ) = δ ( r 1 − r 2 ) r 1 2 ∑ l = 0 ∞ ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1132) \delta(\mathbf{r}_1-\mathbf{r}_2)=\frac{\delta(r_1-r_2)}{r_1^2}\sum_{l=0}^{\infty}\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right) \tag*{(1132)} δ ( r 1 − r 2 ) = r 1 2 δ ( r 1 − r 2 ) l = 0 ∑ ∞ ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1132)
This equation results from the completeness relation for the spherical harmonics (Eqs. (988) ).
An expansion of the Green’s function for the scalar Helmholtz equation may be written in the form
e i k r r = 4 π i k ∑ l = 0 ∞ j l ( k r 1 ) h l ( 1 ) ( k r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) , ( r 1 < r 2 ) (1133) \frac{e^{ikr}}{r}=4\pi ik\sum_{l=0}^{\infty}j_l(kr_1)h_l^{(1)}(kr_2)\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right),\qquad(r_1<r_2) \tag*{(1133)} r e ik r = 4 π ik l = 0 ∑ ∞ j l ( k r 1 ) h l ( 1 ) ( k r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) , ( r 1 < r 2 ) (1133)
where j l ( x ) j_l(x) j l ( x ) and h l ( 1 ) ( x ) h_l^{(1)}(x) h l ( 1 ) ( x ) are the spherical Bessel function and the Hankel function of the first kind; respectively. Equation (5.17.19) is valid at r 1 < r 2 r_1<r_2 r 1 < r 2 ; for r 1 > r 2 r_1>r_2 r 1 > r 2 one should make the replacement r 1 ⇄ r 2 r_1\rightleftarrows r_2 r 1 ⇄ r 2 . Equation (5.17.19) follows from a more general expansion which is given in Ref. [49].
z 0 ( k r ) = 4 π ∑ l = 0 ∞ j l ( k r 1 ) z l ( k r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) , ( r 1 < r 2 ) (1134) z_0(kr)=4\pi\sum_{l=0}^{\infty}j_l(kr_1)z_l(kr_2)\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right),\qquad(r_1<r_2) \tag*{(1134)} z 0 ( k r ) = 4 π l = 0 ∑ ∞ j l ( k r 1 ) z l ( k r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) , ( r 1 < r 2 ) (1134)
Here z l ( x ) = π 2 x Z l + 1 2 ( x ) z_l(x)=\sqrt{\frac{\pi}{2x}}Z_{l+\frac{1}{2}}(x) z l ( x ) = 2 x π Z l + 2 1 ( x ) , and Z l + 1 2 ( x ) Z_{l+\frac{1}{2}}(x) Z l + 2 1 ( x ) is any of the cylinder functions.
When k = 0 k=0 k = 0 Eq. (1133) yields the important expansion of the Green’s function for the scalar Laplace equation
1 r = 4 π r 2 ∑ l = 0 ∞ 1 2 l + 1 ( r 1 r 2 ) l ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) , ( r 1 < r 2 ) (1135) \frac{1}{r}=\frac{4\pi}{r_2}\sum_{l=0}^{\infty}\frac{1}{2l+1}\left(\frac{r_1}{r_2}\right)^l\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right),\qquad(r_1<r_2) \tag*{(1135)} r 1 = r 2 4 π l = 0 ∑ ∞ 2 l + 1 1 ( r 2 r 1 ) l ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) , ( r 1 < r 2 ) (1135)
This result as well as Eq. (1133) are valid when r 1 < r 2 r_1<r_2 r 1 < r 2 , whereas if r 1 > r 2 r_1>r_2 r 1 > r 2 it becomes valid after the interchange r 1 ↔ r 2 r_1\leftrightarrow r_2 r 1 ↔ r 2 . We may also write Eq. (1135) in a symmetric form valid both at r 1 < r 2 r_1<r_2 r 1 < r 2 and at r 1 > r 2 r_1>r_2 r 1 > r 2 :
1 r = 4 π r 1 2 + r 2 2 ∑ l = 0 ∞ { ∑ n = l , l + 2 ∞ ( 2 n − 1 ) ! ! ( n − l ) ! ! ( l + n + 1 ) ! ! ( r 1 r 2 r 1 2 + r 2 2 ) n } ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1136) \frac{1}{r}=\frac{4\pi}{\sqrt{r_1^2+r_2^2}}\sum_{l=0}^{\infty}\left\{\sum_{n=l,l+2}^{\infty}\frac{(2n-1)!!}{(n-l)!!(l+n+1)!!}\left(\frac{r_1r_2}{r_1^2+r_2^2}\right)^n\right\}\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right) \tag*{(1136)} r 1 = r 1 2 + r 2 2 4 π l = 0 ∑ ∞ ⎩ ⎨ ⎧ n = l , l + 2 ∑ ∞ ( n − l )!! ( l + n + 1 )!! ( 2 n − 1 )!! ( r 1 2 + r 2 2 r 1 r 2 ) n ⎭ ⎬ ⎫ ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1136)
Below we give some examples of expansions of r n r^n r n (n = 1 , − 3 , − 5 n=1,-3,-5 n = 1 , − 3 , − 5 )
r = 4 π ∑ l = 0 ∞ 1 2 l + 1 ( r 1 l r 2 l + 1 ) ( r 1 2 2 l + 3 − r 2 2 2 l − 1 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1137) r=4\pi\sum_{l=0}^{\infty}\frac{1}{2l+1}\left(\frac{r_1^l}{r_2^{l+1}}\right)\left(\frac{r_1^2}{2l+3}-\frac{r_2^2}{2l-1}\right)\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right) \tag*{(1137)} r = 4 π l = 0 ∑ ∞ 2 l + 1 1 ( r 2 l + 1 r 1 l ) ( 2 l + 3 r 1 2 − 2 l − 1 r 2 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1137)
1 r 3 = 4 π r 2 2 − r 1 2 ∑ l = 0 ∞ ( r 1 l r 2 l + 1 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1138) \frac{1}{r^3}=\frac{4\pi}{r_2^2-r_1^2}\sum_{l=0}^{\infty}\left(\frac{r_1^l}{r_2^{l+1}}\right)\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right) \tag*{(1138)} r 3 1 = r 2 2 − r 1 2 4 π l = 0 ∑ ∞ ( r 2 l + 1 r 1 l ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1138)
1 r 5 = 4 π 3 ( r 2 2 − r 1 2 ) 3 ∑ l = 0 ∞ ( 2 l − 1 ) ( 2 l + 3 ) ( r 1 l r 2 l + 1 ) ( r 2 2 2 l − 1 − r 1 2 2 l + 3 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1139) \frac{1}{r^5}=\frac{4\pi}{3(r_2^2-r_1^2)^3}\sum_{l=0}^{\infty}(2l-1)(2l+3)\left(\frac{r_1^l}{r_2^{l+1}}\right)\left(\frac{r_2^2}{2l-1}-\frac{r_1^2}{2l+3}\right)\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right) \tag*{(1139)} r 5 1 = 3 ( r 2 2 − r 1 2 ) 3 4 π l = 0 ∑ ∞ ( 2 l − 1 ) ( 2 l + 3 ) ( r 2 l + 1 r 1 l ) ( 2 l − 1 r 2 2 − 2 l + 3 r 1 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) (1139)
These equations are valid when r 1 < r 2 r_1<r_2 r 1 < r 2 , whereas for r 1 > r 2 r_1>r_2 r 1 > r 2 they are valid after interchanging r 1 r_1 r 1 and r 2 r_2 r 2 . Equations (5.17.23)–(5.17.25) follow from the expansions which will be considered below.
Expansions of r n ≡ ∣ r 1 − r 2 ∣ n r^n\equiv|r_1-r_2|^n r n ≡ ∣ r 1 − r 2 ∣ n
Let n n n be an integer. Then the expansion under discussion is as follows
r n = 4 π ∑ l = 0 ∞ 1 2 l + 1 a l n ( r 1 , r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) , (1140) r^n=4\pi\sum_{l=0}^{\infty}\frac{1}{2l+1}a_l^n(r_1,r_2)\left(\mathbf{Y}_l(\Omega_1)\cdot\mathbf{Y}_l(\Omega_2)\right), \tag*{(1140)} r n = 4 π l = 0 ∑ ∞ 2 l + 1 1 a l n ( r 1 , r 2 ) ( Y l ( Ω 1 ) ⋅ Y l ( Ω 2 ) ) , (1140)
where the expansion coefficients a l n ( r 1 , r 2 ) a_l^n(r_1,r_2) a l n ( r 1 , r 2 ) are determined by (see Ref. [50])
a l n ( r 1 , r 2 ) = ( − n 2 ) l ( 1 2 ) l r 2 n ( r 1 r 2 ) l F ( l − n 2 , 1 2 − n 2 ; l + 3 2 ; r 1 2 r 2 2 ) , ( r 1 < r 2 ) , (1141) a_l^n(r_1,r_2)=\frac{\left(-\frac{n}{2}\right)_l}{\left(\frac{1}{2}\right)_l}r_2^n\left(\frac{r_1}{r_2}\right)^lF\left(l-\frac{n}{2},\frac{1}{2}-\frac{n}{2};l+\frac{3}{2};\frac{r_1^2}{r_2^2}\right),\quad(r_1<r_2), \tag*{(1141)} a l n ( r 1 , r 2 ) = ( 2 1 ) l ( − 2 n ) l r 2 n ( r 2 r 1 ) l F ( l − 2 n , 2 1 − 2 n ; l + 2 3 ; r 2 2 r 1 2 ) , ( r 1 < r 2 ) , (1141)
a l n ( r 1 , r 2 ) = ( − n 2 ) l ( 1 2 ) l r 1 l ( r 2 2 − r 1 2 ) n + 2 r 2 l + n + 4 F ( l + 2 + n 2 , 3 2 + n 2 ; l + 3 2 ; r 1 2 r 2 2 ) , ( r 1 < r 2 ) , (1142) a_l^n(r_1,r_2)=\frac{\left(-\frac{n}{2}\right)_l}{\left(\frac{1}{2}\right)_l}\frac{r_1^l(r_2^2-r_1^2)^{n+2}}{r_2^{l+n+4}}F\left(l+2+\frac{n}{2},\frac{3}{2}+\frac{n}{2};l+\frac{3}{2};\frac{r_1^2}{r_2^2}\right),\quad(r_1<r_2), \tag*{(1142)} a l n ( r 1 , r 2 ) = ( 2 1 ) l ( − 2 n ) l r 2 l + n + 4 r 1 l ( r 2 2 − r 1 2 ) n + 2 F ( l + 2 + 2 n , 2 3 + 2 n ; l + 2 3 ; r 2 2 r 1 2 ) , ( r 1 < r 2 ) , (1142)
a l n ( r 1 , r 2 ) = 2 n + 1 ( l + 1 2 ) l ( − n 2 ) l ( 1 + n 2 ) l + 1 r 2 n − l F ( l − n 2 , − 1 2 − n 2 ; − 1 − n ; r 2 2 − r 1 2 r 2 2 ) − 2 l + 1 2 n + 3 ( n + 2 ) ⋅ r 1 l ( r 2 2 − r 1 2 ) n + 2 r 2 l + n + 4 F ( l + n 2 + 2 , 3 2 + n 2 ; n + 3 ; r 2 2 − r 1 2 r 2 2 ) , (1143) \begin{aligned}
a_l^n(r_1,r_2)={}&\frac{2^{n+1}\left(l+\frac{1}{2}\right)_l\left(-\frac{n}{2}\right)_l}{\left(1+\frac{n}{2}\right)_{l+1}r_2^{n-l}F\left(l-\frac{n}{2},-\frac{1}{2}-\frac{n}{2};-1-n;\frac{r_2^2-r_1^2}{r_2^2}\right)}\\
&-\frac{2l+1}{2^{n+3}(n+2)}\cdot\frac{r_1^l(r_2^2-r_1^2)^{n+2}}{r_2^{l+n+4}}F\left(l+\frac{n}{2}+2,\frac{3}{2}+\frac{n}{2};n+3;\frac{r_2^2-r_1^2}{r_2^2}\right), \tag*{(1143)}
\end{aligned} a l n ( r 1 , r 2 ) = ( 1 + 2 n ) l + 1 r 2 n − l F ( l − 2 n , − 2 1 − 2 n ; − 1 − n ; r 2 2 r 2 2 − r 1 2 ) 2 n + 1 ( l + 2 1 ) l ( − 2 n ) l − 2 n + 3 ( n + 2 ) 2 l + 1 ⋅ r 2 l + n + 4 r 1 l ( r 2 2 − r 1 2 ) n + 2 F ( l + 2 n + 2 , 2 3 + 2 n ; n + 3 ; r 2 2 r 2 2 − r 1 2 ) , (1143)
a l n ( r 1 , r 2 ) = ( − n 2 ) l ( − 1 ) n 2 cos n π 2 r 1 − n − l r 2 l F ( l − n 2 , − 1 2 − n 2 ; l + 3 2 ; r 2 2 r 1 2 ) ( 1 2 ) l + Γ ( l + 3 2 ) π ( − 1 ) n + 1 2 r 1 n + l + 1 Γ ( − n 2 ) Γ ( 2 + l + n 2 ) r 2 l + 1 F ( − 1 − l − n 2 , 1 2 − n 2 ; 2 − l ; r 2 2 r 1 2 ) , (1144) \begin{aligned}
a_l^n(r_1,r_2)={}&\frac{\left(-\frac{n}{2}\right)_l(-1)^{\frac{n}{2}}\cos\frac{n\pi}{2}r_1^{-n-l}r_2^lF\left(l-\frac{n}{2},-\frac{1}{2}-\frac{n}{2};l+\frac{3}{2};\frac{r_2^2}{r_1^2}\right)}{\left(\frac{1}{2}\right)_l}\\
&+\frac{\Gamma\left(l+\frac{3}{2}\right)\sqrt{\pi}(-1)^{\frac{n+1}{2}}r_1^{n+l+1}}{\Gamma\left(-\frac{n}{2}\right)\Gamma\left(2+l+\frac{n}{2}\right)r_2^{l+1}}F\left(-1-l-\frac{n}{2},\frac{1}{2}-\frac{n}{2};2-l;\frac{r_2^2}{r_1^2}\right), \tag*{(1144)}
\end{aligned} a l n ( r 1 , r 2 ) = ( 2 1 ) l ( − 2 n ) l ( − 1 ) 2 n cos 2 nπ r 1 − n − l r 2 l F ( l − 2 n , − 2 1 − 2 n ; l + 2 3 ; r 1 2 r 2 2 ) + Γ ( − 2 n ) Γ ( 2 + l + 2 n ) r 2 l + 1 Γ ( l + 2 3 ) π ( − 1 ) 2 n + 1 r 1 n + l + 1 F ( − 1 − l − 2 n , 2 1 − 2 n ; 2 − l ; r 1 2 r 2 2 ) , (1144)
If r 1 > r 2 r_1>r_2 r 1 > r 2 , one should interchange r 1 r_1 r 1 and r 2 r_2 r 2 in Eqs. (1141) –(5.17.30). Let us present also symmetric expressions for a l n ( r 1 , r 2 ) a_l^n(r_1,r_2) a l n ( r 1 , r 2 ) which are valid when r 1 < r 2 r_1<r_2 r 1 < r 2 and r 1 > r 2 r_1>r_2 r 1 > r 2 :
a l n ( r 1 , r 2 ) = ( − n 2 ) l ( 1 2 ) l ( r 1 r 2 ) l ( r 1 2 + r 2 2 ) l − n 2 F ( l 2 − n 4 , l 2 − n 4 + 1 2 ; l + 3 2 ; ( 2 r 1 r 2 r 1 2 + r 2 2 ) 2 ) (1145) a_l^n(r_1,r_2)=\frac{\left(-\frac{n}{2}\right)_l}{\left(\frac{1}{2}\right)_l}\frac{(r_1r_2)^l}{(r_1^2+r_2^2)^{l-\frac{n}{2}}}F\left(\frac{l}{2}-\frac{n}{4},\frac{l}{2}-\frac{n}{4}+\frac{1}{2};l+\frac{3}{2};\left(\frac{2r_1r_2}{r_1^2+r_2^2}\right)^2\right) \tag*{(1145)} a l n ( r 1 , r 2 ) = ( 2 1 ) l ( − 2 n ) l ( r 1 2 + r 2 2 ) l − 2 n ( r 1 r 2 ) l F ( 2 l − 4 n , 2 l − 4 n + 2 1 ; l + 2 3 ; ( r 1 2 + r 2 2 2 r 1 r 2 ) 2 ) (1145)
a l n ( r 1 , r 2 ) = ( − n 2 ) l ( 1 2 ) l ( r 1 r 2 ) l ( r 1 + r 2 ) 2 l − n F ( l − n 2 , 1 + l ; 2 l + 2 ; 4 r 1 r 2 ( r 1 + r 2 ) 2 ) (1146) a_l^n(r_1,r_2)=\frac{\left(-\frac{n}{2}\right)_l}{\left(\frac{1}{2}\right)_l}\frac{(r_1r_2)^l}{(r_1+r_2)^{2l-n}}F\left(l-\frac{n}{2},1+l;2l+2;\frac{4r_1r_2}{(r_1+r_2)^2}\right) \tag*{(1146)} a l n ( r 1 , r 2 ) = ( 2 1 ) l ( − 2 n ) l ( r 1 + r 2 ) 2 l − n ( r 1 r 2 ) l F ( l − 2 n , 1 + l ; 2 l + 2 ; ( r 1 + r 2 ) 2 4 r 1 r 2 ) (1146)
A comparison of Eqs. (1141) and (5.17.28) shows that the expansion coefficient for r n r^n r n and r − n − 4 r^{-n-4} r − n − 4 are related by
a l n ( r 1 , r 2 ) = ( − n 2 ) l ( n 2 + 2 ) l ( r 2 2 − r 1 2 ) n + 2 a l − n − 4 ( r 1 , r 2 ) (1147) a_l^n(r_1,r_2)=\frac{\left(-\frac{n}{2}\right)_l}{\left(\frac{n}{2}+2\right)_l}(r_2^2-r_1^2)^{n+2}a_l^{-n-4}(r_1,r_2) \tag*{(1147)} a l n ( r 1 , r 2 ) = ( 2 n + 2 ) l ( − 2 n ) l ( r 2 2 − r 1 2 ) n + 2 a l − n − 4 ( r 1 , r 2 ) (1147)
Note that all the expansions given in Secs. 5.17.4 and 5.17.5 remain valid after substituting R = ∣ r 1 + r 2 ∣ R=|r_1+r_2| R = ∣ r 1 + r 2 ∣ for r = ∣ r 1 − r 2 ∣ r=|r_1-r_2| r = ∣ r 1 − r 2 ∣ and multiplying the expansion coefficients by ( − 1 ) l (-1)^l ( − 1 ) l .
Expansions of Spherical Waves
The spherical waves have the form z L ( k r ) Y L M ( ϑ , φ ) z_L(kr)Y_{LM}(\vartheta,\varphi) z L ( k r ) Y L M ( ϑ , φ ) , where r ( r , ϑ , φ ) = r 1 − r 2 \mathbf{r}(r,\vartheta,\varphi)=\mathbf{r}_1-\mathbf{r}_2 r ( r , ϑ , φ ) = r 1 − r 2 , z L ( x ) = π 2 x Z L + 1 2 ( x ) z_L(x)=\sqrt{\frac{\pi}{2x}}Z_{L+\frac{1}{2}}(x) z L ( x ) = 2 x π Z L + 2 1 ( x ) and Z L + 1 2 ( x ) Z_{L+\frac{1}{2}}(x) Z L + 2 1 ( x ) is any of the cylinder functions. The spherical waves are expressed in terms of the bipolar spherical harmonics of rank L L L [49] (Sec. 5.16.1)
z L ( k r ) Y L M ( ϑ , φ ) = 4 π 2 L + 1 ∑ l 1 , l 2 = 0 ∞ i l 1 − l 2 − L ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) C l 1 0 l 2 0 L 0 j l 1 ( k r 1 ) z l 2 ( k r 2 ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M , ( r 1 < r 2 ) (1148) z_L(kr)Y_{LM}(\vartheta,\varphi)=\sqrt{\frac{4\pi}{2L+1}}\sum_{l_1,l_2=0}^{\infty}i^{l_1-l_2-L}\sqrt{(2l_1+1)(2l_2+1)}C_{l_1 0 l_2 0}^{L0}j_{l_1}(kr_1)z_{l_2}(kr_2)\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM},\quad(r_1<r_2) \tag*{(1148)} z L ( k r ) Y L M ( ϑ , φ ) = 2 L + 1 4 π l 1 , l 2 = 0 ∑ ∞ i l 1 − l 2 − L ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) C l 1 0 l 2 0 L 0 j l 1 ( k r 1 ) z l 2 ( k r 2 ) { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M , ( r 1 < r 2 ) (1148)
When r 1 > r 2 r_1>r_2 r 1 > r 2 , one should interchange r 1 r_1 r 1 and r 2 r_2 r 2 in Eq. (1148) . If z L ( k r ) = j L ( k r ) z_L(kr)=j_L(kr) z L ( k r ) = j L ( k r ) , j L ( x ) j_L(x) j L ( x ) being a spherical Bessel function, one obtains as k → 0 k\to0 k → 0 [22,96]
r L Y L M ( ϑ , φ ) = 4 π ( 2 L + 1 ) ! ∑ l 1 , l 2 = 0 l 1 + l 2 = L L ( − 1 ) l 2 r 1 l 1 r 2 l 2 ( 2 l 1 + 1 ) ! ( 2 l 2 + 1 ) ! { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1149) r^LY_{LM}(\vartheta,\varphi)=\sqrt{4\pi(2L+1)!}\sum_{\substack{l_1,l_2=0\\l_1+l_2=L}}^L(-1)^{l_2}\frac{r_1^{l_1}r_2^{l_2}}{\sqrt{(2l_1+1)!(2l_2+1)!}}\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM} \tag*{(1149)} r L Y L M ( ϑ , φ ) = 4 π ( 2 L + 1 )! l 1 , l 2 = 0 l 1 + l 2 = L ∑ L ( − 1 ) l 2 ( 2 l 1 + 1 )! ( 2 l 2 + 1 )! r 1 l 1 r 2 l 2 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1149)
Similarly, if z L ( k r ) = n L ( k r ) z_L(kr)=n_L(kr) z L ( k r ) = n L ( k r ) , n L ( k r ) n_L(kr) n L ( k r ) being a spherical Neumann function, one obtains as k → 0 k\to0 k → 0 [22,96]
1 r L + 1 Y L M ( ϑ , φ ) = 4 π ( 2 L ) ! ∑ l 1 , l 2 = 0 l 2 − l 1 = L ∞ ( − 1 ) l 2 ( 2 l 2 ) ! ( 2 l 1 + 1 ) ! r 1 l 1 r 2 l 2 + 1 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M , ( r 1 < r 2 ) (1150) \frac{1}{r^{L+1}}Y_{LM}(\vartheta,\varphi)=\sqrt{\frac{4\pi}{(2L)!}}\sum_{\substack{l_1,l_2=0\\l_2-l_1=L}}^\infty(-1)^{l_2}\sqrt{\frac{(2l_2)!}{(2l_1+1)!}}\frac{r_1^{l_1}}{r_2^{l_2+1}}\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM},\quad(r_1<r_2) \tag*{(1150)} r L + 1 1 Y L M ( ϑ , φ ) = ( 2 L )! 4 π l 1 , l 2 = 0 l 2 − l 1 = L ∑ ∞ ( − 1 ) l 2 ( 2 l 1 + 1 )! ( 2 l 2 )! r 2 l 2 + 1 r 1 l 1 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M , ( r 1 < r 2 ) (1150)
For example, spherical components of r / r 3 \mathbf{r}/r^3 r / r 3 read
r μ r 3 = 4 π ∑ l = 1 ∞ ( − 1 ) l l 3 r 1 l − 1 r 2 l + 1 { Y l − 1 ( Ω 1 ) ⊗ Y l ( Ω 2 ) } 1 μ , ( r 1 < r 2 ) (1151) \frac{r_\mu}{r^3}=4\pi\sum_{l=1}^{\infty}(-1)^l\sqrt{\frac{l}{3}}\frac{r_1^{l-1}}{r_2^{l+1}}\{Y_{l-1}(\Omega_1)\otimes Y_l(\Omega_2)\}_{1\mu},\quad(r_1<r_2) \tag*{(1151)} r 3 r μ = 4 π l = 1 ∑ ∞ ( − 1 ) l 3 l r 2 l + 1 r 1 l − 1 { Y l − 1 ( Ω 1 ) ⊗ Y l ( Ω 2 ) } 1 μ , ( r 1 < r 2 ) (1151)
Expansions of r N Y L M ( ϑ , φ ) r^N Y_{LM}(\vartheta,\varphi) r N Y L M ( ϑ , φ )
The expansion of this function in terms of bipolar harmonics of rank L L L may be written as
r N Y L M ( ϑ , φ ) = 4 π ∑ l 1 , l 2 = 0 ∞ a l 1 l 2 N L ( r 1 , r 2 ) ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) 4 π ( 2 L + 1 ) C l 1 0 l 2 0 L 0 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1152) r^N Y_{LM}(\vartheta,\varphi)=4\pi\sum_{l_1,l_2=0}^{\infty}a_{l_1l_2}^{NL}(r_1,r_2)\sqrt{\frac{(2l_1+1)(2l_2+1)}{4\pi(2L+1)}}C_{l_1 0 l_2 0}^{L0}\{Y_{l_1}(\Omega_1)\otimes Y_{l_2}(\Omega_2)\}_{LM} \tag*{(1152)} r N Y L M ( ϑ , φ ) = 4 π l 1 , l 2 = 0 ∑ ∞ a l 1 l 2 N L ( r 1 , r 2 ) 4 π ( 2 L + 1 ) ( 2 l 1 + 1 ) ( 2 l 2 + 1 ) C l 1 0 l 2 0 L 0 { Y l 1 ( Ω 1 ) ⊗ Y l 2 ( Ω 2 ) } L M (1152)
The sum is over all positive integer l 1 l_1 l 1 and l 2 l_2 l 2 allowed by the momentum addition rules and the parity selection rule (l 1 + l 2 − L l_1+l_2-L l 1 + l 2 − L is even). The expansion coefficients a l 1 l 2 N L ( r 1 , r 2 ) a_{l_1l_2}^{NL}(r_1,r_2) a l 1 l 2 N L ( r 1 , r 2 ) are given by (see Ref. [51])
a l 1 l 2 N L ( r 1 , r 2 ) = ( − 1 ) l 1 − l 2 − L 2 2 l 1 ( 2 l 1 + 1 ) ! ! Γ ( l 1 + l 2 − N 2 ) Γ ( L + N + 3 2 ) Γ ( L − N 2 ) Γ ( − l 1 + l 2 + N + 3 2 ) × r 2 N ( r 1 r 2 ) l 1 F ( l 1 + l 2 − N 2 , l 1 − l 2 − N − 1 2 ; l 1 + 3 2 ; r 1 2 r 2 2 ) , ( r 1 < r 2 ) . (1153) a_{l_1l_2}^{NL}(r_1,r_2)=(-1)^{\frac{l_1-l_2-L}{2}}\frac{2^{l_1}}{(2l_1+1)!!}\frac{\Gamma\left(\frac{l_1+l_2-N}{2}\right)\Gamma\left(\frac{L+N+3}{2}\right)}{\Gamma\left(\frac{L-N}{2}\right)\Gamma\left(\frac{-l_1+l_2+N+3}{2}\right)}
\times r_2^N\left(\frac{r_1}{r_2}\right)^{l_1}F\left(\frac{l_1+l_2-N}{2},\frac{l_1-l_2-N-1}{2};l_1+\frac{3}{2};\frac{r_1^2}{r_2^2}\right),\quad(r_1<r_2). \tag*{(1153)} a l 1 l 2 N L ( r 1 , r 2 ) = ( − 1 ) 2 l 1 − l 2 − L ( 2 l 1 + 1 )!! 2 l 1 Γ ( 2 L − N ) Γ ( 2 − l 1 + l 2 + N + 3 ) Γ ( 2 l 1 + l 2 − N ) Γ ( 2 L + N + 3 ) × r 2 N ( r 2 r 1 ) l 1 F ( 2 l 1 + l 2 − N , 2 l 1 − l 2 − N − 1 ; l 1 + 2 3 ; r 2 2 r 1 2 ) , ( r 1 < r 2 ) . (1153)
If r 1 > r 2 r_1>r_2 r 1 > r 2 , the expression for the expansion coefficients may be obtained from Eq. (1153) by interchanging r 1 r_1 r 1 and r 2 r_2 r 2 , l 1 l_1 l 1 and l 2 l_2 l 2 and introducing the additional factor ( − 1 ) L (-1)^L ( − 1 ) L . The expression (5.17.39) was studied in detail in Ref. [52]. For different values of N N N and L L L the expansion coefficients are determined by the following relations. For N = L , L + 2 , L + 4 , … N=L,L+2,L+4,\ldots N = L , L + 2 , L + 4 , … we have
a l 1 l 2 N L ( r 1 , r 2 ) = { ( − 1 ) l 2 ( N + L + 1 ) ! ! ( N − L ) ! ! ∑ e 1 , e 2 = 0 ∞ r 1 l 1 + 2 e 1 r 2 l 2 + 2 e 2 ( 2 e 1 ) ! ! ( 2 e 1 + 2 l 1 + 1 ) ! ! ( 2 e 2 ) ! ! ( 2 e 2 + 2 l 2 + 1 ) ! ! , e 1 + e 2 = N − l 1 − l 2 2 if l 1 + l 2 ≤ N , 0 , if l 1 + l 2 > N . (1154) a_{l_1l_2}^{NL}(r_1,r_2)=
\begin{cases}
(-1)^{l_2}(N+L+1)!!(N-L)!!\sum_{e_1,e_2=0}^{\infty}\frac{r_1^{l_1+2e_1}r_2^{l_2+2e_2}}{(2e_1)!!(2e_1+2l_1+1)!!(2e_2)!!(2e_2+2l_2+1)!!}, & e_1+e_2=\frac{N-l_1-l_2}{2}\ \text{if }l_1+l_2\le N,\\
0, & \text{if }l_1+l_2>N.
\end{cases} \tag*{(1154)} a l 1 l 2 N L ( r 1 , r 2 ) = { ( − 1 ) l 2 ( N + L + 1 )!! ( N − L )!! ∑ e 1 , e 2 = 0 ∞ ( 2 e 1 )!! ( 2 e 1 + 2 l 1 + 1 )!! ( 2 e 2 )!! ( 2 e 2 + 2 l 2 + 1 )!! r 1 l 1 + 2 e 1 r 2 l 2 + 2 e 2 , 0 , e 1 + e 2 = 2 N − l 1 − l 2 if l 1 + l 2 ≤ N , if l 1 + l 2 > N . (1154)
For N = L − 2 , L − 4 , L − 6 , … N=L-2,L-4,L-6,\ldots N = L − 2 , L − 4 , L − 6 , … one gets
a l 1 l 2 N L ( r 1 , r 2 ) = ( − 1 ) l 2 + N − 2 2 ( N + L + 1 ) ! ! 2 ( L − N − 2 ) ! ! ( N + l 1 + l 2 + 2 ) ! r 1 l 1 r 2 l 2 × ( 1 r 1 ⋅ ∂ ∂ r 1 ) l 1 ( 1 r 2 ⋅ ∂ ∂ r 2 ) l 2 1 r 1 r 2 [ ( r 1 − r 2 ) N + l 1 + l 2 + 2 ln ∣ r 1 − r 2 ∣ − ( r 1 + r 2 ) N + l 1 + l 2 + 2 ln ( r 1 + r 2 ) ] . (1155) a_{l_1l_2}^{NL}(r_1,r_2)=(-1)^{l_2+\frac{N-2}{2}}\frac{(N+L+1)!!}{2(L-N-2)!!(N+l_1+l_2+2)!}r_1^{l_1}r_2^{l_2}
\times\left(\frac{1}{r_1}\cdot\frac{\partial}{\partial r_1}\right)^{l_1}\left(\frac{1}{r_2}\cdot\frac{\partial}{\partial r_2}\right)^{l_2}\frac{1}{r_1r_2}\left[(r_1-r_2)^{N+l_1+l_2+2}\ln|r_1-r_2|-(r_1+r_2)^{N+l_1+l_2+2}\ln(r_1+r_2)\right]. \tag*{(1155)} a l 1 l 2 N L ( r 1 , r 2 ) = ( − 1 ) l 2 + 2 N − 2 2 ( L − N − 2 )!! ( N + l 1 + l 2 + 2 )! ( N + L + 1 )!! r 1 l 1 r 2 l 2 × ( r 1 1 ⋅ ∂ r 1 ∂ ) l 1 ( r 2 1 ⋅ ∂ r 2 ∂ ) l 2 r 1 r 2 1 [ ( r 1 − r 2 ) N + l 1 + l 2 + 2 ln ∣ r 1 − r 2 ∣ − ( r 1 + r 2 ) N + l 1 + l 2 + 2 ln ( r 1 + r 2 ) ] . (1155)
If N N N is integer, N − L N-L N − L is odd, and r 1 < r 2 r_1<r_2 r 1 < r 2 , the expansion coefficient is given by
a l 1 l 2 N L ( r 1 , r 2 ) = { ( − 1 ) N + l 2 + 1 + l 1 ( N + L + 1 ) ! ! ( L − N − 2 ) ! ! r 1 N − l 1 r 2 N − l 1 + l 2 + 1 ∑ e = 0 ( N − l 1 + l 2 + 1 ) / 2 ( r 1 / r 2 ) 2 e ( 2 e ) ! ! ( N − l 1 − l 2 − 2 e ) ! ! ( N − l 1 + l 2 + 1 − 2 e ) ! ! ( 2 l 1 + 2 e + 1 ) ! ! , 0 if N − l 1 + l 2 + 1 ≥ 0 , if N − l 1 + l 2 + 1 < 0. (1156) a^{NL}_{l_1l_2}(r_1,r_2)=
\begin{cases}
(-1)^{N+l_2+1+l_1}\dfrac{(N+L+1)!!}{(L-N-2)!!}r_1^{N-l_1}r_2^{N-l_1+l_2+1}\sum_{e=0}^{(N-l_1+l_2+1)/2}\dfrac{(r_1/r_2)^{2e}}{(2e)!!(N-l_1-l_2-2e)!!(N-l_1+l_2+1-2e)!!(2l_1+2e+1)!!},\\
0\quad\text{if }N-l_1+l_2+1\geq0,\\
\quad\text{if }N-l_1+l_2+1<0.
\end{cases} \tag*{(1156)} a l 1 l 2 N L ( r 1 , r 2 ) = ⎩ ⎨ ⎧ ( − 1 ) N + l 2 + 1 + l 1 ( L − N − 2 )!! ( N + L + 1 )!! r 1 N − l 1 r 2 N − l 1 + l 2 + 1 ∑ e = 0 ( N − l 1 + l 2 + 1 ) /2 ( 2 e )!! ( N − l 1 − l 2 − 2 e )!! ( N − l 1 + l 2 + 1 − 2 e )!! ( 2 l 1 + 2 e + 1 )!! ( r 1 / r 2 ) 2 e , 0 if N − l 1 + l 2 + 1 ≥ 0 , if N − l 1 + l 2 + 1 < 0. (1156)
The expansion coefficient for r 1 > r 2 r_1>r_2 r 1 > r 2 may be obtained from (5.17.42) by interchanging r 1 r_1 r 1 and r 2 r_2 r 2 , l 1 l_1 l 1 and l 2 l_2 l 2 and introducing the additional factor ( − 1 ) L (-1)^L ( − 1 ) L . It should be emphasised that if N ≤ − 3 N\leq-3 N ≤ − 3 the expansion coefficients contain additional terms involving the delta function δ ( r 1 − r 2 ) \delta(r_1-r_2) δ ( r 1 − r 2 ) and its derivatives [52]
a l 1 l 2 N L ( r 1 , r 2 ) δ = ( − 1 ) N + l 1 − 1 ( N + L + 1 ) ! ! ( L − N − 2 ) ! ! ∑ e 1 = 0 l 1 ∑ e 2 = 0 e 1 + e 2 + N ≤ − 3 l 2 ( − 1 ) e 2 ( l 1 + s 1 ) ! ( l 2 + s 2 ) ! ( l 1 − s 1 ) ! ( l 2 − s 2 ) ! a^{NL}_{l_1l_2}(r_1,r_2)_\delta=(-1)^{N+l_1-1}\frac{(N+L+1)!!}{(L-N-2)!!}\sum_{e_1=0}^{l_1}\sum_{\substack{e_2=0\\e_1+e_2+N\leq-3}}^{l_2}(-1)^{e_2}\frac{(l_1+s_1)!(l_2+s_2)!}{(l_1-s_1)!(l_2-s_2)!} a l 1 l 2 N L ( r 1 , r 2 ) δ = ( − 1 ) N + l 1 − 1 ( L − N − 2 )!! ( N + L + 1 )!! e 1 = 0 ∑ l 1 e 2 = 0 e 1 + e 2 + N ≤ − 3 ∑ l 2 ( − 1 ) e 2 ( l 1 − s 1 )! ( l 2 − s 2 )! ( l 1 + s 1 )! ( l 2 + s 2 )!
× 1 ( 2 s 1 ) ! ! ( 2 s 2 ) ! ! 1 r 2 β 2 + 1 ( ∂ ∂ r 2 ) − N − s 1 − e 2 − 3 [ δ ( r 1 − r 2 ) r 1 2 β 1 r 2 β 1 − 1 ] . (1157) {}\times\frac{1}{(2s_1)!!(2s_2)!!}\frac{1}{r_2^{\beta_2+1}}\left(\frac{\partial}{\partial r_2}\right)^{-N-s_1-e_2-3}\left[\frac{\delta(r_1-r_2)}{r_1^{2\beta_1}r_2^{\beta_1-1}}\right]. \tag*{(1157)} × ( 2 s 1 )!! ( 2 s 2 )!! 1 r 2 β 2 + 1 1 ( ∂ r 2 ∂ ) − N − s 1 − e 2 − 3 [ r 1 2 β 1 r 2 β 1 − 1 δ ( r 1 − r 2 ) ] . (1157)
These additional terms give finite contributions into the integrals involving the expansion (5.17.38). Finally, if N N N is not integer, the expansion coefficient is given by
a l 1 l 2 N L ( r 1 , r 2 ) = ( − 1 ) l 1 + L + 1 2 L − 1 Γ ( N + L + 3 2 ) Γ ( N − L + 2 ) Γ ( N − L + 3 2 ) Γ ( N + l 1 + l 2 + 3 ) r 1 l 1 r 2 l 2 a^{NL}_{l_1l_2}(r_1,r_2)=(-1)^{l_1+L+1}2^{L-1}\frac{\Gamma\left(\frac{N+L+3}{2}\right)\Gamma(N-L+2)}{\Gamma\left(\frac{N-L+3}{2}\right)\Gamma(N+l_1+l_2+3)}r_1^{l_1}r_2^{l_2} a l 1 l 2 N L ( r 1 , r 2 ) = ( − 1 ) l 1 + L + 1 2 L − 1 Γ ( 2 N − L + 3 ) Γ ( N + l 1 + l 2 + 3 ) Γ ( 2 N + L + 3 ) Γ ( N − L + 2 ) r 1 l 1 r 2 l 2
× ( 1 r 1 ∂ ∂ r 1 ) l 1 ( 1 r 2 ∂ ∂ r 2 ) l 2 1 r 1 r 2 { ∣ r 1 + r 2 ∣ N + l 1 + l 2 + 2 − ∣ r 1 − r 2 ∣ N + l 1 + l 2 + 2 } . (1158) {}\times\left(\frac{1}{r_1}\frac{\partial}{\partial r_1}\right)^{l_1}\left(\frac{1}{r_2}\frac{\partial}{\partial r_2}\right)^{l_2}\frac{1}{r_1r_2}\left\{\lvert r_1+r_2\rvert^{N+l_1+l_2+2}-\lvert r_1-r_2\rvert^{N+l_1+l_2+2}\right\}. \tag*{(1158)} × ( r 1 1 ∂ r 1 ∂ ) l 1 ( r 2 1 ∂ r 2 ∂ ) l 2 r 1 r 2 1 { ∣ r 1 + r 2 ∣ N + l 1 + l 2 + 2 − ∣ r 1 − r 2 ∣ N + l 1 + l 2 + 2 } . (1158)
Note that all the expansions given in § 24.10.6 and 5.17.7 remain valid after substituting R ( R , Θ , Φ ) = r 1 + r 2 R(R,\Theta,\Phi)=r_1+r_2 R ( R , Θ , Φ ) = r 1 + r 2 for r ( r , ϑ , ϕ ) = r 1 − r 2 r(r,\vartheta,\phi)=r_1-r_2 r ( r , ϑ , ϕ ) = r 1 − r 2 and introducing the additional factor ( − 1 ) l 2 (-1)^{l_2} ( − 1 ) l 2 into the expansion coefficients.
Spin functions
Spin functions of particles with arbitrary spin
Definition
Spin functions describe polarization states of particles1 of definite spin, i.e. definite intrinsic angular momentum. The spin functions χ ( σ ) \chi(\sigma) χ ( σ ) may be treated as functions of a discrete variable σ \sigma σ which is the spin projection on the z z z -axis. The variable σ \sigma σ takes 2 S + 1 2S+1 2 S + 1 values, σ = − S , − S + 1 , … , S − 1 , S \sigma=-S,-S+1,\ldots,S-1,S σ = − S , − S + 1 , … , S − 1 , S , where S S S is the particle spin (integer or half integer non-negative number). The quantity ∣ χ ( σ ) ∣ 2 \lvert\chi(\sigma)\rvert^2 ∣ χ ( σ ) ∣ 2 gives the probability that the spin projection on the z z z -axis in a given state equals σ \sigma σ .
The spin functions are commonly written as column matrices which contain 2 S + 1 2S+1 2 S + 1 elements:
χ = ( χ ( S ) χ ( S − 1 ) ⋮ χ ( − S + 1 ) χ ( − S ) ) . (1159) \chi=
\begin{pmatrix}
\chi(S)\\
\chi(S-1)\\
\vdots\\
\chi(-S+1)\\
\chi(-S)
\end{pmatrix}. \tag*{(1159)} χ = χ ( S ) χ ( S − 1 ) ⋮ χ ( − S + 1 ) χ ( − S ) . (1159)
The elements of this matrix give the values of the spin function χ ( σ ) \chi(\sigma) χ ( σ ) for corresponding values of spin variable σ \sigma σ . The quantities χ ( σ ) \chi(\sigma) χ ( σ ) are called the contravariant components of the spin function χ \chi χ (see § 25.1.4 below).
In such a representation operators acting on the spin variables take the form of square ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrices, and summation over the spin variable is replaced by matrix multiplication. The Hermitian conjugate function χ † \chi^\dagger χ † has the form of a row matrix
χ † = ( χ ∗ ( S ) , χ ∗ ( S − 1 ) , … , χ ∗ ( − S + 1 ) , χ ∗ ( − S ) ) . (1160) \chi^\dagger= \left(\chi^*(S),\chi^*(S-1),\ldots,\chi^*(-S+1),\chi^*(-S)\right). \tag*{(1160)} χ † = ( χ ∗ ( S ) , χ ∗ ( S − 1 ) , … , χ ∗ ( − S + 1 ) , χ ∗ ( − S ) ) . (1160)
The interpretation of ∣ χ ( σ ) ∣ 2 \lvert\chi(\sigma)\rvert^2 ∣ χ ( σ ) ∣ 2 as the probability for the spin projection on the z z z -axis to be equal to σ \sigma σ is possible only if χ ( σ ) \chi(\sigma) χ ( σ ) satisfies the normalization condition
∑ σ = − S S ∣ χ ( σ ) ∣ 2 = 1. (1161) \sum_{\sigma=-S}^{S}\lvert\chi(\sigma)\rvert^2=1. \tag*{(1161)} σ = − S ∑ S ∣ χ ( σ ) ∣ 2 = 1. (1161)
χ † χ = 1. (1162) \chi^\dagger\chi=1. \tag*{(1162)} χ † χ = 1. (1162)
The above representation of spin functions is called the spherical basis representation. When the spin value is integer, one may also use the cartesian basis representation. The latter representation is considered below (§ 25.4 ) for spin-1 particles.
Basis Spin Functions
According to definition, the basis spin functions describe the states with definite spin and spin projection on the z z z -axis. The basis spin functions are eigenfunctions of the operators S ^ 2 \hat{S}^2 S ^ 2 and S ^ z \hat{S}_z S ^ z , where S ^ \hat{S} S ^ is the spin operator (see § 10.3 )
S ^ 2 χ S m = S ( S + 1 ) χ S m , S ^ z χ S m = m χ S m . (1163) \hat{S}^2\chi_{Sm}=S(S+1)\chi_{Sm},\quad\hat{S}_z\chi_{Sm}=m\chi_{Sm}. \tag*{(1163)} S ^ 2 χ S m = S ( S + 1 ) χ S m , S ^ z χ S m = m χ S m . (1163)
As follows from the definition, the dependence of the basis spin functions χ S m ( σ ) \chi_{Sm}(\sigma) χ S m ( σ ) on the spin variable σ \sigma σ is given by
χ S m ( σ ) = δ m σ . (1164) \chi_{Sm}(\sigma)=\delta_{m\sigma}. \tag*{(1164)} χ S m ( σ ) = δ mσ . (1164)
In other words, contravariant components of the basis spin functions χ S m \chi_{Sm} χ S m have the form
[ χ S m ] σ = δ m σ . (1165) [\chi_{Sm}]^{\sigma}=\delta_{m\sigma}. \tag*{(1165)} [ χ S m ] σ = δ mσ . (1165)
If the basis spin functions are written as column matrices, then
χ S S = ( 1 0 ⋮ 0 0 ) , χ S S − 1 = ( 0 1 ⋮ 0 0 ) , … , χ S − S = ( 0 0 ⋮ 0 1 ) . (1166) \chi_{SS}=\begin{pmatrix}1\\0\\\vdots\\0\\0\end{pmatrix},\quad
\chi_{S\,S-1}=\begin{pmatrix}0\\1\\\vdots\\0\\0\end{pmatrix},\quad\ldots,\quad
\chi_{S\,-S}=\begin{pmatrix}0\\0\\\vdots\\0\\1\end{pmatrix}. \tag*{(1166)} χ S S = 1 0 ⋮ 0 0 , χ S S − 1 = 0 1 ⋮ 0 0 , … , χ S − S = 0 0 ⋮ 0 1 . (1166)
The collection of 2 S + 1 2S+1 2 S + 1 basis functions χ S m ( m = S , S − 1 , … , − S ) \chi_{Sm}(m=S,S-1,\ldots,-S) χ S m ( m = S , S − 1 , … , − S ) constitutes a complete orthonormal set of functions. The orthonormality condition for the basis functions is
χ S m † χ S m ′ = δ m m ′ . (1167) \chi_{Sm}^{\dagger}\chi_{Sm'}=\delta_{mm'}. \tag*{(1167)} χ S m † χ S m ′ = δ m m ′ . (1167)
The completeness condition for the basis spin functions may be written in matrix form
∑ m = − S S χ S m χ S m † = I ^ . (1168) \sum_{m=-S}^{S}\chi_{Sm}\chi_{Sm}^{\dagger}=\hat{I}. \tag*{(1168)} m = − S ∑ S χ S m χ S m † = I ^ . (1168)
where I ^ \hat{I} I ^ is the unit ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrix.
Matrix products χ S m χ S m ′ † \chi_{Sm}\chi_{Sm'}^{\dagger} χ S m χ S m ′ † of the basis spin functions are the square ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrices. They may be expanded in terms of the polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) (see 2.4),
χ S m χ S m ′ † = ∑ L 2 L + 1 2 S + 1 C S m ′ L M S m T ^ L M ( S ) . (1169) \chi_{Sm}\chi_{Sm'}^{\dagger}=\sum_{L}\sqrt{\frac{2L+1}{2S+1}}C^{Sm}_{Sm'\,LM}\hat{T}_{LM}(S). \tag*{(1169)} χ S m χ S m ′ † = L ∑ 2 S + 1 2 L + 1 C S m ′ L M S m T ^ L M ( S ) . (1169)
Matrix elements of spherical components of the spin operator S ^ μ \hat{S}_{\mu} S ^ μ (μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ) between the basis spin functions are given by
χ S m ′ † S ^ μ χ S m = S ( S + 1 ) C S m 1 μ S m ′ ; (1170) \chi_{Sm'}^{\dagger}\hat{S}_{\mu}\chi_{Sm}=\sqrt{S(S+1)}C^{Sm'}_{Sm\,1\mu}; \tag*{(1170)} χ S m ′ † S ^ μ χ S m = S ( S + 1 ) C S m 1 μ S m ′ ; (1170)
The only non-vanishing matrix elements are the following:
χ S m + 1 † S ^ + 1 χ S m = − 1 2 ( S − m ) ( S + m + 1 ) , \chi_{S\,m+1}^{\dagger}\hat{S}_{+1}\chi_{Sm}=-\frac{1}{\sqrt{2}}\sqrt{(S-m)(S+m+1)}, χ S m + 1 † S ^ + 1 χ S m = − 2 1 ( S − m ) ( S + m + 1 ) ,
χ S m † S ^ 0 χ S m = m , (1171) \chi_{Sm}^{\dagger}\hat{S}_{0}\chi_{Sm}=m, \tag*{(1171)} χ S m † S ^ 0 χ S m = m , (1171)
χ S m − 1 † S ^ − 1 χ S m = 1 2 ( S + m ) ( S − m + 1 ) . \chi_{S\,m-1}^{\dagger}\hat{S}_{-1}\chi_{Sm}=\frac{1}{\sqrt{2}}\sqrt{(S+m)(S-m+1)}. χ S m − 1 † S ^ − 1 χ S m = 2 1 ( S + m ) ( S − m + 1 ) .
For cartesian components of the spin operator S ^ i \hat{S}_{i} S ^ i (i = x , y , z i=x,y,z i = x , y , z ) the non-vanishing matrix elements are
χ S m ± 1 † S ^ x χ S m = 1 2 ( S ∓ m ) ( S ± m + 1 ) . \chi_{S\,m\pm1}^{\dagger}\hat{S}_{x}\chi_{Sm}=\frac{1}{2}\sqrt{(S\mp m)(S\pm m+1)}. χ S m ± 1 † S ^ x χ S m = 2 1 ( S ∓ m ) ( S ± m + 1 ) .
χ S m ± 1 † S ^ y χ S m = ∓ i 2 ( S ∓ m ) ( S ± m + 1 ) . (1172) \chi_{S\,m\pm1}^{\dagger}\hat{S}_{y}\chi_{Sm}=\mp\frac{i}{2}\sqrt{(S\mp m)(S\pm m+1)}. \tag*{(1172)} χ S m ± 1 † S ^ y χ S m = ∓ 2 i ( S ∓ m ) ( S ± m + 1 ) . (1172)
χ S m † S ^ z χ S m = m . \chi_{Sm}^{\dagger}\hat{S}_{z}\chi_{Sm}=m. χ S m † S ^ z χ S m = m .
The matrix elements of the polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) between the basis spin functions may be written as
χ S m ′ † T ^ L M ( S ) χ S m = 2 L + 1 2 S + 1 C S m L M S m ′ . (1173) \chi_{Sm'}^{\dagger}\hat{T}_{LM}(S)\chi_{Sm}=\sqrt{\frac{2L+1}{2S+1}}C^{Sm'}_{Sm\,LM}. \tag*{(1173)} χ S m ′ † T ^ L M ( S ) χ S m = 2 S + 1 2 L + 1 C S m L M S m ′ . (1173)
Under rotations of the coordinate system the basis spin functions are transformed either by the rotation operator D ^ S ( α , β , γ ) \hat{D}^{S}(\alpha,\beta,\gamma) D ^ S ( α , β , γ ) (Eq. (287) ), if rotations are defined by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ , or by the rotation operator U ^ S ( ω ; θ , Φ ) \hat{U}^{S}(\omega;\theta,\Phi) U ^ S ( ω ; θ , Φ ) (Eq. (288) ) if rotations are described by the rotation axis n ( θ , Φ ) \mathbf{n}(\theta,\Phi) n ( θ , Φ ) and the rotation angle ω \omega ω
χ S m ′ ′ ≡ D ^ S ( α , β , γ ) χ S m ′ = ∑ m D m m ′ S ( α , β , γ ) χ S m , \chi_{Sm'}' \equiv\hat{D}^{S}(\alpha,\beta,\gamma)\chi_{Sm'}=\sum_{m}D^{S}_{mm'}(\alpha,\beta,\gamma)\chi_{Sm}, χ S m ′ ′ ≡ D ^ S ( α , β , γ ) χ S m ′ = m ∑ D m m ′ S ( α , β , γ ) χ S m ,
χ S m ′ ′ ≡ U ^ S ( ω ; Θ , Φ ) χ S m ′ = ∑ m U m m ′ S ( ω ; Θ , Φ ) χ S m . (1174) \chi_{Sm'}' \equiv\hat{U}^{S}(\omega;\Theta,\Phi)\chi_{Sm'}=\sum_{m}U^{S}_{mm'}(\omega;\Theta,\Phi)\chi_{Sm}. \tag*{(1174)} χ S m ′ ′ ≡ U ^ S ( ω ; Θ , Φ ) χ S m ′ = m ∑ U m m ′ S ( ω ; Θ , Φ ) χ S m . (1174)
where D m m ′ S D^{S}_{mm'} D m m ′ S are the Wigner D D D -functions (Chap. 4), and U m m ′ S U^{S}_{mm'} U m m ′ S are the functions defined in Sec. 4.5. The functions χ S m ′ \chi_{Sm'} χ S m ′ describe quantum states with definite spin S S S and spin projection m ′ m' m ′ on the new z ′ z' z ′ -axis. They are eigenfunctions of the operators S ^ ′ 2 \hat{S}'^{2} S ^ ′ 2 and S ^ z ′ \hat{S}'_{z} S ^ z ′ ,
S ^ ′ 2 χ S m ′ = S ( S + 1 ) χ S m ′ , S ^ z ′ χ S m ′ = m ′ χ S m ′ . (1175) \hat{S}'^{2}\chi_{Sm'}=S(S+1)\chi_{Sm'}, \qquad
\hat{S}'_{z}\chi_{Sm'}=m'\chi_{Sm'}. \tag*{(1175)} S ^ ′ 2 χ S m ′ = S ( S + 1 ) χ S m ′ , S ^ z ′ χ S m ′ = m ′ χ S m ′ . (1175)
Here S ^ ′ \hat{\mathbf{S}}' S ^ ′ is the spin operator in the rotated coordinate system
S ^ μ ′ = ∑ ν D ν μ 1 ( α , β , γ ) S ^ ν , ( μ , ν = ± 1 , 0 ) . (1176) \hat{S}'_{\mu}=\sum_{\nu}D^{1}_{\nu\mu}(\alpha,\beta,\gamma)\hat{S}_{\nu}, \qquad(\mu,\nu=\pm1,0). \tag*{(1176)} S ^ μ ′ = ν ∑ D ν μ 1 ( α , β , γ ) S ^ ν , ( μ , ν = ± 1 , 0 ) . (1176)
Helicity Basis Functions
The spin projection on the linear momentum direction of a particle is called the helicity. For particles of spin S S S the helicity λ \lambda λ assumes ( 2 S + 1 ) (2S+1) ( 2 S + 1 ) values, λ = S , S − 1 , … , − S + 1 , S \lambda=S,S-1,\ldots,-S+1,S λ = S , S − 1 , … , − S + 1 , S .
The helicity basis functions χ S λ ( ϑ , φ ) \chi_{S\lambda}(\vartheta,\varphi) χ S λ ( ϑ , φ ) , by definition, are the eigenfunctions of the operators S ^ 2 \hat{\mathbf{S}}^{2} S ^ 2 and S ^ ⋅ n \hat{\mathbf{S}}\cdot\mathbf{n} S ^ ⋅ n where S ^ \hat{\mathbf{S}} S ^ is the spin operator, n = p / p \mathbf{n}=\mathbf{p}/p n = p / p is the momentum direction of the particles, ϑ \vartheta ϑ and φ \varphi φ are the polar angles of the vector n \mathbf{n} n ,
S ^ 2 χ S λ ( ϑ , φ ) = S ( S + 1 ) χ S λ ( ϑ , φ ) , S ^ ⋅ n χ S λ ( ϑ , φ ) = λ χ S λ ( ϑ , φ ) . (1177) \hat{\mathbf{S}}^{2}\chi_{S\lambda}(\vartheta,\varphi)=S(S+1)\chi_{S\lambda}(\vartheta,\varphi),
\qquad
\hat{\mathbf{S}}\cdot\mathbf{n}\chi_{S\lambda}(\vartheta,\varphi)=\lambda\chi_{S\lambda}(\vartheta,\varphi). \tag*{(1177)} S ^ 2 χ S λ ( ϑ , φ ) = S ( S + 1 ) χ S λ ( ϑ , φ ) , S ^ ⋅ n χ S λ ( ϑ , φ ) = λ χ S λ ( ϑ , φ ) . (1177)
The function χ S λ ( ϑ , φ ) \chi_{S\lambda}(\vartheta,\varphi) χ S λ ( ϑ , φ ) describes a quantum state with definite spin S S S and helicity λ \lambda λ .
The helicity basis functions χ S λ ( ϑ , φ ) \chi_{S\lambda}(\vartheta,\varphi) χ S λ ( ϑ , φ ) may be expressed in terms of the basis functions χ S m \chi_{Sm} χ S m by means of the rotation which turns the z z z -axis parallel to n ( ϑ , φ ) \mathbf{n}(\vartheta,\varphi) n ( ϑ , φ ) . Such a rotation is determined by the Euler angles α = φ \alpha=\varphi α = φ , β = ϑ \beta=\vartheta β = ϑ . The third Euler angle γ \gamma γ (the angle of rotation about the new z ′ z' z ′ -axis) is arbitrary. For simplicity, let us adopt γ = 0 \gamma=0 γ = 0 . In this case the rotated cartesian basis vectors e x ′ , e y ′ , e z ′ \mathbf{e}'_x,\mathbf{e}'_y,\mathbf{e}'_z e x ′ , e y ′ , e z ′ will coincide with the polar basis vectors e ϑ , e φ , e ρ \mathbf{e}_{\vartheta},\mathbf{e}_{\varphi},\mathbf{e}_{\rho} e ϑ , e φ , e ρ , respectively (§ 9.1.2 ). Making use of the transformation properties of the basis spin functions χ S m \chi_{Sm} χ S m under the rotation ((6.1.16)) we obtain
χ S λ ( ϑ , φ ) = ∑ m D m λ S ( φ , ϑ , 0 ) χ S m , \chi_{S\lambda}(\vartheta,\varphi)=\sum_m D^{S}_{m\lambda}(\varphi,\vartheta,0)\chi_{Sm}, χ S λ ( ϑ , φ ) = m ∑ D mλ S ( φ , ϑ , 0 ) χ S m ,
χ S m = ∑ λ D − λ − m S ( 0 , ϑ , φ ) χ S λ ( ϑ , φ ) . (1178) \chi_{Sm}=\sum_{\lambda}D^{S}_{-\lambda-m}(0,\vartheta,\varphi)\chi_{S\lambda}(\vartheta,\varphi). \tag*{(1178)} χ S m = λ ∑ D − λ − m S ( 0 , ϑ , φ ) χ S λ ( ϑ , φ ) . (1178)
For Hermitian adjoint functions we have
χ S λ † ( ϑ , φ ) = ∑ m ( − 1 ) λ − m D − m − λ S ( φ , ϑ , 0 ) χ S m † , \chi^\dagger_{S\lambda}(\vartheta,\varphi)=\sum_m(-1)^{\lambda-m}D^{S}_{-m-\lambda}(\varphi,\vartheta,0)\chi^\dagger_{Sm}, χ S λ † ( ϑ , φ ) = m ∑ ( − 1 ) λ − m D − m − λ S ( φ , ϑ , 0 ) χ S m † ,
χ S m † = ∑ λ ( − 1 ) λ − m D λ m S ( 0 , ϑ , φ ) χ S λ † ( ϑ , φ ) . (1179) \chi^\dagger_{Sm}=\sum_{\lambda}(-1)^{\lambda-m}D^{S}_{\lambda m}(0,\vartheta,\varphi)\chi^\dagger_{S\lambda}(\vartheta,\varphi). \tag*{(1179)} χ S m † = λ ∑ ( − 1 ) λ − m D λm S ( 0 , ϑ , φ ) χ S λ † ( ϑ , φ ) . (1179)
It follows from ((6.1.7)), ((6.1.20)), ((6.1.21)) that the contravariant components of the helicity basis functions have the form
[ χ S λ ( ϑ , φ ) ] σ = D σ λ S ( φ , ϑ , 0 ) , [\chi_{S\lambda}(\vartheta,\varphi)]^{\sigma}=D^{S}_{\sigma\lambda}(\varphi,\vartheta,0), [ χ S λ ( ϑ , φ ) ] σ = D σ λ S ( φ , ϑ , 0 ) ,
[ χ S λ † ( ϑ , φ ) ] σ = ( − 1 ) λ − σ D − σ − λ S ( φ , ϑ , 0 ) , ( σ , λ = − S , − S + 1 , … , S − 1 , S ) . (1180) [\chi^\dagger_{S\lambda}(\vartheta,\varphi)]^{\sigma}=(-1)^{\lambda-\sigma}D^{S}_{-\sigma-\lambda}(\varphi,\vartheta,0), \qquad(\sigma,\lambda=-S,-S+1,\ldots,S-1,S). \tag*{(1180)} [ χ S λ † ( ϑ , φ ) ] σ = ( − 1 ) λ − σ D − σ − λ S ( φ , ϑ , 0 ) , ( σ , λ = − S , − S + 1 , … , S − 1 , S ) . (1180)
The collection of 2 S + 1 2S+1 2 S + 1 helicity basis functions χ S λ ( ϑ , φ ) \chi_{S\lambda}(\vartheta,\varphi) χ S λ ( ϑ , φ ) ( λ = S , S − 1 , … , − S ) (\lambda=S,S-1,\ldots,-S) ( λ = S , S − 1 , … , − S ) as well as the set of the functions χ S m \chi_{Sm} χ S m form complete orthonormal sets. The orthonormality condition for the helicity basis functions is
χ S λ † ( ϑ , φ ) χ S λ ′ ( ϑ , φ ) = δ λ λ ′ . (1181) \chi^\dagger_{S\lambda}(\vartheta,\varphi)\chi_{S\lambda'}(\vartheta,\varphi)=\delta_{\lambda\lambda'}. \tag*{(1181)} χ S λ † ( ϑ , φ ) χ S λ ′ ( ϑ , φ ) = δ λ λ ′ . (1181)
The completeness condition for these functions has the following matrix form
∑ λ χ S λ ( ϑ , φ ) χ S λ † ( ϑ , φ ) = I ^ . (1182) \sum_{\lambda}\chi_{S\lambda}(\vartheta,\varphi)\chi^\dagger_{S\lambda}(\vartheta,\varphi)=\hat{I}. \tag*{(1182)} λ ∑ χ S λ ( ϑ , φ ) χ S λ † ( ϑ , φ ) = I ^ . (1182)
Matrix products χ S λ ( ϑ , φ ) χ S λ ′ † ( ϑ , φ ) \chi_{S\lambda}(\vartheta,\varphi)\chi^\dagger_{S\lambda'}(\vartheta,\varphi) χ S λ ( ϑ , φ ) χ S λ ′ † ( ϑ , φ ) of the helicity basis functions are the square ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrices. They may be expanded in terms of the polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) (§ 10.4 ) as
χ S λ ( ϑ , φ ) χ S λ ′ † ( ϑ , φ ) = ∑ L M 2 L + 1 2 S + 1 C S λ L A S λ ′ D M A L ( φ , ϑ , 0 ) T ^ L M ( S ) . (1183) \chi_{S\lambda}(\vartheta,\varphi)\chi^\dagger_{S\lambda'}(\vartheta,\varphi)=\sum_{LM}\sqrt{\frac{2L+1}{2S+1}}C^{S\lambda'}_{S\lambda\,L A}D^{L}_{M A}(\varphi,\vartheta,0)\hat{T}_{LM}(S). \tag*{(1183)} χ S λ ( ϑ , φ ) χ S λ ′ † ( ϑ , φ ) = L M ∑ 2 S + 1 2 L + 1 C S λ L A S λ ′ D M A L ( φ , ϑ , 0 ) T ^ L M ( S ) . (1183)
Matrix elements of the polarization operators between the helicity basis functions are given by
χ S λ ′ † ( ϑ , ϕ ) T ^ L M ( S ) χ S λ ( ϑ , ϕ ) = 2 L + 1 2 S + 1 ( − 1 ) M + Λ C S λ L Λ S λ ′ D M − A L ( ϕ , ϑ , 0 ) . (1184) \chi^{\dagger}_{S\lambda'}(\vartheta,\phi)\hat{T}^{(S)}_{LM}\chi_{S\lambda}(\vartheta,\phi)=\sqrt{\frac{2L+1}{2S+1}}(-1)^{M+\Lambda}C^{S\lambda'}_{S\lambda L\Lambda}D^{L}_{M-A}(\phi,\vartheta,0). \tag*{(1184)} χ S λ ′ † ( ϑ , ϕ ) T ^ L M ( S ) χ S λ ( ϑ , ϕ ) = 2 S + 1 2 L + 1 ( − 1 ) M + Λ C S λ L Λ S λ ′ D M − A L ( ϕ , ϑ , 0 ) . (1184)
In particular, the diagonal matrix elements are
χ S λ † ( ϑ , ϕ ) T ^ L M ( S ) χ S λ ( ϑ , ϕ ) = 4 π 2 S + 1 C S λ L 0 S λ Y L M ( ϑ , ϕ ) . (1185) \chi^{\dagger}_{S\lambda}(\vartheta,\phi)\hat{T}^{(S)}_{LM}\chi_{S\lambda}(\vartheta,\phi)=\sqrt{\frac{4\pi}{2S+1}}C^{S\lambda}_{S\lambda L0}Y_{LM}(\vartheta,\phi). \tag*{(1185)} χ S λ † ( ϑ , ϕ ) T ^ L M ( S ) χ S λ ( ϑ , ϕ ) = 2 S + 1 4 π C S λ L 0 S λ Y L M ( ϑ , ϕ ) . (1185)
General Spin Functions
Any spin function (6.1.1) of particles of spin S S S can be expanded in sum of the basis spin functions χ S m \chi_{Sm} χ S m ,
χ = ∑ m = − S S a m χ S m , (1186) \chi=\sum_{m=-S}^{S}a^{m}\chi_{Sm}, \tag*{(1186)} χ = m = − S ∑ S a m χ S m , (1186)
where a m ≡ χ ( m ) a^{m}\equiv\chi(m) a m ≡ χ ( m ) is a contravariant component of some irreducible tensor of rank S S S . Similarly, the expansion of the Hermitian adjoint function χ † \chi^{\dagger} χ † has the form
χ † = ∑ m = − S S ( a m ) ∗ χ S m † . (1187) \chi^{\dagger}=\sum_{m=-S}^{S}(a^{m})^{*}\chi^{\dagger}_{Sm}. \tag*{(1187)} χ † = m = − S ∑ S ( a m ) ∗ χ S m † . (1187)
In general, ( a m ) ∗ ≠ a m (a^{m})^{*}\ne a_{m} ( a m ) ∗ = a m , i.e., the corresponding irreducible tensor is not real. The normalization condition (6.1.4) imposes a restriction on a m a^{m} a m :
∑ m = − S S ∣ a m ∣ 2 = 1 \sum_{m=-S}^{S}\lvert a^{m}\rvert^{2}=1 m = − S ∑ S ∣ a m ∣ 2 = 1
The product χ χ † \chi\chi^{\dagger} χ χ † of two spin functions represents the square ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrix which may be expanded in terms of the polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) (Sec. 2.4)
χ χ † = ∑ L = 0 2 S ∑ M = − L L ( − 1 ) M P L − M T ^ L M ( S ) , (1188) \chi\chi^{\dagger}=\sum_{L=0}^{2S}\sum_{M=-L}^{L}(-1)^{M}P_{L-M}\hat{T}_{LM}(S), \tag*{(1188)} χ χ † = L = 0 ∑ 2 S M = − L ∑ L ( − 1 ) M P L − M T ^ L M ( S ) , (1188)
where the expansion coefficients may be expressed through a m a^{m} a m as
P L M = 2 L + 1 2 S + 1 ∑ m , n = − S S C S m L M S n ( a n ) ∗ a m . (1189) P_{LM}=\sqrt{\frac{2L+1}{2S+1}}\sum_{m,n=-S}^{S}C^{S n}_{S m L M}(a^{n})^{*}a^{m}. \tag*{(1189)} P L M = 2 S + 1 2 L + 1 m , n = − S ∑ S C S m L M S n ( a n ) ∗ a m . (1189)
In particular, P 00 = 1 / 2 S + 1 P_{00}=1/\sqrt{2S+1} P 00 = 1/ 2 S + 1 .
The quantities P L M P_{LM} P L M (M = − L , − L + 1 , … , L M=-L,-L+1,\ldots,L M = − L , − L + 1 , … , L ) are covariant components of some irreducible tensor of rank L L L . They represent the mean values of the polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) in the states described by the spin functions χ \chi χ :
χ † T ^ L M ( S ) χ = P L M . (1190) \chi^{\dagger}\hat{T}_{LM}(S)\chi=P_{LM}. \tag*{(1190)} χ † T ^ L M ( S ) χ = P L M . (1190)
Under complex conjugation P L M P_{LM} P L M transforms as
( P L M ) ∗ = ( − 1 ) M P L − M . (1191) (P_{LM})^{*}=(-1)^{M}P_{L-M}. \tag*{(1191)} ( P L M ) ∗ = ( − 1 ) M P L − M . (1191)
The equality χ χ † χ χ † = χ χ † \chi\chi^{\dagger}\chi\chi^{\dagger}=\chi\chi^{\dagger} χ χ † χ χ † = χ χ † implies the following conditions for the coefficients P L M P_{LM} P L M ,
P L M = ∑ L 1 M 1 L 2 M 2 ( − 1 ) 2 S + L 1 + L 2 ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) { L 1 L 2 L S S S } C L 1 M 1 L 2 M 2 L M P L 1 M 1 P L 2 M 2 . (1192) P_{LM}=\sum_{L_{1}M_{1}L_{2}M_{2}}(-1)^{2S+L_{1}+L_{2}}\sqrt{(2L_{1}+1)(2L_{2}+1)}
\begin{Bmatrix}
L_{1} & L_{2} & L\\
S & S & S
\end{Bmatrix}
C^{LM}_{L_{1}M_{1}L_{2}M_{2}}P_{L_{1}M_{1}}P_{L_{2}M_{2}}. \tag*{(1192)} P L M = L 1 M 1 L 2 M 2 ∑ ( − 1 ) 2 S + L 1 + L 2 ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) { L 1 S L 2 S L S } C L 1 M 1 L 2 M 2 L M P L 1 M 1 P L 2 M 2 . (1192)
In particular, Eq. (6.1.34) gives the normalization condition
∑ L = 0 2 S ∑ M = − L L ∣ P L M ∣ 2 = 1. (1193) \sum_{L=0}^{2S}\sum_{M=-L}^{L}\lvert P_{LM}\rvert^{2}=1. \tag*{(1193)} L = 0 ∑ 2 S M = − L ∑ L ∣ P L M ∣ 2 = 1. (1193)
Under rotations of the coordinate system specified by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ the spin function χ \chi χ is transformed by the rotation operator D ^ S ( α , β , γ ) \hat{D}^{S}(\alpha,\beta,\gamma) D ^ S ( α , β , γ ) (Eq. (2.4.17)) into
χ ′ = D ^ S ( α , β , γ ) χ . (1194) \chi'=\hat{D}^{S}(\alpha,\beta,\gamma)\chi. \tag*{(1194)} χ ′ = D ^ S ( α , β , γ ) χ . (1194)
The spin function χ ′ \chi' χ ′ in the rotated coordinate system may be expanded in sum of basis spin functions χ S m \chi_{Sm} χ S m or χ S m ′ \chi'_{Sm} χ S m ′ in the original or rotated coordinate systems:
χ ′ = ∑ m = − S S a ′ m χ S m = ∑ m = − S S a m χ S m ′ . (1195) \chi'=\sum_{m=-S}^{S}a'^{m}\chi_{Sm}=\sum_{m=-S}^{S}a^{m}\chi'_{Sm}. \tag*{(1195)} χ ′ = m = − S ∑ S a ′ m χ S m = m = − S ∑ S a m χ S m ′ . (1195)
where a m a^{m} a m and a ′ m a'^{m} a ′ m are components of the spin function in the original and rotated systems, respectively. The relations between the basis functions χ S m \chi_{Sm} χ S m and χ S m ′ \chi'_{Sm} χ S m ′ are given by Eq. (1174) . The transformation properties of a m a^{m} a m under rotations are
a ′ m = ∑ n = − S S D m n S ( α , β , γ ) a n , (1196) a'^{m}=\sum_{n=-S}^{S}D^{S}_{mn}(\alpha,\beta,\gamma)a^{n}, \tag*{(1196)} a ′ m = n = − S ∑ S D mn S ( α , β , γ ) a n , (1196)
a n = ∑ m = − S S D m n S ∗ ( α , β , γ ) a ′ m , a^{n}=\sum_{m=-S}^{S}D^{S*}_{mn}(\alpha,\beta,\gamma)a'^{m}, a n = m = − S ∑ S D mn S ∗ ( α , β , γ ) a ′ m ,
in accordance with general transformation rule for contravariant components (Sec. 4.1.2).
Polarization Density Matrix
Spin functions enable one to describe only the totally polarized (pure) particle states. For describing partially polarized states which are incoherent statistical mixtures of the pure states, the polarization density matrix is used.
The polarization density matrix ρ ^ \hat{\rho} ρ ^ of the particles of spin S S S is the square ( 2 S + 1 ) × ( 2 S + 1 ) (2S+1)\times(2S+1) ( 2 S + 1 ) × ( 2 S + 1 ) matrix defined by
ρ σ σ ′ = ⟨ χ ( σ ) χ ∗ ( σ ′ ) ⟩ ε , (1197) \rho_{\sigma\sigma'}=\langle\chi(\sigma)\chi^{*}(\sigma')\rangle_{\varepsilon}, \tag*{(1197)} ρ σ σ ′ = ⟨ χ ( σ ) χ ∗ ( σ ′ ) ⟩ ε , (1197)
or in a matrix form by
ρ ^ = ⟨ χ χ † ⟩ ε , (1198) \hat{\rho}=\langle\chi\chi^{\dagger}\rangle_{\varepsilon}, \tag*{(1198)} ρ ^ = ⟨ χ χ † ⟩ ε , (1198)
where ⟨ ⟩ ε \langle\rangle_{\varepsilon} ⟨ ⟩ ε denotes the statistical average. In particular, for pure states one gets
ρ ^ = χ χ † . (1199) \hat{\rho}=\chi\chi^{\dagger}. \tag*{(1199)} ρ ^ = χ χ † . (1199)
The density matrix is Hermitian,
ρ ^ † = ρ ^ , i.e., ρ σ σ ′ ∗ = ρ σ ′ σ ; (1200) \hat{\rho}^{\dagger}=\hat{\rho},\qquad\text{i.e.,}\quad\rho^{*}_{\sigma\sigma'}=\rho_{\sigma'\sigma}; \tag*{(1200)} ρ ^ † = ρ ^ , i.e., ρ σ σ ′ ∗ = ρ σ ′ σ ; (1200)
and normalized,
Tr ρ ^ = 1 , i.e., ∑ σ ρ σ σ = 1. (1201) \operatorname{Tr}\hat{\rho}=1,\qquad\text{i.e.,}\quad\sum_{\sigma}\rho_{\sigma\sigma}=1. \tag*{(1201)} Tr ρ ^ = 1 , i.e., σ ∑ ρ σ σ = 1. (1201)
The conditions (6.1.42), (6.1.43) reveal that the density matrix of particles of spin S S S is completely determined by ( 2 S + 1 ) 2 − 1 = 4 S ( S + 1 ) (2S+1)^2-1=4S(S+1) ( 2 S + 1 ) 2 − 1 = 4 S ( S + 1 ) real parameters.
An expectation value of any polarization operator T ^ \hat{T} T ^ in a state described by the density matrix ρ ^ \hat{\rho} ρ ^ may be evaluated by
⟨ T ^ ⟩ = Tr { T ^ ρ ^ } = Tr { ρ ^ T ^ } . (1202) \langle\hat{T}\rangle=\operatorname{Tr}\{\hat{T}\hat{\rho}\}=\operatorname{Tr}\{\hat{\rho}\hat{T}\}. \tag*{(1202)} ⟨ T ^ ⟩ = Tr { T ^ ρ ^ } = Tr { ρ ^ T ^ } . (1202)
Note also the following property of the density matrix
Tr { ρ ^ 2 } ≤ 1 , (1203) \operatorname{Tr}\{\hat{\rho}^{2}\}\leq1, \tag*{(1203)} Tr { ρ ^ 2 } ≤ 1 , (1203)
where the equality holds only for pure states. The density matrix of pure states satisfies the relation
ρ ^ 2 = ρ ^ . (1204) \hat{\rho}^{2}=\hat{\rho}. \tag*{(1204)} ρ ^ 2 = ρ ^ . (1204)
The density matrix ρ ^ \hat{\rho} ρ ^ may be expanded into a sum of the polarisation operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) ,
ρ ^ = ∑ L = 0 2 S ∑ M = − L L ( − 1 ) M t L , − M T ^ L M ( S ) . (1205) \hat{\rho}=\sum_{L=0}^{2S}\sum_{M=-L}^{L}(-1)^{M}t_{L,-M}\hat{T}_{LM}(S). \tag*{(1205)} ρ ^ = L = 0 ∑ 2 S M = − L ∑ L ( − 1 ) M t L , − M T ^ L M ( S ) . (1205)
The expansion coefficients t L M t_{LM} t L M are called the statistical tensors. They represent expectation values of the polarization operators T ^ L M ( S ) \hat{T}_{LM}(S) T ^ L M ( S ) in a state described by the density matrix ρ ^ \hat{\rho} ρ ^ ,
t L M = ⟨ T ^ L M ( S ) ⟩ = Tr { ρ ^ T ^ L M ( S ) } . (1206) t_{LM}=\langle\hat{T}_{LM}(S)\rangle=\operatorname{Tr}\{\hat{\rho}\hat{T}_{LM}(S)\}. \tag*{(1206)} t L M = ⟨ T ^ L M ( S )⟩ = Tr { ρ ^ T ^ L M ( S )} . (1206)
The statistical tensors are related to the density matrix elements by
t L M = 2 L + 1 2 S + 1 ∑ σ , σ ′ = − S S C S σ ′ L M S σ S σ ′ ρ σ σ ′ . (1207) t_{LM}=\sqrt{\frac{2L+1}{2S+1}}\sum_{\sigma,\sigma'=-S}^{S}C^{S\sigma S\sigma'}_{S\sigma' L M}\rho_{\sigma\sigma'}. \tag*{(1207)} t L M = 2 S + 1 2 L + 1 σ , σ ′ = − S ∑ S C S σ ′ L M S σ S σ ′ ρ σ σ ′ . (1207)
ρ σ σ ′ = ∑ L , M 2 L + 1 2 S + 1 C S σ ′ L M S σ S σ ′ t L M . (1208) \rho_{\sigma\sigma'}=\sum_{L,M}\sqrt{\frac{2L+1}{2S+1}}C^{S\sigma S\sigma'}_{S\sigma' L M}t_{LM}. \tag*{(1208)} ρ σ σ ′ = L , M ∑ 2 S + 1 2 L + 1 C S σ ′ L M S σ S σ ′ t L M . (1208)
In particular,
t 00 = 1 2 S + 1 . (1209) t_{00}=\frac{1}{\sqrt{2S+1}}. \tag*{(1209)} t 00 = 2 S + 1 1 . (1209)
Under complex conjugation the statistical tensors transform as
t L M ∗ = ( − 1 ) M t L − M . t_{LM}^{*}=(-1)^{M}t_{L-M}. t L M ∗ = ( − 1 ) M t L − M .
The properties of the statistical tensors for pure (completely polarized) states have been considered above (Eqs. (1189) –(1193) ). For unpolarised states all statistical tensors vanish except t 00 t_{00} t 00 , thus
ρ ^ u n p o l = 1 2 S + 1 I ^ , (1210) \hat{\rho}_{\mathrm{unpol}}=\frac{1}{2S+1}\hat{I}, \tag*{(1210)} ρ ^ unpol = 2 S + 1 1 I ^ , (1210)
where I ^ \hat{I} I ^ is the unit matrix.
The quantities t L M t_{LM} t L M are covariant components of some irreducible tensor of rank L L L . Under rotations they transform in accordance with Eq. (538) .
Two Particles with Arbitrary Spins
Let us consider a system of two particles, 1 and 2, with spins S 1 S_1 S 1 and S 2 S_2 S 2 . Total spin S S S of the system takes the values ∣ S 1 − S 2 ∣ , ∣ S 1 − S 2 ∣ + 1 , … , S 1 + S 2 \lvert S_1-S_2\rvert,\lvert S_1-S_2\rvert+1,\ldots,S_1+S_2 ∣ S 1 − S 2 ∣ , ∣ S 1 − S 2 ∣ + 1 , … , S 1 + S 2 . The spin function χ S m ( 1 , 2 ) \chi_{Sm}(1,2) χ S m ( 1 , 2 ) which describes the system with total spin S S S and spin projection m m m may be written as
χ S m ( 1 , 2 ) = ∑ m 1 m 2 C S 1 m 1 S 2 m 2 S m χ S 1 m 1 ( 1 ) χ S 2 m 2 ( 2 ) . (1211) \chi_{Sm}(1,2)=\sum_{m_1m_2}C^{Sm}_{S_1m_1\,S_2m_2}\chi_{S_1m_1}(1)\chi_{S_2m_2}(2). \tag*{(1211)} χ S m ( 1 , 2 ) = m 1 m 2 ∑ C S 1 m 1 S 2 m 2 S m χ S 1 m 1 ( 1 ) χ S 2 m 2 ( 2 ) . (1211)
The inverse relation is
χ S 1 m 1 ( 1 ) χ S 2 m 2 ( 2 ) = ∑ S C S 1 m 1 S 2 m 2 S m χ S m ( 1 , 2 ) . (1212) \chi_{S_1m_1}(1)\chi_{S_2m_2}(2)=\sum_S C^{Sm}_{S_1m_1\,S_2m_2}\chi_{Sm}(1,2). \tag*{(1212)} χ S 1 m 1 ( 1 ) χ S 2 m 2 ( 2 ) = S ∑ C S 1 m 1 S 2 m 2 S m χ S m ( 1 , 2 ) . (1212)
Spin functions (6.1.53) are orthonormalized
χ S m † ( 1 , 2 ) χ S ′ m ′ ( 1 , 2 ) = δ S S ′ δ m m ′ , (1213) \chi^\dagger_{Sm}(1,2)\chi_{S'm'}(1,2)=\delta_{SS'}\delta_{mm'}, \tag*{(1213)} χ S m † ( 1 , 2 ) χ S ′ m ′ ( 1 , 2 ) = δ S S ′ δ m m ′ , (1213)
and satisfy the completeness condition
∑ S , m χ S m ( 1 , 2 ) χ S m † ( 1 , 2 ) = I ^ . (1214) \sum_{S,m}\chi_{Sm}(1,2)\chi^\dagger_{Sm}(1,2)=\hat{I}. \tag*{(1214)} S , m ∑ χ S m ( 1 , 2 ) χ S m † ( 1 , 2 ) = I ^ . (1214)
Let the polarization operator T ^ L M ( i ) \hat{T}_{LM}(i) T ^ L M ( i ) ( i = 1 , 2 ) (i=1,2) ( i = 1 , 2 ) act on spin variables of a particle i i i only. Then we have
T ^ L 1 M 1 ( 1 ) χ S m ( 1 , 2 ) = ( − 1 ) S 1 + S 2 + S + L 1 ( 2 L 1 + 1 ) ( 2 S + 1 ) ∑ S ′ { S 1 S 2 S S ′ L 1 S 1 } C S m L 1 M 1 S ′ m ′ χ S ′ m ′ ( 1 , 2 ) , (1215) \hat{T}_{L_1M_1}(1)\chi_{Sm}(1,2)=(-1)^{S_1+S_2+S+L_1}\sqrt{(2L_1+1)(2S+1)}
\sum_{S'}\left\{\begin{matrix} S_1&S_2&S\\ S'&L_1&S_1 \end{matrix}\right\}C^{S'm'}_{Sm\,L_1M_1}\chi_{S'm'}(1,2), \tag*{(1215)} T ^ L 1 M 1 ( 1 ) χ S m ( 1 , 2 ) = ( − 1 ) S 1 + S 2 + S + L 1 ( 2 L 1 + 1 ) ( 2 S + 1 ) S ′ ∑ { S 1 S ′ S 2 L 1 S S 1 } C S m L 1 M 1 S ′ m ′ χ S ′ m ′ ( 1 , 2 ) , (1215)
T ^ L 2 M 2 ( 2 ) χ S m ( 1 , 2 ) = ( 2 L 2 + 1 ) ( 2 S + 1 ) ∑ S ′ ( − 1 ) S 1 + S 2 + S ′ + L 2 { S 1 S 2 S L 2 S ′ S 2 } C S m L 2 M 2 S ′ m ′ χ S ′ m ′ ( 1 , 2 ) . (1216) \hat{T}_{L_2M_2}(2)\chi_{Sm}(1,2)=\sqrt{(2L_2+1)(2S+1)}
\sum_{S'}(-1)^{S_1+S_2+S'+L_2}\left\{\begin{matrix} S_1&S_2&S\\ L_2&S'&S_2 \end{matrix}\right\}C^{S'm'}_{Sm\,L_2M_2}\chi_{S'm'}(1,2). \tag*{(1216)} T ^ L 2 M 2 ( 2 ) χ S m ( 1 , 2 ) = ( 2 L 2 + 1 ) ( 2 S + 1 ) S ′ ∑ ( − 1 ) S 1 + S 2 + S ′ + L 2 { S 1 L 2 S 2 S ′ S S 2 } C S m L 2 M 2 S ′ m ′ χ S ′ m ′ ( 1 , 2 ) . (1216)
One may also consider a set of scalar operators Q ^ L ( 1 , 2 ) \hat{Q}_L(1,2) Q ^ L ( 1 , 2 ) with 0 ≤ L ≤ min { 2 S 1 , 2 S 2 } 0\leq L\leq\min\{2S_1,2S_2\} 0 ≤ L ≤ min { 2 S 1 , 2 S 2 }
Q ^ L ( 1 , 2 ) = ∑ M ( − 1 ) M T ^ L M ( 1 ) T ^ L − M ( 2 ) . (1217) \hat{Q}_L(1,2)=\sum_M(-1)^M\hat{T}_{LM}(1)\hat{T}_{L-M}(2). \tag*{(1217)} Q ^ L ( 1 , 2 ) = M ∑ ( − 1 ) M T ^ L M ( 1 ) T ^ L − M ( 2 ) . (1217)
These operators leave the spin of a system and spin projection unchanged:
Q ^ L ( 1 , 2 ) χ S m ( 1 , 2 ) = ( − 1 ) S 1 + S 2 + S ( 2 L + 1 ) { S 1 S 2 S S 2 S 1 L } χ S m ( 1 , 2 ) . (1218) \hat{Q}_L(1,2)\chi_{Sm}(1,2)=(-1)^{S_1+S_2+S}(2L+1)\left\{\begin{matrix} S_1&S_2&S\\ S_2&S_1&L \end{matrix}\right\}\chi_{Sm}(1,2). \tag*{(1218)} Q ^ L ( 1 , 2 ) χ S m ( 1 , 2 ) = ( − 1 ) S 1 + S 2 + S ( 2 L + 1 ) { S 1 S 2 S 2 S 1 S L } χ S m ( 1 , 2 ) . (1218)
Among scalar operators the so-called projection operators P ^ S ( 1 , 2 ) \hat{P}_S(1,2) P ^ S ( 1 , 2 ) (∣ S 1 − S 2 ∣ ≤ S ≤ S 1 + S 2 |S_1-S_2| \le S \le S_1+S_2 ∣ S 1 − S 2 ∣ ≤ S ≤ S 1 + S 2 ) are of special interest. When applied to arbitrary spin function χ ( 1 , 2 ) \chi(1,2) χ ( 1 , 2 ) , the operator P ^ S ( 1 , 2 ) \hat{P}_S(1,2) P ^ S ( 1 , 2 ) gives the component of this function with definite total spin S S S . Thus, the projection operator P ^ S ( 1 , 2 ) \hat{P}_S(1,2) P ^ S ( 1 , 2 ) may be defined by the equations
P ^ S ( 1 , 2 ) χ S ′ m ′ ( 1 , 2 ) = δ S S ′ χ S ′ m ′ ( 1 , 2 ) . (1219) \hat{P}_S(1,2)\chi_{S'm'}(1,2)=\delta_{SS'}\chi_{S'm'}(1,2). \tag*{(1219)} P ^ S ( 1 , 2 ) χ S ′ m ′ ( 1 , 2 ) = δ S S ′ χ S ′ m ′ ( 1 , 2 ) . (1219)
From Eqs. (6.1.61) one may obtain the following properties of the projection operators
P ^ S ( 1 , 2 ) P ^ S ′ ( 1 , 2 ) = δ S S ′ P ^ S ( 1 , 2 ) , (1220) \hat{P}_S(1,2)\hat{P}_{S'}(1,2)=\delta_{SS'}\hat{P}_S(1,2), \tag*{(1220)} P ^ S ( 1 , 2 ) P ^ S ′ ( 1 , 2 ) = δ S S ′ P ^ S ( 1 , 2 ) , (1220)
∑ S P ^ S ( 1 , 2 ) = I ^ . (1221) \sum_S \hat{P}_S(1,2)=\hat{I}. \tag*{(1221)} S ∑ P ^ S ( 1 , 2 ) = I ^ . (1221)
The form of P ^ S ( 1 , 2 ) \hat{P}_S(1,2) P ^ S ( 1 , 2 ) may be written explicitly as
P ^ S ( 1 , 2 ) = ∑ m χ S m ( 1 , 2 ) χ S m † ( 1 , 2 ) , (1222) \hat{P}_S(1,2)=\sum_m\chi_{Sm}(1,2)\chi^\dagger_{Sm}(1,2), \tag*{(1222)} P ^ S ( 1 , 2 ) = m ∑ χ S m ( 1 , 2 ) χ S m † ( 1 , 2 ) , (1222)
or, in terms of scalar operators Q ^ L ( 1 , 2 ) \hat{Q}_L(1,2) Q ^ L ( 1 , 2 ) ,
P ^ S ( 1 , 2 ) = ( − 1 ) S 1 + S 2 + S ( 2 S + 1 ) ∑ L { S 1 S 2 S S 2 S 1 L } Q ^ L ( 1 , 2 ) . (1223) \hat{P}_S(1,2)=(-1)^{S_1+S_2+S}(2S+1)\sum_L
\begin{Bmatrix}
S_1 & S_2 & S\\
S_2 & S_1 & L
\end{Bmatrix}
\hat{Q}_L(1,2). \tag*{(1223)} P ^ S ( 1 , 2 ) = ( − 1 ) S 1 + S 2 + S ( 2 S + 1 ) L ∑ { S 1 S 2 S 2 S 1 S L } Q ^ L ( 1 , 2 ) . (1223)
Spin functions for S = 1 / 2 S=1/2 S = 1/2
Basis Spin Functions
The basis functions χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m (m = ± 1 / 2 m=\pm1/2 m = ± 1/2 ) are the eigenfunctions of the operators S ^ 2 \hat{S}^2 S ^ 2 and S ^ z \hat{S}_z S ^ z :
S ^ 2 χ 1 2 m = 3 4 χ 1 2 m , (1224) \hat{S}^2\chi_{\frac{1}{2}m}=\frac{3}{4}\chi_{\frac{1}{2}m}, \tag*{(1224)} S ^ 2 χ 2 1 m = 4 3 χ 2 1 m , (1224)
S ^ z χ 1 2 m = m χ 1 2 m . \hat{S}_z\chi_{\frac{1}{2}m}=m\chi_{\frac{1}{2}m}. S ^ z χ 2 1 m = m χ 2 1 m .
A function χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m describes the state of a spin-1 2 \frac{1}{2} 2 1 particle with definite spin projection, m m m , on the z z z -axis. The dependence of the basis functions χ 1 2 m ( σ ) \chi_{\frac{1}{2}m}(\sigma) χ 2 1 m ( σ ) on the spin variable σ \sigma σ is given by
χ 1 2 m ( σ ) = δ m σ . (1225) \chi_{\frac{1}{2}m}(\sigma)=\delta_{m\sigma}. \tag*{(1225)} χ 2 1 m ( σ ) = δ mσ . (1225)
According to Eq. (6.1.1), the basis functions χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m may be written as
χ 1 2 1 2 = ( 1 0 ) , χ 1 2 − 1 2 = ( 0 1 ) . (1226) \chi_{\frac{1}{2}\frac{1}{2}}=\begin{pmatrix}1\\0\end{pmatrix},\quad
\chi_{\frac{1}{2}-\frac{1}{2}}=\begin{pmatrix}0\\1\end{pmatrix}. \tag*{(1226)} χ 2 1 2 1 = ( 1 0 ) , χ 2 1 − 2 1 = ( 0 1 ) . (1226)
The Hermitian conjugate functions χ 1 2 m † \chi^\dagger_{\frac{1}{2}m} χ 2 1 m † have the form
χ 1 2 1 2 † = ( 1 , 0 ) , χ 1 2 − 1 2 † = ( 0 , 1 ) . (1227) \chi^\dagger_{\frac{1}{2}\frac{1}{2}}=(1,0),\quad
\chi^\dagger_{\frac{1}{2}-\frac{1}{2}}=(0,1). \tag*{(1227)} χ 2 1 2 1 † = ( 1 , 0 ) , χ 2 1 − 2 1 † = ( 0 , 1 ) . (1227)
The following compact notations for the basis functions are widely used:
χ 1 2 1 2 ≡ α , χ 1 2 − 1 2 ≡ β . (1228) \chi_{\frac{1}{2}\frac{1}{2}}\equiv\alpha,\quad\chi_{\frac{1}{2}-\frac{1}{2}}\equiv\beta. \tag*{(1228)} χ 2 1 2 1 ≡ α , χ 2 1 − 2 1 ≡ β . (1228)
The basis functions χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m satisfy the orthonormality condition
χ 1 2 m ′ † χ 1 2 m = δ m ′ m . (1229) \chi^\dagger_{\frac{1}{2}m'}\chi_{\frac{1}{2}m}=\delta_{m'm}. \tag*{(1229)} χ 2 1 m ′ † χ 2 1 m = δ m ′ m . (1229)
The completeness condition for the basis functions is
∑ m χ 1 2 m χ 1 2 m † = I ^ , (1230) \sum_m\chi_{\frac{1}{2}m}\chi^\dagger_{\frac{1}{2}m}=\hat{I}, \tag*{(1230)} m ∑ χ 2 1 m χ 2 1 m † = I ^ , (1230)
where I ^ \hat{I} I ^ is the unit 2 × 2 2\times2 2 × 2 matrix.
Expansions of Products of Basis Functions
Expansions of Products of Basis Functions
The products χ 1 2 m χ 1 2 m † \chi_{\frac{1}{2}m}\chi_{\frac{1}{2}m}^{\dagger} χ 2 1 m χ 2 1 m † represent 2 × 2 2 \times2 2 × 2 square matrices whose elements are
[ χ 1 2 m χ 1 2 m ′ † ] σ σ ′ = χ 1 2 σ † ( χ 1 2 m χ 1 2 m ′ † ) χ 1 2 σ ′ = δ σ m δ σ ′ m ′ . (1231) \left[\chi_{\frac{1}{2}m}\chi_{\frac{1}{2}m'}^{\dagger}\right]_{\sigma\sigma'}=\chi_{\frac{1}{2}\sigma}^{\dagger}\left(\chi_{\frac{1}{2}m}\chi_{\frac{1}{2}m'}^{\dagger}\right)\chi_{\frac{1}{2}\sigma'}=\delta_{\sigma m}\delta_{\sigma' m'}. \tag*{(1231)} [ χ 2 1 m χ 2 1 m ′ † ] σ σ ′ = χ 2 1 σ † ( χ 2 1 m χ 2 1 m ′ † ) χ 2 1 σ ′ = δ σ m δ σ ′ m ′ . (1231)
The expansions of these products in terms of the spin matrices are given by
χ 1 2 m χ 1 2 m ′ † = 1 2 δ m m ′ I ^ − 3 C 1 2 m 1 2 m ′ 1 μ S ^ μ , (1232) \chi_{\frac{1}{2}m}\chi_{\frac{1}{2}m'}^{\dagger}=\frac{1}{2}\delta_{mm'}\hat{I}-\sqrt{3}C_{\frac{1}{2}m\frac{1}{2}m'}^{1\mu}\hat{S}_{\mu}, \tag*{(1232)} χ 2 1 m χ 2 1 m ′ † = 2 1 δ m m ′ I ^ − 3 C 2 1 m 2 1 m ′ 1 μ S ^ μ , (1232)
where S μ ( μ = ± 1 , 0 ) S_{\mu}(\mu=\pm1,0) S μ ( μ = ± 1 , 0 ) are spherical components of the spin operator (Sec. § 10.6 ). In more detailed form Eq. (1232) may be written as
α α † ≡ χ 1 2 1 2 χ 1 2 1 2 † = 1 2 I ^ + S ^ z = 1 2 I ^ + S ^ 0 , α β † ≡ χ 1 2 1 2 χ 1 2 − 1 2 † = S ^ x + i S ^ y = − 2 S ^ + 1 , β α † ≡ χ 1 2 − 1 2 χ 1 2 1 2 † = S ^ x − i S ^ y = 2 S ^ − 1 , β β † ≡ χ 1 2 − 1 2 χ 1 2 − 1 2 † = 1 2 I ^ − S ^ z = 1 2 I ^ − S ^ 0 . (1233) \begin{aligned}
\alpha\alpha^{\dagger}&\equiv\chi_{\frac{1}{2}\frac{1}{2}}\chi_{\frac{1}{2}\frac{1}{2}}^{\dagger}=\frac{1}{2}\hat{I}+\hat{S}_{z}=\frac{1}{2}\hat{I}+\hat{S}_{0},\\
\alpha\beta^{\dagger}&\equiv\chi_{\frac{1}{2}\frac{1}{2}}\chi_{\frac{1}{2}-\frac{1}{2}}^{\dagger}=\hat{S}_{x}+i\hat{S}_{y}=-\sqrt{2}\hat{S}_{+1},\\
\beta\alpha^{\dagger}&\equiv\chi_{\frac{1}{2}-\frac{1}{2}}\chi_{\frac{1}{2}\frac{1}{2}}^{\dagger}=\hat{S}_{x}-i\hat{S}_{y}=\sqrt{2}\hat{S}_{-1},\\
\beta\beta^{\dagger}&\equiv\chi_{\frac{1}{2}-\frac{1}{2}}\chi_{\frac{1}{2}-\frac{1}{2}}^{\dagger}=\frac{1}{2}\hat{I}-\hat{S}_{z}=\frac{1}{2}\hat{I}-\hat{S}_{0}. \tag*{(1233)}
\end{aligned} α α † α β † β α † β β † ≡ χ 2 1 2 1 χ 2 1 2 1 † = 2 1 I ^ + S ^ z = 2 1 I ^ + S ^ 0 , ≡ χ 2 1 2 1 χ 2 1 − 2 1 † = S ^ x + i S ^ y = − 2 S ^ + 1 , ≡ χ 2 1 − 2 1 χ 2 1 2 1 † = S ^ x − i S ^ y = 2 S ^ − 1 , ≡ χ 2 1 − 2 1 χ 2 1 − 2 1 † = 2 1 I ^ − S ^ z = 2 1 I ^ − S ^ 0 . (1233)
Action of Spin Operators on Basis Functions
Cartesian components of a spin operator S ^ i \hat{S}_{i} S ^ i (see Sec. § 10.6 ) applied to the basis functions yield
S ^ x χ 1 2 1 2 = 1 2 χ 1 2 − 1 2 , S ^ x χ 1 2 − 1 2 = 1 2 χ 1 2 1 2 , S ^ y χ 1 2 1 2 = i 2 χ 1 2 − 1 2 , S ^ y χ 1 2 − 1 2 = − i 2 χ 1 2 1 2 , S ^ z χ 1 2 1 2 = 1 2 χ 1 2 1 2 , S ^ z χ 1 2 − 1 2 = − 1 2 χ 1 2 − 1 2 . (1234) \begin{aligned}
\hat{S}_{x}\chi_{\frac{1}{2}\frac{1}{2}}&=\frac{1}{2}\chi_{\frac{1}{2}-\frac{1}{2}},&\hat{S}_{x}\chi_{\frac{1}{2}-\frac{1}{2}}&=\frac{1}{2}\chi_{\frac{1}{2}\frac{1}{2}},\\
\hat{S}_{y}\chi_{\frac{1}{2}\frac{1}{2}}&=\frac{i}{2}\chi_{\frac{1}{2}-\frac{1}{2}},&\hat{S}_{y}\chi_{\frac{1}{2}-\frac{1}{2}}&=-\frac{i}{2}\chi_{\frac{1}{2}\frac{1}{2}},\\
\hat{S}_{z}\chi_{\frac{1}{2}\frac{1}{2}}&=\frac{1}{2}\chi_{\frac{1}{2}\frac{1}{2}},&\hat{S}_{z}\chi_{\frac{1}{2}-\frac{1}{2}}&=-\frac{1}{2}\chi_{\frac{1}{2}-\frac{1}{2}}. \tag*{(1234)}
\end{aligned} S ^ x χ 2 1 2 1 S ^ y χ 2 1 2 1 S ^ z χ 2 1 2 1 = 2 1 χ 2 1 − 2 1 , = 2 i χ 2 1 − 2 1 , = 2 1 χ 2 1 2 1 , S ^ x χ 2 1 − 2 1 S ^ y χ 2 1 − 2 1 S ^ z χ 2 1 − 2 1 = 2 1 χ 2 1 2 1 , = − 2 i χ 2 1 2 1 , = − 2 1 χ 2 1 − 2 1 . (1234)
Spherical components of the spin operator S ^ μ \hat{S}_{\mu} S ^ μ (see Sec. § 10.6 ) act on the basis functions as follows:
S ^ μ χ 1 2 m = − 3 2 C 1 2 m ′ 1 2 m 1 μ χ 1 2 m ′ . (1235) \hat{S}_{\mu}\chi_{\frac{1}{2}m}=-\frac{\sqrt{3}}{2}C_{\frac{1}{2}m'\frac{1}{2}m}^{1\mu}\chi_{\frac{1}{2}m'}. \tag*{(1235)} S ^ μ χ 2 1 m = − 2 3 C 2 1 m ′ 2 1 m 1 μ χ 2 1 m ′ . (1235)
More explicitly Eq. (1235) is
S ^ + 1 χ 1 2 1 2 = 0 , S ^ + 1 χ 1 2 − 1 2 = − 1 2 χ 1 2 1 2 , S ^ 0 χ 1 2 1 2 = 1 2 χ 1 2 1 2 , S ^ 0 χ 1 2 − 1 2 = − 1 2 χ 1 2 − 1 2 , S ^ − 1 χ 1 2 1 2 = 1 2 χ 1 2 − 1 2 , S ^ − 1 χ 1 2 − 1 2 = 0. (1236) \begin{aligned}
\hat{S}_{+1}\chi_{\frac{1}{2}\frac{1}{2}}&=0,&\hat{S}_{+1}\chi_{\frac{1}{2}-\frac{1}{2}}&=-\frac{1}{\sqrt{2}}\chi_{\frac{1}{2}\frac{1}{2}},\\
\hat{S}_{0}\chi_{\frac{1}{2}\frac{1}{2}}&=\frac{1}{2}\chi_{\frac{1}{2}\frac{1}{2}},&\hat{S}_{0}\chi_{\frac{1}{2}-\frac{1}{2}}&=-\frac{1}{2}\chi_{\frac{1}{2}-\frac{1}{2}},\\
\hat{S}_{-1}\chi_{\frac{1}{2}\frac{1}{2}}&=\frac{1}{\sqrt{2}}\chi_{\frac{1}{2}-\frac{1}{2}},&\hat{S}_{-1}\chi_{\frac{1}{2}-\frac{1}{2}}&=0. \tag*{(1236)}
\end{aligned} S ^ + 1 χ 2 1 2 1 S ^ 0 χ 2 1 2 1 S ^ − 1 χ 2 1 2 1 = 0 , = 2 1 χ 2 1 2 1 , = 2 1 χ 2 1 − 2 1 , S ^ + 1 χ 2 1 − 2 1 S ^ 0 χ 2 1 − 2 1 S ^ − 1 χ 2 1 − 2 1 = − 2 1 χ 2 1 2 1 , = − 2 1 χ 2 1 − 2 1 , = 0. (1236)
According to Eq. (1235) , matrix elements of the operator S ^ ⋅ a \hat{\mathbf{S}}\cdot\mathbf{a} S ^ ⋅ a for any vector a \mathbf{a} a have the form
[ S ^ ⋅ a ] σ σ ′ = χ 1 2 σ † S ^ ⋅ a χ 1 2 σ ′ = ( − 1 ) 1 − μ 3 2 C 1 2 μ 1 2 σ 1 σ ′ a − μ , (1237) [\hat{\mathbf{S}}\cdot\mathbf{a}]_{\sigma\sigma'}=\chi_{\frac{1}{2}\sigma}^{\dagger}\hat{\mathbf{S}}\cdot\mathbf{a}\chi_{\frac{1}{2}\sigma'}=(-1)^{1-\mu}\frac{\sqrt{3}}{2}C_{\frac{1}{2}\mu\frac{1}{2}\sigma}^{1\sigma'}a_{-\mu}, \tag*{(1237)} [ S ^ ⋅ a ] σ σ ′ = χ 2 1 σ † S ^ ⋅ a χ 2 1 σ ′ = ( − 1 ) 1 − μ 2 3 C 2 1 μ 2 1 σ 1 σ ′ a − μ , (1237)
where a − μ ( μ = ± 1 , 0 ) a_{-\mu}(\mu=\pm1,0) a − μ ( μ = ± 1 , 0 ) are covariant spherical components of a \mathbf{a} a (Sec. § 9.6 ).
Transformation of the Basis Functions Under Rotations of the Coordinate System
The basis functions χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m are covariant. Under rotations S → S ′ S\rightarrow S' S → S ′ of the coordinate system these functions are transformed by the rotation operator D ^ 1 2 ( α , β , γ ) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma) D ^ 2 1 ( α , β , γ ) (Eq. (326) ) if rotations are described by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ , or by the rotation operator U ^ 1 2 ( ω ; Θ , Φ ) \hat{U}^{\frac{1}{2}}(\omega;\Theta,\Phi) U ^ 2 1 ( ω ; Θ , Φ ) (Eq. (330) ), if rotations are defined by the rotation axis n ( θ , Φ ) \mathbf{n}(\theta,\Phi) n ( θ , Φ ) and the rotation angle ω \omega ω :
χ 1 2 m ′ ′ = D ^ 1 2 ( α , β , γ ) χ 1 2 m ′ = ∑ m χ 1 2 m D m m ′ 1 2 ( α , β , γ ) , \chi'_{\frac{1}{2}m'}=\hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma)\chi_{\frac{1}{2}m'}=\sum_m\chi_{\frac{1}{2}m}D^{\frac{1}{2}}_{mm'}(\alpha,\beta,\gamma), χ 2 1 m ′ ′ = D ^ 2 1 ( α , β , γ ) χ 2 1 m ′ = m ∑ χ 2 1 m D m m ′ 2 1 ( α , β , γ ) ,
χ 1 2 m ′ ′ = U ^ 1 2 ( ω ; θ , Φ ) χ 1 2 m ′ = ∑ m χ 1 2 m U m m ′ 1 2 ( ω ; θ , Φ ) . (1238) \chi'_{\frac{1}{2}m'}=\hat{U}^{\frac{1}{2}}(\omega;\theta,\Phi)\chi_{\frac{1}{2}m'}=\sum_m\chi_{\frac{1}{2}m}U^{\frac{1}{2}}_{mm'}(\omega;\theta,\Phi). \tag*{(1238)} χ 2 1 m ′ ′ = U ^ 2 1 ( ω ; θ , Φ ) χ 2 1 m ′ = m ∑ χ 2 1 m U m m ′ 2 1 ( ω ; θ , Φ ) . (1238)
where D ˉ m m ′ 1 2 ( α , β , γ ) \bar{D}^{\frac{1}{2}}_{mm'}(\alpha,\beta,\gamma) D ˉ m m ′ 2 1 ( α , β , γ ) are the Wigner D-functions (Chap. 4), and U m m ′ 1 2 U^{\frac{1}{2}}_{mm'} U m m ′ 2 1 are defined in Sec. 4.5.
A function χ 1 2 m ′ ′ \chi'_{\frac{1}{2}m'} χ 2 1 m ′ ′ describes a state in which a particle of spin 1 2 \frac{1}{2} 2 1 has the spin projection m ′ m' m ′ on the new z ′ z' z ′ -axis.
The functions χ 1 2 m ′ ′ \chi'_{\frac{1}{2}m'} χ 2 1 m ′ ′ are eigenfunctions of the operators S ^ ′ 2 \hat{S}'^2 S ^ ′2 and S ^ z ′ \hat{S}'_z S ^ z ′ , S ^ ′ \hat{\mathbf{S}}' S ^ ′ being the spin operator in the rotated coordinate system (Eqs. (334) and (2.5.41)):
S ^ ′ 2 χ 1 2 m ′ ′ = 3 4 χ 1 2 m ′ ′ , S ^ z ′ χ 1 2 m ′ ′ = m ′ χ 1 2 m ′ ′ . (1239) \hat{\mathbf{S}}'^2\chi'_{\frac{1}{2}m'}=\frac{3}{4}\chi'_{\frac{1}{2}m'},\qquad
\hat{S}'_z\chi'_{\frac{1}{2}m'}=m'\chi'_{\frac{1}{2}m'}. \tag*{(1239)} S ^ ′2 χ 2 1 m ′ ′ = 4 3 χ 2 1 m ′ ′ , S ^ z ′ χ 2 1 m ′ ′ = m ′ χ 2 1 m ′ ′ . (1239)
The explicit form of the basis spin functions in the rotated coordinate system is
χ 1 2 1 2 ′ = ( cos β 2 e − i α + γ 2 sin β 2 e i α − γ 2 ) , χ 1 2 − 1 2 ′ = ( − sin β 2 e − i α − γ 2 cos β 2 e i α + γ 2 ) . (1240) \chi'_{\frac{1}{2}\frac{1}{2}}=\begin{pmatrix}\cos\frac{\beta}{2}e^{-i\frac{\alpha+\gamma}{2}}\\ \sin\frac{\beta}{2}e^{i\frac{\alpha-\gamma}{2}}\end{pmatrix},\qquad
\chi'_{\frac{1}{2}-\frac{1}{2}}=\begin{pmatrix}-\sin\frac{\beta}{2}e^{-i\frac{\alpha-\gamma}{2}}\\ \cos\frac{\beta}{2}e^{i\frac{\alpha+\gamma}{2}}\end{pmatrix}. \tag*{(1240)} χ 2 1 2 1 ′ = ( cos 2 β e − i 2 α + γ sin 2 β e i 2 α − γ ) , χ 2 1 − 2 1 ′ = ( − sin 2 β e − i 2 α − γ cos 2 β e i 2 α + γ ) . (1240)
if the rotation is specified by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ and
χ 1 2 1 2 ′ = ( cos ω 2 − i sin ω 2 cos θ − i sin ω 2 sin θ e i ϕ ) , χ 1 2 − 1 2 ′ = ( − i sin ω 2 sin θ e − i ϕ cos ω 2 + i sin ω 2 cos θ ) . (1241) \chi'_{\frac{1}{2}\frac{1}{2}}=\begin{pmatrix}\cos\frac{\omega}{2}-i\sin\frac{\omega}{2}\cos\theta\\ -i\sin\frac{\omega}{2}\sin\theta e^{i\phi}\end{pmatrix},\qquad
\chi'_{\frac{1}{2}-\frac{1}{2}}=\begin{pmatrix}-i\sin\frac{\omega}{2}\sin\theta e^{-i\phi}\\ \cos\frac{\omega}{2}+i\sin\frac{\omega}{2}\cos\theta\end{pmatrix}. \tag*{(1241)} χ 2 1 2 1 ′ = ( cos 2 ω − i sin 2 ω cos θ − i sin 2 ω sin θ e i ϕ ) , χ 2 1 − 2 1 ′ = ( − i sin 2 ω sin θ e − i ϕ cos 2 ω + i sin 2 ω cos θ ) . (1241)
in terms of the angles ω , θ , Φ \omega,\theta,\Phi ω , θ , Φ .
The functions χ 1 2 m ′ \chi'_{\frac{1}{2}m} χ 2 1 m ′ , satisfy the orthonormality and completeness conditions (6.2.6), (6.2.7) as well as the relations (6.2.9)-(6.2.13) in which the spin operator S ^ \hat{\mathbf{S}} S ^ should be replaced by S ^ ′ \hat{\mathbf{S}}' S ^ ′ .
Helicity Basis Functions
The helicity basis functions χ 1 2 λ ( ϑ , ϕ ) \chi_{\frac{1}{2}\lambda}(\vartheta,\phi) χ 2 1 λ ( ϑ , ϕ ) (λ = ± 1 2 \lambda=\pm\frac{1}{2} λ = ± 2 1 ) describe states in which the spin projection on the linear momentum direction n ( ϑ , ϕ ) ≡ p / p \mathbf{n}(\vartheta,\phi)\equiv\mathbf{p}/p n ( ϑ , ϕ ) ≡ p / p is equal to λ \lambda λ . These functions are eigenfunctions of the operators S ^ 2 \hat{\mathbf{S}}^2 S ^ 2 and S ^ ⋅ n \hat{\mathbf{S}}\cdot\mathbf{n} S ^ ⋅ n .
S ^ 2 χ 1 2 λ ( ϑ , ϕ ) = 3 4 χ 1 2 λ ( ϑ , ϕ ) , S ^ ⋅ n χ 1 2 λ ( ϑ , ϕ ) = λ χ 1 2 λ ( ϑ , ϕ ) . (1242) \hat{\mathbf{S}}^2\chi_{\frac{1}{2}\lambda}(\vartheta,\phi)=\frac{3}{4}\chi_{\frac{1}{2}\lambda}(\vartheta,\phi),\qquad
\hat{\mathbf{S}}\cdot\mathbf{n}\chi_{\frac{1}{2}\lambda}(\vartheta,\phi)=\lambda\chi_{\frac{1}{2}\lambda}(\vartheta,\phi). \tag*{(1242)} S ^ 2 χ 2 1 λ ( ϑ , ϕ ) = 4 3 χ 2 1 λ ( ϑ , ϕ ) , S ^ ⋅ n χ 2 1 λ ( ϑ , ϕ ) = λ χ 2 1 λ ( ϑ , ϕ ) . (1242)
According to Eq. (1238) the helicity functions χ 1 2 λ ( ϑ , ϕ ) \chi_{\frac{1}{2}\lambda}(\vartheta,\phi) χ 2 1 λ ( ϑ , ϕ ) are related to the basis functions χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m by
χ 1 2 λ ( ϑ , ϕ ) = ∑ m D m λ 1 2 ( ϕ , ϑ , 0 ) χ 1 2 m , χ 1 2 m = ∑ λ D − λ − m 1 2 ( 0 , ϑ , ϕ ) χ 1 2 λ ( ϑ , ϕ ) . (1243) \chi_{\frac{1}{2}\lambda}(\vartheta,\phi)=\sum_m D^{\frac{1}{2}}_{m\lambda}(\phi,\vartheta,0)\chi_{\frac{1}{2}m},\qquad
\chi_{\frac{1}{2}m}=\sum_\lambda D^{\frac{1}{2}}_{-\lambda-m}(0,\vartheta,\phi)\chi_{\frac{1}{2}\lambda}(\vartheta,\phi). \tag*{(1243)} χ 2 1 λ ( ϑ , ϕ ) = m ∑ D mλ 2 1 ( ϕ , ϑ , 0 ) χ 2 1 m , χ 2 1 m = λ ∑ D − λ − m 2 1 ( 0 , ϑ , ϕ ) χ 2 1 λ ( ϑ , ϕ ) . (1243)
For Hermitian adjoint functions Eqs. (1243) yield
χ 1 2 λ † ( ϑ , ϕ ) = ∑ m ( − 1 ) λ − m D − m − λ 1 2 ( ϕ , ϑ , 0 ) χ 1 2 m † , χ 1 2 m † = ∑ λ ( − 1 ) λ − m D λ m 1 2 ( 0 , ϑ , ϕ ) χ 1 2 λ † ( ϑ , ϕ ) . (1244) \chi^\dagger_{\frac{1}{2}\lambda}(\vartheta,\phi)=\sum_m(-1)^{\lambda-m}D^{\frac{1}{2}}_{-m-\lambda}(\phi,\vartheta,0)\chi^\dagger_{\frac{1}{2}m},
\qquad
\chi^\dagger_{\frac{1}{2}m}=\sum_\lambda(-1)^{\lambda-m}D^{\frac{1}{2}}_{\lambda m}(0,\vartheta,\phi)\chi^\dagger_{\frac{1}{2}\lambda}(\vartheta,\phi). \tag*{(1244)} χ 2 1 λ † ( ϑ , ϕ ) = m ∑ ( − 1 ) λ − m D − m − λ 2 1 ( ϕ , ϑ , 0 ) χ 2 1 m † , χ 2 1 m † = λ ∑ ( − 1 ) λ − m D λm 2 1 ( 0 , ϑ , ϕ ) χ 2 1 λ † ( ϑ , ϕ ) . (1244)
The explicit form of the helicity functions is given by
χ 1 2 1 2 ( ϑ , ϕ ) = ( cos ϑ 2 e − i ϕ 2 sin ϑ 2 e i ϕ 2 ) , χ 1 2 − 1 2 ( ϑ , ϕ ) = ( − sin ϑ 2 e − i ϕ 2 cos ϑ 2 e i ϕ 2 ) . (1245) \chi_{\frac{1}{2}\frac{1}{2}}(\vartheta,\phi)=\begin{pmatrix}\cos\frac{\vartheta}{2}e^{-i\frac{\phi}{2}}\\ \sin\frac{\vartheta}{2}e^{i\frac{\phi}{2}}\end{pmatrix},\qquad
\chi_{\frac{1}{2}-\frac{1}{2}}(\vartheta,\phi)=\begin{pmatrix}-\sin\frac{\vartheta}{2}e^{-i\frac{\phi}{2}}\\ \cos\frac{\vartheta}{2}e^{i\frac{\phi}{2}}\end{pmatrix}. \tag*{(1245)} χ 2 1 2 1 ( ϑ , ϕ ) = ( cos 2 ϑ e − i 2 ϕ sin 2 ϑ e i 2 ϕ ) , χ 2 1 − 2 1 ( ϑ , ϕ ) = ( − sin 2 ϑ e − i 2 ϕ cos 2 ϑ e i 2 ϕ ) . (1245)
χ 1 2 1 2 † ( ϑ , ϕ ) = ( cos ϑ 2 e i ϕ 2 , sin ϑ 2 e − i ϕ 2 ) , χ 1 2 − 1 2 † ( ϑ , ϕ ) = ( − sin ϑ 2 e i ϕ 2 , cos ϑ 2 e − i ϕ 2 ) . (1246) \chi^\dagger_{\frac{1}{2}\frac{1}{2}}(\vartheta,\phi)=\left(\cos\frac{\vartheta}{2}e^{i\frac{\phi}{2}},\sin\frac{\vartheta}{2}e^{-i\frac{\phi}{2}}\right),\qquad
\chi^\dagger_{\frac{1}{2}-\frac{1}{2}}(\vartheta,\phi)=\left(-\sin\frac{\vartheta}{2}e^{i\frac{\phi}{2}},\cos\frac{\vartheta}{2}e^{-i\frac{\phi}{2}}\right). \tag*{(1246)} χ 2 1 2 1 † ( ϑ , ϕ ) = ( cos 2 ϑ e i 2 ϕ , sin 2 ϑ e − i 2 ϕ ) , χ 2 1 − 2 1 † ( ϑ , ϕ ) = ( − sin 2 ϑ e i 2 ϕ , cos 2 ϑ e − i 2 ϕ ) . (1246)
The helicity functions χ 1 2 λ ( ϑ , ϕ ) \chi_{\frac{1}{2}\lambda}(\vartheta,\phi) χ 2 1 λ ( ϑ , ϕ ) as well as χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m constitute an orthonormalised basis. The orthonormality condition for the helicity functions may be written as
χ 1 2 λ ′ † ( ϑ , ϕ ) χ 1 2 λ ( ϑ , ϕ ) = δ λ ′ λ . (1247) \chi^\dagger_{\frac{1}{2}\lambda'}(\vartheta,\phi)\chi_{\frac{1}{2}\lambda}(\vartheta,\phi)=\delta_{\lambda'\lambda}. \tag*{(1247)} χ 2 1 λ ′ † ( ϑ , ϕ ) χ 2 1 λ ( ϑ , ϕ ) = δ λ ′ λ . (1247)
The completeness condition is
∑ λ χ 1 2 λ ( ϑ , ϕ ) χ 1 2 λ † ( ϑ , ϕ ) = I ^ . (1248) \sum_\lambda\chi_{\frac{1}{2}\lambda}(\vartheta,\phi)\chi^\dagger_{\frac{1}{2}\lambda}(\vartheta,\phi)=\hat{I}. \tag*{(1248)} λ ∑ χ 2 1 λ ( ϑ , ϕ ) χ 2 1 λ † ( ϑ , ϕ ) = I ^ . (1248)
The expansion of products of the helicity functions has the form
χ 1 2 λ † ( ϑ , φ ) χ 1 2 λ ′ ( ϑ , φ ) = 1 2 δ λ λ ′ I ^ + 3 ∑ μ , ν C 1 2 1 2 λ λ ′ μ D μ ν 1 ( φ , ϑ , 0 ) S ^ ν . (1249) \chi_{\frac{1}{2}\lambda}^{\dagger}(\vartheta,\varphi)\chi_{\frac{1}{2}\lambda'}(\vartheta,\varphi)=\frac{1}{2}\delta_{\lambda\lambda'}\hat{I}+\sqrt{3}\sum_{\mu,\nu}C_{\frac{1}{2}\frac{1}{2}}^{\lambda\lambda'\mu}D_{\mu\nu}^{1}(\varphi,\vartheta,0)\hat{S}_{\nu}. \tag*{(1249)} χ 2 1 λ † ( ϑ , φ ) χ 2 1 λ ′ ( ϑ , φ ) = 2 1 δ λ λ ′ I ^ + 3 μ , ν ∑ C 2 1 2 1 λ λ ′ μ D μν 1 ( φ , ϑ , 0 ) S ^ ν . (1249)
Explicitly, this gives
χ 1 2 1 2 ( ϑ , φ ) χ 1 2 1 2 † ( ϑ , φ ) = 1 2 ( 1 + cos ϑ sin ϑ e − i φ sin ϑ e i φ 1 − cos ϑ ) = 1 2 I ^ + n ⋅ S ^ , \chi_{\frac{1}{2}\frac{1}{2}}(\vartheta,\varphi)\chi_{\frac{1}{2}\frac{1}{2}}^{\dagger}(\vartheta,\varphi)=\frac{1}{2}\begin{pmatrix}1+\cos\vartheta&\sin\vartheta e^{-i\varphi}\\\sin\vartheta e^{i\varphi}&1-\cos\vartheta\end{pmatrix}=\frac{1}{2}\hat{I}+\mathbf{n}\cdot\hat{\mathbf{S}}, χ 2 1 2 1 ( ϑ , φ ) χ 2 1 2 1 † ( ϑ , φ ) = 2 1 ( 1 + cos ϑ sin ϑ e i φ sin ϑ e − i φ 1 − cos ϑ ) = 2 1 I ^ + n ⋅ S ^ ,
χ 1 2 1 2 ( ϑ , φ ) χ 1 2 − 1 2 † ( ϑ , φ ) = 1 2 ( ( cos ϑ + 1 ) e − i φ − sin ϑ ( cos ϑ − 1 ) e i φ sin ϑ ) = − 2 ∑ ν D ν 1 1 ( φ , ϑ , 0 ) S ^ ν , \chi_{\frac{1}{2}\frac{1}{2}}(\vartheta,\varphi)\chi_{\frac{1}{2}-\frac{1}{2}}^{\dagger}(\vartheta,\varphi)=\frac{1}{2}\begin{pmatrix}(\cos\vartheta+1)e^{-i\varphi}&-\sin\vartheta\\(\cos\vartheta-1)e^{i\varphi}&\sin\vartheta\end{pmatrix}=-\sqrt{2}\sum_{\nu}D_{\nu1}^{1}(\varphi,\vartheta,0)\hat{S}_{\nu}, χ 2 1 2 1 ( ϑ , φ ) χ 2 1 − 2 1 † ( ϑ , φ ) = 2 1 ( ( cos ϑ + 1 ) e − i φ ( cos ϑ − 1 ) e i φ − sin ϑ sin ϑ ) = − 2 ν ∑ D ν 1 1 ( φ , ϑ , 0 ) S ^ ν ,
χ 1 2 − 1 2 ( ϑ , φ ) χ 1 2 1 2 † ( ϑ , φ ) = 1 2 ( − sin ϑ ( cos ϑ − 1 ) e − i φ ( cos ϑ + 1 ) e i φ sin ϑ ) = 2 ∑ ν D ν − 1 1 ( φ , ϑ , 0 ) S ^ ν , \chi_{\frac{1}{2}-\frac{1}{2}}(\vartheta,\varphi)\chi_{\frac{1}{2}\frac{1}{2}}^{\dagger}(\vartheta,\varphi)=\frac{1}{2}\begin{pmatrix}-\sin\vartheta&(\cos\vartheta-1)e^{-i\varphi}\\(\cos\vartheta+1)e^{i\varphi}&\sin\vartheta\end{pmatrix}=\sqrt{2}\sum_{\nu}D_{\nu-1}^{1}(\varphi,\vartheta,0)\hat{S}_{\nu}, χ 2 1 − 2 1 ( ϑ , φ ) χ 2 1 2 1 † ( ϑ , φ ) = 2 1 ( − sin ϑ ( cos ϑ + 1 ) e i φ ( cos ϑ − 1 ) e − i φ sin ϑ ) = 2 ν ∑ D ν − 1 1 ( φ , ϑ , 0 ) S ^ ν ,
χ 1 2 − 1 2 ( ϑ , φ ) χ 1 2 − 1 2 † ( ϑ , φ ) = 1 2 ( 1 − cos ϑ − sin ϑ e − i φ − sin ϑ e i φ 1 + cos ϑ ) = 1 2 I ^ − n ⋅ S ^ . (1250) \chi_{\frac{1}{2}-\frac{1}{2}}(\vartheta,\varphi)\chi_{\frac{1}{2}-\frac{1}{2}}^{\dagger}(\vartheta,\varphi)=\frac{1}{2}\begin{pmatrix}1-\cos\vartheta&-\sin\vartheta e^{-i\varphi}\\-\sin\vartheta e^{i\varphi}&1+\cos\vartheta\end{pmatrix}=\frac{1}{2}\hat{I}-\mathbf{n}\cdot\hat{\mathbf{S}}. \tag*{(1250)} χ 2 1 − 2 1 ( ϑ , φ ) χ 2 1 − 2 1 † ( ϑ , φ ) = 2 1 ( 1 − cos ϑ − sin ϑ e i φ − sin ϑ e − i φ 1 + cos ϑ ) = 2 1 I ^ − n ⋅ S ^ . (1250)
Matrix elements of the spherical components of the spin operator S ^ μ ( μ = ± 1 , 0 ) \hat{S}_{\mu}(\mu=\pm1,0) S ^ μ ( μ = ± 1 , 0 ) between the helicity states are given by
χ 1 2 λ ′ † ( ϑ , φ ) S ^ μ χ 1 2 λ ( ϑ , φ ) = ( − 1 ) μ + ν 3 2 C 1 2 1 2 λ λ ′ μ D 1 μ − ν 1 ( φ , ϑ , 0 ) . (1251) \chi_{\frac{1}{2}\lambda'}^{\dagger}(\vartheta,\varphi)\hat{S}_{\mu}\chi_{\frac{1}{2}\lambda}(\vartheta,\varphi)=(-1)^{\mu+\nu}\frac{\sqrt{3}}{2}C_{\frac{1}{2}\frac{1}{2}}^{\lambda\lambda'\mu}D_{1\mu-\nu}^{1}(\varphi,\vartheta,0). \tag*{(1251)} χ 2 1 λ ′ † ( ϑ , φ ) S ^ μ χ 2 1 λ ( ϑ , φ ) = ( − 1 ) μ + ν 2 3 C 2 1 2 1 λ λ ′ μ D 1 μ − ν 1 ( φ , ϑ , 0 ) . (1251)
or, in a detailed form
⟨ + ∣ S ^ + 1 ∣ + ⟩ = − sin ϑ 2 2 e i φ , ⟨ + ∣ S ^ + 1 ∣ − ⟩ = − 1 + cos ϑ 2 2 e i φ , ⟨ + ∣ S ^ 0 ∣ + ⟩ = 1 2 cos ϑ , ⟨ + ∣ S ^ 0 ∣ − ⟩ = − 1 2 sin ϑ , ⟨ + ∣ S ^ − 1 ∣ + ⟩ = sin ϑ 2 2 e − i φ , ⟨ + ∣ S ^ − 1 ∣ − ⟩ = − 1 − cos ϑ 2 2 e − i φ , ⟨ − ∣ S ^ + 1 ∣ + ⟩ = 1 − cos ϑ 2 2 e i φ , ⟨ − ∣ S ^ + 1 ∣ − ⟩ = sin ϑ 2 2 e i φ , ⟨ − ∣ S ^ 0 ∣ + ⟩ = − 1 2 sin ϑ , ⟨ − ∣ S ^ 0 ∣ − ⟩ = − 1 2 cos ϑ , ⟨ − ∣ S ^ − 1 ∣ + ⟩ = 1 + cos ϑ 2 2 e − i φ , ⟨ − ∣ S ^ − 1 ∣ − ⟩ = − sin ϑ 2 2 e − i φ . (1252) \begin{aligned}
\langle+|\hat{S}_{+1}|+\rangle&=-\frac{\sin\vartheta}{2\sqrt{2}}e^{i\varphi},&
\langle+|\hat{S}_{+1}|-\rangle&=-\frac{1+\cos\vartheta}{2\sqrt{2}}e^{i\varphi},\\
\langle+|\hat{S}_{0}|+\rangle&=\frac{1}{2}\cos\vartheta,&
\langle+|\hat{S}_{0}|-\rangle&=-\frac{1}{2}\sin\vartheta,\\
\langle+|\hat{S}_{-1}|+\rangle&=\frac{\sin\vartheta}{2\sqrt{2}}e^{-i\varphi},&
\langle+|\hat{S}_{-1}|-\rangle&=-\frac{1-\cos\vartheta}{2\sqrt{2}}e^{-i\varphi},\\
\langle-|\hat{S}_{+1}|+\rangle&=\frac{1-\cos\vartheta}{2\sqrt{2}}e^{i\varphi},&
\langle-|\hat{S}_{+1}|-\rangle&=\frac{\sin\vartheta}{2\sqrt{2}}e^{i\varphi},\\
\langle-|\hat{S}_{0}|+\rangle&=-\frac{1}{2}\sin\vartheta,&
\langle-|\hat{S}_{0}|-\rangle&=-\frac{1}{2}\cos\vartheta,\\
\langle-|\hat{S}_{-1}|+\rangle&=\frac{1+\cos\vartheta}{2\sqrt{2}}e^{-i\varphi},&
\langle-|\hat{S}_{-1}|-\rangle&=-\frac{\sin\vartheta}{2\sqrt{2}}e^{-i\varphi}. \tag*{(1252)}
\end{aligned} ⟨ + ∣ S ^ + 1 ∣ + ⟩ ⟨ + ∣ S ^ 0 ∣ + ⟩ ⟨ + ∣ S ^ − 1 ∣ + ⟩ ⟨ − ∣ S ^ + 1 ∣ + ⟩ ⟨ − ∣ S ^ 0 ∣ + ⟩ ⟨ − ∣ S ^ − 1 ∣ + ⟩ = − 2 2 sin ϑ e i φ , = 2 1 cos ϑ , = 2 2 sin ϑ e − i φ , = 2 2 1 − cos ϑ e i φ , = − 2 1 sin ϑ , = 2 2 1 + cos ϑ e − i φ , ⟨ + ∣ S ^ + 1 ∣ − ⟩ ⟨ + ∣ S ^ 0 ∣ − ⟩ ⟨ + ∣ S ^ − 1 ∣ − ⟩ ⟨ − ∣ S ^ + 1 ∣ − ⟩ ⟨ − ∣ S ^ 0 ∣ − ⟩ ⟨ − ∣ S ^ − 1 ∣ − ⟩ = − 2 2 1 + cos ϑ e i φ , = − 2 1 sin ϑ , = − 2 2 1 − cos ϑ e − i φ , = 2 2 sin ϑ e i φ , = − 2 1 cos ϑ , = − 2 2 sin ϑ e − i φ . (1252)
Here we use the abbreviations
⟨ + ∣ S ^ μ ∣ + ⟩ ≡ χ 1 2 1 2 † ( ϑ , φ ) S ^ μ χ 1 2 1 2 ( ϑ , φ ) \langle+|\hat{S}_{\mu}|+\rangle\equiv\chi_{\frac{1}{2}\frac{1}{2}}^{\dagger}(\vartheta,\varphi)\hat{S}_{\mu}\chi_{\frac{1}{2}\frac{1}{2}}(\vartheta,\varphi) ⟨ + ∣ S ^ μ ∣ + ⟩ ≡ χ 2 1 2 1 † ( ϑ , φ ) S ^ μ χ 2 1 2 1 ( ϑ , φ )
⟨ + ∣ S ^ μ ∣ − ⟩ ≡ χ 1 2 1 2 † ( ϑ , φ ) S ^ μ χ 1 2 − 1 2 ( ϑ , φ ) , etc. \langle+|\hat{S}_{\mu}|-\rangle\equiv\chi_{\frac{1}{2}\frac{1}{2}}^{\dagger}(\vartheta,\varphi)\hat{S}_{\mu}\chi_{\frac{1}{2}-\frac{1}{2}}(\vartheta,\varphi),\ \text{etc.} ⟨ + ∣ S ^ μ ∣ − ⟩ ≡ χ 2 1 2 1 † ( ϑ , φ ) S ^ μ χ 2 1 − 2 1 ( ϑ , φ ) , etc.
Matrix elements of cartesian components of the spin matrices S ^ i ( i = x , y , z ) \hat{S}_{i}(i=x,y,z) S ^ i ( i = x , y , z ) have the form
⟨ + ∣ S ^ x ∣ + ⟩ = 1 2 sin ϑ cos φ , ⟨ + ∣ S ^ x ∣ − ⟩ = 1 2 ( cos ϑ cos φ + i sin φ ) , ⟨ + ∣ S ^ y ∣ + ⟩ = 1 2 sin ϑ sin φ , ⟨ + ∣ S ^ y ∣ − ⟩ = 1 2 ( cos ϑ sin φ − i cos φ ) , ⟨ + ∣ S ^ z ∣ + ⟩ = 1 2 cos ϑ , ⟨ + ∣ S ^ z ∣ − ⟩ = − 1 2 sin ϑ , ⟨ − ∣ S ^ x ∣ + ⟩ = 1 2 ( cos ϑ cos φ − i sin φ ) , ⟨ − ∣ S ^ x ∣ − ⟩ = − 1 2 sin ϑ cos φ , ⟨ − ∣ S ^ y ∣ + ⟩ = 1 2 ( cos ϑ sin φ + i cos φ ) , ⟨ − ∣ S ^ y ∣ − ⟩ = − 1 2 sin ϑ sin φ , ⟨ − ∣ S ^ z ∣ + ⟩ = − 1 2 sin ϑ , ⟨ − ∣ S ^ z ∣ − ⟩ = − 1 2 cos ϑ (1253) \begin{aligned}
\langle+|\hat{S}_{x}|+\rangle&=\frac{1}{2}\sin\vartheta\cos\varphi,&
\langle+|\hat{S}_{x}|-\rangle&=\frac{1}{2}(\cos\vartheta\cos\varphi+i\sin\varphi),\\
\langle+|\hat{S}_{y}|+\rangle&=\frac{1}{2}\sin\vartheta\sin\varphi,&
\langle+|\hat{S}_{y}|-\rangle&=\frac{1}{2}(\cos\vartheta\sin\varphi-i\cos\varphi),\\
\langle+|\hat{S}_{z}|+\rangle&=\frac{1}{2}\cos\vartheta,&
\langle+|\hat{S}_{z}|-\rangle&=-\frac{1}{2}\sin\vartheta,\\
\langle-|\hat{S}_{x}|+\rangle&=\frac{1}{2}(\cos\vartheta\cos\varphi-i\sin\varphi),&
\langle-|\hat{S}_{x}|-\rangle&=-\frac{1}{2}\sin\vartheta\cos\varphi,\\
\langle-|\hat{S}_{y}|+\rangle&=\frac{1}{2}(\cos\vartheta\sin\varphi+i\cos\varphi),&
\langle-|\hat{S}_{y}|-\rangle&=-\frac{1}{2}\sin\vartheta\sin\varphi,\\
\langle-|\hat{S}_{z}|+\rangle&=-\frac{1}{2}\sin\vartheta,&
\langle-|\hat{S}_{z}|-\rangle&=-\frac{1}{2}\cos\vartheta \tag*{(1253)}
\end{aligned} ⟨ + ∣ S ^ x ∣ + ⟩ ⟨ + ∣ S ^ y ∣ + ⟩ ⟨ + ∣ S ^ z ∣ + ⟩ ⟨ − ∣ S ^ x ∣ + ⟩ ⟨ − ∣ S ^ y ∣ + ⟩ ⟨ − ∣ S ^ z ∣ + ⟩ = 2 1 sin ϑ cos φ , = 2 1 sin ϑ sin φ , = 2 1 cos ϑ , = 2 1 ( cos ϑ cos φ − i sin φ ) , = 2 1 ( cos ϑ sin φ + i cos φ ) , = − 2 1 sin ϑ , ⟨ + ∣ S ^ x ∣ − ⟩ ⟨ + ∣ S ^ y ∣ − ⟩ ⟨ + ∣ S ^ z ∣ − ⟩ ⟨ − ∣ S ^ x ∣ − ⟩ ⟨ − ∣ S ^ y ∣ − ⟩ ⟨ − ∣ S ^ z ∣ − ⟩ = 2 1 ( cos ϑ cos φ + i sin φ ) , = 2 1 ( cos ϑ sin φ − i cos φ ) , = − 2 1 sin ϑ , = − 2 1 sin ϑ cos φ , = − 2 1 sin ϑ sin φ , = − 2 1 cos ϑ (1253)
Note that the diagonal matrix elements of the spin operator may be written as
χ 1 2 λ † ( ϑ , φ ) S ^ χ 1 2 λ ( ϑ , φ ) = λ n ( ϑ , φ ) . (1254) \chi_{\frac{1}{2}\lambda}^{\dagger}(\vartheta,\varphi)\hat{S}\chi_{\frac{1}{2}\lambda}(\vartheta,\varphi)=\lambda\mathbf{n}(\vartheta,\varphi). \tag*{(1254)} χ 2 1 λ † ( ϑ , φ ) S ^ χ 2 1 λ ( ϑ , φ ) = λ n ( ϑ , φ ) . (1254)
General Spin Functions for S = 1 2 S=\frac{1}{2} S = 2 1
An arbitrary spin function χ 1 2 \chi_{\frac{1}{2}} χ 2 1 of a particle of spin 1 2 \frac{1}{2} 2 1 may be expanded in terms of the basis spin functions χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m ,
χ 1 2 = ∑ m = − 1 2 1 2 a m χ 1 2 m = ( a 1 2 a − 1 2 ) . (1255) \chi_{\frac{1}{2}}=\sum_{m=-\frac{1}{2}}^{\frac{1}{2}}a^{m}\chi_{\frac{1}{2}m}=\begin{pmatrix}a^{\frac{1}{2}}\\a^{-\frac{1}{2}}\end{pmatrix}. \tag*{(1255)} χ 2 1 = m = − 2 1 ∑ 2 1 a m χ 2 1 m = ( a 2 1 a − 2 1 ) . (1255)
An analogous expansion of the Hermitian conjugate function χ 1 2 † \chi_{\frac{1}{2}}^\dagger χ 2 1 † is
χ 1 2 † = ∑ m = − 1 2 1 2 a m ∗ χ 1 2 m † = ( a 1 2 ∗ , a − 1 2 ∗ ) . (1256) \chi_{\frac{1}{2}}^\dagger=\sum_{m=-\frac{1}{2}}^{\frac{1}{2}}a^{m*}\chi_{\frac{1}{2}m}^\dagger=\left(a^{\frac{1}{2}*},a^{-\frac{1}{2}*}\right). \tag*{(1256)} χ 2 1 † = m = − 2 1 ∑ 2 1 a m ∗ χ 2 1 m † = ( a 2 1 ∗ , a − 2 1 ∗ ) . (1256)
Here a m a^m a m (m = ± 1 2 m=\pm\frac{1}{2} m = ± 2 1 ) is the contravariant component of the spinor χ 1 2 \chi_{\frac{1}{2}} χ 2 1 (see below). The normalization condition
χ 1 2 † χ 1 2 = 1. (1257) \chi_{\frac{1}{2}}^\dagger\chi_{\frac{1}{2}}=1. \tag*{(1257)} χ 2 1 † χ 2 1 = 1. (1257)
imposes the following restriction
∣ a 1 2 ∣ 2 + ∣ a − 1 2 ∣ 2 = 1. \left|a^{\frac{1}{2}}\right|^2+\left|a^{-\frac{1}{2}}\right|^2=1. a 2 1 2 + a − 2 1 2 = 1.
The coefficient a m a^m a m may be interpreted as the probability amplitude that the spin projection on the z z z -axis for the particle in a given state χ 1 2 \chi_{\frac{1}{2}} χ 2 1 is equal to m m m .
(a) Special cases
(b) Spin is directed along the z z z -axis
a 1 2 = 1 , a − 1 2 = 0 , χ 1 2 = χ 1 2 1 2 = ( 1 0 ) . (1258) a^{\frac{1}{2}}=1,\ a^{-\frac{1}{2}}=0,\quad\chi_{\frac{1}{2}}=\chi_{\frac{1}{2}\frac{1}{2}}=\begin{pmatrix}1\\0\end{pmatrix}. \tag*{(1258)} a 2 1 = 1 , a − 2 1 = 0 , χ 2 1 = χ 2 1 2 1 = ( 1 0 ) . (1258)
(c) Spin is directed along the negative z z z -axis
a 1 2 = 0 , a − 1 2 = 1 , χ 1 2 = χ 1 2 − 1 2 = ( 0 1 ) . (1259) a^{\frac{1}{2}}=0,\ a^{-\frac{1}{2}}=1,\quad\chi_{\frac{1}{2}}=\chi_{\frac{1}{2}-\frac{1}{2}}=\begin{pmatrix}0\\1\end{pmatrix}. \tag*{(1259)} a 2 1 = 0 , a − 2 1 = 1 , χ 2 1 = χ 2 1 − 2 1 = ( 0 1 ) . (1259)
(d) Spin is parallel to n ( ϑ , φ ) \mathbf{n}(\vartheta,\varphi) n ( ϑ , φ )
a 1 2 = cos ϑ 2 e − i φ 2 , a − 1 2 = sin ϑ 2 e i φ 2 , χ 1 2 = χ 1 2 1 2 ( ϑ , φ ) = ( cos ϑ 2 e − i φ 2 sin ϑ 2 e i φ 2 ) . (1260) a^{\frac{1}{2}}=\cos\frac{\vartheta}{2}e^{-i\frac{\varphi}{2}},\ a^{-\frac{1}{2}}=\sin\frac{\vartheta}{2}e^{i\frac{\varphi}{2}},\quad\chi_{\frac{1}{2}}=\chi_{\frac{1}{2}\frac{1}{2}}(\vartheta,\varphi)=\begin{pmatrix}\cos\frac{\vartheta}{2}e^{-i\frac{\varphi}{2}}\\ \sin\frac{\vartheta}{2}e^{i\frac{\varphi}{2}}\end{pmatrix}. \tag*{(1260)} a 2 1 = cos 2 ϑ e − i 2 φ , a − 2 1 = sin 2 ϑ e i 2 φ , χ 2 1 = χ 2 1 2 1 ( ϑ , φ ) = ( cos 2 ϑ e − i 2 φ sin 2 ϑ e i 2 φ ) . (1260)
(e) Spin is antiparallel to n ( ϑ , φ ) \mathbf{n}(\vartheta,\varphi) n ( ϑ , φ )
a 1 2 = − sin ϑ 2 e − i φ 2 , a − 1 2 = cos ϑ 2 e i φ 2 , χ 1 2 = χ 1 2 − 1 2 ( ϑ , φ ) = ( − sin ϑ 2 e − i φ 2 cos ϑ 2 e i φ 2 ) . (1261) a^{\frac{1}{2}}=-\sin\frac{\vartheta}{2}e^{-i\frac{\varphi}{2}},\ a^{-\frac{1}{2}}=\cos\frac{\vartheta}{2}e^{i\frac{\varphi}{2}},\quad\chi_{\frac{1}{2}}=\chi_{\frac{1}{2}-\frac{1}{2}}(\vartheta,\varphi)=\begin{pmatrix}-\sin\frac{\vartheta}{2}e^{-i\frac{\varphi}{2}}\\ \cos\frac{\vartheta}{2}e^{i\frac{\varphi}{2}}\end{pmatrix}. \tag*{(1261)} a 2 1 = − sin 2 ϑ e − i 2 φ , a − 2 1 = cos 2 ϑ e i 2 φ , χ 2 1 = χ 2 1 − 2 1 ( ϑ , φ ) = ( − sin 2 ϑ e − i 2 φ cos 2 ϑ e i 2 φ ) . (1261)
In general, the spin function (6.2.32) describes a spin-1 2 \frac{1}{2} 2 1 particle whose spin is directed along some unit vector n \mathbf{n} n with cartesian components
n x = 2 Re ( a 1 2 ∗ a − 1 2 ) , n y = 2 Im ( a 1 2 ∗ a − 1 2 ) , n z = ∣ a 1 2 ∣ 2 − ∣ a − 1 2 ∣ 2 . (1262) n_x=2\operatorname{Re}\left(a^{\frac{1}{2}*}a^{-\frac{1}{2}}\right),\ n_y=2\operatorname{Im}\left(a^{\frac{1}{2}*}a^{-\frac{1}{2}}\right),\ n_z=\left|a^{\frac{1}{2}}\right|^2-\left|a^{-\frac{1}{2}}\right|^2. \tag*{(1262)} n x = 2 Re ( a 2 1 ∗ a − 2 1 ) , n y = 2 Im ( a 2 1 ∗ a − 2 1 ) , n z = a 2 1 2 − a − 2 1 2 . (1262)
The spherical components of this unit vector n \mathbf{n} n are given by
n μ = 3 ∑ m , m ′ C 1 2 m 1 2 m ′ 1 μ a m ′ ∗ a m , (1263) n_\mu=\sqrt{3}\sum_{m,m'}C_{\frac{1}{2}m\,\frac{1}{2}m'}^{1\mu}a^{m'*}a^m, \tag*{(1263)} n μ = 3 m , m ′ ∑ C 2 1 m 2 1 m ′ 1 μ a m ′ ∗ a m , (1263)
or more explicitly
n + 1 = − 2 a 1 2 ∗ a − 1 2 , n 0 = ∣ a 1 2 ∣ 2 − ∣ a − 1 2 ∣ 2 , n − 1 = 2 a − 1 2 ∗ a 1 2 . (1264) n_{+1}=-\sqrt{2}a^{\frac{1}{2}*}a^{-\frac{1}{2}},\ n_0=\left|a^{\frac{1}{2}}\right|^2-\left|a^{-\frac{1}{2}}\right|^2,\ n_{-1}=\sqrt{2}a^{-\frac{1}{2}*}a^{\frac{1}{2}}. \tag*{(1264)} n + 1 = − 2 a 2 1 ∗ a − 2 1 , n 0 = a 2 1 2 − a − 2 1 2 , n − 1 = 2 a − 2 1 ∗ a 2 1 . (1264)
The product of the spin functions χ 1 2 χ 1 2 † \chi_{\frac{1}{2}}\chi_{\frac{1}{2}}^\dagger χ 2 1 χ 2 1 † may be expressed in terms of the spin matrices
χ 1 2 χ 1 2 † = ( ∣ a 1 2 ∣ 2 a − 1 2 ∗ a 1 2 a 1 2 ∗ a − 1 2 ∣ a − 1 2 ∣ 2 ) = 1 2 I ^ + n ⋅ S ^ . (1265) \chi_{\frac{1}{2}}\chi_{\frac{1}{2}}^\dagger=
\begin{pmatrix}
\left|a^{\frac{1}{2}}\right|^2 & a^{-\frac{1}{2}*}a^{\frac{1}{2}}\\
a^{\frac{1}{2}*}a^{-\frac{1}{2}} & \left|a^{-\frac{1}{2}}\right|^2
\end{pmatrix}
=\frac{1}{2}\widehat{I}+\mathbf{n}\cdot\widehat{\mathbf{S}}. \tag*{(1265)} χ 2 1 χ 2 1 † = a 2 1 2 a 2 1 ∗ a − 2 1 a − 2 1 ∗ a 2 1 a − 2 1 2 = 2 1 I + n ⋅ S . (1265)
where n \mathbf{n} n specifies the spin direction (6.2.39)–(6.2.41).
Matrix elements of the spin operator S ^ \widehat{\mathbf{S}} S may be written in terms of the vector n \mathbf{n} n as
χ 1 2 † S ^ χ 1 2 = 1 2 n . (1266) \chi_{\frac{1}{2}}^\dagger\widehat{\mathbf{S}}\chi_{\frac{1}{2}}=\frac{1}{2}\mathbf{n}. \tag*{(1266)} χ 2 1 † S χ 2 1 = 2 1 n . (1266)
The transformation of spin functions under rotations of coordinate system is effected by the rotation operators D ^ 1 2 ( α , β , γ ) \hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma) D ^ 2 1 ( α , β , γ ) (Eq. (326) ) or U ^ 1 2 ( ω ; θ , Φ ) \hat{U}^{\frac{1}{2}}(\omega;\theta,\Phi) U ^ 2 1 ( ω ; θ , Φ ) (Eq. (330) ), depending on the choice of the parameters to describe rotations:
χ 1 2 ′ = D ^ 1 2 ( α , β , γ ) χ 1 2 = U ^ 1 2 ( ω ; θ , Φ ) χ 1 2 . (1267) \chi'_{\frac{1}{2}}=\hat{D}^{\frac{1}{2}}(\alpha,\beta,\gamma)\chi_{\frac{1}{2}}=\hat{U}^{\frac{1}{2}}(\omega;\theta,\Phi)\chi_{\frac{1}{2}}. \tag*{(1267)} χ 2 1 ′ = D ^ 2 1 ( α , β , γ ) χ 2 1 = U ^ 2 1 ( ω ; θ , Φ ) χ 2 1 . (1267)
The spin functions χ 1 2 ′ \chi'_{\frac{1}{2}} χ 2 1 ′ in the rotated coordinate system may be expressed in terms of the basis spin functions, χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m or χ 1 2 m ′ ′ \chi'_{\frac{1}{2}m'} χ 2 1 m ′ ′ , referred to the original or rotated coordinate systems, respectively.
χ 1 2 ′ = ( a ′ 1 2 a ′ − 1 2 ) = ∑ m a m χ 1 2 m = ∑ m ′ a m ′ χ 1 2 m ′ ′ . (1268) \chi'_{\frac{1}{2}}=
\begin{pmatrix}
a'^{\frac{1}{2}}\\
a'^{-\frac{1}{2}}
\end{pmatrix}
=\sum_m a^m\chi_{\frac{1}{2}m}
=\sum_{m'}a^{m'}\chi'_{\frac{1}{2}m'}. \tag*{(1268)} χ 2 1 ′ = ( a ′ 2 1 a ′ − 2 1 ) = m ∑ a m χ 2 1 m = m ′ ∑ a m ′ χ 2 1 m ′ ′ . (1268)
In this case a m a^m a m and a m ′ a^{m'} a m ′ are spinor components in the original and rotated coordinate systems. The relation between the basis functions χ 1 2 m ′ \chi'_{\frac{1}{2}m} χ 2 1 m ′ and χ 1 2 m \chi_{\frac{1}{2}m} χ 2 1 m is given by Eq. (1238) . The transformation properties of a m a^m a m are as follows:
\begin{aligned}
a'^{m}&=\sum_n D^{\frac{1}{2}}_{m n}(\alpha,\beta,\gamma)a^n,\\
a^n&=\sum_m D^{\frac{1}{2}*}_{m n}(\alpha,\beta,\gamma)a'^{m}.
\end{aligned}
\qquad\left(m,n=\pm\frac{1}{2}\right) \tag*{(1269)}\end{aligned}
Thus, a m a^m a m are contravariant spinor components (see Eq. (538) ).
Polarization Density Matrix
The polarisation density matrix for particles of spin 1 2 \frac{1}{2} 2 1 may be written in the form
ρ ^ = 1 2 { I ^ + 2 P S ^ } , (1270) \hat{\rho}=\frac{1}{2}\{\hat{I}+2\mathbf{P}\hat{\mathbf{S}}\}, \tag*{(1270)} ρ ^ = 2 1 { I ^ + 2 P S ^ } , (1270)
where the real vector P \mathbf{P} P is called the polarisation vector. This vector gives the expectation value of the spin operator multiplied by 2,
P = 2 ⟨ S ⟩ = 2 Tr { ρ ^ S ^ } . (1271) \mathbf{P}=2\langle\mathbf{S}\rangle=2\operatorname{Tr}\{\hat{\rho}\hat{\mathbf{S}}\}. \tag*{(1271)} P = 2 ⟨ S ⟩ = 2 Tr { ρ ^ S ^ } . (1271)
The absolute value of the vector P \mathbf{P} P is called the polarization degree; it ranges from 0 (for unpolarised states) to 1 (for pure, i.e., totally polarized states):
0 ≤ ∣ P ∣ ≤ 1. (1272) 0\leq|\mathbf{P}|\leq1. \tag*{(1272)} 0 ≤ ∣ P ∣ ≤ 1. (1272)
The spherical components of the polarization vector are related to the elements of the density matrix by
P μ = 3 ∑ σ σ ′ C 1 2 σ 1 μ 1 2 σ ′ ρ σ σ ′ , (1273) P_\mu=\sqrt{3}\sum_{\sigma\sigma'}C^{\frac{1}{2}\sigma'}_{\frac{1}{2}\sigma\,1\mu}\rho_{\sigma\sigma'}, \tag*{(1273)} P μ = 3 σ σ ′ ∑ C 2 1 σ 1 μ 2 1 σ ′ ρ σ σ ′ , (1273)
or in an expanded form
P + 1 = − 2 ρ − 1 2 1 2 , P 0 = ρ 1 2 1 2 − ρ − 1 2 − 1 2 , P − 1 = 2 ρ 1 2 − 1 2 . (1274) P_{+1}=-\sqrt{2}\rho_{-\frac{1}{2}\frac{1}{2}},\quad P_0=\rho_{\frac{1}{2}\frac{1}{2}}-\rho_{-\frac{1}{2}-\frac{1}{2}},\quad P_{-1}=\sqrt{2}\rho_{\frac{1}{2}-\frac{1}{2}}. \tag*{(1274)} P + 1 = − 2 ρ − 2 1 2 1 , P 0 = ρ 2 1 2 1 − ρ − 2 1 − 2 1 , P − 1 = 2 ρ 2 1 − 2 1 . (1274)
Cartesian components of the polarization vector are given by
P x = ρ 1 2 − 1 2 + ρ − 1 2 1 2 , P y = i ( ρ 1 2 − 1 2 − ρ − 1 2 1 2 ) , P z = ρ 1 2 1 2 − ρ − 1 2 − 1 2 . (1275) P_x=\rho_{\frac{1}{2}-\frac{1}{2}}+\rho_{-\frac{1}{2}\frac{1}{2}},\quad P_y=i\left(\rho_{\frac{1}{2}-\frac{1}{2}}-\rho_{-\frac{1}{2}\frac{1}{2}}\right),\quad P_z=\rho_{\frac{1}{2}\frac{1}{2}}-\rho_{-\frac{1}{2}-\frac{1}{2}}. \tag*{(1275)} P x = ρ 2 1 − 2 1 + ρ − 2 1 2 1 , P y = i ( ρ 2 1 − 2 1 − ρ − 2 1 2 1 ) , P z = ρ 2 1 2 1 − ρ − 2 1 − 2 1 . (1275)
For a pure state described by a spin function χ 1 2 \chi_{\frac{1}{2}} χ 2 1 the density matrix ρ ^ \hat{\rho} ρ ^ has the form
ρ ^ p u r e = χ 1 2 χ 1 2 † , (1276) \hat{\rho}_{\mathrm{pure}}=\chi_{\frac{1}{2}}\chi_{\frac{1}{2}}^\dagger, \tag*{(1276)} ρ ^ pure = χ 2 1 χ 2 1 † , (1276)
and the polarization vector P \mathbf{P} P coincides with the unit vector n \mathbf{n} n (see Eqs. (1262) –(1264) ). For an unpolarized state
ρ ^ u n p o l = 1 2 I ^ . (1277) \hat{\rho}_{\mathrm{unpol}}=\frac{1}{2}\hat{I}. \tag*{(1277)} ρ ^ unpol = 2 1 I ^ . (1277)
Spin functions for S = 1 S=1 S = 1
Basis Spin Functions
The basis functions χ 1 m ( m = ± 1 , 0 ) \chi_{1m}(m=\pm1,0) χ 1 m ( m = ± 1 , 0 ) are eigenfunctions of the operators S ^ 2 \hat{S}^{2} S ^ 2 and S ^ z \hat{S}_{z} S ^ z
S ^ 2 χ 1 m = 2 χ 1 m , (1278) \hat{S}^{2}\chi_{1m}=2\chi_{1m}, \tag*{(1278)} S ^ 2 χ 1 m = 2 χ 1 m , (1278)
S ^ z χ 1 m = m χ 1 m . \hat{S}_{z}\chi_{1m}=m\chi_{1m}. S ^ z χ 1 m = m χ 1 m .
Spin functions for S = 1 S=1 S = 1
The function χ 1 m \chi_{1m} χ 1 m describes the state of a spin-1 particle with definite spin projection m m m on the z z z -axis. The three basis functions χ 1 m \chi_{1m} χ 1 m (m = ± 1 , 0 m=\pm1,0 m = ± 1 , 0 ) may be treated as spherical covariant basis vectors e m e_m e m (see § 9.1 ) written in column form. Instead of χ 1 m \chi_{1m} χ 1 m one may use the basis functions χ i \chi_i χ i (i = x , y , z i=x,y,z i = x , y , z ) which are the cartesian basis vectors e i e_i e i :
χ 11 = − 1 2 ( χ x + i χ y ) , χ x = 1 2 ( χ 1 − 1 − χ 11 ) χ 10 = χ x , χ y = i 2 ( χ 1 − 1 + χ 11 ) , χ 1 − 1 = 1 2 ( χ x − i χ y ) , χ z = χ 10 . (1279) \begin{aligned}
\chi_{11} &= -\frac{1}{\sqrt{2}}(\chi_x+i\chi_y), & \chi_x &= \frac{1}{\sqrt{2}}(\chi_{1-1}-\chi_{11}) \\
\chi_{10} &= \chi_x, & \chi_y &= \frac{i}{\sqrt{2}}(\chi_{1-1}+\chi_{11}), \\
\chi_{1-1} &= \frac{1}{\sqrt{2}}(\chi_x-i\chi_y), & \chi_z &= \chi_{10}. \tag*{(1279)}
\end{aligned} χ 11 χ 10 χ 1 − 1 = − 2 1 ( χ x + i χ y ) , = χ x , = 2 1 ( χ x − i χ y ) , χ x χ y χ z = 2 1 ( χ 1 − 1 − χ 11 ) = 2 i ( χ 1 − 1 + χ 11 ) , = χ 10 . (1279)
The functions χ 1 m \chi_{1m} χ 1 m (m = ± 1 , 0 m=\pm1,0 m = ± 1 , 0 ) as well as χ i \chi_i χ i (i = x , y , z i=x,y,z i = x , y , z ) constitute an orthonormal basis. The orthonormality conditions read
χ 1 m ′ † χ 1 m = δ m ′ m , χ i † χ k = δ i k . (1280) \chi^\dagger_{1m'}\chi_{1m}=\delta_{m'm},\qquad\chi^\dagger_i\chi_k=\delta_{ik}. \tag*{(1280)} χ 1 m ′ † χ 1 m = δ m ′ m , χ i † χ k = δ ik . (1280)
The completeness condition for basis functions may be written as
∑ m = ± 1 , 0 χ 1 m χ 1 m † = I ^ , ∑ i = x , y , z χ i χ i † = I ^ , (1281) \sum_{m=\pm1,0}\chi_{1m}\chi^\dagger_{1m}=\hat{I},\qquad\sum_{i=x,y,z}\chi_i\chi^\dagger_i=\hat{I}, \tag*{(1281)} m = ± 1 , 0 ∑ χ 1 m χ 1 m † = I ^ , i = x , y , z ∑ χ i χ i † = I ^ , (1281)
where I ^ \hat{I} I ^ is the unit 3 × 3 3 \times3 3 × 3 matrix. To describe the states of particles with spin 1 one may use the spherical basis representation or the cartesian basis representation.
(a) Spherical basis representation
In the spherical basis representation the dependence of the functions χ 1 m ( σ ) \chi_{1m}(\sigma) χ 1 m ( σ ) on the spin variable σ \sigma σ is given by
χ 1 m ( σ ) = δ m σ . (1282) \chi_{1m}(\sigma)=\delta_{m\sigma}. \tag*{(1282)} χ 1 m ( σ ) = δ mσ . (1282)
According to Eq. (1159) , the basis spin functions χ 1 m \chi_{1m} χ 1 m may be written as
χ 11 = ( 1 0 0 ) , χ 10 = ( 0 1 0 ) , χ 1 − 1 = ( 0 0 1 ) , (1283) \chi_{11}=\begin{pmatrix}1\\0\\0\end{pmatrix},\qquad\chi_{10}=\begin{pmatrix}0\\1\\0\end{pmatrix},\qquad\chi_{1-1}=\begin{pmatrix}0\\0\\1\end{pmatrix}, \tag*{(1283)} χ 11 = 1 0 0 , χ 10 = 0 1 0 , χ 1 − 1 = 0 0 1 , (1283)
and the Hermitian conjugate functions χ 1 m † \chi^\dagger_{1m} χ 1 m † read
χ 11 † = ( 1 , 0 , 0 ) , χ 10 † = ( 0 , 1 , 0 ) , χ 1 − 1 † = ( 0 , 0 , 1 ) . (1284) \chi^\dagger_{11}=(1,0,0),\qquad\chi^\dagger_{10}=(0,1,0),\qquad\chi^\dagger_{1-1}=(0,0,1). \tag*{(1284)} χ 11 † = ( 1 , 0 , 0 ) , χ 10 † = ( 0 , 1 , 0 ) , χ 1 − 1 † = ( 0 , 0 , 1 ) . (1284)
In the spherical basis representation the functions χ i \chi_i χ i (i = x , y , z i=x,y,z i = x , y , z ) have the form
χ x = 1 2 ( − 1 0 1 ) , χ y = i 2 ( 1 0 1 ) , χ z = ( 0 1 0 ) , (1285) \chi_x=\frac{1}{\sqrt{2}}\begin{pmatrix}-1\\0\\1\end{pmatrix},\qquad\chi_y=\frac{i}{\sqrt{2}}\begin{pmatrix}1\\0\\1\end{pmatrix},\qquad\chi_z=\begin{pmatrix}0\\1\\0\end{pmatrix}, \tag*{(1285)} χ x = 2 1 − 1 0 1 , χ y = 2 i 1 0 1 , χ z = 0 1 0 , (1285)
χ x † = 1 2 ( − 1 , 0 , 1 ) , χ y † = − i 2 ( 1 , 0 , 1 ) , χ z † = ( 0 , 1 , 0 ) . (1286) \chi^\dagger_x=\frac{1}{\sqrt{2}}(-1,0,1),\qquad\chi^\dagger_y=-\frac{i}{\sqrt{2}}(1,0,1),\qquad\chi^\dagger_z=(0,1,0). \tag*{(1286)} χ x † = 2 1 ( − 1 , 0 , 1 ) , χ y † = − 2 i ( 1 , 0 , 1 ) , χ z † = ( 0 , 1 , 0 ) . (1286)
It follows from (6.3.6), (6.3.8) that in the spherical basis representation the functions χ 1 m \chi_{1m} χ 1 m are real,
χ 1 m ∗ = χ 1 m , ( m = ± 1 , 0 ) , (1287) \chi^*_{1m}=\chi_{1m},\qquad(m=\pm1,0), \tag*{(1287)} χ 1 m ∗ = χ 1 m , ( m = ± 1 , 0 ) , (1287)
and the functions χ i \chi_i χ i satisfy the relations
χ x ∗ = χ x , χ y ∗ = − χ y , χ z ∗ = χ z . (1288) \chi^*_x=\chi_x,\qquad\chi^*_y=-\chi_y,\qquad\chi^*_z=\chi_z. \tag*{(1288)} χ x ∗ = χ x , χ y ∗ = − χ y , χ z ∗ = χ z . (1288)
Explicit forms of the spin matrices and the polarization operators in the spherical basis representation are given by Eqs. (352) –(365) .
(b) Cartesian basis representation
In the cartesian basis representation the spin variable σ \sigma σ assumes three possible values, σ = x , y , z \sigma=x,y,z σ = x , y , z . The dependence of the functions χ i ( σ ) \chi_i(\sigma) χ i ( σ ) on σ \sigma σ is given by
χ i ( σ ) = δ i σ . (1289) \chi_i(\sigma)=\delta_{i\sigma}. \tag*{(1289)} χ i ( σ ) = δ iσ . (1289)
Thus, the basis spin functions χ i \chi_i χ i (i = x , y , z i=x,y,z i = x , y , z ) in the cartesian basis representation may be written in the following form (see also Eq. (166) )
χ x = ( 1 0 0 ) , χ y = ( 0 1 0 ) , χ z = ( 0 0 1 ) . (1290) \chi_x=\begin{pmatrix}1\\0\\0\end{pmatrix},\qquad\chi_y=\begin{pmatrix}0\\1\\0\end{pmatrix},\qquad\chi_z=\begin{pmatrix}0\\0\\1\end{pmatrix}. \tag*{(1290)} χ x = 1 0 0 , χ y = 0 1 0 , χ z = 0 0 1 . (1290)
χ x † = ( 1 , 0 , 0 ) , χ y † = ( 0 , 1 , 0 ) , χ z † = ( 0 , 0 , 1 ) . (1291) \chi_x^\dagger=(1,0,0),\qquad\chi_y^\dagger=(0,1,0),\qquad\chi_z^\dagger=(0,0,1). \tag*{(1291)} χ x † = ( 1 , 0 , 0 ) , χ y † = ( 0 , 1 , 0 ) , χ z † = ( 0 , 0 , 1 ) . (1291)
The basis functions χ 1 m \chi_{1m} χ 1 m (m = ± 1 , 0 m=\pm1,0 m = ± 1 , 0 ) read
χ 11 = − 1 2 ( 1 i 0 ) , χ 10 = ( 0 0 1 ) , χ 1 − 1 = 1 2 ( 1 − i 0 ) , (1292) \chi_{11}=-\frac{1}{\sqrt{2}}\begin{pmatrix}1\\i\\0\end{pmatrix},\qquad\chi_{10}=\begin{pmatrix}0\\0\\1\end{pmatrix},\qquad\chi_{1-1}=\frac{1}{\sqrt{2}}\begin{pmatrix}1\\-i\\0\end{pmatrix}, \tag*{(1292)} χ 11 = − 2 1 1 i 0 , χ 10 = 0 0 1 , χ 1 − 1 = 2 1 1 − i 0 , (1292)
χ 11 † = − 1 2 ( 1 , − i , 0 ) , χ 10 † = ( 0 , 0 , 1 ) , χ 1 − 1 † = 1 2 ( 1 , i , 0 ) . (1293) \chi_{11}^\dagger=-\frac{1}{\sqrt{2}}(1,-i,0),\qquad\chi_{10}^\dagger=(0,0,1),\qquad\chi_{1-1}^\dagger=\frac{1}{\sqrt{2}}(1,i,0). \tag*{(1293)} χ 11 † = − 2 1 ( 1 , − i , 0 ) , χ 10 † = ( 0 , 0 , 1 ) , χ 1 − 1 † = 2 1 ( 1 , i , 0 ) . (1293)
It follows from Eqs. (1290) , (1292) that in the cartesian basis representation the functions χ i \chi_i χ i are real,
χ i ∗ = χ i ( i = x , y , z ) , (1294) \chi_i^*=\chi_i\qquad(i=x,y,z), \tag*{(1294)} χ i ∗ = χ i ( i = x , y , z ) , (1294)
and the functions χ 1 m \chi_{1m} χ 1 m satisfy the relations
χ 1 m ∗ = ( − 1 ) m χ 1 − m ( m = ± 1 , 0 ) . (1295) \chi_{1m}^*=(-1)^m\chi_{1-m}\qquad(m=\pm1,0). \tag*{(1295)} χ 1 m ∗ = ( − 1 ) m χ 1 − m ( m = ± 1 , 0 ) . (1295)
The spin matrices and polarization operators in the cartesian basis representation are given by Eqs. (366) and (2.6.35).
The direct and reverse transformations from a spherical basis to a cartesian one can be performed by the use of the unitary 3 × 3 3\times3 3 × 3 matrix U U U ,
χ ( spherical basis ) = U χ ( cartesian basis ) , \chi(\text{spherical basis})=U\chi(\text{cartesian basis}), χ ( spherical basis ) = U χ ( cartesian basis ) ,
χ ( cartesian basis ) = U − 1 χ ( spherical basis ) . (1296) \chi(\text{cartesian basis})=U^{-1}\chi(\text{spherical basis}). \tag*{(1296)} χ ( cartesian basis ) = U − 1 χ ( spherical basis ) . (1296)
The explicit form of U U U is given by Eq. (380) . The expressions below are independent of the representation used unless the contrary is indicated.
Expansions of Products of Spin Functions
Products of the basis functions χ 1 m χ 1 m ′ † \chi_{1m}\chi_{1m'}^\dagger χ 1 m χ 1 m ′ † and χ i χ k † \chi_i\chi_k^\dagger χ i χ k † are square 3 × 3 3\times3 3 × 3 matrices which may be expanded in terms of the polarization matrices I ^ \hat I I ^ , S ^ μ \hat S_\mu S ^ μ (μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ), T ^ 2 M \hat T_{2M} T ^ 2 M (M = ± 2 , ± 1 , 0 M=\pm2,\pm1,0 M = ± 2 , ± 1 , 0 ) or, equivalently, of the matrices I ^ \hat I I ^ , S ^ i \hat S_i S ^ i , Q ^ i k \hat Q_{ik} Q ^ ik (i , k = x , y , z i,k=x,y,z i , k = x , y , z ) (§ 11 ). These expansions are given by
χ 1 m χ 1 m ′ † = 1 3 δ m m ′ I ^ + 1 2 C 1 m ′ 1 μ 1 m S ^ μ + 5 3 C 1 m ′ 2 M 1 m T ^ 2 M , (1297) \chi_{1m}\chi_{1m'}^\dagger=\frac{1}{3}\delta_{mm'}\hat I+\frac{1}{\sqrt{2}}C^{1m}_{1m'1\mu}\hat S_\mu+\sqrt{\frac{5}{3}}C^{1m}_{1m'2M}\hat T_{2M}, \tag*{(1297)} χ 1 m χ 1 m ′ † = 3 1 δ m m ′ I ^ + 2 1 C 1 m ′ 1 μ 1 m S ^ μ + 3 5 C 1 m ′ 2 M 1 m T ^ 2 M , (1297)
χ i χ k † = 1 3 δ i k I ^ + i 2 ϵ i k l S ^ l − Q ^ i k . (1298) \chi_i\chi_k^\dagger=\frac{1}{3}\delta_{ik}\hat I+\frac{i}{2}\epsilon_{ikl}\hat S_l-\hat Q_{ik}. \tag*{(1298)} χ i χ k † = 3 1 δ ik I ^ + 2 i ϵ ik l S ^ l − Q ^ ik . (1298)
Written in component form these expansions yield
χ 11 χ 11 † = 1 3 I ^ + 1 2 S ^ 0 + 1 6 T ^ 20 = 1 3 I ^ + 1 2 S ^ z + 1 2 Q ^ z z , χ 11 χ 10 † = − 1 2 S ^ + 1 − 1 2 T ^ 21 = 1 2 2 ( S ^ x + i S ^ y + 2 Q ^ x z + 2 i Q ^ y z ) , χ 11 χ 1 − 1 † = T ^ 22 = 1 2 ( Q ^ x x − Q ^ y y + 2 i Q ^ x y ) , χ 10 χ 11 † = 1 2 S ^ − 1 + 1 2 T ^ 2 − 1 = 1 2 2 ( S ^ x − i S ^ y + 2 Q ^ x z − 2 i Q ^ y z ) , χ 10 χ 10 † = 1 3 I ^ − 2 3 T ^ 20 = 1 3 I ^ − Q ^ z z , χ 10 χ 1 − 1 † = − 1 2 S ^ + 1 + 1 2 T ^ 21 = 1 2 2 ( S ^ x + i S ^ y − 2 Q ^ x z − 2 i Q ^ y z ) , χ 1 − 1 χ 11 † = T ^ 2 − 2 = 1 2 ( Q ^ x x − Q ^ y y − 2 i Q ^ x y ) , χ 1 − 1 χ 10 † = 1 2 S ^ − 1 − 1 2 T ^ 2 − 1 = 1 2 2 ( S ^ x − i S ^ y − 2 Q ^ x z + 2 i Q ^ y z ) , χ 1 − 1 χ 1 − 1 † = 1 3 I ^ − 1 2 S ^ 0 + 1 6 T ^ 20 = 1 3 I ^ − 1 2 S ^ z + 1 2 Q ^ z z . (1299) \begin{aligned}
\chi_{11}\chi_{11}^{\dagger} &= \frac{1}{3}\hat{I}+\frac{1}{2}\hat{S}_0+\frac{1}{\sqrt{6}}\hat{T}_{20}=\frac{1}{3}\hat{I}+\frac{1}{2}\hat{S}_z+\frac{1}{2}\hat{Q}_{zz},\\
\chi_{11}\chi_{10}^{\dagger} &= -\frac{1}{2}\hat{S}_{+1}-\frac{1}{\sqrt{2}}\hat{T}_{21}=\frac{1}{2\sqrt{2}}\left(\hat{S}_x+i\hat{S}_y+2\hat{Q}_{xz}+2i\hat{Q}_{yz}\right),\\
\chi_{11}\chi_{1-1}^{\dagger} &= \hat{T}_{22}=\frac{1}{2}\left(\hat{Q}_{xx}-\hat{Q}_{yy}+2i\hat{Q}_{xy}\right),\\
\chi_{10}\chi_{11}^{\dagger} &= \frac{1}{2}\hat{S}_{-1}+\frac{1}{\sqrt{2}}\hat{T}_{2-1}=\frac{1}{2\sqrt{2}}\left(\hat{S}_x-i\hat{S}_y+2\hat{Q}_{xz}-2i\hat{Q}_{yz}\right),\\
\chi_{10}\chi_{10}^{\dagger} &= \frac{1}{3}\hat{I}-\sqrt{\frac{2}{3}}\hat{T}_{20}=\frac{1}{3}\hat{I}-\hat{Q}_{zz},\\
\chi_{10}\chi_{1-1}^{\dagger} &= -\frac{1}{2}\hat{S}_{+1}+\frac{1}{\sqrt{2}}\hat{T}_{21}=\frac{1}{2\sqrt{2}}\left(\hat{S}_x+i\hat{S}_y-2\hat{Q}_{xz}-2i\hat{Q}_{yz}\right),\\
\chi_{1-1}\chi_{11}^{\dagger} &= \hat{T}_{2-2}=\frac{1}{2}\left(\hat{Q}_{xx}-\hat{Q}_{yy}-2i\hat{Q}_{xy}\right),\\
\chi_{1-1}\chi_{10}^{\dagger} &= \frac{1}{2}\hat{S}_{-1}-\frac{1}{\sqrt{2}}\hat{T}_{2-1}=\frac{1}{2\sqrt{2}}\left(\hat{S}_x-i\hat{S}_y-2\hat{Q}_{xz}+2i\hat{Q}_{yz}\right),\\
\chi_{1-1}\chi_{1-1}^{\dagger} &= \frac{1}{3}\hat{I}-\frac{1}{2}\hat{S}_0+\frac{1}{\sqrt{6}}\hat{T}_{20}=\frac{1}{3}\hat{I}-\frac{1}{2}\hat{S}_z+\frac{1}{2}\hat{Q}_{zz}. \tag*{(1299)}
\end{aligned} χ 11 χ 11 † χ 11 χ 10 † χ 11 χ 1 − 1 † χ 10 χ 11 † χ 10 χ 10 † χ 10 χ 1 − 1 † χ 1 − 1 χ 11 † χ 1 − 1 χ 10 † χ 1 − 1 χ 1 − 1 † = 3 1 I ^ + 2 1 S ^ 0 + 6 1 T ^ 20 = 3 1 I ^ + 2 1 S ^ z + 2 1 Q ^ z z , = − 2 1 S ^ + 1 − 2 1 T ^ 21 = 2 2 1 ( S ^ x + i S ^ y + 2 Q ^ x z + 2 i Q ^ y z ) , = T ^ 22 = 2 1 ( Q ^ xx − Q ^ y y + 2 i Q ^ x y ) , = 2 1 S ^ − 1 + 2 1 T ^ 2 − 1 = 2 2 1 ( S ^ x − i S ^ y + 2 Q ^ x z − 2 i Q ^ y z ) , = 3 1 I ^ − 3 2 T ^ 20 = 3 1 I ^ − Q ^ z z , = − 2 1 S ^ + 1 + 2 1 T ^ 21 = 2 2 1 ( S ^ x + i S ^ y − 2 Q ^ x z − 2 i Q ^ y z ) , = T ^ 2 − 2 = 2 1 ( Q ^ xx − Q ^ y y − 2 i Q ^ x y ) , = 2 1 S ^ − 1 − 2 1 T ^ 2 − 1 = 2 2 1 ( S ^ x − i S ^ y − 2 Q ^ x z + 2 i Q ^ y z ) , = 3 1 I ^ − 2 1 S ^ 0 + 6 1 T ^ 20 = 3 1 I ^ − 2 1 S ^ z + 2 1 Q ^ z z . (1299)
χ x χ x † = 1 3 I ^ − 1 2 T ^ 2 − 2 − 1 2 T ^ 22 + 1 6 T ^ 20 = 1 3 I ^ − Q ^ x x , χ x χ y † = i 2 ( S ^ 0 − T ^ 2 − 2 + T ^ 22 ) = i 2 S ^ z − Q ^ x y , χ x χ z † = 1 2 2 ( S ^ − 1 + S ^ + 1 ) − 1 2 ( T ^ 2 − 1 − T ^ 21 ) = − i 2 S ^ y − Q ^ x z , χ y χ x † = − i 2 ( S ^ 0 + T ^ 2 − 2 − T ^ 22 ) = − i 2 S ^ z − Q ^ y x , χ y χ y † = 1 3 I ^ + 1 2 T ^ 2 − 2 + 1 2 T ^ 22 + 1 6 T ^ 20 = 1 3 I ^ − Q ^ y y , χ y χ z † = i 2 2 ( S ^ − 1 − S ^ + 1 ) − i 2 ( T ^ 2 − 1 + T ^ 21 ) = i 2 S ^ x − Q ^ y z , χ z χ x † = − 1 2 2 ( S ^ − 1 + S ^ + 1 ) − 1 2 ( T ^ 2 − 1 − T ^ 21 ) = i 2 S ^ y − Q ^ z x , χ z χ y † = − i 2 2 ( S ^ − 1 − S ^ + 1 ) − i 2 ( T ^ 2 − 1 + T ^ 21 ) = − i 2 S ^ x − Q ^ z y , χ z χ z † = 1 3 I ^ − 2 3 T ^ 20 = 1 3 I ^ − Q ^ z z . (1300) \begin{aligned}
\chi_x\chi_x^{\dagger} &= \frac{1}{3}\hat{I}-\frac{1}{2}\hat{T}_{2-2}-\frac{1}{2}\hat{T}_{22}+\frac{1}{\sqrt{6}}\hat{T}_{20}=\frac{1}{3}\hat{I}-\hat{Q}_{xx},\\
\chi_x\chi_y^{\dagger} &= \frac{i}{2}\left(\hat{S}_0-\hat{T}_{2-2}+\hat{T}_{22}\right)=\frac{i}{2}\hat{S}_z-\hat{Q}_{xy},\\
\chi_x\chi_z^{\dagger} &= \frac{1}{2\sqrt{2}}\left(\hat{S}_{-1}+\hat{S}_{+1}\right)-\frac{1}{2}\left(\hat{T}_{2-1}-\hat{T}_{21}\right)=-\frac{i}{2}\hat{S}_y-\hat{Q}_{xz},\\
\chi_y\chi_x^{\dagger} &= -\frac{i}{2}\left(\hat{S}_0+\hat{T}_{2-2}-\hat{T}_{22}\right)=-\frac{i}{2}\hat{S}_z-\hat{Q}_{yx},\\
\chi_y\chi_y^{\dagger} &= \frac{1}{3}\hat{I}+\frac{1}{2}\hat{T}_{2-2}+\frac{1}{2}\hat{T}_{22}+\frac{1}{\sqrt{6}}\hat{T}_{20}=\frac{1}{3}\hat{I}-\hat{Q}_{yy},\\
\chi_y\chi_z^{\dagger} &= \frac{i}{2\sqrt{2}}\left(\hat{S}_{-1}-\hat{S}_{+1}\right)-\frac{i}{2}\left(\hat{T}_{2-1}+\hat{T}_{21}\right)=\frac{i}{2}\hat{S}_x-\hat{Q}_{yz},\\
\chi_z\chi_x^{\dagger} &= -\frac{1}{2\sqrt{2}}\left(\hat{S}_{-1}+\hat{S}_{+1}\right)-\frac{1}{2}\left(\hat{T}_{2-1}-\hat{T}_{21}\right)=\frac{i}{2}\hat{S}_y-\hat{Q}_{zx},\\
\chi_z\chi_y^{\dagger} &= -\frac{i}{2\sqrt{2}}\left(\hat{S}_{-1}-\hat{S}_{+1}\right)-\frac{i}{2}\left(\hat{T}_{2-1}+\hat{T}_{21}\right)=-\frac{i}{2}\hat{S}_x-\hat{Q}_{zy},\\
\chi_z\chi_z^{\dagger} &= \frac{1}{3}\hat{I}-\sqrt{\frac{2}{3}}\hat{T}_{20}=\frac{1}{3}\hat{I}-\hat{Q}_{zz}. \tag*{(1300)}
\end{aligned} χ x χ x † χ x χ y † χ x χ z † χ y χ x † χ y χ y † χ y χ z † χ z χ x † χ z χ y † χ z χ z † = 3 1 I ^ − 2 1 T ^ 2 − 2 − 2 1 T ^ 22 + 6 1 T ^ 20 = 3 1 I ^ − Q ^ xx , = 2 i ( S ^ 0 − T ^ 2 − 2 + T ^ 22 ) = 2 i S ^ z − Q ^ x y , = 2 2 1 ( S ^ − 1 + S ^ + 1 ) − 2 1 ( T ^ 2 − 1 − T ^ 21 ) = − 2 i S ^ y − Q ^ x z , = − 2 i ( S ^ 0 + T ^ 2 − 2 − T ^ 22 ) = − 2 i S ^ z − Q ^ y x , = 3 1 I ^ + 2 1 T ^ 2 − 2 + 2 1 T ^ 22 + 6 1 T ^ 20 = 3 1 I ^ − Q ^ y y , = 2 2 i ( S ^ − 1 − S ^ + 1 ) − 2 i ( T ^ 2 − 1 + T ^ 21 ) = 2 i S ^ x − Q ^ y z , = − 2 2 1 ( S ^ − 1 + S ^ + 1 ) − 2 1 ( T ^ 2 − 1 − T ^ 21 ) = 2 i S ^ y − Q ^ z x , = − 2 2 i ( S ^ − 1 − S ^ + 1 ) − 2 i ( T ^ 2 − 1 + T ^ 21 ) = − 2 i S ^ x − Q ^ z y , = 3 1 I ^ − 3 2 T ^ 20 = 3 1 I ^ − Q ^ z z . (1300)
Action of Spin Operators on Basis Functions
Spherical components of the spin operator S ^ μ ( μ = ± 1 , 0 ) \hat{S}_{\mu}(\mu=\pm1,0) S ^ μ ( μ = ± 1 , 0 ) and its cartesian components S ^ i ( i = x , y , z ) \hat{S}_i(i=x,y,z) S ^ i ( i = x , y , z ) act on the basis functions as follows
S ^ μ χ 1 m = 2 C 1 m , 1 μ 1 m ′ χ 1 m ′ , (1301) \hat{S}_{\mu}\chi_{1m}=\sqrt{2}C^{1m'}_{1m,1\mu}\chi_{1m'}, \tag*{(1301)} S ^ μ χ 1 m = 2 C 1 m , 1 μ 1 m ′ χ 1 m ′ , (1301)
S ^ i χ k = i ε i k l χ l . \hat{S}_i\chi_k=i\varepsilon_{ikl}\chi_l. S ^ i χ k = i ε ik l χ l .
In a more detailed form
S ^ + 1 χ 11 = 0 , S ^ + 1 χ 10 = − χ 11 , S ^ + 1 χ 1 − 1 = − χ 10 , S ^ 0 χ 11 = χ 11 , S ^ 0 χ 10 = 0 , S ^ 0 χ 1 − 1 = − χ 1 − 1 , S ^ − 1 χ 11 = χ 10 , S ^ − 1 χ 10 = χ 1 − 1 , S ^ − 1 χ 1 − 1 = 0. (1302) \begin{aligned}
\hat{S}_{+1}\chi_{11}&=0, & \hat{S}_{+1}\chi_{10}&=-\chi_{11}, & \hat{S}_{+1}\chi_{1-1}&=-\chi_{10},\\
\hat{S}_{0}\chi_{11}&=\chi_{11}, & \hat{S}_{0}\chi_{10}&=0, & \hat{S}_{0}\chi_{1-1}&=-\chi_{1-1},\\
\hat{S}_{-1}\chi_{11}&=\chi_{10}, & \hat{S}_{-1}\chi_{10}&=\chi_{1-1}, & \hat{S}_{-1}\chi_{1-1}&=0. \tag*{(1302)}
\end{aligned} S ^ + 1 χ 11 S ^ 0 χ 11 S ^ − 1 χ 11 = 0 , = χ 11 , = χ 10 , S ^ + 1 χ 10 S ^ 0 χ 10 S ^ − 1 χ 10 = − χ 11 , = 0 , = χ 1 − 1 , S ^ + 1 χ 1 − 1 S ^ 0 χ 1 − 1 S ^ − 1 χ 1 − 1 = − χ 10 , = − χ 1 − 1 , = 0. (1302)
S ^ + 1 χ x = − 1 2 χ z , S ^ + 1 χ y = − i 2 χ z , S ^ + 1 χ z = 1 2 ( χ x + i χ y ) , S ^ 0 χ x = i χ y , S ^ 0 χ y = − i χ x , S ^ 0 χ z = 0 , S ^ − 1 χ x = − 1 2 χ z , S ^ − 1 χ y = i 2 χ z , S ^ − 1 χ z = 1 2 ( χ x − i χ y ) . (1303) \begin{aligned}
\hat{S}_{+1}\chi_x&=-\frac{1}{\sqrt{2}}\chi_z, & \hat{S}_{+1}\chi_y&=-\frac{i}{\sqrt{2}}\chi_z, & \hat{S}_{+1}\chi_z&=\frac{1}{\sqrt{2}}(\chi_x+i\chi_y),\\
\hat{S}_{0}\chi_x&=i\chi_y, & \hat{S}_{0}\chi_y&=-i\chi_x, & \hat{S}_{0}\chi_z&=0,\\
\hat{S}_{-1}\chi_x&=-\frac{1}{\sqrt{2}}\chi_z, & \hat{S}_{-1}\chi_y&=\frac{i}{\sqrt{2}}\chi_z, & \hat{S}_{-1}\chi_z&=\frac{1}{\sqrt{2}}(\chi_x-i\chi_y). \tag*{(1303)}
\end{aligned} S ^ + 1 χ x S ^ 0 χ x S ^ − 1 χ x = − 2 1 χ z , = i χ y , = − 2 1 χ z , S ^ + 1 χ y S ^ 0 χ y S ^ − 1 χ y = − 2 i χ z , = − i χ x , = 2 i χ z , S ^ + 1 χ z S ^ 0 χ z S ^ − 1 χ z = 2 1 ( χ x + i χ y ) , = 0 , = 2 1 ( χ x − i χ y ) . (1303)
S ^ x χ 11 = 1 2 χ 10 , S ^ x χ 10 = 1 2 ( χ 1 − 1 + χ 11 ) , S ^ x χ 1 − 1 = 1 2 χ 10 , S ^ y χ 11 = i 2 χ 10 , S ^ y χ 10 = i 2 ( χ 1 − 1 − χ 11 ) , S ^ y χ 1 − 1 = − i 2 χ 10 , S ^ z χ 11 = χ 11 , S ^ z χ 10 = 0 , S ^ z χ 1 − 1 = − χ 1 − 1 . (1304) \begin{aligned}
\hat{S}_x\chi_{11}&=\frac{1}{\sqrt{2}}\chi_{10}, & \hat{S}_x\chi_{10}&=\frac{1}{\sqrt{2}}(\chi_{1-1}+\chi_{11}), & \hat{S}_x\chi_{1-1}&=\frac{1}{\sqrt{2}}\chi_{10},\\
\hat{S}_y\chi_{11}&=\frac{i}{\sqrt{2}}\chi_{10}, & \hat{S}_y\chi_{10}&=\frac{i}{\sqrt{2}}(\chi_{1-1}-\chi_{11}), & \hat{S}_y\chi_{1-1}&=-\frac{i}{\sqrt{2}}\chi_{10},\\
\hat{S}_z\chi_{11}&=\chi_{11}, & \hat{S}_z\chi_{10}&=0, & \hat{S}_z\chi_{1-1}&=-\chi_{1-1}. \tag*{(1304)}
\end{aligned} S ^ x χ 11 S ^ y χ 11 S ^ z χ 11 = 2 1 χ 10 , = 2 i χ 10 , = χ 11 , S ^ x χ 10 S ^ y χ 10 S ^ z χ 10 = 2 1 ( χ 1 − 1 + χ 11 ) , = 2 i ( χ 1 − 1 − χ 11 ) , = 0 , S ^ x χ 1 − 1 S ^ y χ 1 − 1 S ^ z χ 1 − 1 = 2 1 χ 10 , = − 2 i χ 10 , = − χ 1 − 1 . (1304)
S ^ x χ x = 0 , S ^ x χ y = i χ z , S ^ x χ z = − i χ y , S ^ y χ x = − i χ z , S ^ y χ y = 0 , S ^ y χ z = i χ x , S ^ z χ x = i χ y , S ^ z χ y = − i χ x , S ^ z χ z = 0. (1305) \begin{aligned}
\hat{S}_x\chi_x&=0, & \hat{S}_x\chi_y&=i\chi_z, & \hat{S}_x\chi_z&=-i\chi_y,\\
\hat{S}_y\chi_x&=-i\chi_z, & \hat{S}_y\chi_y&=0, & \hat{S}_y\chi_z&=i\chi_x,\\
\hat{S}_z\chi_x&=i\chi_y, & \hat{S}_z\chi_y&=-i\chi_x, & \hat{S}_z\chi_z&=0. \tag*{(1305)}
\end{aligned} S ^ x χ x S ^ y χ x S ^ z χ x = 0 , = − i χ z , = i χ y , S ^ x χ y S ^ y χ y S ^ z χ y = i χ z , = 0 , = − i χ x , S ^ x χ z S ^ y χ z S ^ z χ z = − i χ y , = i χ x , = 0. (1305)
Action of Quadrupole Operators on Basis Functions
The operators T ^ 2 M ( M = ± 2 , ± 1 , 0 ) \hat{T}_{2M}(M=\pm2,\pm1,0) T ^ 2 M ( M = ± 2 , ± 1 , 0 ) and Q ^ i k ( i , k = x , y , z ) \hat{Q}_{ik}(i,k=x,y,z) Q ^ ik ( i , k = x , y , z ) act on basis functions in accordance with
T ^ 2 M χ 1 m = 5 3 ∑ m ′ C 1 m 2 M 1 m ′ χ 1 m ′ , (1306) \hat{T}_{2M}\chi_{1m}=\sqrt{\frac{5}{3}}\sum_{m'}C^{1m'}_{1m\,2M}\chi_{1m'}, \tag*{(1306)} T ^ 2 M χ 1 m = 3 5 m ′ ∑ C 1 m 2 M 1 m ′ χ 1 m ′ , (1306)
Q ^ i k χ l = 1 2 ( 2 3 δ i k χ l − δ i l χ k − δ k l χ i ) . \hat{Q}_{ik}\chi_l=\frac{1}{2}\left(\frac{2}{3}\delta_{ik}\chi_l-\delta_{il}\chi_k-\delta_{kl}\chi_i\right). Q ^ ik χ l = 2 1 ( 3 2 δ ik χ l − δ i l χ k − δ k l χ i ) .
In detailed form we have
T ^ 22 χ 11 = 0 , T ^ 22 χ 10 = 0 , T ^ 22 χ 1 − 1 = χ 11 , T ^ 21 χ 11 = 0 , T ^ 21 χ 10 = − 1 2 χ 11 , T ^ 21 χ 1 − 1 = 1 2 χ 10 , T ^ 20 χ 11 = 1 6 χ 11 , T ^ 20 χ 10 = − 2 3 χ 10 , T ^ 20 χ 1 − 1 = 1 6 χ 1 − 1 , T ^ 2 − 1 χ 11 = 1 2 χ 10 , T ^ 2 − 1 χ 10 = − 1 2 χ 1 − 1 , T ^ 2 − 1 χ 1 − 1 = 0 , T ^ 2 − 2 χ 11 = χ 1 − 1 , T ^ 2 − 2 χ 10 = 0 , T ^ 2 − 2 χ 1 − 1 = 0. (1307) \begin{aligned}
\hat{T}_{22}\chi_{11}&=0, & \hat{T}_{22}\chi_{10}&=0, & \hat{T}_{22}\chi_{1-1}&=\chi_{11},\\
\hat{T}_{21}\chi_{11}&=0, & \hat{T}_{21}\chi_{10}&=-\frac{1}{\sqrt{2}}\chi_{11}, & \hat{T}_{21}\chi_{1-1}&=\frac{1}{\sqrt{2}}\chi_{10},\\
\hat{T}_{20}\chi_{11}&=\frac{1}{\sqrt{6}}\chi_{11}, & \hat{T}_{20}\chi_{10}&=-\sqrt{\frac{2}{3}}\chi_{10}, & \hat{T}_{20}\chi_{1-1}&=\frac{1}{\sqrt{6}}\chi_{1-1},\\
\hat{T}_{2-1}\chi_{11}&=\frac{1}{\sqrt{2}}\chi_{10}, & \hat{T}_{2-1}\chi_{10}&=-\frac{1}{\sqrt{2}}\chi_{1-1}, & \hat{T}_{2-1}\chi_{1-1}&=0,\\
\hat{T}_{2-2}\chi_{11}&=\chi_{1-1}, & \hat{T}_{2-2}\chi_{10}&=0, & \hat{T}_{2-2}\chi_{1-1}&=0. \tag*{(1307)}
\end{aligned} T ^ 22 χ 11 T ^ 21 χ 11 T ^ 20 χ 11 T ^ 2 − 1 χ 11 T ^ 2 − 2 χ 11 = 0 , = 0 , = 6 1 χ 11 , = 2 1 χ 10 , = χ 1 − 1 , T ^ 22 χ 10 T ^ 21 χ 10 T ^ 20 χ 10 T ^ 2 − 1 χ 10 T ^ 2 − 2 χ 10 = 0 , = − 2 1 χ 11 , = − 3 2 χ 10 , = − 2 1 χ 1 − 1 , = 0 , T ^ 22 χ 1 − 1 T ^ 21 χ 1 − 1 T ^ 20 χ 1 − 1 T ^ 2 − 1 χ 1 − 1 T ^ 2 − 2 χ 1 − 1 = χ 11 , = 2 1 χ 10 , = 6 1 χ 1 − 1 , = 0 , = 0. (1307)
T ^ 22 χ x = − 1 2 ( χ x + i χ y ) , T ^ 22 χ y = 1 2 ( χ y − i χ x ) , T ^ 22 χ z = 0 , T ^ 21 χ x = 1 2 χ z , T ^ 21 χ y = 1 2 χ z , T ^ 21 χ z = 1 2 ( χ x + i χ y ) , T ^ 20 χ x = 1 6 χ x , T ^ 20 χ y = 1 6 χ y , T ^ 20 χ z = − 2 3 χ z , T ^ 2 − 1 χ x = − 1 2 χ z , T ^ 2 − 1 χ y = 1 2 χ z , T ^ 2 − 1 χ z = − 1 2 ( χ x − i χ y ) , T ^ 2 − 2 χ x = − 1 2 ( χ x − i χ y ) , T ^ 2 − 2 χ y = 1 2 ( χ y + i χ x ) , T ^ 2 − 2 χ z = 0. (1308) \begin{aligned}
\hat{T}_{22}\chi_x&=-\frac{1}{2}(\chi_x+i\chi_y), & \hat{T}_{22}\chi_y&=\frac{1}{2}(\chi_y-i\chi_x), & \hat{T}_{22}\chi_z&=0,\\
\hat{T}_{21}\chi_x&=\frac{1}{2}\chi_z, & \hat{T}_{21}\chi_y&=\frac{1}{2}\chi_z, & \hat{T}_{21}\chi_z&=\frac{1}{2}(\chi_x+i\chi_y),\\
\hat{T}_{20}\chi_x&=\frac{1}{\sqrt{6}}\chi_x, & \hat{T}_{20}\chi_y&=\frac{1}{\sqrt{6}}\chi_y, & \hat{T}_{20}\chi_z&=-\sqrt{\frac{2}{3}}\chi_z,\\
\hat{T}_{2-1}\chi_x&=-\frac{1}{2}\chi_z, & \hat{T}_{2-1}\chi_y&=\frac{1}{2}\chi_z, & \hat{T}_{2-1}\chi_z&=-\frac{1}{2}(\chi_x-i\chi_y),\\
\hat{T}_{2-2}\chi_x&=-\frac{1}{2}(\chi_x-i\chi_y), & \hat{T}_{2-2}\chi_y&=\frac{1}{2}(\chi_y+i\chi_x), & \hat{T}_{2-2}\chi_z&=0. \tag*{(1308)}
\end{aligned} T ^ 22 χ x T ^ 21 χ x T ^ 20 χ x T ^ 2 − 1 χ x T ^ 2 − 2 χ x = − 2 1 ( χ x + i χ y ) , = 2 1 χ z , = 6 1 χ x , = − 2 1 χ z , = − 2 1 ( χ x − i χ y ) , T ^ 22 χ y T ^ 21 χ y T ^ 20 χ y T ^ 2 − 1 χ y T ^ 2 − 2 χ y = 2 1 ( χ y − i χ x ) , = 2 1 χ z , = 6 1 χ y , = 2 1 χ z , = 2 1 ( χ y + i χ x ) , T ^ 22 χ z T ^ 21 χ z T ^ 20 χ z T ^ 2 − 1 χ z T ^ 2 − 2 χ z = 0 , = 2 1 ( χ x + i χ y ) , = − 3 2 χ z , = − 2 1 ( χ x − i χ y ) , = 0. (1308)
Q ^ x x χ 11 = 1 2 χ 1 − 1 − 1 6 χ 11 , Q ^ x x χ 10 = 1 3 χ 10 , Q ^ x x χ 1 − 1 = 1 2 χ 11 − 1 6 χ 1 − 1 , Q ^ y y χ 11 = − 1 2 χ 1 − 1 − 1 6 χ 11 , Q ^ y y χ 10 = 1 3 χ 10 , Q ^ y y χ 1 − 1 = − 1 2 χ 11 − 1 6 χ 1 − 1 , Q ^ z z χ 11 = 1 3 χ 11 , Q ^ z z χ 10 = − 2 3 χ 10 , Q ^ z z χ 1 − 1 = 1 3 χ 1 − 1 , Q ^ x y χ 11 = i 2 χ 1 − 1 , Q ^ x y χ 10 = 0 , Q ^ x y χ 1 − 1 = − i 2 χ 11 , Q ^ x z χ 11 = 1 2 χ 10 , Q ^ x z χ 10 = 1 2 2 ( χ 11 − χ 1 − 1 ) , Q ^ x z χ 1 − 1 = − 1 2 χ 10 , Q ^ y z χ 11 = i 2 2 χ 10 , Q ^ y z χ 10 = − i 2 2 ( χ 11 + χ 1 − 1 ) , Q ^ y z χ 1 − 1 = i 2 2 χ 10 . (1309) \begin{aligned}
\hat{Q}_{xx}\chi_{11}&=\frac{1}{2}\chi_{1-1}-\frac{1}{6}\chi_{11}, & \hat{Q}_{xx}\chi_{10}&=\frac{1}{3}\chi_{10}, & \hat{Q}_{xx}\chi_{1-1}&=\frac{1}{2}\chi_{11}-\frac{1}{6}\chi_{1-1},\\
\hat{Q}_{yy}\chi_{11}&=-\frac{1}{2}\chi_{1-1}-\frac{1}{6}\chi_{11}, & \hat{Q}_{yy}\chi_{10}&=\frac{1}{3}\chi_{10}, & \hat{Q}_{yy}\chi_{1-1}&=-\frac{1}{2}\chi_{11}-\frac{1}{6}\chi_{1-1},\\
\hat{Q}_{zz}\chi_{11}&=\frac{1}{3}\chi_{11}, & \hat{Q}_{zz}\chi_{10}&=-\frac{2}{3}\chi_{10}, & \hat{Q}_{zz}\chi_{1-1}&=\frac{1}{3}\chi_{1-1},\\
\hat{Q}_{xy}\chi_{11}&=\frac{i}{2}\chi_{1-1}, & \hat{Q}_{xy}\chi_{10}&=0, & \hat{Q}_{xy}\chi_{1-1}&=-\frac{i}{2}\chi_{11},\\
\hat{Q}_{xz}\chi_{11}&=\frac{1}{2}\chi_{10}, & \hat{Q}_{xz}\chi_{10}&=\frac{1}{2\sqrt{2}}(\chi_{11}-\chi_{1-1}), & \hat{Q}_{xz}\chi_{1-1}&=-\frac{1}{2}\chi_{10},\\
\hat{Q}_{yz}\chi_{11}&=\frac{i}{2\sqrt{2}}\chi_{10}, & \hat{Q}_{yz}\chi_{10}&=-\frac{i}{2\sqrt{2}}(\chi_{11}+\chi_{1-1}), & \hat{Q}_{yz}\chi_{1-1}&=\frac{i}{2\sqrt{2}}\chi_{10}. \tag*{(1309)}
\end{aligned} Q ^ xx χ 11 Q ^ y y χ 11 Q ^ z z χ 11 Q ^ x y χ 11 Q ^ x z χ 11 Q ^ y z χ 11 = 2 1 χ 1 − 1 − 6 1 χ 11 , = − 2 1 χ 1 − 1 − 6 1 χ 11 , = 3 1 χ 11 , = 2 i χ 1 − 1 , = 2 1 χ 10 , = 2 2 i χ 10 , Q ^ xx χ 10 Q ^ y y χ 10 Q ^ z z χ 10 Q ^ x y χ 10 Q ^ x z χ 10 Q ^ y z χ 10 = 3 1 χ 10 , = 3 1 χ 10 , = − 3 2 χ 10 , = 0 , = 2 2 1 ( χ 11 − χ 1 − 1 ) , = − 2 2 i ( χ 11 + χ 1 − 1 ) , Q ^ xx χ 1 − 1 Q ^ y y χ 1 − 1 Q ^ z z χ 1 − 1 Q ^ x y χ 1 − 1 Q ^ x z χ 1 − 1 Q ^ y z χ 1 − 1 = 2 1 χ 11 − 6 1 χ 1 − 1 , = − 2 1 χ 11 − 6 1 χ 1 − 1 , = 3 1 χ 1 − 1 , = − 2 i χ 11 , = − 2 1 χ 10 , = 2 2 i χ 10 . (1309)
Q ^ x x χ x = − 2 3 χ x , Q ^ x x χ y = 1 3 χ y , Q ^ x x χ z = 1 3 χ z , Q ^ y y χ x = 1 3 χ x , Q ^ y y χ y = − 2 3 χ y , Q ^ y y χ z = 1 3 χ z , Q ^ z z χ x = 1 3 χ x , Q ^ z z χ y = 1 3 χ y , Q ^ z z χ z = − 2 3 χ z , Q ^ x y χ x = − 1 2 χ y , Q ^ x y χ y = − 1 2 χ x , Q ^ x y χ z = 0 , Q ^ x z χ x = − 1 2 χ z , Q ^ x z χ y = 0 , Q ^ x z χ z = − 1 2 χ x , Q ^ y z χ x = 0 , Q ^ y z χ y = − 1 2 χ z , Q ^ y z χ z = − 1 2 χ y . (1310) \begin{aligned}
\hat{Q}_{xx}\chi_x&=-\frac{2}{3}\chi_x, & \hat{Q}_{xx}\chi_y&=\frac{1}{3}\chi_y, & \hat{Q}_{xx}\chi_z&=\frac{1}{3}\chi_z,\\
\hat{Q}_{yy}\chi_x&=\frac{1}{3}\chi_x, & \hat{Q}_{yy}\chi_y&=-\frac{2}{3}\chi_y, & \hat{Q}_{yy}\chi_z&=\frac{1}{3}\chi_z,\\
\hat{Q}_{zz}\chi_x&=\frac{1}{3}\chi_x, & \hat{Q}_{zz}\chi_y&=\frac{1}{3}\chi_y, & \hat{Q}_{zz}\chi_z&=-\frac{2}{3}\chi_z,\\
\hat{Q}_{xy}\chi_x&=-\frac{1}{2}\chi_y, & \hat{Q}_{xy}\chi_y&=-\frac{1}{2}\chi_x, & \hat{Q}_{xy}\chi_z&=0,\\
\hat{Q}_{xz}\chi_x&=-\frac{1}{2}\chi_z, & \hat{Q}_{xz}\chi_y&=0, & \hat{Q}_{xz}\chi_z&=-\frac{1}{2}\chi_x,\\
\hat{Q}_{yz}\chi_x&=0, & \hat{Q}_{yz}\chi_y&=-\frac{1}{2}\chi_z, & \hat{Q}_{yz}\chi_z&=-\frac{1}{2}\chi_y. \tag*{(1310)}
\end{aligned} Q ^ xx χ x Q ^ y y χ x Q ^ z z χ x Q ^ x y χ x Q ^ x z χ x Q ^ y z χ x = − 3 2 χ x , = 3 1 χ x , = 3 1 χ x , = − 2 1 χ y , = − 2 1 χ z , = 0 , Q ^ xx χ y Q ^ y y χ y Q ^ z z χ y Q ^ x y χ y Q ^ x z χ y Q ^ y z χ y = 3 1 χ y , = − 3 2 χ y , = 3 1 χ y , = − 2 1 χ x , = 0 , = − 2 1 χ z , Q ^ xx χ z Q ^ y y χ z Q ^ z z χ z Q ^ x y χ z Q ^ x z χ z Q ^ y z χ z = 3 1 χ z , = 3 1 χ z , = − 3 2 χ z , = 0 , = − 2 1 χ x , = − 2 1 χ y . (1310)
Transformation of Basis Functions Under Rotations of the Coordinate Systems
The spin functions χ 1 m \chi_{1m} χ 1 m (m = ± 1 , 0 m=\pm1,0 m = ± 1 , 0 ) and χ i \chi_i χ i (i = x , y , z i=x,y,z i = x , y , z ) constitute the covariant spherical basis and cartesian basis, respectively. Under rotations they transform through the rotation operator D ^ 1 ( α , β , γ ) \hat{D}^{1}(\alpha,\beta,\gamma) D ^ 1 ( α , β , γ ) (Eqs. (418) , (421) ) if rotations are specified by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ or through the rotation operator U ^ 1 ( ω ; Θ , Φ ) \hat{U}^{1}(\omega;\Theta,\Phi) U ^ 1 ( ω ; Θ , Φ ) (Eqs. (422) –(426) ) if rotations are described by the rotation angle ω \omega ω and the rotation axis n ( Θ , Φ ) \boldsymbol{n}(\Theta,\Phi) n ( Θ , Φ ) .
χ 1 m ′ ′ = D ^ 1 ( α , β , γ ) χ 1 m ′ = ∑ m = ± 1 , 0 D m m ′ 1 ( α , β , γ ) χ 1 m , \chi'_{1m'}=\hat{D}^{1}(\alpha,\beta,\gamma)\chi_{1m'}=\sum_{m=\pm1,0}D^{1}_{mm'}(\alpha,\beta,\gamma)\chi_{1m}, χ 1 m ′ ′ = D ^ 1 ( α , β , γ ) χ 1 m ′ = m = ± 1 , 0 ∑ D m m ′ 1 ( α , β , γ ) χ 1 m ,
χ i ′ = D ^ 1 ( α , β , γ ) χ i = ∑ k = x , y , z a k i χ k . (1311) \chi'_i=\hat{D}^{1}(\alpha,\beta,\gamma)\chi_i=\sum_{k=x,y,z}a_{ki}\chi_k. \tag*{(1311)} χ i ′ = D ^ 1 ( α , β , γ ) χ i = k = x , y , z ∑ a k i χ k . (1311)
Here D m m ′ 1 ( α , β , γ ) D^{1}_{mm'}(\alpha,\beta,\gamma) D m m ′ 1 ( α , β , γ ) are the Wigner D D D -functions (Chap. 4) and a i k a_{ik} a ik are elements of the rotation matrix (§ 9.10.2 ).
The functions χ 1 m ′ ′ \chi'_{1m'} χ 1 m ′ ′ describe quantum states in which a particle of spin 1 has the spin projection m ′ m' m ′ on the new z ′ z' z ′ -axis. These functions are eigenfunctions of the operators S ^ ′ 2 \hat{S}'^{2} S ^ ′ 2 and S ^ z ′ \hat{S}'_z S ^ z ′ where S ^ ′ \hat{\boldsymbol{S}}' S ^ ′ is the spin operator in the rotated coordinate system (Eqs. (428) and (430) )
S ^ ′ 2 χ 1 m ′ ′ = 2 χ 1 m ′ ′ , \hat{S}'^{2}\chi'_{1m'}=2\chi'_{1m'}, S ^ ′ 2 χ 1 m ′ ′ = 2 χ 1 m ′ ′ ,
S ^ z ′ χ 1 m ′ ′ = m ′ χ 1 m ′ ′ . (1312) \hat{S}'_z\chi'_{1m'}=m'\chi'_{1m'}. \tag*{(1312)} S ^ z ′ χ 1 m ′ ′ = m ′ χ 1 m ′ ′ . (1312)
In the spherical basis representation the functions χ 1 m ′ ′ \chi'_{1m'} χ 1 m ′ ′ have the form
χ 11 ′ = ( 1 + cos β 2 e − i ( α + γ ) sin β 2 e − i γ 1 − cos β 2 e i ( α − γ ) ) , χ 10 ′ = ( − sin β 2 e − i α cos β sin β 2 e i α ) , χ 1 − 1 ′ = ( 1 − cos β 2 e − i ( α − γ ) − sin β 2 e i γ 1 + cos β 2 e i ( α + γ ) ) . (1313) \chi'_{11}=
\begin{pmatrix}
\dfrac{1+\cos\beta}{2}e^{-i(\alpha+\gamma)}\\
\dfrac{\sin\beta}{\sqrt{2}}e^{-i\gamma}\\
\dfrac{1-\cos\beta}{2}e^{i(\alpha-\gamma)}
\end{pmatrix},
\qquad
\chi'_{10}=
\begin{pmatrix}
-\dfrac{\sin\beta}{\sqrt{2}}e^{-i\alpha}\\
\cos\beta\\
\dfrac{\sin\beta}{\sqrt{2}}e^{i\alpha}
\end{pmatrix},
\qquad
\chi'_{1-1}=
\begin{pmatrix}
\dfrac{1-\cos\beta}{2}e^{-i(\alpha-\gamma)}\\
-\dfrac{\sin\beta}{\sqrt{2}}e^{i\gamma}\\
\dfrac{1+\cos\beta}{2}e^{i(\alpha+\gamma)}
\end{pmatrix}. \tag*{(1313)} χ 11 ′ = 2 1 + cos β e − i ( α + γ ) 2 sin β e − iγ 2 1 − cos β e i ( α − γ ) , χ 10 ′ = − 2 sin β e − i α cos β 2 sin β e i α , χ 1 − 1 ′ = 2 1 − cos β e − i ( α − γ ) − 2 sin β e iγ 2 1 + cos β e i ( α + γ ) . (1313)
and the functions χ i ′ \chi'_i χ i ′ are given by
χ x ′ = ( − cos β cos γ − i sin γ 2 e − i α − sin β cos γ cos β cos γ + i sin γ 2 e i α ) , χ y ′ = ( cos β sin γ + i cos γ 2 e − i α sin β sin γ − cos β sin γ − i cos γ 2 e i α ) , χ z ′ = ( − sin β 2 e − i α cos β sin β 2 e i α ) . (1314) \chi'_x=
\begin{pmatrix}
-\dfrac{\cos\beta\cos\gamma-i\sin\gamma}{\sqrt{2}}e^{-i\alpha}\\
-\sin\beta\cos\gamma\\
\dfrac{\cos\beta\cos\gamma+i\sin\gamma}{\sqrt{2}}e^{i\alpha}
\end{pmatrix},
\qquad
\chi'_y=
\begin{pmatrix}
\dfrac{\cos\beta\sin\gamma+i\cos\gamma}{\sqrt{2}}e^{-i\alpha}\\
\sin\beta\sin\gamma\\
-\dfrac{\cos\beta\sin\gamma-i\cos\gamma}{\sqrt{2}}e^{i\alpha}
\end{pmatrix},
\qquad
\chi'_z=
\begin{pmatrix}
-\dfrac{\sin\beta}{\sqrt{2}}e^{-i\alpha}\\
\cos\beta\\
\dfrac{\sin\beta}{\sqrt{2}}e^{i\alpha}
\end{pmatrix}. \tag*{(1314)} χ x ′ = − 2 cos β cos γ − i sin γ e − i α − sin β cos γ 2 cos β cos γ + i sin γ e i α , χ y ′ = 2 cos β sin γ + i cos γ e − i α sin β sin γ − 2 cos β sin γ − i cos γ e i α , χ z ′ = − 2 sin β e − i α cos β 2 sin β e i α . (1314)
In the cartesian basis representation these functions read
χ 11 ′ = ( − cos β cos α − i sin α 2 e − i γ − cos β sin α + i cos α 2 e − i γ sin β 2 e − i γ ) , χ 10 ′ = ( sin β cos α sin β sin α cos β ) , χ 1 − 1 ′ = ( cos β cos α + i sin α 2 e i γ cos β sin α − i cos α 2 e i γ − sin β 2 e i γ ) . (1315) \chi'_{11}=
\begin{pmatrix}
-\dfrac{\cos\beta\cos\alpha-i\sin\alpha}{\sqrt{2}}e^{-i\gamma}\\
-\dfrac{\cos\beta\sin\alpha+i\cos\alpha}{\sqrt{2}}e^{-i\gamma}\\
\dfrac{\sin\beta}{\sqrt{2}}e^{-i\gamma}
\end{pmatrix},
\qquad
\chi'_{10}=
\begin{pmatrix}
\sin\beta\cos\alpha\\
\sin\beta\sin\alpha\\
\cos\beta
\end{pmatrix},
\qquad
\chi'_{1-1}=
\begin{pmatrix}
\dfrac{\cos\beta\cos\alpha+i\sin\alpha}{\sqrt{2}}e^{i\gamma}\\
\dfrac{\cos\beta\sin\alpha-i\cos\alpha}{\sqrt{2}}e^{i\gamma}\\
-\dfrac{\sin\beta}{\sqrt{2}}e^{i\gamma}
\end{pmatrix}. \tag*{(1315)} χ 11 ′ = − 2 cos β cos α − i sin α e − iγ − 2 cos β sin α + i cos α e − iγ 2 sin β e − iγ , χ 10 ′ = sin β cos α sin β sin α cos β , χ 1 − 1 ′ = 2 cos β cos α + i sin α e iγ 2 cos β sin α − i cos α e iγ − 2 sin β e iγ . (1315)
χ x ′ = ( cos β cos α cos γ − sin α sin γ cos β sin α cos γ + cos α sin γ − sin β cos γ ) , χ y ′ = ( − cos β cos α sin γ − sin α cos γ − cos β sin α sin γ + cos α cos γ sin β sin γ ) , χ z ′ = ( sin β cos α sin β sin α cos β ) . (1316) \chi'_x=
\begin{pmatrix}
\cos\beta\cos\alpha\cos\gamma-\sin\alpha\sin\gamma\\
\cos\beta\sin\alpha\cos\gamma+\cos\alpha\sin\gamma\\
-\sin\beta\cos\gamma
\end{pmatrix},
\qquad
\chi'_y=
\begin{pmatrix}
-\cos\beta\cos\alpha\sin\gamma-\sin\alpha\cos\gamma\\
-\cos\beta\sin\alpha\sin\gamma+\cos\alpha\cos\gamma\\
\sin\beta\sin\gamma
\end{pmatrix},
\qquad
\chi'_z=
\begin{pmatrix}
\sin\beta\cos\alpha\\
\sin\beta\sin\alpha\\
\cos\beta
\end{pmatrix}. \tag*{(1316)} χ x ′ = cos β cos α cos γ − sin α sin γ cos β sin α cos γ + cos α sin γ − sin β cos γ , χ y ′ = − cos β cos α sin γ − sin α cos γ − cos β sin α sin γ + cos α cos γ sin β sin γ , χ z ′ = sin β cos α sin β sin α cos β . (1316)
The expansions for the basis functions in the rotated coordinate system in terms of the angles ω \omega ω , Θ \Theta Θ , Φ \Phi Φ may be obtained by the use of the rotation operator U ^ 1 ( ω ; Θ , Φ ) \hat{U}^{1}(\omega;\Theta,\Phi) U ^ 1 ( ω ; Θ , Φ ) (Eqs. (423) –(424) ).
Helicity Basis Functions for S = 1 S=1 S = 1
The helicity basis functions χ 1 λ ( ϑ , φ ) \chi_{1\lambda}(\vartheta,\varphi) χ 1 λ ( ϑ , φ ) (λ = ± 1 , 0 \lambda=\pm1,0 λ = ± 1 , 0 ) describe the states in which the spin projection on the linear momentum direction n ( ϑ , φ ) ≡ p / ∣ p ∣ \boldsymbol{n}(\vartheta,\varphi)\equiv\boldsymbol{p}/|\boldsymbol{p}| n ( ϑ , φ ) ≡ p /∣ p ∣ is equal to λ \lambda λ . The functions χ 1 λ ( ϑ , φ ) \chi_{1\lambda}(\vartheta,\varphi) χ 1 λ ( ϑ , φ ) are eigenfunctions of the operators S ^ 2 \hat{S}^{2} S ^ 2 and S ^ ⋅ n \hat{\boldsymbol{S}}\cdot\boldsymbol{n} S ^ ⋅ n
S ^ 2 χ 1 λ ( ϑ , φ ) = 2 χ 1 λ ( ϑ , φ ) , \hat{S}^{2}\chi_{1\lambda}(\vartheta,\varphi)=2\chi_{1\lambda}(\vartheta,\varphi), S ^ 2 χ 1 λ ( ϑ , φ ) = 2 χ 1 λ ( ϑ , φ ) ,
S ^ ⋅ n χ 1 λ ( ϑ , φ ) = λ χ 1 λ ( ϑ , φ ) . (1317) \hat{\boldsymbol{S}}\cdot\boldsymbol{n}\chi_{1\lambda}(\vartheta,\varphi)=\lambda\chi_{1\lambda}(\vartheta,\varphi). \tag*{(1317)} S ^ ⋅ n χ 1 λ ( ϑ , φ ) = λ χ 1 λ ( ϑ , φ ) . (1317)
According to Eq. (1178) , the helicity basis functions may be derived from the functions χ 1 m \chi_{1m} χ 1 m by a coordinate rotation
χ 1 λ ( ϑ , φ ) = ∑ m D m λ 1 ( φ , ϑ , 0 ) χ 1 m , \chi_{1\lambda}(\vartheta,\varphi)=\sum_m D^{1}_{m\lambda}(\varphi,\vartheta,0)\chi_{1m}, χ 1 λ ( ϑ , φ ) = m ∑ D mλ 1 ( φ , ϑ , 0 ) χ 1 m ,
χ 1 m = ∑ λ D − λ − m 1 ( 0 , ϑ , φ ) χ 1 λ ( ϑ , φ ) . (1318) \chi_{1m}=\sum_\lambda D^{1}_{-\lambda-m}(0,\vartheta,\varphi)\chi_{1\lambda}(\vartheta,\varphi). \tag*{(1318)} χ 1 m = λ ∑ D − λ − m 1 ( 0 , ϑ , φ ) χ 1 λ ( ϑ , φ ) . (1318)
For Hermitian adjoint functions Eqs. (1318) assume the form
χ 1 λ † ( ϑ , φ ) = ∑ m ( − 1 ) λ − m D − m − λ 1 ( φ , ϑ , 0 ) χ 1 m † , \chi^{\dagger}_{1\lambda}(\vartheta,\varphi)=\sum_{m}(-1)^{\lambda-m}D^{1}_{-m-\lambda}(\varphi,\vartheta,0)\chi^{\dagger}_{1m}, χ 1 λ † ( ϑ , φ ) = m ∑ ( − 1 ) λ − m D − m − λ 1 ( φ , ϑ , 0 ) χ 1 m † ,
χ 1 m † = ∑ λ ( − 1 ) λ − m D λ m 1 ( 0 , ϑ , φ ) χ 1 λ † ( ϑ , φ ) . (1319) \chi^{\dagger}_{1m}=\sum_{\lambda}(-1)^{\lambda-m}D^{1}_{\lambda m}(0,\vartheta,\varphi)\chi^{\dagger}_{1\lambda}(\vartheta,\varphi). \tag*{(1319)} χ 1 m † = λ ∑ ( − 1 ) λ − m D λm 1 ( 0 , ϑ , φ ) χ 1 λ † ( ϑ , φ ) . (1319)
The helicity functions χ 1 λ ( ϑ , φ ) \chi_{1\lambda}(\vartheta,\varphi) χ 1 λ ( ϑ , φ ) may be treated as the covariant helicity basis vectors e λ ′ \mathbf{e}'_{\lambda} e λ ′ (see § 9.1.4 ) written as column matrices
χ 1 λ ( ϑ , φ ) = e λ ′ ( λ = ± 1 , 0 ) . (1320) \chi_{1\lambda}(\vartheta,\varphi)=\mathbf{e}'_{\lambda}\qquad(\lambda=\pm1,0). \tag*{(1320)} χ 1 λ ( ϑ , φ ) = e λ ′ ( λ = ± 1 , 0 ) . (1320)
One can construct the following linear combinations χ i ( ϑ , φ ) \chi_i(\vartheta,\varphi) χ i ( ϑ , φ ) (i = x , y , z i=x,y,z i = x , y , z ) of the helicity functions χ 1 λ ( ϑ , φ ) \chi_{1\lambda}(\vartheta,\varphi) χ 1 λ ( ϑ , φ ) :
χ x ( ϑ , φ ) = 1 2 { χ 1 − 1 ( ϑ , φ ) − χ 11 ( ϑ , φ ) } , \chi_x(\vartheta,\varphi)=\frac{1}{\sqrt{2}}\{\chi_{1-1}(\vartheta,\varphi)-\chi_{11}(\vartheta,\varphi)\}, χ x ( ϑ , φ ) = 2 1 { χ 1 − 1 ( ϑ , φ ) − χ 11 ( ϑ , φ )} ,
χ y ( ϑ , φ ) = i 2 { χ 1 − 1 ( ϑ , φ ) + χ 11 ( ϑ , φ ) } , \chi_y(\vartheta,\varphi)=\frac{i}{\sqrt{2}}\{\chi_{1-1}(\vartheta,\varphi)+\chi_{11}(\vartheta,\varphi)\}, χ y ( ϑ , φ ) = 2 i { χ 1 − 1 ( ϑ , φ ) + χ 11 ( ϑ , φ )} ,
χ z ( ϑ , φ ) = χ 10 ( ϑ , φ ) . (1321) \chi_z(\vartheta,\varphi)=\chi_{10}(\vartheta,\varphi). \tag*{(1321)} χ z ( ϑ , φ ) = χ 10 ( ϑ , φ ) . (1321)
The functions χ i ( ϑ , φ ) \chi_i(\vartheta,\varphi) χ i ( ϑ , φ ) (i = x , y , z i=x,y,z i = x , y , z ) coincide with the polar basis vectors (see § 9.1.3 ):
χ x ( ϑ , φ ) ≡ e ϑ , χ y ( ϑ , φ ) ≡ e φ , χ z ( ϑ , φ ) = e r . (1322) \chi_x(\vartheta,\varphi)\equiv\mathbf{e}_{\vartheta},\quad\chi_y(\vartheta,\varphi)\equiv\mathbf{e}_{\varphi},\quad\chi_z(\vartheta,\varphi)=\mathbf{e}_{r}. \tag*{(1322)} χ x ( ϑ , φ ) ≡ e ϑ , χ y ( ϑ , φ ) ≡ e φ , χ z ( ϑ , φ ) = e r . (1322)
It should be emphasized that the functions χ i ( ϑ , φ ) \chi_i(\vartheta,\varphi) χ i ( ϑ , φ ) describe the states with no definite helicity. These functions are related to the basis functions χ k \chi_k χ k (k = x , y , z k=x,y,z k = x , y , z ) (§ 25.4.1 ) by
χ i ( ϑ , φ ) = ∑ k a k i ( φ , ϑ , 0 ) χ k , \chi_i(\vartheta,\varphi)=\sum_k a_{ki}(\varphi,\vartheta,0)\chi_k, χ i ( ϑ , φ ) = k ∑ a k i ( φ , ϑ , 0 ) χ k ,
χ k = ∑ i a i k ( φ , ϑ , 0 ) χ i ( ϑ , φ ) . (1323) \chi_k=\sum_i a_{ik}(\varphi,\vartheta,0)\chi_i(\vartheta,\varphi). \tag*{(1323)} χ k = i ∑ a ik ( φ , ϑ , 0 ) χ i ( ϑ , φ ) . (1323)
where a i k ( α , β , γ ) a_{ik}(\alpha,\beta,\gamma) a ik ( α , β , γ ) are elements of the rotation matrix given by Eqs. (183) . The relations between the Hermitian conjugate functions χ i † ( ϑ , φ ) \chi_i^\dagger(\vartheta,\varphi) χ i † ( ϑ , φ ) and χ k † \chi_k^\dagger χ k † are also given by Eq. (1323) because the elements a i k a_{ik} a ik are real.
Three helicity functions χ 1 λ ( ϑ , φ ) \chi_{1\lambda}(\vartheta,\varphi) χ 1 λ ( ϑ , φ ) with λ = ± 1 , 0 \lambda=\pm1,0 λ = ± 1 , 0 or, equivalently, three functions χ i ( ϑ , φ ) \chi_i(\vartheta,\varphi) χ i ( ϑ , φ ) with i = x , y , z i=x,y,z i = x , y , z constitute an orthonormalized basis. The orthogonality and normalization conditions have the form
χ 1 λ † ( ϑ , φ ) χ 1 λ ′ ( ϑ , φ ) = δ λ λ ′ , \chi^\dagger_{1\lambda}(\vartheta,\varphi)\chi_{1\lambda'}(\vartheta,\varphi)=\delta_{\lambda\lambda'}, χ 1 λ † ( ϑ , φ ) χ 1 λ ′ ( ϑ , φ ) = δ λ λ ′ ,
χ i † ( ϑ , φ ) χ k ( ϑ , φ ) = δ i k . (1324) \chi^\dagger_i(\vartheta,\varphi)\chi_k(\vartheta,\varphi)=\delta_{ik}. \tag*{(1324)} χ i † ( ϑ , φ ) χ k ( ϑ , φ ) = δ ik . (1324)
The completeness conditions are
∑ λ = ± 1 , 0 χ 1 λ ( ϑ , φ ) χ 1 λ † ( ϑ , φ ) = I ^ , \sum_{\lambda=\pm1,0}\chi_{1\lambda}(\vartheta,\varphi)\chi^\dagger_{1\lambda}(\vartheta,\varphi)=\hat{I}, λ = ± 1 , 0 ∑ χ 1 λ ( ϑ , φ ) χ 1 λ † ( ϑ , φ ) = I ^ ,
∑ i = x , y , z χ i ( ϑ , φ ) χ i † ( ϑ , φ ) = I ^ . (1325) \sum_{i=x,y,z}\chi_i(\vartheta,\varphi)\chi^\dagger_i(\vartheta,\varphi)=\hat{I}. \tag*{(1325)} i = x , y , z ∑ χ i ( ϑ , φ ) χ i † ( ϑ , φ ) = I ^ . (1325)
Explicit form of the helicity functions χ 1 λ ( ϑ , φ ) \chi_{1\lambda}(\vartheta,\varphi) χ 1 λ ( ϑ , φ ) and the functions χ i ( ϑ , φ ) \chi_i(\vartheta,\varphi) χ i ( ϑ , φ )
Spherical basis representation
χ 11 ( ϑ , φ ) = ( 1 + cos ϑ 2 e − i φ sin ϑ 2 1 − cos ϑ 2 e i φ ) , χ 10 ( ϑ , φ ) = ( − sin ϑ 2 e − i φ cos ϑ sin ϑ 2 e i φ ) , χ 1 − 1 ( ϑ , φ ) = ( 1 − cos ϑ 2 e − i φ − sin ϑ 2 1 + cos ϑ 2 e i φ ) . (1326) \chi_{11}(\vartheta,\varphi)=
\begin{pmatrix}
\frac{1+\cos\vartheta}{2}e^{-i\varphi}\\
\frac{\sin\vartheta}{\sqrt{2}}\\
\frac{1-\cos\vartheta}{2}e^{i\varphi}
\end{pmatrix},
\quad
\chi_{10}(\vartheta,\varphi)=
\begin{pmatrix}
-\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}\\
\cos\vartheta\\
\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}
\end{pmatrix},
\quad
\chi_{1-1}(\vartheta,\varphi)=
\begin{pmatrix}
\frac{1-\cos\vartheta}{2}e^{-i\varphi}\\
-\frac{\sin\vartheta}{\sqrt{2}}\\
\frac{1+\cos\vartheta}{2}e^{i\varphi}
\end{pmatrix}. \tag*{(1326)} χ 11 ( ϑ , φ ) = 2 1 + c o s ϑ e − i φ 2 s i n ϑ 2 1 − c o s ϑ e i φ , χ 10 ( ϑ , φ ) = − 2 s i n ϑ e − i φ cos ϑ 2 s i n ϑ e i φ , χ 1 − 1 ( ϑ , φ ) = 2 1 − c o s ϑ e − i φ − 2 s i n ϑ 2 1 + c o s ϑ e i φ . (1326)
χ x ( ϑ , φ ) = ( − cos ϑ 2 e − i φ − sin ϑ cos ϑ 2 e i φ ) , χ y ( ϑ , φ ) = ( i 2 e − i φ 0 i 2 e i φ ) , χ z ( ϑ , φ ) = ( − sin ϑ 2 e − i φ cos ϑ sin ϑ 2 e i φ ) . (1327) \chi_x(\vartheta,\varphi)=
\begin{pmatrix}
-\frac{\cos\vartheta}{\sqrt{2}}e^{-i\varphi}\\
-\sin\vartheta\\
\frac{\cos\vartheta}{\sqrt{2}}e^{i\varphi}
\end{pmatrix},
\quad
\chi_y(\vartheta,\varphi)=
\begin{pmatrix}
\frac{i}{\sqrt{2}}e^{-i\varphi}\\
0\\
\frac{i}{\sqrt{2}}e^{i\varphi}
\end{pmatrix},
\quad
\chi_z(\vartheta,\varphi)=
\begin{pmatrix}
-\frac{\sin\vartheta}{\sqrt{2}}e^{-i\varphi}\\
\cos\vartheta\\
\frac{\sin\vartheta}{\sqrt{2}}e^{i\varphi}
\end{pmatrix}. \tag*{(1327)} χ x ( ϑ , φ ) = − 2 c o s ϑ e − i φ − sin ϑ 2 c o s ϑ e i φ , χ y ( ϑ , φ ) = 2 i e − i φ 0 2 i e i φ , χ z ( ϑ , φ ) = − 2 s i n ϑ e − i φ cos ϑ 2 s i n ϑ e i φ . (1327)
General Spin Functions for S = 1 S=1 S = 1
(b) Cartesian basis representation
χ 11 ( ϑ , φ ) = − 1 2 ( cos ϑ cos φ − i sin φ cos ϑ sin φ + i cos φ − sin ϑ ) , χ 10 ( ϑ , φ ) = ( sin ϑ cos φ sin ϑ sin φ cos ϑ ) , \chi_{11}(\vartheta,\varphi)=-\frac{1}{\sqrt{2}}
\begin{pmatrix}
\cos\vartheta\cos\varphi-i\sin\varphi\\
\cos\vartheta\sin\varphi+i\cos\varphi\\
-\sin\vartheta
\end{pmatrix},\qquad
\chi_{10}(\vartheta,\varphi)=
\begin{pmatrix}
\sin\vartheta\cos\varphi\\
\sin\vartheta\sin\varphi\\
\cos\vartheta
\end{pmatrix}, χ 11 ( ϑ , φ ) = − 2 1 cos ϑ cos φ − i sin φ cos ϑ sin φ + i cos φ − sin ϑ , χ 10 ( ϑ , φ ) = sin ϑ cos φ sin ϑ sin φ cos ϑ ,
χ 1 − 1 ( ϑ , φ ) = 1 2 ( cos ϑ cos φ + i sin φ cos ϑ sin φ − i cos φ − sin ϑ ) (1328) \chi_{1-1}(\vartheta,\varphi)=\frac{1}{\sqrt{2}}
\begin{pmatrix}
\cos\vartheta\cos\varphi+i\sin\varphi\\
\cos\vartheta\sin\varphi-i\cos\varphi\\
-\sin\vartheta
\end{pmatrix} \tag*{(1328)} χ 1 − 1 ( ϑ , φ ) = 2 1 cos ϑ cos φ + i sin φ cos ϑ sin φ − i cos φ − sin ϑ (1328)
χ x ( ϑ , φ ) = ( cos ϑ cos φ cos ϑ sin φ − sin ϑ ) , χ y ( ϑ , φ ) = ( − sin φ cos φ 0 ) , χ z ( ϑ , φ ) = ( sin ϑ cos φ sin ϑ sin φ cos ϑ ) . (1329) \chi_x(\vartheta,\varphi)=
\begin{pmatrix}
\cos\vartheta\cos\varphi\\
\cos\vartheta\sin\varphi\\
-\sin\vartheta
\end{pmatrix},\qquad
\chi_y(\vartheta,\varphi)=
\begin{pmatrix}
-\sin\varphi\\
\cos\varphi\\
0
\end{pmatrix},\qquad
\chi_z(\vartheta,\varphi)=
\begin{pmatrix}
\sin\vartheta\cos\varphi\\
\sin\vartheta\sin\varphi\\
\cos\vartheta
\end{pmatrix}. \tag*{(1329)} χ x ( ϑ , φ ) = cos ϑ cos φ cos ϑ sin φ − sin ϑ , χ y ( ϑ , φ ) = − sin φ cos φ 0 , χ z ( ϑ , φ ) = sin ϑ cos φ sin ϑ sin φ cos ϑ . (1329)
In the cartesian basis representation these functions satisfy the relations
χ 1 λ ∗ ( ϑ , φ ) = ( − 1 ) λ χ 1 − λ ( ϑ , φ ) , ( λ = ± 1 , 0 ) , (1330) \chi_{1\lambda}^{*}(\vartheta,\varphi)=(-1)^{\lambda}\chi_{1-\lambda}(\vartheta,\varphi),\qquad(\lambda=\pm1,0), \tag*{(1330)} χ 1 λ ∗ ( ϑ , φ ) = ( − 1 ) λ χ 1 − λ ( ϑ , φ ) , ( λ = ± 1 , 0 ) , (1330)
χ i ∗ ( ϑ , φ ) = χ i ( ϑ , φ ) , ( i = x , y , z ) . \chi_i^{*}(\vartheta,\varphi)=\chi_i(\vartheta,\varphi),\qquad(i=x,y,z). χ i ∗ ( ϑ , φ ) = χ i ( ϑ , φ ) , ( i = x , y , z ) .
The expansions of products of the helicity basis functions in series of spin matrices and quadrupole operators (§ 11 ) are given by
χ 1 λ ( ϑ , φ ) χ 1 λ ′ † ( ϑ , φ ) = 1 3 δ λ λ ′ I ^ + 1 2 C 1 λ 1 λ ′ 1 ν ∑ μ = − 1 1 D μ ν 1 ( φ , ϑ , 0 ) S ^ μ + 5 3 C 1 λ 1 λ ′ 2 N ∑ M = − 2 2 D M N 2 ( φ , ϑ , 0 ) T ^ 2 M . (1331) \chi_{1\lambda}(\vartheta,\varphi)\chi_{1\lambda'}^{\dagger}(\vartheta,\varphi)
=\frac{1}{3}\delta_{\lambda\lambda'}\hat{I}
+\frac{1}{\sqrt{2}}C_{1\lambda\,1\lambda'}^{1\nu}
\sum_{\mu=-1}^{1}D_{\mu\nu}^{1}(\varphi,\vartheta,0)\hat{S}_{\mu}
+\sqrt{\frac{5}{3}}C_{1\lambda\,1\lambda'}^{2N}
\sum_{M=-2}^{2}D_{MN}^{2}(\varphi,\vartheta,0)\hat{T}_{2M}. \tag*{(1331)} χ 1 λ ( ϑ , φ ) χ 1 λ ′ † ( ϑ , φ ) = 3 1 δ λ λ ′ I ^ + 2 1 C 1 λ 1 λ ′ 1 ν μ = − 1 ∑ 1 D μν 1 ( φ , ϑ , 0 ) S ^ μ + 3 5 C 1 λ 1 λ ′ 2 N M = − 2 ∑ 2 D M N 2 ( φ , ϑ , 0 ) T ^ 2 M . (1331)
Analogous expansions for χ i ( ϑ , φ ) \chi_i(\vartheta,\varphi) χ i ( ϑ , φ ) can be written as
χ i ( ϑ , φ ) χ k † ( ϑ , φ ) = 1 3 δ i k I ^ + i 2 ϵ i k l ∑ m = x , y , z a m l ( φ , ϑ , 0 ) S ^ m − ∑ m , n = x , y , z a m i ( φ , ϑ , 0 ) a n k ( φ , ϑ , 0 ) Q ^ m n , (1332) \chi_i(\vartheta,\varphi)\chi_k^{\dagger}(\vartheta,\varphi)
=\frac{1}{3}\delta_{ik}\hat{I}
+\frac{i}{2}\epsilon_{ikl}\sum_{m=x,y,z}a_{ml}(\varphi,\vartheta,0)\hat{S}_m
-\sum_{m,n=x,y,z}a_{mi}(\varphi,\vartheta,0)a_{nk}(\varphi,\vartheta,0)\hat{Q}_{mn}, \tag*{(1332)} χ i ( ϑ , φ ) χ k † ( ϑ , φ ) = 3 1 δ ik I ^ + 2 i ϵ ik l m = x , y , z ∑ a m l ( φ , ϑ , 0 ) S ^ m − m , n = x , y , z ∑ a mi ( φ , ϑ , 0 ) a nk ( φ , ϑ , 0 ) Q ^ mn , (1332)
where a m l ( α , β , γ ) a_{ml}(\alpha,\beta,\gamma) a m l ( α , β , γ ) are elements of the rotation matrix (Eq. (183) ).
Matrix elements of the spin operator between the helicity states χ 1 λ ( ϑ , φ ) \chi_{1\lambda}(\vartheta,\varphi) χ 1 λ ( ϑ , φ ) or between the states χ i ( ϑ , φ ) \chi_i(\vartheta,\varphi) χ i ( ϑ , φ ) are
χ 1 λ † ( ϑ , φ ) S ^ μ χ 1 λ ′ ( ϑ , φ ) = ( − 1 ) μ + ν 2 C 1 λ ′ 1 ν 1 λ D − μ − ν 1 ( φ , ϑ , 0 ) , (1333) \chi_{1\lambda}^{\dagger}(\vartheta,\varphi)\hat{S}_{\mu}\chi_{1\lambda'}(\vartheta,\varphi)
=(-1)^{\mu+\nu}\sqrt{2}C_{1\lambda'\,1\nu}^{1\lambda}D_{-\mu-\nu}^{1}(\varphi,\vartheta,0), \tag*{(1333)} χ 1 λ † ( ϑ , φ ) S ^ μ χ 1 λ ′ ( ϑ , φ ) = ( − 1 ) μ + ν 2 C 1 λ ′ 1 ν 1 λ D − μ − ν 1 ( φ , ϑ , 0 ) , (1333)
χ i † ( ϑ , φ ) S ^ l χ k ( ϑ , φ ) = − i ∑ m ϵ i k m a l m ( φ , ϑ , 0 ) . (1334) \chi_i^{\dagger}(\vartheta,\varphi)\hat{S}_l\chi_k(\vartheta,\varphi)
=-i\sum_m\epsilon_{ikm}a_{lm}(\varphi,\vartheta,0). \tag*{(1334)} χ i † ( ϑ , φ ) S ^ l χ k ( ϑ , φ ) = − i m ∑ ϵ ik m a l m ( φ , ϑ , 0 ) . (1334)
In particular, the expectation value of the spin operator S ^ \hat{S} S ^ in a state with the helicity λ \lambda λ is given by
χ 1 λ † ( ϑ , φ ) S ^ χ 1 λ ( ϑ , φ ) = λ n ( ϑ , φ ) , ( λ = ± 1 , 0 ) , (1335) \chi_{1\lambda}^{\dagger}(\vartheta,\varphi)\hat{S}\chi_{1\lambda}(\vartheta,\varphi)=\lambda\mathbf{n}(\vartheta,\varphi),\qquad(\lambda=\pm1,0), \tag*{(1335)} χ 1 λ † ( ϑ , φ ) S ^ χ 1 λ ( ϑ , φ ) = λ n ( ϑ , φ ) , ( λ = ± 1 , 0 ) , (1335)
The expectation value of the operator S ^ \hat{S} S ^ in the states χ i ( ϑ , φ ) \chi_i(\vartheta,\varphi) χ i ( ϑ , φ ) is zero.
χ i † ( ϑ , φ ) S ^ χ i ( ϑ , φ ) = 0 , ( i = x , y , z ) . (1336) \chi_i^{\dagger}(\vartheta,\varphi)\hat{S}\chi_i(\vartheta,\varphi)=0,\qquad(i=x,y,z). \tag*{(1336)} χ i † ( ϑ , φ ) S ^ χ i ( ϑ , φ ) = 0 , ( i = x , y , z ) . (1336)
The evaluation of matrix elements of the quadrupole operators T ^ 2 M \hat{T}_{2M} T ^ 2 M (M = ± 2 , ± 1 , 0 M=\pm2,\pm1,0 M = ± 2 , ± 1 , 0 ) and Q ^ i k \hat{Q}_{ik} Q ^ ik (i , k = x , y , z i,k=x,y,z i , k = x , y , z ) gives
χ 1 λ † ( ϑ , φ ) T ^ 2 M χ 1 λ ′ ( ϑ , φ ) = ( − 1 ) M + N 5 3 C 1 λ ′ 2 N 1 λ D − M − N 2 ( φ , ϑ , 0 ) , (1337) \chi_{1\lambda}^{\dagger}(\vartheta,\varphi)\hat{T}_{2M}\chi_{1\lambda'}(\vartheta,\varphi)
=(-1)^{M+N}\sqrt{\frac{5}{3}}C_{1\lambda'\,2N}^{1\lambda}D_{-M-N}^{2}(\varphi,\vartheta,0), \tag*{(1337)} χ 1 λ † ( ϑ , φ ) T ^ 2 M χ 1 λ ′ ( ϑ , φ ) = ( − 1 ) M + N 3 5 C 1 λ ′ 2 N 1 λ D − M − N 2 ( φ , ϑ , 0 ) , (1337)
χ i † ( ϑ , φ ) Q ^ l m χ k ( ϑ , φ ) = − 1 2 { a l i ( φ , ϑ , 0 ) a m k ( φ , ϑ , 0 ) + a l k ( φ , ϑ , 0 ) a m i ( φ , ϑ , 0 ) − 2 3 δ i k δ l m } . (1338) \chi_i^{\dagger}(\vartheta,\varphi)\hat{Q}_{lm}\chi_k(\vartheta,\varphi)
=-\frac{1}{2}\left\{a_{li}(\varphi,\vartheta,0)a_{mk}(\varphi,\vartheta,0)+a_{lk}(\varphi,\vartheta,0)a_{mi}(\varphi,\vartheta,0)-\frac{2}{3}\delta_{ik}\delta_{lm}\right\}. \tag*{(1338)} χ i † ( ϑ , φ ) Q ^ l m χ k ( ϑ , φ ) = − 2 1 { a l i ( φ , ϑ , 0 ) a mk ( φ , ϑ , 0 ) + a l k ( φ , ϑ , 0 ) a mi ( φ , ϑ , 0 ) − 3 2 δ ik δ l m } . (1338)
In particular, the expectation values of the quadrupole operators T ^ 2 M \hat{T}_{2M} T ^ 2 M in states with helicity λ \lambda λ are equal to
χ 1 λ † ( ϑ , φ ) T ^ 2 M χ 1 λ ( ϑ , φ ) = ( − 1 ) 1 + λ 1 + ∣ λ ∣ 8 π 15 Y 2 M ( ϑ , φ ) , \chi_{1\lambda}^{\dagger}(\vartheta,\varphi)\hat{T}_{2M}\chi_{1\lambda}(\vartheta,\varphi)
=\frac{(-1)^{1+\lambda}}{1+|\lambda|}\sqrt{\frac{8\pi}{15}}Y_{2M}(\vartheta,\varphi), χ 1 λ † ( ϑ , φ ) T ^ 2 M χ 1 λ ( ϑ , φ ) = 1 + ∣ λ ∣ ( − 1 ) 1 + λ 15 8 π Y 2 M ( ϑ , φ ) ,
( λ = ± 1 , 0 ; M = ± 2 , ± 1 , 0 ) . (1339) (\lambda=\pm1,0;\ M=\pm2,\pm1,0). \tag*{(1339)} ( λ = ± 1 , 0 ; M = ± 2 , ± 1 , 0 ) . (1339)
General Spin Functions for S = 1 S = 1 S = 1
An arbitrary spin function χ 1 \chi_1 χ 1 of a particle of spin 1 may be expanded in terms of basis spin functions, i.e., written in the following form:
χ 1 = ∑ m = ± 1 , 0 a m χ 1 m = ∑ m = ± 1 , 0 ( − 1 ) m a − m χ 1 m = ∑ i = x , y , z a i χ i . (1340) \chi_1 = \sum_{m=\pm1,0} a^m \chi_{1m} = \sum_{m=\pm1,0} (-1)^m a_{-m}\chi_{1m} = \sum_{i=x,y,z} a_i\chi_i. \tag*{(1340)} χ 1 = m = ± 1 , 0 ∑ a m χ 1 m = m = ± 1 , 0 ∑ ( − 1 ) m a − m χ 1 m = i = x , y , z ∑ a i χ i . (1340)
The expansion of the Hermitian conjugate function χ 1 † \chi_1^\dagger χ 1 † is given by
χ 1 † = ∑ m = ± 1 , 0 ( a m ) ∗ χ 1 m † = ∑ m = ± 1 , 0 ( − 1 ) m ( a − m ) ∗ χ 1 m † = ∑ i = x , y , z a i ∗ χ i † . (1341) \chi_1^\dagger= \sum_{m=\pm1,0} (a^m)^* \chi_{1m}^\dagger= \sum_{m=\pm1,0} (-1)^m (a_{-m})^* \chi_{1m}^\dagger= \sum_{i=x,y,z} a_i^* \chi_i^\dagger. \tag*{(1341)} χ 1 † = m = ± 1 , 0 ∑ ( a m ) ∗ χ 1 m † = m = ± 1 , 0 ∑ ( − 1 ) m ( a − m ) ∗ χ 1 m † = i = x , y , z ∑ a i ∗ χ i † . (1341)
In a spherical basis the function χ 1 \chi_1 χ 1 has the form
χ 1 = ( a + 1 a 0 a − 1 ) = ( − a − 1 a 0 − a + 1 ) = ( − 1 2 ( a x − i a y ) a z 1 2 ( a x + i a y ) ) , (1342) \chi_1 =
\begin{pmatrix}
a^{+1}\\
a^0\\
a^{-1}
\end{pmatrix}
=
\begin{pmatrix}
-a_{-1}\\
a_0\\
-a_{+1}
\end{pmatrix}
=
\begin{pmatrix}
-\frac{1}{\sqrt{2}}(a_x-i a_y)\\
a_z\\
\frac{1}{\sqrt{2}}(a_x+i a_y)
\end{pmatrix}, \tag*{(1342)} χ 1 = a + 1 a 0 a − 1 = − a − 1 a 0 − a + 1 = − 2 1 ( a x − i a y ) a z 2 1 ( a x + i a y ) , (1342)
and in a cartesian basis this function reads
χ 1 = ( 1 2 ( a − 1 − a + 1 ) − i 2 ( a − 1 + a + 1 ) a 0 ) = ( 1 2 ( a − 1 − a + 1 ) i 2 ( a − 1 + a + 1 ) a 0 ) = ( a x a y a z ) . (1343) \chi_1 =
\begin{pmatrix}
\frac{1}{\sqrt{2}}(a^{-1}-a^{+1})\\
-\frac{i}{\sqrt{2}}(a^{-1}+a^{+1})\\
a^0
\end{pmatrix}
=
\begin{pmatrix}
\frac{1}{\sqrt{2}}(a_{-1}-a_{+1})\\
\frac{i}{\sqrt{2}}(a_{-1}+a_{+1})\\
a_0
\end{pmatrix}
=
\begin{pmatrix}
a_x\\
a_y\\
a_z
\end{pmatrix}. \tag*{(1343)} χ 1 = 2 1 ( a − 1 − a + 1 ) − 2 i ( a − 1 + a + 1 ) a 0 = 2 1 ( a − 1 − a + 1 ) 2 i ( a − 1 + a + 1 ) a 0 = a x a y a z . (1343)
The expansion coefficients a m a^m a m are contravariant spherical components of some generally complex vector a \mathbf{a} a , a m a_m a m are covariant spherical components of a \mathbf{a} a and a i a_i a i are cartesian components. Sometimes a \mathbf{a} a is called the polarization vector of particles of spin 1.
Under complex conjugation components of this vector transform as
( a m ) ∗ = ( a m ∗ ) , ( a m ) ∗ = ( a ∗ ) m , ( a i ) ∗ = ( a ∗ ) i . (1344) (a^m)^*=(a_m^*) ,\quad(a_m)^*=(a^*)^m,\quad(a_i)^*=(a^*)_i. \tag*{(1344)} ( a m ) ∗ = ( a m ∗ ) , ( a m ) ∗ = ( a ∗ ) m , ( a i ) ∗ = ( a ∗ ) i . (1344)
The normalization condition
χ 1 † χ 1 = 1 (1345) \chi_1^\dagger\chi_1=1 \tag*{(1345)} χ 1 † χ 1 = 1 (1345)
imposes a restriction to the components of a \mathbf{a} a
∣ a + 1 ∣ 2 + ∣ a 0 ∣ 2 + ∣ a − 1 ∣ 2 = ∣ a + 1 ∣ 2 + ∣ a 0 ∣ 2 + ∣ a − 1 ∣ 2 = ∣ a x ∣ 2 + ∣ a y ∣ 2 + ∣ a z ∣ 2 = 1 , (1346) |a^{+1}|^2+|a^0|^2+|a^{-1}|^2=|a_{+1}|^2+|a_0|^2+|a_{-1}|^2=|a_x|^2+|a_y|^2+|a_z|^2=1, \tag*{(1346)} ∣ a + 1 ∣ 2 + ∣ a 0 ∣ 2 + ∣ a − 1 ∣ 2 = ∣ a + 1 ∣ 2 + ∣ a 0 ∣ 2 + ∣ a − 1 ∣ 2 = ∣ a x ∣ 2 + ∣ a y ∣ 2 + ∣ a z ∣ 2 = 1 , (1346)
or, in compact form
∣ a ∣ 2 = a ∗ ⋅ a = 1. (1347) |\mathbf{a}|^2=\mathbf{a}^*\cdot\mathbf{a}=1. \tag*{(1347)} ∣ a ∣ 2 = a ∗ ⋅ a = 1. (1347)
The spin function product χ 1 χ 1 † \chi_1\chi_1^\dagger χ 1 χ 1 † may be expanded in series of the spin matrices S ^ \hat{S} S ^ and quadrupole momentum operators, T ^ 2 M \hat{T}_{2M} T ^ 2 M or Q ^ i k \hat{Q}_{ik} Q ^ ik (see § 11 ),
χ 1 χ 1 † = 1 3 I ^ + 1 2 ∑ μ = − 1 1 ( − 1 ) μ P − μ S ^ μ + ∑ M = − 2 2 ( − 1 ) M P 2 − M T ^ 2 M , \chi_1\chi_1^\dagger=\frac{1}{3}\hat{I}+\frac{1}{2}\sum_{\mu=-1}^{1}(-1)^\mu P_{-\mu}\hat{S}_\mu+\sum_{M=-2}^{2}(-1)^M P_{2-M}\hat{T}_{2M}, χ 1 χ 1 † = 3 1 I ^ + 2 1 μ = − 1 ∑ 1 ( − 1 ) μ P − μ S ^ μ + M = − 2 ∑ 2 ( − 1 ) M P 2 − M T ^ 2 M ,
χ 1 χ 1 † = 1 3 I ^ + 1 2 ∑ i = x , y , z P i S ^ i + ∑ i , k = x , y , z P i k Q ^ i k , (1348) \chi_1\chi_1^\dagger=\frac{1}{3}\hat{I}+\frac{1}{2}\sum_{i=x,y,z}P_i\hat{S}_i+\sum_{i,k=x,y,z}P_{ik}\hat{Q}_{ik}, \tag*{(1348)} χ 1 χ 1 † = 3 1 I ^ + 2 1 i = x , y , z ∑ P i S ^ i + i , k = x , y , z ∑ P ik Q ^ ik , (1348)
or in more compact form,
χ 1 χ 1 † = 1 3 I ^ + 1 2 P ⋅ S ^ + ( P 2 ⋅ T ^ 2 ) , (1349) \chi_1\chi_1^\dagger=\frac{1}{3}\hat{I}+\frac{1}{2}\mathbf{P}\cdot\hat{\mathbf{S}}+(\mathbf{P}_2\cdot\hat{\mathbf{T}}_2), \tag*{(1349)} χ 1 χ 1 † = 3 1 I ^ + 2 1 P ⋅ S ^ + ( P 2 ⋅ T ^ 2 ) , (1349)
where ( P 2 ⋅ T ^ 2 ) (\mathbf{P}_2\cdot\hat{\mathbf{T}}_2) ( P 2 ⋅ T ^ 2 ) is scalar product to two irreducible tensors of rank 2 (see Eq. (465) ).
The real vector P \mathbf{P} P can be expressed in terms of the polarization vector a \mathbf{a} a as
P = i [ a × a ∗ ] , (1350) \mathbf{P}=i[\mathbf{a}\times\mathbf{a}^*], \tag*{(1350)} P = i [ a × a ∗ ] , (1350)
or, in components,
P μ = 2 ∑ m , n C 1 m 1 μ 1 n ( a n ) ∗ a m , ( m , n , μ = ± 1 , 0 ) , P_\mu=\sqrt{2}\sum_{m,n}C_{1m\,1\mu}^{1n}(a^n)^*a^m,\quad(m,n,\mu=\pm1,0), P μ = 2 m , n ∑ C 1 m 1 μ 1 n ( a n ) ∗ a m , ( m , n , μ = ± 1 , 0 ) ,
P i = i ∑ k , l ε i k l a k a l ∗ , ( i , k , l = x , y , z ) . (1351) P_i=i\sum_{k,l}\varepsilon_{ikl}a_k a_l^*,\quad(i,k,l=x,y,z). \tag*{(1351)} P i = i k , l ∑ ε ik l a k a l ∗ , ( i , k , l = x , y , z ) . (1351)
Spin functions for S = 1 S = 1 S = 1
Spherical and cartesian components of a real irreducible tensor P 2 \mathbf{P}_2 P 2 of rank 2 may be expressed in terms of components of a \mathbf{a} a by
P 2 M = 5 3 ∑ m , n C 1 m 2 M 1 n ( a n ) ∗ a m , ( m , n = ± 1 , 0 ; M = ± 2 , ± 1 , 0 ) , P_{2M} = \sqrt{\frac{5}{3}}\sum_{m,n} C^{1n}_{1m\,2M}(a^n)^*a^m,\qquad(m,n=\pm1,0; M=\pm2,\pm1,0), P 2 M = 3 5 m , n ∑ C 1 m 2 M 1 n ( a n ) ∗ a m , ( m , n = ± 1 , 0 ; M = ± 2 , ± 1 , 0 ) ,
P i k = − 1 2 { a i a k ∗ + a k a i ∗ − 2 3 δ i k } , ( i , k = x , y , z ) , (1352) P_{ik}=-\frac{1}{2}\left\{a_i a_k^*+a_k a_i^*-\frac{2}{3}\delta_{ik}\right\},\qquad(i,k=x,y,z), \tag*{(1352)} P ik = − 2 1 { a i a k ∗ + a k a i ∗ − 3 2 δ ik } , ( i , k = x , y , z ) , (1352)
P i k = P k i , ∑ i P i i = 0. P_{ik}=P_{ki},\qquad\sum_i P_{ii}=0. P ik = P k i , i ∑ P ii = 0.
The vector P \mathbf{P} P and tensor P 2 \mathbf{P}_2 P 2 give the following expectation values of the spin operator and the quadrupole momentum operator, respectively:
⟨ S ⟩ = χ 1 † S ^ χ 1 = P , \langle\mathbf{S}\rangle=\chi_1^\dagger\hat{\mathbf{S}}\chi_1=\mathbf{P}, ⟨ S ⟩ = χ 1 † S ^ χ 1 = P ,
⟨ T ^ 2 M ⟩ = χ 1 † T ^ 2 M χ 1 = P 2 M , \langle\hat{T}_{2M}\rangle=\chi_1^\dagger\hat{T}_{2M}\chi_1=P_{2M}, ⟨ T ^ 2 M ⟩ = χ 1 † T ^ 2 M χ 1 = P 2 M ,
⟨ Q ^ i k ⟩ = χ 1 † Q ^ i k χ 1 = P i k . (1353) \langle\hat{Q}_{ik}\rangle=\chi_1^\dagger\hat{Q}_{ik}\chi_1=P_{ik}. \tag*{(1353)} ⟨ Q ^ ik ⟩ = χ 1 † Q ^ ik χ 1 = P ik . (1353)
The transformation of the spin function χ 1 \chi_1 χ 1 under rotations of coordinate systems is performed by the rotation operators D ^ 1 ( α , β , γ ) \hat{D}^{1}(\alpha,\beta,\gamma) D ^ 1 ( α , β , γ ) (Eqs. (419) –(420) ) or U ^ 1 ( ω ; Θ , Φ ) \hat{U}^{1}(\omega;\Theta,\Phi) U ^ 1 ( ω ; Θ , Φ ) (Eqs. (423) –(424) ) depending on the choice of the parameters to describe rotations
χ 1 ′ = D ^ 1 ( α , β , γ ) χ 1 = U ^ 1 ( ω ; Θ , Φ ) χ 1 . (1354) \chi'_1=\hat{D}^{1}(\alpha,\beta,\gamma)\chi_1=\hat{U}^{1}(\omega;\Theta,\Phi)\chi_1. \tag*{(1354)} χ 1 ′ = D ^ 1 ( α , β , γ ) χ 1 = U ^ 1 ( ω ; Θ , Φ ) χ 1 . (1354)
Spin functions in a rotated coordinate system may be expanded into sums of basis spin functions χ 1 m , χ i \chi_{1m},\chi_i χ 1 m , χ i in the original coordinate system as well as into sums of functions χ 1 m ′ , χ i ′ \chi'_{1m},\chi'_i χ 1 m ′ , χ i ′ in the rotated coordinate system:
χ 1 ′ = ∑ m = ± 1 , 0 a ′ m χ 1 m = ∑ m = ± 1 , 0 a m χ 1 m ′ , \chi'_1=\sum_{m=\pm1,0}a'^{m}\chi_{1m}=\sum_{m=\pm1,0}a^m\chi'_{1m}, χ 1 ′ = m = ± 1 , 0 ∑ a ′ m χ 1 m = m = ± 1 , 0 ∑ a m χ 1 m ′ ,
χ 1 ′ = ∑ m = ± 1 , 0 ( − 1 ) m a − m ′ χ 1 m = ∑ m = ± 1 , 0 ( − 1 ) m a − m χ 1 m ′ , \chi'_1=\sum_{m=\pm1,0}(-1)^m a'_{-m}\chi_{1m}=\sum_{m=\pm1,0}(-1)^m a_{-m}\chi'_{1m}, χ 1 ′ = m = ± 1 , 0 ∑ ( − 1 ) m a − m ′ χ 1 m = m = ± 1 , 0 ∑ ( − 1 ) m a − m χ 1 m ′ ,
χ 1 ′ = ∑ i = x , y , z a i ′ χ i = ∑ i = x , y , z a i χ i ′ . (1355) \chi'_1=\sum_{i=x,y,z}a'_i\chi_i=\sum_{i=x,y,z}a_i\chi'_i. \tag*{(1355)} χ 1 ′ = i = x , y , z ∑ a i ′ χ i = i = x , y , z ∑ a i χ i ′ . (1355)
The relations between basis functions in the rotated and original coordinate systems are given by Eqs. (1311) . The components of a \mathbf{a} a in the above systems are related by
a ′ m = ∑ n D m n 1 ( α , β , γ ) a n , a n = ∑ m D m n 1 ∗ ( α , β , γ ) a ′ m , a'^{m}=\sum_n D^1_{mn}(\alpha,\beta,\gamma)a^n,\qquad a^n=\sum_m D^{1*}_{mn}(\alpha,\beta,\gamma)a'^{m}, a ′ m = n ∑ D mn 1 ( α , β , γ ) a n , a n = m ∑ D mn 1 ∗ ( α , β , γ ) a ′ m ,
a m ′ = ∑ n D n m 1 ( α , β , γ ) a n , a n = ∑ m D n m 1 ∗ ( α , β , γ ) a m ′ , a'_m=\sum_n D^1_{nm}(\alpha,\beta,\gamma)a_n,\qquad a_n=\sum_m D^{1*}_{nm}(\alpha,\beta,\gamma)a'_m, a m ′ = n ∑ D nm 1 ( α , β , γ ) a n , a n = m ∑ D nm 1 ∗ ( α , β , γ ) a m ′ ,
a i ′ = ∑ k a i k a k , a k = ∑ i a i k a i ′ , a'_i=\sum_k a_{ik}a_k,\qquad a_k=\sum_i a_{ik}a'_i, a i ′ = k ∑ a ik a k , a k = i ∑ a ik a i ′ ,
( m , n = ± 1 , 0 ; i , k = x , y , z ) (1356) (m,n=\pm1,0;i,k=x,y,z) \tag*{(1356)} ( m , n = ± 1 , 0 ; i , k = x , y , z ) (1356)
Tensor spherical harmonics
General properties of tensor spherical harmonics
Definition
The tensor spherical harmonics Y J M L S ( ϑ , φ ) Y_{JM}^{LS}(\vartheta,\varphi) Y J M L S ( ϑ , φ ) , by definition, are eigenfunctions of the operators J ^ 2 \hat{J}^{2} J ^ 2 , J ^ z \hat{J}_{z} J ^ z , L ^ 2 \hat{L}^{2} L ^ 2 and S ^ 2 \hat{S}^{2} S ^ 2 where L ^ \hat{L} L ^ is the operator of orbital angular momentum (§ 10.2 ), S ^ \hat{S} S ^ is the spin operator (§ 10.3 ), and J ^ = L ^ + S ^ \hat{J}=\hat{L}+\hat{S} J ^ = L ^ + S ^ is the operator of total angular momentum (§ 10.1 ):
J ^ 2 Y J M L S ( ϑ , φ ) = J ( J + 1 ) Y J M L S ( ϑ , φ ) , J ^ z Y J M L S ( ϑ , φ ) = M Y J M L S ( ϑ , φ ) , L ^ 2 Y J M L S ( ϑ , φ ) = L ( L + 1 ) Y J M L S ( ϑ , φ ) , S ^ 2 Y J M L S ( ϑ , φ ) = S ( S + 1 ) Y J M L S ( ϑ , φ ) . (1357) \hat{J}^{2}Y_{JM}^{LS}(\vartheta,\varphi)=J(J+1)Y_{JM}^{LS}(\vartheta,\varphi),
\\
\hat{J}_{z}Y_{JM}^{LS}(\vartheta,\varphi)=MY_{JM}^{LS}(\vartheta,\varphi),
\\
\hat{L}^{2}Y_{JM}^{LS}(\vartheta,\varphi)=L(L+1)Y_{JM}^{LS}(\vartheta,\varphi),
\\
\hat{S}^{2}Y_{JM}^{LS}(\vartheta,\varphi)=S(S+1)Y_{JM}^{LS}(\vartheta,\varphi). \tag*{(1357)} J ^ 2 Y J M L S ( ϑ , φ ) = J ( J + 1 ) Y J M L S ( ϑ , φ ) , J ^ z Y J M L S ( ϑ , φ ) = M Y J M L S ( ϑ , φ ) , L ^ 2 Y J M L S ( ϑ , φ ) = L ( L + 1 ) Y J M L S ( ϑ , φ ) , S ^ 2 Y J M L S ( ϑ , φ ) = S ( S + 1 ) Y J M L S ( ϑ , φ ) . (1357)
A tensor spherical harmonic describes the angular distribution and polarization of spin-S S S particles in a state with definite total angular momentum J J J , projection M M M , and orbital angular momentum L L L . The spin value S S S is sometimes called the rank of the tensor spherical harmonic. Accordingly, one often uses such names as spinor spherical harmonics (S = 1 2 S=\frac{1}{2} S = 2 1 ), vector spherical harmonics (S = 1 S=1 S = 1 ), etc. However, strictly speaking, these terms are not entirely adequate because, in fact, the transformation properties of the tensor spherical harmonics under rotation of coordinate system are determined by J J J , but not by S S S (see § 27.2.2 ).
The tensor spherical harmonics may be constructed from the spherical harmonics Y L M ( ϑ , φ ) Y_{LM}(\vartheta,\varphi) Y L M ( ϑ , φ ) (eigenfunctions of L ^ 2 \hat{L}^{2} L ^ 2 and L ^ z \hat{L}_{z} L ^ z ) and the spin functions χ S σ \chi_{S\sigma} χ S σ (eigenfunctions of S ^ 2 \hat{S}^{2} S ^ 2 and S ^ z \hat{S}_{z} S ^ z ) in accordance with the coupling scheme of two angular momenta
Y J M L S ( ϑ , φ ) = ∑ m , σ C L m S σ J M Y L m ( ϑ , φ ) χ S σ . (1358) Y_{JM}^{LS}(\vartheta,\varphi)=\sum_{m,\sigma}C_{LmS\sigma}^{JM}Y_{Lm}(\vartheta,\varphi)\chi_{S\sigma}. \tag*{(1358)} Y J M L S ( ϑ , φ ) = m , σ ∑ C L m S σ J M Y L m ( ϑ , φ ) χ S σ . (1358)
Thus, the tensor spherical harmonics are irreducible tensor products (of rank J J J ) of scalar spherical harmonics and spin functions:
Y J M L S ( ϑ , φ ) = { Y L ⊗ χ S } J M . (1359) Y_{JM}^{LS}(\vartheta,\varphi)=\{Y_{L}\otimes\chi_{S}\}_{JM}. \tag*{(1359)} Y J M L S ( ϑ , φ ) = { Y L ⊗ χ S } J M . (1359)
The indices J J J and S S S are integer or half-integer nonnegative numbers, and L L L is always an integer nonnegative number. For given J J J and S S S , the momentum L L L takes the values L = ∣ J − S ∣ , ∣ J − S ∣ + 1 , … , J + S L=|J-S|,|J-S|+1,\ldots,J+S L = ∣ J − S ∣ , ∣ J − S ∣ + 1 , … , J + S . The possible values of M M M are M = − J , − J + 1 , … , J − 1 , J M=-J,-J+1,\ldots,J-1,J M = − J , − J + 1 , … , J − 1 , J .
The tensor spherical harmonics with fixed indices are functions of three variables: two polar angles ϑ , φ \vartheta,\varphi ϑ , φ (0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π ) and spin variable ξ \xi ξ (ξ = − S , − S + 1 , … , S − 1 , S \xi=-S,-S+1,\ldots,S-1,S ξ = − S , − S + 1 , … , S − 1 , S ) which is the argument of spin function χ S σ \chi_{S\sigma} χ S σ . The spin variable ξ \xi ξ is usually not mentioned as an argument of tensor spherical harmonics. Instead, the tensor spherical harmonics are written in matrix form by analogy with spin functions. More precisely, Y J M L S ( ϑ , φ ) Y_{JM}^{LS}(\vartheta,\varphi) Y J M L S ( ϑ , φ ) is represented by a column matrix with 2 S + 1 2S+1 2 S + 1 elements, and Y J M L S + ( ϑ , φ ) Y_{JM}^{LS+}(\vartheta,\varphi) Y J M L S + ( ϑ , φ ) is represented by a row matrix. Summation over spin variable is replaced by matrix multiplication.
The tensor spherical harmonics Y J M L S ( ϑ , φ ) Y_{JM}^{LS}(\vartheta,\varphi) Y J M L S ( ϑ , φ ) constitute a complete orthonormal set for series expansion of rank S S S tensor functions within the domain of arguments 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π (see § 27.2.6 ).
Note also the following orthogonality relations for Y J M L S ( ϑ , φ ) Y_{JM}^{LS}(\vartheta,\varphi) Y J M L S ( ϑ , φ ) which have the same J J J , M M M and S S S but different L L L :
∑ M Y J M L ′ S + ( ϑ , φ ) ⋅ Y J M L S ( ϑ , φ ) = 0 , if L ′ ≠ L . (1360) \sum_{M}Y_{JM}^{L'S+}(\vartheta,\varphi)\cdot Y_{JM}^{LS}(\vartheta,\varphi)=0,\qquad\text{if }L'\ne L. \tag*{(1360)} M ∑ Y J M L ′ S + ( ϑ , φ ) ⋅ Y J M L S ( ϑ , φ ) = 0 , if L ′ = L . (1360)
Components of Tensor Spherical Harmonics
(a) Spherical contravariant components of tensor spherical harmonics may be written, in accordance with (7.1.2), as
[ Y J M L S ( ϑ , φ ) ] μ = C L M − μ S μ J M Y L M − μ ( ϑ , φ ) . (1361) \left[Y^{LS}_{JM}(\vartheta,\varphi)\right]^\mu=C^{JM}_{LM-\mu\,S\mu}Y_{LM-\mu}(\vartheta,\varphi). \tag*{(1361)} [ Y J M L S ( ϑ , φ ) ] μ = C L M − μ S μ J M Y L M − μ ( ϑ , φ ) . (1361)
( μ = − S , … , S − 1 , S ) (\mu=-S,\ldots,S-1,S) ( μ = − S , … , S − 1 , S ) .
Covariant spherical components may be defined by the relation
[ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) S − μ [ Y J M L S ( ϑ , φ ) ] − μ , (1362) \left[Y^{LS}_{JM}(\vartheta,\varphi)\right]_\mu=(-1)^{S-\mu}\left[Y^{LS}_{JM}(\vartheta,\varphi)\right]^{-\mu}, \tag*{(1362)} [ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) S − μ [ Y J M L S ( ϑ , φ ) ] − μ , (1362)
from which one obtains
[ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) S − μ C L M + μ S − μ J M Y L M + μ ( ϑ , φ ) . (1363) \left[Y^{LS}_{JM}(\vartheta,\varphi)\right]_\mu=(-1)^{S-\mu}C^{JM}_{LM+\mu\,S-\mu}Y_{LM+\mu}(\vartheta,\varphi). \tag*{(1363)} [ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) S − μ C L M + μ S − μ J M Y L M + μ ( ϑ , φ ) . (1363)
If S S S is integer, an alternative definition of the covariant spherical components that is widely used is (see § 27.4.2 )
[ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) μ [ Y J M L S ( ϑ , φ ) ] − μ , (1364) \left[Y^{LS}_{JM}(\vartheta,\varphi)\right]_\mu=(-1)^\mu\left[Y^{LS}_{JM}(\vartheta,\varphi)\right]^{-\mu}, \tag*{(1364)} [ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) μ [ Y J M L S ( ϑ , φ ) ] − μ , (1364)
so that
[ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) μ C L M + μ S − μ J M Y L M + μ ( ϑ , φ ) . (1365) \left[Y^{LS}_{JM}(\vartheta,\varphi)\right]_\mu=(-1)^\mu C^{JM}_{LM+\mu\,S-\mu}Y_{LM+\mu}(\vartheta,\varphi). \tag*{(1365)} [ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) μ C L M + μ S − μ J M Y L M + μ ( ϑ , φ ) . (1365)
Note that the equations of this chapter are independent of definition of covariant components, unless the contrary is indicated.
(b) The expansion of tensor spherical harmonics in terms of helicity spin functions may be written as
Y J M L S ( ϑ , φ ) = ∑ λ = − S S [ Y J M L S ( ϑ , φ ) ] λ χ S λ ( ϑ , φ ) , (1366) Y^{LS}_{JM}(\vartheta,\varphi)=\sum_{\lambda=-S}^{S}\left[Y^{LS}_{JM}(\vartheta,\varphi)\right]^\lambda\chi_{S\lambda}(\vartheta,\varphi), \tag*{(1366)} Y J M L S ( ϑ , φ ) = λ = − S ∑ S [ Y J M L S ( ϑ , φ ) ] λ χ S λ ( ϑ , φ ) , (1366)
where contravariant helicity components are given by
[ Y J M L S ( ϑ , φ ) ] λ = 2 L + 1 4 π C L 0 S λ J λ D − λ − M J ( 0 , ϑ , φ ) . (1367) \left[Y^{LS}_{JM}(\vartheta,\varphi)\right]^\lambda=\sqrt{\frac{2L+1}{4\pi}}C^{J\lambda}_{L0\,S\lambda}D^J_{-\lambda-M}(0,\vartheta,\varphi). \tag*{(1367)} [ Y J M L S ( ϑ , φ ) ] λ = 4 π 2 L + 1 C L 0 S λ J λ D − λ − M J ( 0 , ϑ , φ ) . (1367)
If covariant helicity components are defined by
\left[Y^{LS}_{JM}(\vartheta,\varphi)\right]'_\lambda=(-1)^{S-\lambda}\left[Y^{LS}_{JM}(\vartheta,\varphi)\right'^{-\lambda}, \tag*{(1368)}
then
[ Y J M L S ( ϑ , φ ) ] λ ′ = ( − 1 ) J + L − λ 2 L + 1 4 π C L 0 S λ J λ D λ − M J ( 0 , ϑ , φ ) . (1369) \left[Y^{LS}_{JM}(\vartheta,\varphi)\right]'_\lambda=(-1)^{J+L-\lambda}\sqrt{\frac{2L+1}{4\pi}}C^{J\lambda}_{L0\,S\lambda}D^J_{\lambda-M}(0,\vartheta,\varphi). \tag*{(1369)} [ Y J M L S ( ϑ , φ ) ] λ ′ = ( − 1 ) J + L − λ 4 π 2 L + 1 C L 0 S λ J λ D λ − M J ( 0 , ϑ , φ ) . (1369)
An alternative definition of covariant components (for integer S S S )
\left[Y^{LS}_{JM}(\vartheta,\varphi)\right]'_\lambda=(-1)^\lambda\left[Y^{LS}_{JM}(\vartheta,\varphi)\right'^{-\lambda}, \tag*{(1370)}
yields
[ Y J M L S ( ϑ , φ ) ] λ ′ = ( − 1 ) J + L − S + λ 2 L + 1 4 π C L 0 S λ J λ D λ − M J ( 0 , ϑ , φ ) . (1371) \left[Y^{LS}_{JM}(\vartheta,\varphi)\right]'_\lambda=(-1)^{J+L-S+\lambda}\sqrt{\frac{2L+1}{4\pi}}C^{J\lambda}_{L0\,S\lambda}D^J_{\lambda-M}(0,\vartheta,\varphi). \tag*{(1371)} [ Y J M L S ( ϑ , φ ) ] λ ′ = ( − 1 ) J + L − S + λ 4 π 2 L + 1 C L 0 S λ J λ D λ − M J ( 0 , ϑ , φ ) . (1371)
Complex Conjugation
Spherical components of a tensor spherical harmonic transform under complex conjugation according to
[ Y J M L S ( ϑ , φ ) ] μ ∗ = ( − 1 ) J + L + S − M − μ [ Y J − M L S ( ϑ , φ ) ] − μ , [ Y J M L S ( ϑ , φ ) ] μ ∗ = ( − 1 ) J + L + S − M − μ [ Y J − M L S ( ϑ , φ ) ] − μ . (1372) \begin{aligned}
\left[Y^{LS}_{JM}(\vartheta,\varphi)\right]^{\mu*}&=(-1)^{J+L+S-M-\mu}\left[Y^{LS}_{J-M}(\vartheta,\varphi)\right]^{-\mu},\\
\left[Y^{LS}_{JM}(\vartheta,\varphi)\right]_\mu^*&=(-1)^{J+L+S-M-\mu}\left[Y^{LS}_{J-M}(\vartheta,\varphi)\right]_{-\mu}. \tag*{(1372)}
\end{aligned} [ Y J M L S ( ϑ , φ ) ] μ ∗ [ Y J M L S ( ϑ , φ ) ] μ ∗ = ( − 1 ) J + L + S − M − μ [ Y J − M L S ( ϑ , φ ) ] − μ , = ( − 1 ) J + L + S − M − μ [ Y J − M L S ( ϑ , φ ) ] − μ . (1372)
These transformation rules are valid also for the corresponding helicity components.
Transformations of Coordinate Systems
(a) Coordinate inversion
The operator for coordinate inversion P ^ r \hat{P}_{r} P ^ r , when applied to a tensor spherical harmonic, gives
P ^ r Y J M L S ( ϑ , φ ) = η p Y J M L S ( π − ϑ , π + φ ) = η p ( − 1 ) L Y J M L S ( ϑ , φ ) , (1373) \hat{P}_{r}Y^{LS}_{JM}(\vartheta,\varphi)=\eta_{p}Y^{LS}_{JM}(\pi-\vartheta,\pi+\varphi)=\eta_{p}(-1)^{L}Y^{LS}_{JM}(\vartheta,\varphi), \tag*{(1373)} P ^ r Y J M L S ( ϑ , φ ) = η p Y J M L S ( π − ϑ , π + φ ) = η p ( − 1 ) L Y J M L S ( ϑ , φ ) , (1373)
where η p \eta_{p} η p is a phase factor which describes the intrinsic parity of the tensor.
(b) Rotations of coordinate systems
Under rotations of the coordinate system specified by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ the tensor spherical harmonics transform according to
Y J M ′ L S ( ϑ ′ , φ ′ ) ≡ D ^ ( α , β , γ ) Y J M ′ L S ( ϑ , φ ) = ∑ M = − J J D M M ′ J ( α , β , γ ) Y J M L S ( ϑ , φ ) , (1374) Y^{LS}_{JM'}(\vartheta',\varphi')\equiv\hat{D}(\alpha,\beta,\gamma)Y^{LS}_{JM'}(\vartheta,\varphi)=\sum_{M=-J}^{J}D^{J}_{MM'}(\alpha,\beta,\gamma)Y^{LS}_{JM}(\vartheta,\varphi), \tag*{(1374)} Y J M ′ L S ( ϑ ′ , φ ′ ) ≡ D ^ ( α , β , γ ) Y J M ′ L S ( ϑ , φ ) = M = − J ∑ J D M M ′ J ( α , β , γ ) Y J M L S ( ϑ , φ ) , (1374)
where M M M is the projection of total angular momentum on the initial z z z -axis, M ′ M' M ′ is its projection on the final z ′ z' z ′ axis; ϑ , φ \vartheta,\varphi ϑ , φ and ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ are polar angles of the unit vector n \mathbf{n} n in the initial and final coordinate systems, respectively. The relations between ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ and ϑ , φ \vartheta,\varphi ϑ , φ are given by Eqs. (131) , (132) . The coefficients D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) are matrix elements of the rotation operator, i.e., the Wigner D D D -functions (Chap. 4). Note that the transformation properties of tensor spherical harmonics are determined by the total angular momentum J \mathbf{J} J , but not by the spin S \mathbf{S} S . Thus, it is J J J that determines true rank of a tensor spherical harmonic.
Differential Equations
(a) Tensor spherical harmonics are eigenfunctions of the operator L ^ 2 \hat{\mathbf{L}}^{2} L ^ 2 ; hence, they satisfy the equation
{ Δ Ω + L ( L + 1 ) } Y J M L S ( ϑ , φ ) = 0 , (1375) \left\{\Delta_{\Omega}+L(L+1)\right\}Y^{LS}_{JM}(\vartheta,\varphi)=0, \tag*{(1375)} { Δ Ω + L ( L + 1 ) } Y J M L S ( ϑ , φ ) = 0 , (1375)
where Δ Ω \Delta_{\Omega} Δ Ω is the angular part of the Laplace operator (see § 9.8.2 ). The expanded form of (7.1.19) is
1 sin 2 ϑ ∂ ∂ ϑ { sin ϑ ∂ ∂ ϑ Y J M L S ( ϑ , φ ) } + 1 sin 2 ϑ ∂ 2 ∂ φ 2 Y J M L S ( ϑ , φ ) + L ( L + 1 ) Y J M L S ( ϑ , φ ) = 0. (1376) \frac{1}{\sin^{2}\vartheta}\frac{\partial}{\partial\vartheta}\left\{\sin\vartheta\frac{\partial}{\partial\vartheta}Y^{LS}_{JM}(\vartheta,\varphi)\right\}+\frac{1}{\sin^{2}\vartheta}\frac{\partial^{2}}{\partial\varphi^{2}}Y^{LS}_{JM}(\vartheta,\varphi)+L(L+1)Y^{LS}_{JM}(\vartheta,\varphi)=0. \tag*{(1376)} sin 2 ϑ 1 ∂ ϑ ∂ { sin ϑ ∂ ϑ ∂ Y J M L S ( ϑ , φ ) } + sin 2 ϑ 1 ∂ φ 2 ∂ 2 Y J M L S ( ϑ , φ ) + L ( L + 1 ) Y J M L S ( ϑ , φ ) = 0. (1376)
(b) Components of the tensors r L Y J M L S ( ϑ , φ ) r^{L}Y^{LS}_{JM}(\vartheta,\varphi) r L Y J M L S ( ϑ , φ ) are harmonic polynomials of degree L L L and satisfy the Laplace equation
Δ { r L Y J M L S ( ϑ , φ ) } = 0. (1377) \Delta\left\{r^{L}Y^{LS}_{JM}(\vartheta,\varphi)\right\}=0. \tag*{(1377)} Δ { r L Y J M L S ( ϑ , φ ) } = 0. (1377)
(c) The quantities z L ( k r ) Y J M L S ( ϑ , φ ) z_{L}(kr)Y^{LS}_{JM}(\vartheta,\varphi) z L ( k r ) Y J M L S ( ϑ , φ ) obey the Helmholtz equation
( Δ + k 2 ) { z L ( k r ) Y J M L S ( ϑ , φ ) } = 0 , (1378) \left(\Delta+k^{2}\right)\left\{z_{L}(kr)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=0, \tag*{(1378)} ( Δ + k 2 ) { z L ( k r ) Y J M L S ( ϑ , φ ) } = 0 , (1378)
where
z L ( k r ) = π 2 k r Z L + 1 2 ( k r ) . (1379) z_{L}(kr)=\sqrt{\frac{\pi}{2kr}}Z_{L+\frac{1}{2}}(kr). \tag*{(1379)} z L ( k r ) = 2 k r π Z L + 2 1 ( k r ) . (1379)
Z L + 1 2 Z_{L+\frac{1}{2}} Z L + 2 1 is any of the cylinder functions of half-integer order.
Action of Operators ∇ \nabla ∇ , n \mathbf{n} n and Angular Momentum Operators
In the equations given below r = { r , ϑ , φ } \mathbf{r}=\{r,\vartheta,\varphi\} r = { r , ϑ , φ } , n = r / r \mathbf{n}=\mathbf{r}/r n = r / r , Φ ( r ) \Phi(r) Φ ( r ) is an arbitrary function of r = ∣ r ∣ r=|\mathbf{r}| r = ∣ r ∣ , L ^ \hat{\mathbf{L}} L ^ is the operator of orbital angular momentum, S ^ \hat{\mathbf{S}} S ^ is the spin operator, J ^ = L ^ + S ^ \hat{\mathbf{J}}=\hat{\mathbf{L}}+\hat{\mathbf{S}} J ^ = L ^ + S ^ is the operator of total angular momentum, and n μ , ∇ μ , L ^ μ n_{\mu},\nabla_{\mu},\hat{L}_{\mu} n μ , ∇ μ , L ^ μ , etc. are spherical components of the corresponding operators (μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ).
(a)
∇ μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = ( − 1 ) J + L + S ( 2 J + 1 ) ( L + 1 ) { d Φ ( r ) d r − L r Φ ( r ) } × ∑ J ′ { J J ′ 1 L + 1 L S } C J M 1 μ J ′ M + μ Y J ′ M + μ L + 1 , S ( ϑ , φ ) + ( − 1 ) J + L + S + 1 ( 2 J + 1 ) L { d Φ ( r ) d r + L + 1 r Φ ( r ) } × ∑ J ′ { J J ′ 1 L − 1 L S } C J M 1 μ J ′ M + μ Y J ′ M + μ L − 1 , S ( ϑ , φ ) . (1380) \begin{aligned}
\nabla_{\mu}\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}
={}&(-1)^{J+L+S}\sqrt{(2J+1)(L+1)}
\left\{\frac{d\Phi(r)}{dr}-\frac{L}{r}\Phi(r)\right\}\\
&\times\sum_{J'}\left\{\begin{matrix}J&J'&1\\L+1&L&S\end{matrix}\right\}C^{J'M+\mu}_{JM\,1\mu}Y^{L+1,S}_{J'M+\mu}(\vartheta,\varphi)\\
&+(-1)^{J+L+S+1}\sqrt{(2J+1)L}
\left\{\frac{d\Phi(r)}{dr}+\frac{L+1}{r}\Phi(r)\right\}\\
&\times\sum_{J'}\left\{\begin{matrix}J&J'&1\\L-1&L&S\end{matrix}\right\}C^{J'M+\mu}_{JM\,1\mu}Y^{L-1,S}_{J'M+\mu}(\vartheta,\varphi). \tag*{(1380)}
\end{aligned} ∇ μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = ( − 1 ) J + L + S ( 2 J + 1 ) ( L + 1 ) { d r d Φ ( r ) − r L Φ ( r ) } × J ′ ∑ { J L + 1 J ′ L 1 S } C J M 1 μ J ′ M + μ Y J ′ M + μ L + 1 , S ( ϑ , φ ) + ( − 1 ) J + L + S + 1 ( 2 J + 1 ) L { d r d Φ ( r ) + r L + 1 Φ ( r ) } × J ′ ∑ { J L − 1 J ′ L 1 S } C J M 1 μ J ′ M + μ Y J ′ M + μ L − 1 , S ( ϑ , φ ) . (1380)
In particular,
r ∇ μ Y J M L S ( ϑ , φ ) = ( − 1 ) J + L + S + 1 2 J + 1 ∑ J ′ L ′ A L ′ ( L ) { J J ′ 1 L ′ L S } C J M 1 μ J ′ M + μ Y J ′ M + μ L ′ S ( ϑ , φ ) , (1381) r\nabla_{\mu}Y^{LS}_{JM}(\vartheta,\varphi)=(-1)^{J+L+S+1}\sqrt{2J+1}\sum_{J'L'}A_{L'}(L)\left\{\begin{matrix}J&J'&1\\L'&L&S\end{matrix}\right\}C^{J'M+\mu}_{JM\,1\mu}Y^{L'S}_{J'M+\mu}(\vartheta,\varphi), \tag*{(1381)} r ∇ μ Y J M L S ( ϑ , φ ) = ( − 1 ) J + L + S + 1 2 J + 1 J ′ L ′ ∑ A L ′ ( L ) { J L ′ J ′ L 1 S } C J M 1 μ J ′ M + μ Y J ′ M + μ L ′ S ( ϑ , φ ) , (1381)
where
A L ′ ( L ) = { L L + 1 , if L ′ = L + 1 , ( L + 1 ) L , if L ′ = L − 1 , 0 , if L ′ ≠ L ± 1. A_{L'}(L)=\begin{cases}
L\sqrt{L+1},&\text{if }L'=L+1,\\
(L+1)\sqrt{L},&\text{if }L'=L-1,\\
0,&\text{if }L'\ne L\pm1.
\end{cases} A L ′ ( L ) = ⎩ ⎨ ⎧ L L + 1 , ( L + 1 ) L , 0 , if L ′ = L + 1 , if L ′ = L − 1 , if L ′ = L ± 1.
n μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = ( − 1 ) J + S + 1 Φ ( r ) ( 2 J + 1 ) ( 2 L + 1 ) ∑ J ′ L ′ ( − 1 ) L ′ { J J ′ 1 L ′ L S } C L 0 10 L ′ 0 C J M 1 μ J ′ M ′ Y J ′ M ′ L ′ S ( ϑ , φ ) , (1382) n_{\mu}\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=(-1)^{J+S+1}\Phi(r)\sqrt{(2J+1)(2L+1)}\sum_{J'L'}(-1)^{L'}\left\{\begin{matrix}J&J'&1\\L'&L&S\end{matrix}\right\}C^{L'0}_{L0\,10}C^{J'M'}_{JM\,1\mu}Y^{L'S}_{J'M'}(\vartheta,\varphi), \tag*{(1382)} n μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = ( − 1 ) J + S + 1 Φ ( r ) ( 2 J + 1 ) ( 2 L + 1 ) J ′ L ′ ∑ ( − 1 ) L ′ { J L ′ J ′ L 1 S } C L 0 10 L ′ 0 C J M 1 μ J ′ M ′ Y J ′ M ′ L ′ S ( ϑ , φ ) , (1382)
L ^ μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = Φ ( r ) L ^ μ { Y J M L S ( ϑ , φ ) } = ( − 1 ) J + L + S + 1 Φ ( r ) ( 2 J + 1 ) L ( L + 1 ) ( 2 L + 1 ) ∑ J ′ { J J ′ 1 L L S } C J M 1 μ J ′ M + μ Y J ′ M + μ L S ( ϑ , φ ) , (1383) \begin{aligned}
\widehat{L}_{\mu}\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=\Phi(r)\widehat{L}_{\mu}\left\{Y^{LS}_{JM}(\vartheta,\varphi)\right\} \\
&=(-1)^{J+L+S+1}\Phi(r)\sqrt{(2J+1)L(L+1)(2L+1)}\sum_{J'}\left\{\begin{matrix}J&J'&1\\L&L&S\end{matrix}\right\}C^{J'M+\mu}_{JM\,1\mu}Y^{LS}_{J'M+\mu}(\vartheta,\varphi), \tag*{(1383)}
\end{aligned} L μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = Φ ( r ) L μ { Y J M L S ( ϑ , φ ) } = ( − 1 ) J + L + S + 1 Φ ( r ) ( 2 J + 1 ) L ( L + 1 ) ( 2 L + 1 ) J ′ ∑ { J L J ′ L 1 S } C J M 1 μ J ′ M + μ Y J ′ M + μ L S ( ϑ , φ ) , (1383)
S ^ μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = Φ ( r ) S ^ μ { Y J M L S ( ϑ , φ ) } = ( − 1 ) L + 1 Φ ( r ) ( 2 J + 1 ) S ( S + 1 ) ( 2 S + 1 ) ∑ J ′ ( − 1 ) J + S { J J ′ 1 S S L } C J M 1 μ J ′ M + μ Y J ′ M + μ L S ( ϑ , φ ) (1384) \begin{aligned}
\widehat{S}_{\mu}\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=\Phi(r)\widehat{S}_{\mu}\left\{Y^{LS}_{JM}(\vartheta,\varphi)\right\} \\
&=(-1)^{L+1}\Phi(r)\sqrt{(2J+1)S(S+1)(2S+1)}\sum_{J'}(-1)^{J+S}\left\{\begin{matrix}J&J'&1\\S&S&L\end{matrix}\right\}C^{J'M+\mu}_{JM\,1\mu}Y^{LS}_{J'M+\mu}(\vartheta,\varphi) \tag*{(1384)}
\end{aligned} S μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = Φ ( r ) S μ { Y J M L S ( ϑ , φ ) } = ( − 1 ) L + 1 Φ ( r ) ( 2 J + 1 ) S ( S + 1 ) ( 2 S + 1 ) J ′ ∑ ( − 1 ) J + S { J S J ′ S 1 L } C J M 1 μ J ′ M + μ Y J ′ M + μ L S ( ϑ , φ ) (1384)
J ^ μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = Φ ( r ) J ^ μ { Y J M L S ( ϑ , φ ) } = Φ ( r ) J ( J + 1 ) C J M 1 μ J M + μ Y J M + μ L S ( ϑ , φ ) . (1385) \widehat{J}_{\mu}\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=\Phi(r)\widehat{J}_{\mu}\left\{Y^{LS}_{JM}(\vartheta,\varphi)\right\}=\Phi(r)\sqrt{J(J+1)}C^{JM+\mu}_{JM\,1\mu}Y^{LS}_{JM+\mu}(\vartheta,\varphi). \tag*{(1385)} J μ { Φ ( r ) Y J M L S ( ϑ , φ ) } = Φ ( r ) J μ { Y J M L S ( ϑ , φ ) } = Φ ( r ) J ( J + 1 ) C J M 1 μ J M + μ Y J M + μ L S ( ϑ , φ ) . (1385)
(b)
( r ⋅ ∇ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = r d Φ ( r ) d r Y J M L S ( ϑ , φ ) , (1386) (\mathbf r\cdot\nabla)\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=r\frac{d\Phi(r)}{dr}Y^{LS}_{JM}(\vartheta,\varphi), \tag*{(1386)} ( r ⋅ ∇ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = r d r d Φ ( r ) Y J M L S ( ϑ , φ ) , (1386)
( r ⋅ L ^ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = 0 , (1387) (\mathbf r\cdot\widehat{\mathbf L})\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=0, \tag*{(1387)} ( r ⋅ L ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = 0 , (1387)
( r ⋅ S ^ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = r Φ ( r ) ( − 1 ) J + L + S S ( S + 1 ) ( 2 S + 1 ) ( 2 L + 1 ) ∑ L ′ { L L ′ 1 S S L } C L 0 10 L ′ 0 Y J M L ′ S ( ϑ , φ ) . (1388) (\mathbf r\cdot\widehat{\mathbf S})\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=r\Phi(r)(-1)^{J+L+S}\sqrt{S(S+1)(2S+1)(2L+1)}\sum_{L'}\left\{\begin{matrix}L&L'&1\\S&S&L\end{matrix}\right\}C^{L'0}_{L0\,10}Y^{L'S}_{JM}(\vartheta,\varphi). \tag*{(1388)} ( r ⋅ S ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = r Φ ( r ) ( − 1 ) J + L + S S ( S + 1 ) ( 2 S + 1 ) ( 2 L + 1 ) L ′ ∑ { L S L ′ S 1 L } C L 0 10 L ′ 0 Y J M L ′ S ( ϑ , φ ) . (1388)
An expanded form of (7.1.32) is
( S ^ ⋅ n ) Y J M L S ( ϑ , φ ) = − 1 2 { ( J + L + S + 2 ) ( J + L − S + 1 ) ( J − L + S ) ( − J + L + S + 1 ) ( 2 L + 1 ) ( 2 L + 3 ) Y J M L + 1 , S ( ϑ , φ ) + ( J + L + S + 1 ) ( J + L − S ) ( J − L + S + 1 ) ( − J + L + S ) ( 2 L − 1 ) ( 2 L + 1 ) Y J M L − 1 , S ( ϑ , φ ) } , (1389) (\widehat{\mathbf S}\cdot\mathbf n)Y^{LS}_{JM}(\vartheta,\varphi)=-\frac{1}{2}\left\{\sqrt{\frac{(J+L+S+2)(J+L-S+1)(J-L+S)(-J+L+S+1)}{(2L+1)(2L+3)}}Y^{L+1,S}_{JM}(\vartheta,\varphi)+\sqrt{\frac{(J+L+S+1)(J+L-S)(J-L+S+1)(-J+L+S)}{(2L-1)(2L+1)}}Y^{L-1,S}_{JM}(\vartheta,\varphi)\right\}, \tag*{(1389)} ( S ⋅ n ) Y J M L S ( ϑ , φ ) = − 2 1 { ( 2 L + 1 ) ( 2 L + 3 ) ( J + L + S + 2 ) ( J + L − S + 1 ) ( J − L + S ) ( − J + L + S + 1 ) Y J M L + 1 , S ( ϑ , φ ) + ( 2 L − 1 ) ( 2 L + 1 ) ( J + L + S + 1 ) ( J + L − S ) ( J − L + S + 1 ) ( − J + L + S ) Y J M L − 1 , S ( ϑ , φ ) } , (1389)
( r ⋅ J ^ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = ( r ⋅ S ^ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } . (1390) (\mathbf r\cdot\widehat{\mathbf J})\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=(\mathbf r\cdot\widehat{\mathbf S})\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}. \tag*{(1390)} ( r ⋅ J ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = ( r ⋅ S ) { Φ ( r ) Y J M L S ( ϑ , φ ) } . (1390)
(c)
( S ^ ⋅ ∇ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = = − 1 2 { d Φ ( r ) d r − L r Φ ( r ) } ( J + L + S + 2 ) ( J + L − S + 1 ) ( J − L + S ) ( − J + L + S + 1 ) ( 2 L + 1 ) ( 2 L + 3 ) Y J M L + 1 , S ( ϑ , φ ) \begin{aligned}
(\widehat{\mathbf S}\cdot\nabla)\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}= \\
&=-\frac{1}{2}\left\{\frac{d\Phi(r)}{dr}-\frac{L}{r}\Phi(r)\right\}\sqrt{\frac{(J+L+S+2)(J+L-S+1)(J-L+S)(-J+L+S+1)}{(2L+1)(2L+3)}}Y^{L+1,S}_{JM}(\vartheta,\varphi)
\end{aligned} ( S ⋅ ∇ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = = − 2 1 { d r d Φ ( r ) − r L Φ ( r ) } ( 2 L + 1 ) ( 2 L + 3 ) ( J + L + S + 2 ) ( J + L − S + 1 ) ( J − L + S ) ( − J + L + S + 1 ) Y J M L + 1 , S ( ϑ , φ )
− 1 2 { d Φ ( r ) d r + L + 1 r Φ ( r ) } ( J + L + S + 1 ) ( J + L − S ) ( J − L + S + 1 ) ( − J + L + S ) ( 2 L − 1 ) ( 2 L + 1 ) Y J M L − 1 , S ( ϑ , φ ) . (1391) -\frac{1}{2}\left\{\frac{d\Phi(r)}{dr}+\frac{L+1}{r}\Phi(r)\right\}\sqrt{\frac{(J+L+S+1)(J+L-S)(J-L+S+1)(-J+L+S)}{(2L-1)(2L+1)}}Y^{L-1,S}_{JM}(\vartheta,\varphi). \tag*{(1391)} − 2 1 { d r d Φ ( r ) + r L + 1 Φ ( r ) } ( 2 L − 1 ) ( 2 L + 1 ) ( J + L + S + 1 ) ( J + L − S ) ( J − L + S + 1 ) ( − J + L + S ) Y J M L − 1 , S ( ϑ , φ ) . (1391)
( S ^ ⋅ L ^ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = 1 2 { J ( J + 1 ) − L ( L + 1 ) − S ( S + 1 ) } Φ ( r ) Y J M L S ( ϑ , φ ) . (1392) (\widehat{\mathbf S}\cdot\widehat{\mathbf L})\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=\frac{1}{2}\{J(J+1)-L(L+1)-S(S+1)\}\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi). \tag*{(1392)} ( S ⋅ L ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = 2 1 { J ( J + 1 ) − L ( L + 1 ) − S ( S + 1 )} Φ ( r ) Y J M L S ( ϑ , φ ) . (1392)
( S ^ ⋅ J ^ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = 1 2 { J ( J + 1 ) − L ( L + 1 ) + S ( S + 1 ) } Φ ( r ) Y J M L S ( ϑ , φ ) . (1393) (\widehat{\mathbf S}\cdot\widehat{\mathbf J})\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=\frac{1}{2}\{J(J+1)-L(L+1)+S(S+1)\}\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi). \tag*{(1393)} ( S ⋅ J ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = 2 1 { J ( J + 1 ) − L ( L + 1 ) + S ( S + 1 )} Φ ( r ) Y J M L S ( ϑ , φ ) . (1393)
( L ^ ⋅ J ^ ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = 1 2 { J ( J + 1 ) + L ( L + 1 ) − S ( S + 1 ) } Φ ( r ) Y J M L S ( ϑ , φ ) . (1394) (\widehat{\mathbf L}\cdot\widehat{\mathbf J})\left\{\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi)\right\}=\frac{1}{2}\{J(J+1)+L(L+1)-S(S+1)\}\Phi(r)Y^{LS}_{JM}(\vartheta,\varphi). \tag*{(1394)} ( L ⋅ J ) { Φ ( r ) Y J M L S ( ϑ , φ ) } = 2 1 { J ( J + 1 ) + L ( L + 1 ) − S ( S + 1 )} Φ ( r ) Y J M L S ( ϑ , φ ) . (1394)
Sums of Tensor Spherical Harmonics
In what follows (7.1.39)–(7.1.42) S S S is integer, and covariant components of tensor spherical harmonics are defined in accordance with (7.1.8), (7.1.9):
∑ μ = − S S Y S μ ( ϑ , φ ) [ Y J M L S ( ϑ , φ ) ] μ = ( 2 S + 1 ) ( 2 L + 1 ) 4 π ( 2 J + 1 ) C L 0 S 0 J 0 Y J M ( ϑ , φ ) . (1395) \sum_{\mu=-S}^{S}Y_{S\mu}(\vartheta,\varphi)\left[Y_{JM}^{LS}(\vartheta,\varphi)\right]^{\mu}=\sqrt{\frac{(2S+1)(2L+1)}{4\pi(2J+1)}}C_{L0S0}^{J0}Y_{JM}(\vartheta,\varphi). \tag*{(1395)} μ = − S ∑ S Y S μ ( ϑ , φ ) [ Y J M L S ( ϑ , φ ) ] μ = 4 π ( 2 J + 1 ) ( 2 S + 1 ) ( 2 L + 1 ) C L 0 S 0 J 0 Y J M ( ϑ , φ ) . (1395)
∑ μ = − S S Y S μ ∗ ( ϑ , φ ) [ Y J M L S ( ϑ , φ ) ] μ = ( 2 S + 1 ) ( 2 L + 1 ) 4 π ( 2 J + 1 ) C L 0 S 0 J 0 Y J M ∗ ( ϑ , φ ) . (1396) \sum_{\mu=-S}^{S}Y_{S\mu}^{*}(\vartheta,\varphi)\left[Y_{JM}^{LS}(\vartheta,\varphi)\right]_{\mu}=\sqrt{\frac{(2S+1)(2L+1)}{4\pi(2J+1)}}C_{L0S0}^{J0}Y_{JM}^{*}(\vartheta,\varphi). \tag*{(1396)} μ = − S ∑ S Y S μ ∗ ( ϑ , φ ) [ Y J M L S ( ϑ , φ ) ] μ = 4 π ( 2 J + 1 ) ( 2 S + 1 ) ( 2 L + 1 ) C L 0 S 0 J 0 Y J M ∗ ( ϑ , φ ) . (1396)
∑ M = − J J Y J M ∗ ( ϑ , φ ) [ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) μ ( 2 J + 1 ) ( 2 L + 1 ) 4 π ( 2 S + 1 ) C S 0 L 0 J 0 Y S − μ ( ϑ , φ ) . (1397) \sum_{M=-J}^{J}Y_{JM}^{*}(\vartheta,\varphi)\left[Y_{JM}^{LS}(\vartheta,\varphi)\right]^{\mu}=(-1)^{\mu}\sqrt{\frac{(2J+1)(2L+1)}{4\pi(2S+1)}}C_{S0L0}^{J0}Y_{S-\mu}(\vartheta,\varphi). \tag*{(1397)} M = − J ∑ J Y J M ∗ ( ϑ , φ ) [ Y J M L S ( ϑ , φ ) ] μ = ( − 1 ) μ 4 π ( 2 S + 1 ) ( 2 J + 1 ) ( 2 L + 1 ) C S 0 L 0 J 0 Y S − μ ( ϑ , φ ) . (1397)
∑ M = − J J Y J M ( ϑ , φ ) [ Y J M L S ( ϑ , φ ) ] μ = ( 2 J + 1 ) ( 2 L + 1 ) 4 π ( 2 S + 1 ) C S 0 L 0 J 0 Y S μ ( ϑ , φ ) . (1398) \sum_{M=-J}^{J}Y_{JM}(\vartheta,\varphi)\left[Y_{JM}^{LS}(\vartheta,\varphi)\right]_{\mu}=\sqrt{\frac{(2J+1)(2L+1)}{4\pi(2S+1)}}C_{S0L0}^{J0}Y_{S\mu}(\vartheta,\varphi). \tag*{(1398)} M = − J ∑ J Y J M ( ϑ , φ ) [ Y J M L S ( ϑ , φ ) ] μ = 4 π ( 2 S + 1 ) ( 2 J + 1 ) ( 2 L + 1 ) C S 0 L 0 J 0 Y S μ ( ϑ , φ ) . (1398)
Orthogonality, Normalization and Completeness
A collection of tensor spherical harmonics with the same S S S and all possible J J J , L L L , M M M forms a complete orthonormal set of tensor functions in the domain of the arguments 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π .
The orthonormality condition for tensor spherical harmonics has the form
∫ 0 π ∫ 0 2 π Y J ′ M ′ L S + ( ϑ , φ ) Y J M L S ( ϑ , φ ) sin ϑ d ϑ d φ = δ J J ′ δ M M ′ δ L L ′ . (1399) \int_{0}^{\pi}\int_{0}^{2\pi}Y_{J'M'}^{LS+}(\vartheta,\varphi)Y_{JM}^{LS}(\vartheta,\varphi)\sin\vartheta\,\mathrm{d}\vartheta\,\mathrm{d}\varphi=\delta_{JJ'}\delta_{MM'}\delta_{LL'}. \tag*{(1399)} ∫ 0 π ∫ 0 2 π Y J ′ M ′ L S + ( ϑ , φ ) Y J M L S ( ϑ , φ ) sin ϑ d ϑ d φ = δ J J ′ δ M M ′ δ L L ′ . (1399)
The completeness of these functions is given by
∑ J ∑ L = ∣ J − S ∣ J + S ∑ M = − J J [ Y J M L S ( ϑ , φ ) ] μ [ Y J M L S ( ϑ ′ , φ ′ ) ] ν ∗ = δ μ ν δ ( cos ϑ − cos ϑ ′ ) δ ( φ − φ ′ ) , (1400) \sum_{J}\sum_{L=\lvert J-S\rvert}^{J+S}\sum_{M=-J}^{J}\left[Y_{JM}^{LS}(\vartheta,\varphi)\right]^{\mu}\left[Y_{JM}^{LS}(\vartheta',\varphi')\right]^{\nu*}=\delta_{\mu\nu}\delta(\cos\vartheta-\cos\vartheta')\delta(\varphi-\varphi'), \tag*{(1400)} J ∑ L = ∣ J − S ∣ ∑ J + S M = − J ∑ J [ Y J M L S ( ϑ , φ ) ] μ [ Y J M L S ( ϑ ′ , φ ′ ) ] ν ∗ = δ μν δ ( cos ϑ − cos ϑ ′ ) δ ( φ − φ ′ ) , (1400)
( μ ; ν = − S , … , S − 1 , S ) (\mu;\nu=-S,\ldots,S-1,S) ( μ ; ν = − S , … , S − 1 , S ) .
Expansion in a Series of Tensor Spherical Harmonics
Any tensor F S ( ϑ , φ ) F_S(\vartheta,\varphi) F S ( ϑ , φ ) of rank S S S which depends on polar angles ϑ , φ \vartheta,\varphi ϑ , φ and possesses components which satisfy the conditions
∫ ∣ F S μ ( ϑ , φ ) ∣ 2 d Ω < ∞ , ( μ = − S , … , S − 1 , S ) , (1401) \int\left|F_{S\mu}(\vartheta,\varphi)\right|^2\,\mathrm{d}\Omega<\infty,\qquad(\mu=-S,\ldots,S-1,S), \tag*{(1401)} ∫ ∣ F S μ ( ϑ , φ ) ∣ 2 d Ω < ∞ , ( μ = − S , … , S − 1 , S ) , (1401)
may be expanded in a series of tensor spherical harmonics Y J M L S ( ϑ , φ ) Y_{JM}^{LS}(\vartheta,\varphi) Y J M L S ( ϑ , φ ) , i.e., written in the form
F S ( ϑ , φ ) = ∑ J L M A J L M Y J M L S ( ϑ , φ ) . (1402) F_S(\vartheta,\varphi)=\sum_{JLM}A_{JLM}Y_{JM}^{LS}(\vartheta,\varphi). \tag*{(1402)} F S ( ϑ , φ ) = J L M ∑ A J L M Y J M L S ( ϑ , φ ) . (1402)
The expansion coefficients are given by
A J L M = ∫ d Ω Y J M L S + ( ϑ , φ ) F S ( ϑ , φ ) . (1403) A_{JLM}=\int\mathrm{d}\Omega Y_{JM}^{LS+}(\vartheta,\varphi)F_S(\vartheta,\varphi). \tag*{(1403)} A J L M = ∫ d Ω Y J M L S + ( ϑ , φ ) F S ( ϑ , φ ) . (1403)
These coefficients satisfy the relation
∑ J L M ∣ A J L M ∣ 2 = ∫ d Ω ∣ F S ( ϑ , φ ) ∣ 2 . (1404) \sum_{JLM}\left|A_{JLM}\right|^2=\int\mathrm{d}\Omega\left|F_S(\vartheta,\varphi)\right|^2. \tag*{(1404)} J L M ∑ ∣ A J L M ∣ 2 = ∫ d Ω ∣ F S ( ϑ , φ ) ∣ 2 . (1404)
Spinor spherical harmonics
Definition
The spinor spherical harmonics Ω J M L ( ϑ , φ ) \Omega_{JM}^{L}(\vartheta,\varphi) Ω J M L ( ϑ , φ ) are defined as tensor spherical harmonics with S = 1 2 S=\frac{1}{2} S = 2 1 (§ 27.1.1 )
Ω J M L ( ϑ , φ ) ≡ Y J M L 1 2 ( ϑ , φ ) . (1405) \Omega_{JM}^{L}(\vartheta,\varphi)\equiv Y_{JM}^{L\frac{1}{2}}(\vartheta,\varphi). \tag*{(1405)} Ω J M L ( ϑ , φ ) ≡ Y J M L 2 1 ( ϑ , φ ) . (1405)
They are eigenfunctions of the operators J ^ 2 \hat{J}^{2} J ^ 2 , J ^ z \hat{J}_{z} J ^ z , L ^ 2 \hat{L}^{2} L ^ 2 , S ^ 2 \hat{S}^{2} S ^ 2 where L ^ \hat{L} L ^ is the operator of orbital angular momentum (§ 10.2 ), S ^ \hat{S} S ^ is the spin operator for S = 1 2 S=\frac{1}{2} S = 2 1 (§ 10.6 ) and J ^ \hat{J} J ^ is the operator of total angular momentum (§ 10.1 )
J ^ 2 Ω J M L ( ϑ , φ ) = J ( J + 1 ) Ω J M L ( ϑ , φ ) , J ^ z Ω J M L ( ϑ , φ ) = M Ω J M L ( ϑ , φ ) , L ^ 2 Ω J M L ( ϑ , φ ) = L ( L + 1 ) Ω J M L ( ϑ , φ ) , S ^ 2 Ω J M L ( ϑ , φ ) = 3 4 Ω J M L ( ϑ , φ ) . (1406) \begin{aligned}
\hat{J}^{2}\Omega^{L}_{JM}(\vartheta,\varphi)&=J(J+1)\Omega^{L}_{JM}(\vartheta,\varphi),\\
\hat{J}_{z}\Omega^{L}_{JM}(\vartheta,\varphi)&=M\Omega^{L}_{JM}(\vartheta,\varphi),\\
\hat{L}^{2}\Omega^{L}_{JM}(\vartheta,\varphi)&=L(L+1)\Omega^{L}_{JM}(\vartheta,\varphi),\\
\hat{S}^{2}\Omega^{L}_{JM}(\vartheta,\varphi)&=\frac{3}{4}\Omega^{L}_{JM}(\vartheta,\varphi). \tag*{(1406)}
\end{aligned} J ^ 2 Ω J M L ( ϑ , φ ) J ^ z Ω J M L ( ϑ , φ ) L ^ 2 Ω J M L ( ϑ , φ ) S ^ 2 Ω J M L ( ϑ , φ ) = J ( J + 1 ) Ω J M L ( ϑ , φ ) , = M Ω J M L ( ϑ , φ ) , = L ( L + 1 ) Ω J M L ( ϑ , φ ) , = 4 3 Ω J M L ( ϑ , φ ) . (1406)
The spinor spherical harmonics may be expressed in terms of scalar spherical harmonics Y L M ( ϑ , φ ) Y_{LM}(\vartheta,\varphi) Y L M ( ϑ , φ ) (Chap. 5) and spin functions χ 1 2 σ \chi_{\frac{1}{2}\sigma} χ 2 1 σ (§ 25.2 ) as
Ω J M L ( ϑ , φ ) = ∑ m σ C L m 1 2 σ J M Y L m ( ϑ , φ ) χ 1 2 σ . (1407) \Omega^{L}_{JM}(\vartheta,\varphi)=\sum_{m\sigma}C^{JM}_{Lm\frac{1}{2}\sigma}Y_{Lm}(\vartheta,\varphi)\chi_{\frac{1}{2}\sigma}. \tag*{(1407)} Ω J M L ( ϑ , φ ) = mσ ∑ C L m 2 1 σ J M Y L m ( ϑ , φ ) χ 2 1 σ . (1407)
For the spinor spherical harmonics J J J is a half-integer nonnegative number because L L L is always integer. For given J J J only two values of L L L are possible, L = J ± 1 2 L=J\pm\frac{1}{2} L = J ± 2 1 , while M M M assumes 2 J + 1 2J+1 2 J + 1 values: M = − J , − J + 1 , … , J − 1 , J M=-J,-J+1,\ldots,J-1,J M = − J , − J + 1 , … , J − 1 , J . The spinor spherical harmonics are orthonormalized in the domain of angles 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π (§ 27.3.8 ).
Components of Spinor Spherical Harmonics
(a) The spinor spherical harmonics may be expanded in terms of basis spin functions χ 1 2 μ \chi_{\frac{1}{2}\mu} χ 2 1 μ as
Ω J M L ( ϑ , φ ) = ∑ μ = − 1 2 1 2 [ Ω J M L ( ϑ , φ ) ] μ χ 1 2 μ . (1408) \Omega^{L}_{JM}(\vartheta,\varphi)=\sum_{\mu=-\frac{1}{2}}^{\frac{1}{2}}\left[\Omega^{L}_{JM}(\vartheta,\varphi)\right]^{\mu}\chi_{\frac{1}{2}\mu}. \tag*{(1408)} Ω J M L ( ϑ , φ ) = μ = − 2 1 ∑ 2 1 [ Ω J M L ( ϑ , φ ) ] μ χ 2 1 μ . (1408)
where contravariant components [ Ω J M L ( ϑ , φ ) ] μ \left[\Omega^{L}_{JM}(\vartheta,\varphi)\right]^{\mu} [ Ω J M L ( ϑ , φ ) ] μ are given by
[ Ω J M L ( ϑ , φ ) ] μ = C L M − μ 1 2 μ J M Y L M − μ ( ϑ , φ ) . (1409) \left[\Omega^{L}_{JM}(\vartheta,\varphi)\right]^{\mu}=C^{JM}_{L\,M-\mu\,\frac{1}{2}\mu}Y_{L\,M-\mu}(\vartheta,\varphi). \tag*{(1409)} [ Ω J M L ( ϑ , φ ) ] μ = C L M − μ 2 1 μ J M Y L M − μ ( ϑ , φ ) . (1409)
In more detailed form of (7.2.5) reads
[ Ω J M J + 1 2 ( ϑ , φ ) ] 1 2 = − J − M + 1 2 ( J + 1 ) Y J + 1 2 M − 1 2 ( ϑ , φ ) , \left[\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi)\right]^{\frac{1}{2}}=-\sqrt{\frac{J-M+1}{2(J+1)}}Y_{J+\frac{1}{2}\,M-\frac{1}{2}}(\vartheta,\varphi), [ Ω J M J + 2 1 ( ϑ , φ ) ] 2 1 = − 2 ( J + 1 ) J − M + 1 Y J + 2 1 M − 2 1 ( ϑ , φ ) ,
[ Ω J M J + 1 2 ( ϑ , φ ) ] − 1 2 = J + M + 1 2 ( J + 1 ) Y J + 1 2 M + 1 2 ( ϑ , φ ) , (1410) \left[\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi)\right]^{-\frac{1}{2}}=\sqrt{\frac{J+M+1}{2(J+1)}}Y_{J+\frac{1}{2}\,M+\frac{1}{2}}(\vartheta,\varphi), \tag*{(1410)} [ Ω J M J + 2 1 ( ϑ , φ ) ] − 2 1 = 2 ( J + 1 ) J + M + 1 Y J + 2 1 M + 2 1 ( ϑ , φ ) , (1410)
[ Ω J M J − 1 2 ( ϑ , φ ) ] 1 2 = J + M 2 J Y J − 1 2 M − 1 2 ( ϑ , φ ) , \left[\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi)\right]^{\frac{1}{2}}=\sqrt{\frac{J+M}{2J}}Y_{J-\frac{1}{2}\,M-\frac{1}{2}}(\vartheta,\varphi), [ Ω J M J − 2 1 ( ϑ , φ ) ] 2 1 = 2 J J + M Y J − 2 1 M − 2 1 ( ϑ , φ ) ,
[ Ω J M J − 1 2 ( ϑ , φ ) ] − 1 2 = J − M 2 J Y J − 1 2 M + 1 2 ( ϑ , φ ) . (1411) \left[\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi)\right]^{-\frac{1}{2}}=\sqrt{\frac{J-M}{2J}}Y_{J-\frac{1}{2}\,M+\frac{1}{2}}(\vartheta,\varphi). \tag*{(1411)} [ Ω J M J − 2 1 ( ϑ , φ ) ] − 2 1 = 2 J J − M Y J − 2 1 M + 2 1 ( ϑ , φ ) . (1411)
Covariant components of spinor spherical harmonics are defined by
[ Ω J M L ( ϑ , φ ) ] μ = ( − 1 ) 1 2 − μ [ Ω J M L ( ϑ , φ ) ] − μ , (1412) \left[\Omega^{L}_{JM}(\vartheta,\varphi)\right]_{\mu}=(-1)^{\frac{1}{2}-\mu}\left[\Omega^{L}_{JM}(\vartheta,\varphi)\right]^{-\mu}, \tag*{(1412)} [ Ω J M L ( ϑ , φ ) ] μ = ( − 1 ) 2 1 − μ [ Ω J M L ( ϑ , φ ) ] − μ , (1412)
and have the form
[ Ω J M L ( ϑ , φ ) ] μ = ( − 1 ) 1 2 − μ C L M + μ 1 2 − μ J M Y L M + μ ( ϑ , φ ) . (1413) \left[\Omega^{L}_{JM}(\vartheta,\varphi)\right]_{\mu}=(-1)^{\frac{1}{2}-\mu}C^{JM}_{L\,M+\mu\,\frac{1}{2}-\mu}Y_{L\,M+\mu}(\vartheta,\varphi). \tag*{(1413)} [ Ω J M L ( ϑ , φ ) ] μ = ( − 1 ) 2 1 − μ C L M + μ 2 1 − μ J M Y L M + μ ( ϑ , φ ) . (1413)
A more detailed form of (7.2.9) is
[ Ω J M J + 1 2 ( ϑ , φ ) ] 1 2 = J + M + 1 2 ( J + 1 ) Y J + 1 2 M + 1 2 ( ϑ , φ ) , \left[\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi)\right]_{\frac{1}{2}}=\sqrt{\frac{J+M+1}{2(J+1)}}Y_{J+\frac{1}{2}\,M+\frac{1}{2}}(\vartheta,\varphi), [ Ω J M J + 2 1 ( ϑ , φ ) ] 2 1 = 2 ( J + 1 ) J + M + 1 Y J + 2 1 M + 2 1 ( ϑ , φ ) ,
[ Ω J M J + 1 2 ( ϑ , φ ) ] − 1 2 = J − M + 1 2 ( J + 1 ) Y J + 1 2 M − 1 2 ( ϑ , φ ) , (1414) \left[\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi)\right]_{-\frac{1}{2}}=\sqrt{\frac{J-M+1}{2(J+1)}}Y_{J+\frac{1}{2}\,M-\frac{1}{2}}(\vartheta,\varphi), \tag*{(1414)} [ Ω J M J + 2 1 ( ϑ , φ ) ] − 2 1 = 2 ( J + 1 ) J − M + 1 Y J + 2 1 M − 2 1 ( ϑ , φ ) , (1414)
[ Ω J M J − 1 2 ( ϑ , φ ) ] 1 2 = J − M 2 J Y J − 1 2 M + 1 2 ( ϑ , φ ) , \left[\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi)\right]_{\frac{1}{2}}=\sqrt{\frac{J-M}{2J}}Y_{J-\frac{1}{2}\,M+\frac{1}{2}}(\vartheta,\varphi), [ Ω J M J − 2 1 ( ϑ , φ ) ] 2 1 = 2 J J − M Y J − 2 1 M + 2 1 ( ϑ , φ ) ,
[ Ω J M J − 1 2 ( ϑ , φ ) ] − 1 2 = − J + M 2 J Y J − 1 2 M − 1 2 ( ϑ , φ ) . (1415) \left[\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi)\right]_{-\frac{1}{2}}=-\sqrt{\frac{J+M}{2J}}Y_{J-\frac{1}{2}\,M-\frac{1}{2}}(\vartheta,\varphi). \tag*{(1415)} [ Ω J M J − 2 1 ( ϑ , φ ) ] − 2 1 = − 2 J J + M Y J − 2 1 M − 2 1 ( ϑ , φ ) . (1415)
(b) If spinors χ 1 2 σ \chi_{\frac{1}{2}\sigma} χ 2 1 σ are written as column matrices, the spinor spherical harmonics assume the form
Ω J M J + 1 2 ( ϑ , φ ) = ( − J − M + 1 2 ( J + 1 ) Y J + 1 2 , M − 1 2 ( ϑ , φ ) J + M + 1 2 ( J + 1 ) Y J + 1 2 , M + 1 2 ( ϑ , φ ) ) (1416) \Omega_{JM}^{J+\frac{1}{2}}(\vartheta,\varphi)=\begin{pmatrix}
-\sqrt{\frac{J-M+1}{2(J+1)}}Y_{J+\frac{1}{2},M-\frac{1}{2}}(\vartheta,\varphi)\\
\sqrt{\frac{J+M+1}{2(J+1)}}Y_{J+\frac{1}{2},M+\frac{1}{2}}(\vartheta,\varphi)
\end{pmatrix} \tag*{(1416)} Ω J M J + 2 1 ( ϑ , φ ) = − 2 ( J + 1 ) J − M + 1 Y J + 2 1 , M − 2 1 ( ϑ , φ ) 2 ( J + 1 ) J + M + 1 Y J + 2 1 , M + 2 1 ( ϑ , φ ) (1416)
Ω J M J − 1 2 ( ϑ , φ ) = ( J + M 2 J Y J − 1 2 , M − 1 2 ( ϑ , φ ) J − M 2 J Y J − 1 2 , M + 1 2 ( ϑ , φ ) ) . (1417) \Omega_{JM}^{J-\frac{1}{2}}(\vartheta,\varphi)=\begin{pmatrix}
\sqrt{\frac{J+M}{2J}}Y_{J-\frac{1}{2},M-\frac{1}{2}}(\vartheta,\varphi)\\
\sqrt{\frac{J-M}{2J}}Y_{J-\frac{1}{2},M+\frac{1}{2}}(\vartheta,\varphi)
\end{pmatrix}. \tag*{(1417)} Ω J M J − 2 1 ( ϑ , φ ) = 2 J J + M Y J − 2 1 , M − 2 1 ( ϑ , φ ) 2 J J − M Y J − 2 1 , M + 2 1 ( ϑ , φ ) . (1417)
(c) An expansion of Ω J M L ( ϑ , φ ) \Omega_{JM}^{L}(\vartheta,\varphi) Ω J M L ( ϑ , φ ) in terms of helicity basis spin functions χ 1 2 λ ( ϑ , φ ) \chi_{\frac{1}{2}\lambda}(\vartheta,\varphi) χ 2 1 λ ( ϑ , φ ) (§ 25.3.3 ) may be written as
Ω J M L ( ϑ , φ ) = 2 L + 1 4 π ∑ λ C L 0 1 2 λ J λ D λ − M J ( 0 , ϑ , φ ) χ 1 2 λ ( ϑ , φ ) , (1418) \Omega_{JM}^{L}(\vartheta,\varphi)=\sqrt{\frac{2L+1}{4\pi}}\sum_{\lambda}C_{L0\frac{1}{2}\lambda}^{J\lambda}D_{\lambda-M}^{J}(0,\vartheta,\varphi)\chi_{\frac{1}{2}\lambda}(\vartheta,\varphi), \tag*{(1418)} Ω J M L ( ϑ , φ ) = 4 π 2 L + 1 λ ∑ C L 0 2 1 λ J λ D λ − M J ( 0 , ϑ , φ ) χ 2 1 λ ( ϑ , φ ) , (1418)
or, in a detailed form
Ω J M J + 1 2 ( ϑ , φ ) = 2 J + 1 8 π ∑ λ ( − 1 ) 1 2 + λ D − λ − M J ( 0 , ϑ , φ ) χ 1 2 λ ( ϑ , φ ) , (1419) \Omega_{JM}^{J+\frac{1}{2}}(\vartheta,\varphi)=\sqrt{\frac{2J+1}{8\pi}}\sum_{\lambda}(-1)^{\frac{1}{2}+\lambda}D_{-\lambda-M}^{J}(0,\vartheta,\varphi)\chi_{\frac{1}{2}\lambda}(\vartheta,\varphi), \tag*{(1419)} Ω J M J + 2 1 ( ϑ , φ ) = 8 π 2 J + 1 λ ∑ ( − 1 ) 2 1 + λ D − λ − M J ( 0 , ϑ , φ ) χ 2 1 λ ( ϑ , φ ) , (1419)
Ω J M J − 1 2 ( ϑ , φ ) = 2 J + 1 8 π ∑ λ D − λ − M J ( 0 , ϑ , φ ) χ 1 2 λ ( ϑ , φ ) . \Omega_{JM}^{J-\frac{1}{2}}(\vartheta,\varphi)=\sqrt{\frac{2J+1}{8\pi}}\sum_{\lambda}D_{-\lambda-M}^{J}(0,\vartheta,\varphi)\chi_{\frac{1}{2}\lambda}(\vartheta,\varphi). Ω J M J − 2 1 ( ϑ , φ ) = 8 π 2 J + 1 λ ∑ D − λ − M J ( 0 , ϑ , φ ) χ 2 1 λ ( ϑ , φ ) .
Complex Conjugation. Time Reversal
(a) Here we use the representation in which the basis spin functions are real, χ 1 2 σ ∗ = χ 1 2 σ \chi_{\frac{1}{2}\sigma}^{*}=\chi_{\frac{1}{2}\sigma} χ 2 1 σ ∗ = χ 2 1 σ . For this reason, the transformation properties of spinor spherical harmonics with respect to complex conjugation may be written as
Ω J M L ∗ ( ϑ , φ ) = ( − 1 ) J + L − M i σ ^ y Ω J , − M L ( ϑ , φ ) , (1420) \Omega_{JM}^{L*}(\vartheta,\varphi)=(-1)^{J+L-M}i\hat{\sigma}_{y}\Omega_{J,-M}^{L}(\vartheta,\varphi), \tag*{(1420)} Ω J M L ∗ ( ϑ , φ ) = ( − 1 ) J + L − M i σ ^ y Ω J , − M L ( ϑ , φ ) , (1420)
where 2 × 2 2\times2 2 × 2 matrix i σ ^ y i\hat{\sigma}_{y} i σ ^ y has the form
i σ ^ y = ( 0 1 − 1 0 ) . (1421) i\hat{\sigma}_{y}=\begin{pmatrix}
0 & 1\\
-1 & 0
\end{pmatrix}. \tag*{(1421)} i σ ^ y = ( 0 − 1 1 0 ) . (1421)
Components of spinor spherical harmonics transform under complex conjugation according to
[ Ω J M L ( ϑ , φ ) ] μ ∗ = [ Ω J M L ∗ ( ϑ , φ ) ] μ = ( − 1 ) J + L + M [ Ω J , − M L ( ϑ , φ ) ] μ = ( − 1 ) J + L − M − μ + 1 2 [ Ω J , − M L ( ϑ , φ ) ] − μ , (1422) \left[\Omega_{JM}^{L}(\vartheta,\varphi)\right]_{\mu}^{*}=\left[\Omega_{JM}^{L*}(\vartheta,\varphi)\right]_{\mu}=(-1)^{J+L+M}\left[\Omega_{J,-M}^{L}(\vartheta,\varphi)\right]^{\mu}=(-1)^{J+L-M-\mu+\frac{1}{2}}\left[\Omega_{J,-M}^{L}(\vartheta,\varphi)\right]_{-\mu}, \tag*{(1422)} [ Ω J M L ( ϑ , φ ) ] μ ∗ = [ Ω J M L ∗ ( ϑ , φ ) ] μ = ( − 1 ) J + L + M [ Ω J , − M L ( ϑ , φ ) ] μ = ( − 1 ) J + L − M − μ + 2 1 [ Ω J , − M L ( ϑ , φ ) ] − μ , (1422)
[ Ω J M L ( ϑ , φ ) ] μ ∗ = [ Ω J M L ∗ ( ϑ , φ ) ] μ = ( − 1 ) J + L − M [ Ω J , − M L ( ϑ , φ ) ] μ = ( − 1 ) J + L − M − μ + 1 2 [ Ω J , − M L ( ϑ , φ ) ] − μ . (1423) \left[\Omega_{JM}^{L}(\vartheta,\varphi)\right]^{\mu*}=\left[\Omega_{JM}^{L*}(\vartheta,\varphi)\right]^{\mu}=(-1)^{J+L-M}\left[\Omega_{J,-M}^{L}(\vartheta,\varphi)\right]_{\mu}=(-1)^{J+L-M-\mu+\frac{1}{2}}\left[\Omega_{J,-M}^{L}(\vartheta,\varphi)\right]^{-\mu}. \tag*{(1423)} [ Ω J M L ( ϑ , φ ) ] μ ∗ = [ Ω J M L ∗ ( ϑ , φ ) ] μ = ( − 1 ) J + L − M [ Ω J , − M L ( ϑ , φ ) ] μ = ( − 1 ) J + L − M − μ + 2 1 [ Ω J , − M L ( ϑ , φ ) ] − μ . (1423)
(b) The time-reversal operator P ^ t \hat{P}_{t} P ^ t (as defined by Wigner) acts on a spinor spherical harmonic in the following manner
P ^ t Ω J M L ( ϑ , φ ) = σ ^ y Ω J M L ∗ ( ϑ , φ ) = ( − 1 ) J + L − M + 1 2 Ω J , − M L ( ϑ , φ ) . (1424) \hat{P}_{t}\Omega_{JM}^{L}(\vartheta,\varphi)=\hat{\sigma}_{y}\Omega_{JM}^{L*}(\vartheta,\varphi)=(-1)^{J+L-M+\frac{1}{2}}\Omega_{J,-M}^{L}(\vartheta,\varphi). \tag*{(1424)} P ^ t Ω J M L ( ϑ , φ ) = σ ^ y Ω J M L ∗ ( ϑ , φ ) = ( − 1 ) J + L − M + 2 1 Ω J , − M L ( ϑ , φ ) . (1424)
Transformation of Coordinate Systems.
(a) Coordinate inversion
P ^ R Ω J M L ( ϑ , φ ) = Ω J M L ( π − ϑ , π + φ ) = ( − 1 ) L Ω J M L ( ϑ , φ ) . (1425) \hat{P}_{R}\Omega_{JM}^{L}(\vartheta,\varphi)=\Omega_{JM}^{L}(\pi-\vartheta,\pi+\varphi)=(-1)^{L}\Omega_{JM}^{L}(\vartheta,\varphi). \tag*{(1425)} P ^ R Ω J M L ( ϑ , φ ) = Ω J M L ( π − ϑ , π + φ ) = ( − 1 ) L Ω J M L ( ϑ , φ ) . (1425)
(b) Rotations of coordinate system
Under rotations specified by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ , the spinor spherical harmonics transform according to
Ω J M ′ L ( ϑ ′ , φ ′ ) = D ^ ( α , β , γ ) Ω J M ′ L ( ϑ , φ ) = ∑ M = − J J D M M ′ J ( α , β , γ ) Ω J M L ( ϑ , φ ) , (1426) \Omega_{JM'}^{L}(\vartheta',\varphi')=\hat{D}(\alpha,\beta,\gamma)\Omega_{JM'}^{L}(\vartheta,\varphi)=\sum_{M=-J}^{J}D_{MM'}^{J}(\alpha,\beta,\gamma)\Omega_{JM}^{L}(\vartheta,\varphi), \tag*{(1426)} Ω J M ′ L ( ϑ ′ , φ ′ ) = D ^ ( α , β , γ ) Ω J M ′ L ( ϑ , φ ) = M = − J ∑ J D M M ′ J ( α , β , γ ) Ω J M L ( ϑ , φ ) , (1426)
where D M M ′ J D_{MM'}^{J} D M M ′ J are the Wigner D D D -functions (Chap. 4), M ′ M' M ′ is the projection of total angular momentum on the rotated z ′ z' z ′ -axis; ϑ , φ \vartheta,\varphi ϑ , φ and ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ are polar angles of the position vector r \mathbf{r} r in the initial and final coordinate systems. The relations between ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ and ϑ , φ \vartheta,\varphi ϑ , φ are given by Eqs. (1.4.2)–(1.4.3).
Action of and Angular Momentum Operators
Action of ∇ \nabla ∇ and Angular Momentum Operators
The spinor spherical harmonics satisfy Eqs. (1380) –(1394) for S = 1 2 S=\frac{1}{2} S = 2 1 . Note also the following relations in which S ^ \hat{\mathbf{S}} S ^ is the spin operator (§ 10.6 ), L ^ \hat{\mathbf{L}} L ^ is the operator of orbital angular momentum (§ 10.2 ), J ^ = L ^ + S ^ \hat{\mathbf{J}}=\hat{\mathbf{L}}+\hat{\mathbf{S}} J ^ = L ^ + S ^ is the operator of total angular momentum, and n ( ϑ , φ ) = r / r \mathbf{n}(\vartheta,\varphi)=\mathbf{r}/r n ( ϑ , φ ) = r / r .
(a)
( S ^ ⋅ n ) Ω J M L ( ϑ , φ ) = − 1 2 Ω J M L ′ ( ϑ , φ ) , (1427) (\hat{\mathbf{S}}\cdot\mathbf{n})\Omega^L_{JM}(\vartheta,\varphi)=-\frac{1}{2}\Omega^{L'}_{JM}(\vartheta,\varphi), \tag*{(1427)} ( S ^ ⋅ n ) Ω J M L ( ϑ , φ ) = − 2 1 Ω J M L ′ ( ϑ , φ ) , (1427)
where
L ′ = 2 J − L = { J − 1 2 , if L = J + 1 2 , J + 1 2 , if L = J − 1 2 . (1428) L'=2J-L=\begin{cases}
J-\frac{1}{2}, & \text{if }L=J+\frac{1}{2},\\
J+\frac{1}{2}, & \text{if }L=J-\frac{1}{2}.
\end{cases} \tag*{(1428)} L ′ = 2 J − L = { J − 2 1 , J + 2 1 , if L = J + 2 1 , if L = J − 2 1 . (1428)
An expanded form of (7.2.23) is
( S ^ ⋅ n ) Ω J M J + 1 2 ( ϑ , φ ) = − 1 2 Ω J M J − 1 2 ( ϑ , φ ) (\hat{\mathbf{S}}\cdot\mathbf{n})\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi)=-\frac{1}{2}\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi) ( S ^ ⋅ n ) Ω J M J + 2 1 ( ϑ , φ ) = − 2 1 Ω J M J − 2 1 ( ϑ , φ )
( S ^ ⋅ n ) Ω J M J − 1 2 ( ϑ , φ ) = − 1 2 Ω J M J + 1 2 ( ϑ , φ ) . (1429) (\hat{\mathbf{S}}\cdot\mathbf{n})\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi)=-\frac{1}{2}\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi). \tag*{(1429)} ( S ^ ⋅ n ) Ω J M J − 2 1 ( ϑ , φ ) = − 2 1 Ω J M J + 2 1 ( ϑ , φ ) . (1429)
(b)
r ( S ^ ⋅ ∇ ) Ω J M L ( ϑ , φ ) = − 1 + κ 2 Ω J M L ′ ( ϑ , φ ) , (1430) r(\hat{\mathbf{S}}\cdot\nabla)\Omega^L_{JM}(\vartheta,\varphi)=-\frac{1+\kappa}{2}\Omega^{L'}_{JM}(\vartheta,\varphi), \tag*{(1430)} r ( S ^ ⋅ ∇ ) Ω J M L ( ϑ , φ ) = − 2 1 + κ Ω J M L ′ ( ϑ , φ ) , (1430)
where L ′ = 2 J − L L'=2J-L L ′ = 2 J − L ,
κ = { J + 1 2 = L , if L = J + 1 2 , − ( J + 1 2 ) = − ( L + 1 ) , if L = J − 1 2 . (1431) \kappa=\begin{cases}
J+\frac{1}{2}=L, & \text{if }L=J+\frac{1}{2},\\
-\left(J+\frac{1}{2}\right)=-(L+1), & \text{if }L=J-\frac{1}{2}.
\end{cases} \tag*{(1431)} κ = { J + 2 1 = L , − ( J + 2 1 ) = − ( L + 1 ) , if L = J + 2 1 , if L = J − 2 1 . (1431)
In a more detailed form (7.2.26) gives
r ( S ^ ⋅ ∇ ) Ω J M J + 1 2 ( ϑ , φ ) = − 2 J + 3 4 Ω J M J − 1 2 ( ϑ , φ ) r(\hat{\mathbf{S}}\cdot\nabla)\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi)=-\frac{2J+3}{4}\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi) r ( S ^ ⋅ ∇ ) Ω J M J + 2 1 ( ϑ , φ ) = − 4 2 J + 3 Ω J M J − 2 1 ( ϑ , φ )
r ( S ^ ⋅ ∇ ) Ω J M J − 1 2 ( ϑ , φ ) = 2 J − 1 4 Ω J M J + 1 2 ( ϑ , φ ) . (1432) r(\hat{\mathbf{S}}\cdot\nabla)\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi)=\frac{2J-1}{4}\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi). \tag*{(1432)} r ( S ^ ⋅ ∇ ) Ω J M J − 2 1 ( ϑ , φ ) = 4 2 J − 1 Ω J M J + 2 1 ( ϑ , φ ) . (1432)
(c)
( L ^ ⋅ S ^ ) Ω J M L ( ϑ , φ ) = 1 2 { J ( J + 1 ) − L ( L + 1 ) − 3 4 } Ω J M L ( ϑ , φ ) = − 1 + κ 2 Ω J M L ( ϑ , φ ) , (1433) (\hat{\mathbf{L}}\cdot\hat{\mathbf{S}})\Omega^L_{JM}(\vartheta,\varphi)=\frac{1}{2}\left\{J(J+1)-L(L+1)-\frac{3}{4}\right\}\Omega^L_{JM}(\vartheta,\varphi)=-\frac{1+\kappa}{2}\Omega^L_{JM}(\vartheta,\varphi), \tag*{(1433)} ( L ^ ⋅ S ^ ) Ω J M L ( ϑ , φ ) = 2 1 { J ( J + 1 ) − L ( L + 1 ) − 4 3 } Ω J M L ( ϑ , φ ) = − 2 1 + κ Ω J M L ( ϑ , φ ) , (1433)
or, in detailed form
( L ^ ⋅ S ^ ) Ω J M J + 1 2 ( ϑ , φ ) = − 2 J + 3 4 Ω J M J + 1 2 ( ϑ , φ ) (\hat{\mathbf{L}}\cdot\hat{\mathbf{S}})\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi)=-\frac{2J+3}{4}\Omega^{J+\frac{1}{2}}_{JM}(\vartheta,\varphi) ( L ^ ⋅ S ^ ) Ω J M J + 2 1 ( ϑ , φ ) = − 4 2 J + 3 Ω J M J + 2 1 ( ϑ , φ )
( L ^ ⋅ S ^ ) Ω J M J − 1 2 ( ϑ , φ ) = 2 J − 1 4 Ω J M J − 1 2 ( ϑ , φ ) . (1434) (\hat{\mathbf{L}}\cdot\hat{\mathbf{S}})\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi)=\frac{2J-1}{4}\Omega^{J-\frac{1}{2}}_{JM}(\vartheta,\varphi). \tag*{(1434)} ( L ^ ⋅ S ^ ) Ω J M J − 2 1 ( ϑ , φ ) = 4 2 J − 1 Ω J M J − 2 1 ( ϑ , φ ) . (1434)
Recursion Relations
cos ϑ Ω J M L ( ϑ , φ ) = ( J − M + 1 ) ( J + M + 1 ) 2 ( J + 1 ) Ω J + 1 M L + 1 ( ϑ , φ ) − M 2 J ( J + 1 ) Ω J M L ′ ( ϑ , φ ) + ( J − M ) ( J + M ) 2 J Ω J − 1 M L − 1 ( ϑ , φ ) (1435) \cos\vartheta\Omega^L_{JM}(\vartheta,\varphi)=\frac{\sqrt{(J-M+1)(J+M+1)}}{2(J+1)}\Omega^{L+1}_{J+1M}(\vartheta,\varphi)-\frac{M}{2J(J+1)}\Omega^{L'}_{JM}(\vartheta,\varphi)+\frac{\sqrt{(J-M)(J+M)}}{2J}\Omega^{L-1}_{J-1M}(\vartheta,\varphi) \tag*{(1435)} cos ϑ Ω J M L ( ϑ , φ ) = 2 ( J + 1 ) ( J − M + 1 ) ( J + M + 1 ) Ω J + 1 M L + 1 ( ϑ , φ ) − 2 J ( J + 1 ) M Ω J M L ′ ( ϑ , φ ) + 2 J ( J − M ) ( J + M ) Ω J − 1 M L − 1 ( ϑ , φ ) (1435)
sin ϑ e i φ Ω J M L ( ϑ , φ ) = − ( J + M + 1 ) ( J + M + 2 ) 2 ( J + 1 ) Ω J + 1 M + 1 L + 1 ( ϑ , φ ) − ( J + M + 1 ) ( J − M ) 2 J ( J + 1 ) Ω J M + 1 L ′ ( ϑ , φ ) − ( J − M ) ( J − M − 1 ) 2 J Ω J − 1 M + 1 L − 1 ( ϑ , φ ) (1436) \sin\vartheta e^{i\varphi}\Omega^L_{JM}(\vartheta,\varphi)=-\frac{\sqrt{(J+M+1)(J+M+2)}}{2(J+1)}\Omega^{L+1}_{J+1M+1}(\vartheta,\varphi)-\frac{\sqrt{(J+M+1)(J-M)}}{2J(J+1)}\Omega^{L'}_{JM+1}(\vartheta,\varphi)-\frac{\sqrt{(J-M)(J-M-1)}}{2J}\Omega^{L-1}_{J-1M+1}(\vartheta,\varphi) \tag*{(1436)} sin ϑ e i φ Ω J M L ( ϑ , φ ) = − 2 ( J + 1 ) ( J + M + 1 ) ( J + M + 2 ) Ω J + 1 M + 1 L + 1 ( ϑ , φ ) − 2 J ( J + 1 ) ( J + M + 1 ) ( J − M ) Ω J M + 1 L ′ ( ϑ , φ ) − 2 J ( J − M ) ( J − M − 1 ) Ω J − 1 M + 1 L − 1 ( ϑ , φ ) (1436)
sin ϑ e − i φ Ω J M L ( ϑ , φ ) = ( J − M + 1 ) ( J − M + 2 ) 2 ( J + 1 ) Ω J + 1 M − 1 L + 1 ( ϑ , φ ) − ( J + M ) ( J − M + 1 ) 2 J ( J + 1 ) Ω J M − 1 L ′ ( ϑ , φ ) − ( J + M ) ( J + M − 1 ) 2 J Ω J − 1 M − 1 L − 1 ( ϑ , φ ) (1437) \sin\vartheta e^{-i\varphi}\Omega^L_{JM}(\vartheta,\varphi)=\frac{\sqrt{(J-M+1)(J-M+2)}}{2(J+1)}\Omega^{L+1}_{J+1M-1}(\vartheta,\varphi)-\frac{\sqrt{(J+M)(J-M+1)}}{2J(J+1)}\Omega^{L'}_{JM-1}(\vartheta,\varphi)-\frac{\sqrt{(J+M)(J+M-1)}}{2J}\Omega^{L-1}_{J-1M-1}(\vartheta,\varphi) \tag*{(1437)} sin ϑ e − i φ Ω J M L ( ϑ , φ ) = 2 ( J + 1 ) ( J − M + 1 ) ( J − M + 2 ) Ω J + 1 M − 1 L + 1 ( ϑ , φ ) − 2 J ( J + 1 ) ( J + M ) ( J − M + 1 ) Ω J M − 1 L ′ ( ϑ , φ ) − 2 J ( J + M ) ( J + M − 1 ) Ω J − 1 M − 1 L − 1 ( ϑ , φ ) (1437)
where L ′ L' L ′ is defined by (7.2.24). Note that Eqs. (1435) –(1437) may be written in a compact form as
n μ Ω J M L ( ϑ , φ ) = ( − 1 ) J + L + 1 2 ( 2 J + 1 ) ( 2 L + 1 ) ∑ J ′ L ′ { J ′ J 1 2 L L ′ 1 } C L 0 1 0 L ′ 0 C J M 1 μ J ′ M + μ Ω J ′ M + μ L ′ ( ϑ , φ ) (1438) n_{\mu}\Omega^{L}_{JM}(\vartheta,\varphi)=(-1)^{J+L+\frac{1}{2}}\sqrt{(2J+1)(2L+1)}\sum_{J'L'}\begin{Bmatrix}J'&J&\frac{1}{2}\\L&L'&1\end{Bmatrix}C^{L'0}_{L0\,1\,0}C^{J'M+\mu}_{JM\,1\mu}\Omega^{L'}_{J'M+\mu}(\vartheta,\varphi) \tag*{(1438)} n μ Ω J M L ( ϑ , φ ) = ( − 1 ) J + L + 2 1 ( 2 J + 1 ) ( 2 L + 1 ) J ′ L ′ ∑ { J ′ L J L ′ 2 1 1 } C L 0 1 0 L ′ 0 C J M 1 μ J ′ M + μ Ω J ′ M + μ L ′ ( ϑ , φ ) (1438)
where n μ n_{\mu} n μ (μ = ± 1 , 0 \mu=\pm1,0 μ = ± 1 , 0 ) are spherical components of the unit vector n ( ϑ , φ ) \mathbf{n}(\vartheta,\varphi) n ( ϑ , φ ) .
Orthonormality and Completeness. Expansion in a Series of Spinor Spherical Harmonics
The collection of spinor spherical harmonics Ω J M L ( ϑ , φ ) \Omega^{L}_{JM}(\vartheta,\varphi) Ω J M L ( ϑ , φ ) with half-integer nonnegative J J J (1 2 ≤ J < ∞ \frac{1}{2}\leq J<\infty 2 1 ≤ J < ∞ ), L = J ± 1 2 L=J\pm\frac{1}{2} L = J ± 2 1 and M = − J , − J + 1 , … , J M=-J,-J+1,\ldots,J M = − J , − J + 1 , … , J forms complete set of spinors for 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π .
The orthonormality condition has the form
∫ 0 π d ϑ sin ϑ ∫ 0 2 π d φ Ω J ′ M ′ L ′ ( ϑ , φ ) Ω J M L ( ϑ , φ ) = δ J ′ J δ L ′ L δ M ′ M . (1439) \int_{0}^{\pi}d\vartheta\sin\vartheta\int_{0}^{2\pi}d\varphi\,\Omega^{L'}_{J'M'}(\vartheta,\varphi)\Omega^{L}_{JM}(\vartheta,\varphi)=\delta_{J'J}\delta_{L'L}\delta_{M'M}. \tag*{(1439)} ∫ 0 π d ϑ sin ϑ ∫ 0 2 π d φ Ω J ′ M ′ L ′ ( ϑ , φ ) Ω J M L ( ϑ , φ ) = δ J ′ J δ L ′ L δ M ′ M . (1439)
The completeness condition may be written as
∑ J = 1 2 ∞ ∑ L = J − 1 2 J + 1 2 ∑ M = − J J Ω J M L ( ϑ , φ ) Ω J M L + ( ϑ ′ , φ ′ ) = I ^ δ b ( cos ϑ − cos ϑ ′ ) δ ( φ − φ ′ ) , (1440) \sum_{J=\frac{1}{2}}^{\infty}\sum_{L=J-\frac{1}{2}}^{J+\frac{1}{2}}\sum_{M=-J}^{J}\Omega^{L}_{JM}(\vartheta,\varphi)\Omega^{L+}_{JM}(\vartheta',\varphi')=\widehat{I}\delta_{\mathrm{b}}(\cos\vartheta-\cos\vartheta')\delta(\varphi-\varphi'), \tag*{(1440)} J = 2 1 ∑ ∞ L = J − 2 1 ∑ J + 2 1 M = − J ∑ J Ω J M L ( ϑ , φ ) Ω J M L + ( ϑ ′ , φ ′ ) = I δ b ( cos ϑ − cos ϑ ′ ) δ ( φ − φ ′ ) , (1440)
where I ^ \widehat{I} I is the unit 2 × 2 2\times2 2 × 2 matrix. Any spinor χ ( ϑ , φ ) \chi(\vartheta,\varphi) χ ( ϑ , φ ) which satisfies
∫ χ + ( ϑ , φ ) χ ( ϑ , φ ) d Ω < ∞ , (1441) \int\chi^{+}(\vartheta,\varphi)\chi(\vartheta,\varphi)d\Omega<\infty, \tag*{(1441)} ∫ χ + ( ϑ , φ ) χ ( ϑ , φ ) d Ω < ∞ , (1441)
may be expanded in a series of spinor spherical harmonics
χ ( ϑ , φ ) = ∑ J L M A J M L Ω J M L ( ϑ , φ ) , (1442) \chi(\vartheta,\varphi)=\sum_{JLM}A^{L}_{JM}\Omega^{L}_{JM}(\vartheta,\varphi), \tag*{(1442)} χ ( ϑ , φ ) = J L M ∑ A J M L Ω J M L ( ϑ , φ ) , (1442)
where the expansion coefficients A J M L A^{L}_{JM} A J M L are given by
A J M L = ∫ 0 π d ϑ sin ϑ ∫ 0 2 π d φ Ω J M L + ( ϑ , φ ) χ ( ϑ , φ ) . (1443) A^{L}_{JM}=\int_{0}^{\pi}d\vartheta\sin\vartheta\int_{0}^{2\pi}d\varphi\,\Omega^{L+}_{JM}(\vartheta,\varphi)\chi(\vartheta,\varphi). \tag*{(1443)} A J M L = ∫ 0 π d ϑ sin ϑ ∫ 0 2 π d φ Ω J M L + ( ϑ , φ ) χ ( ϑ , φ ) . (1443)
The expansion (7.2.38) is valid for the range 0 ≤ ϑ ≤ π 0\leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0\leq\varphi<2\pi 0 ≤ φ < 2 π .
Clebsch-Gordan Series
Ω J 1 M 1 L 1 + ( ϑ , φ ) Ω J 2 M 2 L 2 ( ϑ , φ ) = ∑ L ( − 1 ) J 1 + M 1 + J 2 + L + 1 2 { L 1 L 2 L J 2 J 1 1 2 } ( 2 J 1 + 1 ) ( 2 J 2 + 1 ) ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) 4 π ( 2 L + 1 ) C L 1 0 L 2 0 L 0 C J 1 − M 1 J 2 M 2 J M Y L M ( ϑ , φ ) (1444) \Omega^{L_{1}+}_{J_{1}M_{1}}(\vartheta,\varphi)\Omega^{L_{2}}_{J_{2}M_{2}}(\vartheta,\varphi)=\sum_{L}(-1)^{J_{1}+M_{1}+J_{2}+L+\frac{1}{2}}\begin{Bmatrix}L_{1}&L_{2}&L\\J_{2}&J_{1}&\frac{1}{2}\end{Bmatrix}\sqrt{\frac{(2J_{1}+1)(2J_{2}+1)(2L_{1}+1)(2L_{2}+1)}{4\pi(2L+1)}}C^{L0}_{L_{1}0\,L_{2}0}C^{JM}_{J_{1}-M_{1}\,J_{2}M_{2}}Y_{LM}(\vartheta,\varphi) \tag*{(1444)} Ω J 1 M 1 L 1 + ( ϑ , φ ) Ω J 2 M 2 L 2 ( ϑ , φ ) = L ∑ ( − 1 ) J 1 + M 1 + J 2 + L + 2 1 { L 1 J 2 L 2 J 1 L 2 1 } 4 π ( 2 L + 1 ) ( 2 J 1 + 1 ) ( 2 J 2 + 1 ) ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) C L 1 0 L 2 0 L 0 C J 1 − M 1 J 2 M 2 J M Y L M ( ϑ , φ ) (1444)
Ω J 1 M 1 L 1 + ( ϑ , φ ) S ^ Ω J 2 M 2 L 2 ( ϑ , φ ) = ( − 1 ) J 2 + L 1 + M 1 3 ( 2 J 1 + 1 ) ( 2 J 2 + 1 ) ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) 8 π ∑ J L ( − 1 ) J C L 1 0 L 2 0 L 0 { L 1 J 1 1 2 L 2 J 2 1 2 L J 1 } C J 1 − M 1 J 2 M 2 J M Y J M L ( ϑ , φ ) , (1445) \Omega^{L_{1}+}_{J_{1}M_{1}}(\vartheta,\varphi)\widehat{S}\Omega^{L_{2}}_{J_{2}M_{2}}(\vartheta,\varphi)=(-1)^{J_{2}+L_{1}+M_{1}}\sqrt{\frac{3(2J_{1}+1)(2J_{2}+1)(2L_{1}+1)(2L_{2}+1)}{8\pi}}\sum_{JL}(-1)^{J}C^{L0}_{L_{1}0\,L_{2}0}\begin{Bmatrix}L_{1}&J_{1}&\frac{1}{2}\\L_{2}&J_{2}&\frac{1}{2}\\L&J&1\end{Bmatrix}C^{JM}_{J_{1}-M_{1}\,J_{2}M_{2}}Y^{L}_{JM}(\vartheta,\varphi), \tag*{(1445)} Ω J 1 M 1 L 1 + ( ϑ , φ ) S Ω J 2 M 2 L 2 ( ϑ , φ ) = ( − 1 ) J 2 + L 1 + M 1 8 π 3 ( 2 J 1 + 1 ) ( 2 J 2 + 1 ) ( 2 L 1 + 1 ) ( 2 L 2 + 1 ) J L ∑ ( − 1 ) J C L 1 0 L 2 0 L 0 ⎩ ⎨ ⎧ L 1 L 2 L J 1 J 2 J 2 1 2 1 1 ⎭ ⎬ ⎫ C J 1 − M 1 J 2 M 2 J M Y J M L ( ϑ , φ ) , (1445)
where S ^ \widehat{S} S is the spin operator, Y J M L Y^{L}_{JM} Y J M L is a vector spherical harmonic (see § 27.4 ).
Addition Theorems
Let n 1 \mathbf{n}_{1} n 1 and n 2 \mathbf{n}_{2} n 2 be unit vectors with polar angles ϑ 1 , φ 1 \vartheta_{1},\varphi_{1} ϑ 1 , φ 1 and ϑ 2 , φ 2 \vartheta_{2},\varphi_{2} ϑ 2 , φ 2 , cos ω 12 ≡ n 1 ⋅ n 2 = cos ϑ 1 cos ϑ 2 + sin ϑ 1 sin ϑ 2 cos ( φ 1 − φ 2 ) \cos\omega_{12}\equiv\mathbf{n}_{1}\cdot\mathbf{n}_{2}=\cos\vartheta_{1}\cos\vartheta_{2}+\sin\vartheta_{1}\sin\vartheta_{2}\cos(\varphi_{1}-\varphi_{2}) cos ω 12 ≡ n 1 ⋅ n 2 = cos ϑ 1 cos ϑ 2 + sin ϑ 1 sin ϑ 2 cos ( φ 1 − φ 2 ) . The addition theorem for the spinor spherical harmonics may be written in the form
4 π ∑ M = − J J Ω J M L + ( ϑ 1 , φ 1 ) Ω J M L ′ ( ϑ 2 , φ 2 ) = δ L L ′ ( 2 J + 1 ) P L ( cos ω 12 ) , (1446) 4\pi\sum_{M=-J}^{J}\Omega^{L+}_{JM}(\vartheta_{1},\varphi_{1})\Omega^{L'}_{JM}(\vartheta_{2},\varphi_{2})=\delta_{LL'}(2J+1)P_{L}(\cos\omega_{12}), \tag*{(1446)} 4 π M = − J ∑ J Ω J M L + ( ϑ 1 , φ 1 ) Ω J M L ′ ( ϑ 2 , φ 2 ) = δ L L ′ ( 2 J + 1 ) P L ( cos ω 12 ) , (1446)
4 π ∑ M = − J J Ω J M L + ( ϑ 1 , φ 1 ) S ^ Ω J M L ′ ( ϑ 2 , φ 2 ) = { i ∣ n 1 × n 2 ∣ P L ′ ( cos ω 12 ) , if L ′ = L , ( L ′ − L ) { n 1 P L ′ ′ ( cos ω 12 ) − n 2 P L ′ ′ ( cos ω 12 ) } , if L ′ = 2 J − L . (1447) 4\pi\sum_{M=-J}^{J}\Omega^{L+}_{JM}(\vartheta_{1},\varphi_{1})\widehat{S}\Omega^{L'}_{JM}(\vartheta_{2},\varphi_{2})=
\begin{cases}
i\lvert\mathbf{n}_{1}\times\mathbf{n}_{2}\rvert P'_{L}(\cos\omega_{12}),&\text{if }L'=L,\\
(L'-L)\{\mathbf{n}_{1}P'_{L'}(\cos\omega_{12})-\mathbf{n}_{2}P'_{L'}(\cos\omega_{12})\},&\text{if }L'=2J-L.
\end{cases} \tag*{(1447)} 4 π M = − J ∑ J Ω J M L + ( ϑ 1 , φ 1 ) S Ω J M L ′ ( ϑ 2 , φ 2 ) = { i ∣ n 1 × n 2 ∣ P L ′ ( cos ω 12 ) , ( L ′ − L ) { n 1 P L ′ ′ ( cos ω 12 ) − n 2 P L ′ ′ ( cos ω 12 )} , if L ′ = L , if L ′ = 2 J − L . (1447)
where P L ( x ) P_L(x) P L ( x ) is a Legendre polynomial, P L ′ ( x ) = d P L ( x ) / d x P'_L(x)=dP_L(x)/dx P L ′ ( x ) = d P L ( x ) / d x , S ^ \hat{\mathbf{S}} S ^ is the spin operator (§ 10.6 ). Another form of the addition theorem may be obtained for matrix products of the type
4 π ∑ M = − J J Ω J M L ( ϑ 1 , φ 1 ) Ω J M L + ( ϑ 2 , φ 2 ) = ( J + 1 2 ) I ^ P L ( cos ω 12 ) − i 4 ( J − L ) S ^ ⋅ [ n 1 × n 2 ] P L ′ ( cos ω 12 ) , (1448) 4\pi\sum_{M=-J}^{J}\Omega^{L}_{JM}(\vartheta_1,\varphi_1)\Omega^{L+}_{JM}(\vartheta_2,\varphi_2)
=\left(J+\frac{1}{2}\right)\hat{I}P_L(\cos\omega_{12})-i4(J-L)\hat{\mathbf{S}}\cdot[\mathbf{n}_1\times\mathbf{n}_2]P'_L(\cos\omega_{12}), \tag*{(1448)} 4 π M = − J ∑ J Ω J M L ( ϑ 1 , φ 1 ) Ω J M L + ( ϑ 2 , φ 2 ) = ( J + 2 1 ) I ^ P L ( cos ω 12 ) − i 4 ( J − L ) S ^ ⋅ [ n 1 × n 2 ] P L ′ ( cos ω 12 ) , (1448)
4 π ∑ M = − J J Ω J M L ( ϑ 1 , φ 1 ) Ω J M L ′ ( ϑ 2 , φ 2 ) = 2 ( L ′ − L ) { ( S ^ ⋅ n 1 ) P L ′ ( cos ω 12 ) − ( S ^ ⋅ n 2 ) P L ′ ′ ( cos ω 12 ) } , ( L ′ = 2 J − L ) . (1449) 4\pi\sum_{M=-J}^{J}\Omega^{L}_{JM}(\vartheta_1,\varphi_1)\Omega^{L'}_{JM}(\vartheta_2,\varphi_2)
=2(L'-L)\left\{(\hat{\mathbf{S}}\cdot\mathbf{n}_1)P'_L(\cos\omega_{12})-(\hat{\mathbf{S}}\cdot\mathbf{n}_2)P'_{L'}(\cos\omega_{12})\right\},\qquad(L'=2J-L). \tag*{(1449)} 4 π M = − J ∑ J Ω J M L ( ϑ 1 , φ 1 ) Ω J M L ′ ( ϑ 2 , φ 2 ) = 2 ( L ′ − L ) { ( S ^ ⋅ n 1 ) P L ′ ( cos ω 12 ) − ( S ^ ⋅ n 2 ) P L ′ ′ ( cos ω 12 ) } , ( L ′ = 2 J − L ) . (1449)
Quadratic Forms of Spinor Spherical Harmonics
The quadratic forms Ω J M L + ( ϑ , φ ) Ω J M L ( ϑ , φ ) \Omega^{L+}_{JM}(\vartheta,\varphi)\Omega^{L}_{JM}(\vartheta,\varphi) Ω J M L + ( ϑ , φ ) Ω J M L ( ϑ , φ ) describe angular distributions of spin-1 2 \frac{1}{2} 2 1 particles in states with definite total angular momentum J J J , angular momentum projection M M M and orbital angular momentum L L L . Note that in fact these forms are independent of L L L and angle φ \varphi φ .
Let us denote
W J M ( ϑ ) ≡ Ω J M L + ( ϑ , φ ) Ω J M L ( ϑ , φ ) . (1450) W_{JM}(\vartheta)\equiv\Omega^{L+}_{JM}(\vartheta,\varphi)\Omega^{L}_{JM}(\vartheta,\varphi). \tag*{(1450)} W J M ( ϑ ) ≡ Ω J M L + ( ϑ , φ ) Ω J M L ( ϑ , φ ) . (1450)
Then one obtains
W J M ( ϑ ) = 1 2 ( J + 1 ) { ( J + M + 1 ) ∣ Y J + 1 2 , M + 1 2 ( ϑ , φ ) ∣ 2 + ( J − M + 1 ) ∣ Y J + 1 2 , M − 1 2 ( ϑ , φ ) ∣ 2 } = 1 2 J { ( J + M ) ∣ Y J − 1 2 , M − 1 2 ( ϑ , φ ) ∣ 2 + ( J − M ) ∣ Y J − 1 2 , M + 1 2 ( ϑ , φ ) ∣ 2 } . (1451) \begin{aligned}
W_{JM}(\vartheta)&=\frac{1}{2(J+1)}\left\{(J+M+1)\left|Y_{J+\frac{1}{2},M+\frac{1}{2}}(\vartheta,\varphi)\right|^2+(J-M+1)\left|Y_{J+\frac{1}{2},M-\frac{1}{2}}(\vartheta,\varphi)\right|^2\right\}\\
&=\frac{1}{2J}\left\{(J+M)\left|Y_{J-\frac{1}{2},M-\frac{1}{2}}(\vartheta,\varphi)\right|^2+(J-M)\left|Y_{J-\frac{1}{2},M+\frac{1}{2}}(\vartheta,\varphi)\right|^2\right\}. \tag*{(1451)}
\end{aligned} W J M ( ϑ ) = 2 ( J + 1 ) 1 { ( J + M + 1 ) Y J + 2 1 , M + 2 1 ( ϑ , φ ) 2 + ( J − M + 1 ) Y J + 2 1 , M − 2 1 ( ϑ , φ ) 2 } = 2 J 1 { ( J + M ) Y J − 2 1 , M − 2 1 ( ϑ , φ ) 2 + ( J − M ) Y J − 2 1 , M + 2 1 ( ϑ , φ ) 2 } . (1451)
An expansion of W J M ( ϑ ) W_{JM}(\vartheta) W J M ( ϑ ) in terms of the Legendre polynomials may be written as
W J M ( ϑ ) = ∑ n = 0 J − 1 2 a n ( J , M ) P 2 n ( cos ϑ ) , (1452) W_{JM}(\vartheta)=\sum_{n=0}^{J-\frac{1}{2}}a_n(J,M)P_{2n}(\cos\vartheta), \tag*{(1452)} W J M ( ϑ ) = n = 0 ∑ J − 2 1 a n ( J , M ) P 2 n ( cos ϑ ) , (1452)
where
a n ( J , M ) = − 2 J ( 2 J + 1 ) 4 π ( 4 n + 1 ) { J J 2 n J − 1 2 J − 1 2 1 2 } C J − 1 2 0 2 n 0 J M C J − 1 2 0 2 n 0 J M = ( − 1 ) n ( 4 n + 1 ) ( 2 J + 2 n + 1 ) 4 π 2 J + 1 ( J − 1 2 + n ) ! ( 2 n ) ! ( J − 1 2 − n ) ! ( n ! ) 2 ( 2 J − 2 n ) ! ( 2 J + 2 n + 1 ) ! C J − 1 2 0 2 n 0 J M . (1453) \begin{aligned}
a_n(J,M)&=-\frac{\sqrt{2J(2J+1)}}{4\pi}(4n+1)
\begin{Bmatrix}J&J&2n\\J-\frac{1}{2}&J-\frac{1}{2}&\frac{1}{2}\end{Bmatrix}
C^{J\,M}_{J-\frac{1}{2}\,0\,2n\,0}C^{J\,M}_{J-\frac{1}{2}\,0\,2n\,0}\\
&=(-1)^n\frac{(4n+1)(2J+2n+1)}{4\pi\sqrt{2J+1}}\frac{(J-\frac{1}{2}+n)!(2n)!}{(J-\frac{1}{2}-n)!(n!)^2}\sqrt{\frac{(2J-2n)!}{(2J+2n+1)!}}C^{J\,M}_{J-\frac{1}{2}\,0\,2n\,0}. \tag*{(1453)}
\end{aligned} a n ( J , M ) = − 4 π 2 J ( 2 J + 1 ) ( 4 n + 1 ) { J J − 2 1 J J − 2 1 2 n 2 1 } C J − 2 1 0 2 n 0 J M C J − 2 1 0 2 n 0 J M = ( − 1 ) n 4 π 2 J + 1 ( 4 n + 1 ) ( 2 J + 2 n + 1 ) ( J − 2 1 − n )! ( n ! ) 2 ( J − 2 1 + n )! ( 2 n )! ( 2 J + 2 n + 1 )! ( 2 J − 2 n )! C J − 2 1 0 2 n 0 J M . (1453)
In particular,
a 0 ( J , M ) = 1 4 π , a 1 ( J , M ) = 5 16 π J ( J + 1 ) − 3 M 2 J ( J + 1 ) , (1454) a_0(J,M)=\frac{1}{4\pi},\qquad a_1(J,M)=\frac{5}{16\pi}\frac{J(J+1)-3M^2}{J(J+1)}, \tag*{(1454)} a 0 ( J , M ) = 4 π 1 , a 1 ( J , M ) = 16 π 5 J ( J + 1 ) J ( J + 1 ) − 3 M 2 , (1454)
a 2 ( J , M ) = 27 256 π 3 ( J 2 + 2 J − 5 M 2 ) ( J 2 − 5 M 2 − 1 ) − 10 M 2 ( 4 M 2 − 1 ) ( J − 1 ) J ( J + 1 ) ( J + 2 ) . a_2(J,M)=\frac{27}{256\pi}\frac{3(J^2+2J-5M^2)(J^2-5M^2-1)-10M^2(4M^2-1)}{(J-1)J(J+1)(J+2)}. a 2 ( J , M ) = 256 π 27 ( J − 1 ) J ( J + 1 ) ( J + 2 ) 3 ( J 2 + 2 J − 5 M 2 ) ( J 2 − 5 M 2 − 1 ) − 10 M 2 ( 4 M 2 − 1 ) .
The functions W J M ( ϑ ) W_{JM}(\vartheta) W J M ( ϑ ) are normalized according to
∫ W J M ( ϑ ) d Ω = 1 , (1455) \int W_{JM}(\vartheta)d\Omega=1, \tag*{(1455)} ∫ W J M ( ϑ ) d Ω = 1 , (1455)
∑ M = − J J W J M ( ϑ ) = 2 J + 1 4 π . \sum_{M=-J}^{J}W_{JM}(\vartheta)=\frac{2J+1}{4\pi}. M = − J ∑ J W J M ( ϑ ) = 4 π 2 J + 1 .
The symmetries of W J M ( ϑ ) W_{JM}(\vartheta) W J M ( ϑ ) are
W J M ( ϑ ) = W J − M ( ϑ ) = W J M ( π − ϑ ) = W J − M ( π − ϑ ) . (1456) W_{JM}(\vartheta)=W_{J-M}(\vartheta)=W_{JM}(\pi-\vartheta)=W_{J-M}(\pi-\vartheta). \tag*{(1456)} W J M ( ϑ ) = W J − M ( ϑ ) = W J M ( π − ϑ ) = W J − M ( π − ϑ ) . (1456)
For ϑ = 0 \vartheta=0 ϑ = 0 and ϑ = π \vartheta=\pi ϑ = π we have
W J M ( 0 ) = W J M ( π ) = { 2 J + 1 8 π , if M = ± 1 2 , 0 , otherwise . (1457) W_{JM}(0)=W_{JM}(\pi)=
\begin{cases}
\dfrac{2J+1}{8\pi},&\text{if }M=\pm\frac{1}{2},\\
0,&\text{otherwise}.
\end{cases} \tag*{(1457)} W J M ( 0 ) = W J M ( π ) = ⎩ ⎨ ⎧ 8 π 2 J + 1 , 0 , if M = ± 2 1 , otherwise . (1457)
For special M M M values one obtains
W J ± 1 2 ( ϑ ) = 1 2 π ( 2 J + 1 ) { sin 2 ϑ [ P J − 1 2 ′ ( cos ϑ ) ] 2 + ( J + 1 2 ) 2 [ P J − 1 2 ( cos ϑ ) ] 2 } (1458) W_{J\pm\frac{1}{2}}(\vartheta)=\frac{1}{2\pi(2J+1)}\left\{\sin^2\vartheta\left[P'_{J-\frac{1}{2}}(\cos\vartheta)\right]^2+\left(J+\frac{1}{2}\right)^2\left[P_{J-\frac{1}{2}}(\cos\vartheta)\right]^2\right\} \tag*{(1458)} W J ± 2 1 ( ϑ ) = 2 π ( 2 J + 1 ) 1 { sin 2 ϑ [ P J − 2 1 ′ ( cos ϑ ) ] 2 + ( J + 2 1 ) 2 [ P J − 2 1 ( cos ϑ ) ] 2 } (1458)
W J ± ( J − 1 ) ( ϑ ) = ( 2 J − 1 ) ! ( sin ϑ ) 2 J − 3 π 2 2 J + 1 [ ( J − 1 2 ) ! ] 2 { 1 + 4 J ( J − 1 ) cos 2 ϑ } (1459) W_{J\pm(J-1)}(\vartheta)=\frac{(2J-1)!(\sin\vartheta)^{2J-3}}{\pi2^{2J+1}\left[\left(J-\frac{1}{2}\right)!\right]^2}\left\{1+4J(J-1)\cos^2\vartheta\right\} \tag*{(1459)} W J ± ( J − 1 ) ( ϑ ) = π 2 2 J + 1 [ ( J − 2 1 ) ! ] 2 ( 2 J − 1 )! ( sin ϑ ) 2 J − 3 { 1 + 4 J ( J − 1 ) cos 2 ϑ } (1459)
W J ± J ( ϑ ) = ( 2 J ) ! π 2 2 J + 1 [ ( J − 1 2 ) ! ] 2 sin 2 J − 1 ϑ . (1460) W_{J\pm J}(\vartheta)=\frac{(2J)!}{\pi2^{2J+1}\left[\left(J-\frac{1}{2}\right)!\right]^2}\sin^{2J-1}\vartheta. \tag*{(1460)} W J ± J ( ϑ ) = π 2 2 J + 1 [ ( J − 2 1 ) ! ] 2 ( 2 J )! sin 2 J − 1 ϑ . (1460)
The explicit forms of W J M ( ϑ ) W_{JM}(\vartheta) W J M ( ϑ ) for J = 1 2 , 3 2 , 5 2 , 7 2 , 9 2 , 11 2 J=\frac{1}{2},\frac{3}{2},\frac{5}{2},\frac{7}{2},\frac{9}{2},\frac{11}{2} J = 2 1 , 2 3 , 2 5 , 2 7 , 2 9 , 2 11 (1 2 ≤ M ≤ J \frac{1}{2}\leq M\leq J 2 1 ≤ M ≤ J ) are given in Table 7.1 . For negative M M M one may use the relation W J M ( ϑ ) = W J − M ( ϑ ) W_{JM}(\vartheta)=W_{J-M}(\vartheta) W J M ( ϑ ) = W J − M ( ϑ ) .
Table 7.1. Explicit forms of W J M ( ϑ ) W_{JM}(\vartheta) W J M ( ϑ ) .
Vector spherical harmonics
Definition
The vector spherical harmonics Y J M L ( ϑ , φ ) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi) Y J M L ( ϑ , φ ) are defined as the tensor spherical harmonics at S = 1 S=1 S = 1 (§ 27.1.1 ).
Y J M L ( ϑ , φ ) ≡ Y J M L 1 ( ϑ , φ ) . (1461) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\equiv\mathbf{Y}^{L1}_{JM}(\vartheta,\varphi). \tag*{(1461)} Y J M L ( ϑ , φ ) ≡ Y J M L 1 ( ϑ , φ ) . (1461)
They are eigenfunctions of the operators J ^ 2 \hat{\mathbf{J}}^2 J ^ 2 , J ^ z \hat{J}_z J ^ z , L ^ 2 \hat{\mathbf{L}}^2 L ^ 2 , S ^ 2 \hat{\mathbf{S}}^2 S ^ 2 , where L ^ \hat{\mathbf{L}} L ^ is the orbital angular momentum operator (§ 10.2 ), S ^ \hat{\mathbf{S}} S ^ is the spin operator for S = 1 S=1 S = 1 (§ 11 ), and J ^ = L ^ + S ^ \hat{\mathbf{J}}=\hat{\mathbf{L}}+\hat{\mathbf{S}} J ^ = L ^ + S ^ is the total angular momentum operator (§ 10.1 ).
J ^ 2 Y J M L ( ϑ , φ ) = J ( J + 1 ) Y J M L ( ϑ , φ ) , \hat{\mathbf{J}}^2\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=J(J+1)\mathbf{Y}^{L}_{JM}(\vartheta,\varphi), J ^ 2 Y J M L ( ϑ , φ ) = J ( J + 1 ) Y J M L ( ϑ , φ ) ,
J ^ z Y J M L ( ϑ , φ ) = M Y J M L ( ϑ , φ ) , \hat{J}_z\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=M\mathbf{Y}^{L}_{JM}(\vartheta,\varphi), J ^ z Y J M L ( ϑ , φ ) = M Y J M L ( ϑ , φ ) ,
L ^ 2 Y J M L ( ϑ , φ ) = L ( L + 1 ) Y J M L ( ϑ , φ ) , \hat{\mathbf{L}}^2\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=L(L+1)\mathbf{Y}^{L}_{JM}(\vartheta,\varphi), L ^ 2 Y J M L ( ϑ , φ ) = L ( L + 1 ) Y J M L ( ϑ , φ ) ,
S ^ 2 Y J M L ( ϑ , φ ) = 2 Y J M L ( ϑ , φ ) . (1462) \hat{\mathbf{S}}^2\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=2\mathbf{Y}^{L}_{JM}(\vartheta,\varphi). \tag*{(1462)} S ^ 2 Y J M L ( ϑ , φ ) = 2 Y J M L ( ϑ , φ ) . (1462)
According to (7.1.2), the vector spherical harmonics Y J M L \mathbf{Y}^{L}_{JM} Y J M L may be written as
Y J M L ( ϑ , φ ) = ∑ m , σ C L m 1 σ J M Y L m ( ϑ , φ ) e σ . (1463) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=\sum_{m,\sigma}C^{JM}_{Lm1\sigma}Y_{Lm}(\vartheta,\varphi)\mathbf{e}_{\sigma}. \tag*{(1463)} Y J M L ( ϑ , φ ) = m , σ ∑ C L m 1 σ J M Y L m ( ϑ , φ ) e σ . (1463)
where C L m 1 σ J M C^{JM}_{Lm1\sigma} C L m 1 σ J M is the Clebsch-Gordan coefficient (Chap. 8), Y L M Y_{LM} Y L M is a scalar spherical harmonic (Chap. 5), and e σ \mathbf{e}_{\sigma} e σ is a covariant spherical basis vector (spin function for an S = 1 S=1 S = 1 particle, § 9.1.3 ); J J J and L L L are nonnegative integers. For a given J J J , three values of L L L are possible: L = J , J ± 1 L=J,J\pm1 L = J , J ± 1 (with the only exception that L = 1 L=1 L = 1 for J = 0 J=0 J = 0 ), and M M M takes the values M = − J , − J + 1 , … , J − 1 , J M=-J,-J+1,\ldots,J-1,J M = − J , − J + 1 , … , J − 1 , J .
The vectors Y J M L ( ϑ , φ ) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi) Y J M L ( ϑ , φ ) which have the same J , M J,M J , M but different L L L satisfy the orthogonality relations
∑ M Y J M L ′ ( ϑ , φ ) ⋅ Y J M L ( ϑ , φ ) = 0 , if L ≠ L ′ . (1464) \sum_M\mathbf{Y}^{L'}_{JM}(\vartheta,\varphi)\cdot\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=0,\qquad\text{if }L\ne L'. \tag*{(1464)} M ∑ Y J M L ′ ( ϑ , φ ) ⋅ Y J M L ( ϑ , φ ) = 0 , if L = L ′ . (1464)
Along with Y J M L \mathbf{Y}^{L}_{JM} Y J M L , other vector spherical harmonics Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) (λ = 0 , ± 1 \lambda=0,\pm1 λ = 0 , ± 1 ) are widely used. The Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) 's (unlike Y J M L \mathbf{Y}^{L}_{JM} Y J M L ) are not eigenfunctions of the operator L ^ 2 \hat{\mathbf{L}}^2 L ^ 2 but are suitably oriented with respect to the vector n ≡ r / r \mathbf{n}\equiv\mathbf{r}/r n ≡ r / r specified by polar angles ϑ , φ \vartheta,\varphi ϑ , φ . The vector spherical harmonics Y J M ( 1 ) ( ϑ , φ ) \mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi) Y J M ( 1 ) ( ϑ , φ ) and Y J M ( 0 ) ( ϑ , φ ) \mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi) Y J M ( 0 ) ( ϑ , φ ) are transverse, and Y J M ( − 1 ) ( ϑ , φ ) \mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi) Y J M ( − 1 ) ( ϑ , φ ) is longitudinal with respect to n ( ϑ , φ ) \mathbf{n}(\vartheta,\varphi) n ( ϑ , φ ) .
n ⋅ Y J M ( 1 ) ( ϑ , φ ) = n ⋅ Y J M ( 0 ) ( ϑ , φ ) = 0 , \mathbf{n}\cdot\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)=\mathbf{n}\cdot\mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi)=0, n ⋅ Y J M ( 1 ) ( ϑ , φ ) = n ⋅ Y J M ( 0 ) ( ϑ , φ ) = 0 ,
n × Y J M ( − 1 ) ( ϑ , φ ) = 0. (1465) \mathbf{n}\times\mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi)=0. \tag*{(1465)} n × Y J M ( − 1 ) ( ϑ , φ ) = 0. (1465)
Sometimes Y J M ( 1 ) ( ϑ , φ ) \mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi) Y J M ( 1 ) ( ϑ , φ ) and Y J M ( 0 ) ( ϑ , φ ) \mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi) Y J M ( 0 ) ( ϑ , φ ) are called the electric and magnetic multipoles, respectively. They have the form
Y J M ( 1 ) ( ϑ , φ ) = 1 J ( J + 1 ) ∇ Ω Y J M ( ϑ , φ ) , \mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)=\frac{1}{\sqrt{J(J+1)}}\nabla_{\Omega}Y_{JM}(\vartheta,\varphi), Y J M ( 1 ) ( ϑ , φ ) = J ( J + 1 ) 1 ∇ Ω Y J M ( ϑ , φ ) ,
Y J M ( 0 ) ( ϑ , φ ) = − i J ( J + 1 ) ( n × ∇ Ω ) Y J M ( ϑ , φ ) = L ^ J ( J + 1 ) Y J M ( ϑ , φ ) . (1466) \mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi)=\frac{-i}{\sqrt{J(J+1)}}\left(\mathbf{n}\times\nabla_{\Omega}\right)Y_{JM}(\vartheta,\varphi)=\frac{\hat{\mathbf{L}}}{\sqrt{J(J+1)}}Y_{JM}(\vartheta,\varphi). \tag*{(1466)} Y J M ( 0 ) ( ϑ , φ ) = J ( J + 1 ) − i ( n × ∇ Ω ) Y J M ( ϑ , φ ) = J ( J + 1 ) L ^ Y J M ( ϑ , φ ) . (1466)
The longitudinal vector Y J M ( − 1 ) ( ϑ , φ ) \mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi) Y J M ( − 1 ) ( ϑ , φ ) is given by
Y J M ( − 1 ) ( ϑ , φ ) = n Y J M ( ϑ , φ ) . (1467) \mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi)=\mathbf{n}Y_{JM}(\vartheta,\varphi). \tag*{(1467)} Y J M ( − 1 ) ( ϑ , φ ) = n Y J M ( ϑ , φ ) . (1467)
In (7.3.6) ∇ Ω \nabla_{\Omega} ∇ Ω denotes the angular part of the ∇ \nabla ∇ operator (see Eqs. (103) -(104) ), and L ^ \hat{\mathbf{L}} L ^ is the orbital angular momentum operator. Three vectors Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) with different λ \lambda λ are mutually orthogonal:
Y J M ( λ ′ ) ( ϑ , φ ) ⋅ Y J M ( λ ) ( ϑ , φ ) = 0 , if λ ′ ≠ λ . (1468) \mathbf{Y}^{(\lambda')}_{JM}(\vartheta,\varphi)\cdot\mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi)=0,\qquad\text{if }\lambda'\ne\lambda. \tag*{(1468)} Y J M ( λ ′ ) ( ϑ , φ ) ⋅ Y J M ( λ ) ( ϑ , φ ) = 0 , if λ ′ = λ . (1468)
The vector spherical harmonics Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) are linear superpositions of Y J M L ( ϑ , φ ) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi) Y J M L ( ϑ , φ ) with different L L L . Thus
Y J M ( 1 ) ( ϑ , φ ) = J + 1 2 J + 1 Y J M J − 1 ( ϑ , φ ) + J 2 J + 1 Y J M J + 1 ( ϑ , φ ) , \mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)=\sqrt{\frac{J+1}{2J+1}}\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)+\sqrt{\frac{J}{2J+1}}\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi), Y J M ( 1 ) ( ϑ , φ ) = 2 J + 1 J + 1 Y J M J − 1 ( ϑ , φ ) + 2 J + 1 J Y J M J + 1 ( ϑ , φ ) ,
Y J M ( 0 ) ( ϑ , φ ) = Y J M J ( ϑ , φ ) , \mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi)=\mathbf{Y}^{J}_{JM}(\vartheta,\varphi), Y J M ( 0 ) ( ϑ , φ ) = Y J M J ( ϑ , φ ) ,
Y J M ( − 1 ) ( ϑ , φ ) = J 2 J + 1 Y J M J − 1 ( ϑ , φ ) − J + 1 2 J + 1 Y J M J + 1 ( ϑ , φ ) . (1469) \mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi)=\sqrt{\frac{J}{2J+1}}\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)-\sqrt{\frac{J+1}{2J+1}}\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi). \tag*{(1469)} Y J M ( − 1 ) ( ϑ , φ ) = 2 J + 1 J Y J M J − 1 ( ϑ , φ ) − 2 J + 1 J + 1 Y J M J + 1 ( ϑ , φ ) . (1469)
The inverse relations are written as
Y J M J + 1 ( ϑ , φ ) = J 2 J + 1 Y J M ( 1 ) ( ϑ , φ ) − J + 1 2 J + 1 Y J M ( − 1 ) ( ϑ , φ ) , \mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)=\sqrt{\frac{J}{2J+1}}\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)-\sqrt{\frac{J+1}{2J+1}}\mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi), Y J M J + 1 ( ϑ , φ ) = 2 J + 1 J Y J M ( 1 ) ( ϑ , φ ) − 2 J + 1 J + 1 Y J M ( − 1 ) ( ϑ , φ ) ,
Y J M J ( ϑ , φ ) = Y J M ( 0 ) ( ϑ , φ ) , (1470) \mathbf{Y}^{J}_{JM}(\vartheta,\varphi)=\mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi), \tag*{(1470)} Y J M J ( ϑ , φ ) = Y J M ( 0 ) ( ϑ , φ ) , (1470)
Y J M J − 1 ( ϑ , φ ) = J + 1 2 J + 1 Y J M ( 1 ) ( ϑ , φ ) + J 2 J + 1 Y J M ( − 1 ) ( ϑ , φ ) . \mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)=\sqrt{\frac{J+1}{2J+1}}\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)+\sqrt{\frac{J}{2J+1}}\mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi). Y J M J − 1 ( ϑ , φ ) = 2 J + 1 J + 1 Y J M ( 1 ) ( ϑ , φ ) + 2 J + 1 J Y J M ( − 1 ) ( ϑ , φ ) .
The vector spherical harmonics Y J M L ( ϑ , φ ) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi) Y J M L ( ϑ , φ ) , as well as Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) constitute a complete orthonormal vector set for the range 0 ≤ ϑ ≤ π 0 \leq\vartheta\leq\pi 0 ≤ ϑ ≤ π , 0 ≤ φ < 2 π 0 \leq\varphi< 2\pi 0 ≤ φ < 2 π (see Sec. 7.3.13 below).
Components of Vector Spherical Harmonics
(a) The vector spherical harmonics Y J M L ( ϑ , φ ) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi) Y J M L ( ϑ , φ ) may be expanded in terms of the spherical covariant basis vectors e μ ( μ = ± 1 , 0 ) \mathbf{e}_{\mu}(\mu=\pm1,0) e μ ( μ = ± 1 , 0 ) as
Y J M L ( ϑ , φ ) = ∑ μ [ Y J M L ( ϑ , φ ) ] μ e μ = ∑ μ ( − 1 ) μ [ Y J M L ( ϑ , φ ) ] − μ e μ . (1471) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=\sum_{\mu}\left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]^{\mu}\mathbf{e}_{\mu}=\sum_{\mu}(-1)^{\mu}\left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]_{-\mu}\mathbf{e}_{\mu}. \tag*{(1471)} Y J M L ( ϑ , φ ) = μ ∑ [ Y J M L ( ϑ , φ ) ] μ e μ = μ ∑ ( − 1 ) μ [ Y J M L ( ϑ , φ ) ] − μ e μ . (1471)
where [ Y J M L ( ϑ , φ ) ] μ \left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]^{\mu} [ Y J M L ( ϑ , φ ) ] μ and [ Y J M L ( ϑ , φ ) ] μ \left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]_{\mu} [ Y J M L ( ϑ , φ ) ] μ are contravariant and covariant components, respectively.
The contravariant spherical components of Y J M L ( ϑ , φ ) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi) Y J M L ( ϑ , φ ) are given, according to (7.1.5), by
[ Y J M L ( ϑ , φ ) ] μ = C L M − μ 1 μ J M Y L M − μ ( ϑ , φ ) . (1472) \left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]^{\mu}=C^{JM}_{L\,M-\mu\,1\,\mu}Y_{L\,M-\mu}(\vartheta,\varphi). \tag*{(1472)} [ Y J M L ( ϑ , φ ) ] μ = C L M − μ 1 μ J M Y L M − μ ( ϑ , φ ) . (1472)
The covariant spherical components of Y J M L ( ϑ , φ ) \mathbf{Y}^{L}_{JM}(\vartheta,\varphi) Y J M L ( ϑ , φ ) are related to the contravariant ones by
[ Y J M L ( ϑ , φ ) ] μ = ( − 1 ) μ [ Y J M L ( ϑ , φ ) ] − μ , (1473) \left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]_{\mu}=(-1)^{\mu}\left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]^{-\mu}, \tag*{(1473)} [ Y J M L ( ϑ , φ ) ] μ = ( − 1 ) μ [ Y J M L ( ϑ , φ ) ] − μ , (1473)
and are given by
[ Y J M L ( ϑ , φ ) ] μ = ( − 1 ) μ C L M + μ 1 − μ J M Y L M + μ ( ϑ , φ ) . (1474) \left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]_{\mu}=(-1)^{\mu}C^{JM}_{L\,M+\mu\,1\,-\mu}Y_{L\,M+\mu}(\vartheta,\varphi). \tag*{(1474)} [ Y J M L ( ϑ , φ ) ] μ = ( − 1 ) μ C L M + μ 1 − μ J M Y L M + μ ( ϑ , φ ) . (1474)
In more detailed form Eqs. (7.3.12), (7.3.14) may be written as
[ Y J M J + 1 ( ϑ , φ ) ] + 1 = − [ Y J M J + 1 ( ϑ , φ ) ] − 1 = [ ( J − M + 1 ) ( J − M + 2 ) 2 ( J + 1 ) ( 2 J + 3 ) ] 1 2 Y J + 1 M − 1 ( ϑ , φ ) , \left[\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right]^{+1}=-\left[\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right]_{-1}=\left[\frac{(J-M+1)(J-M+2)}{2(J+1)(2J+3)}\right]^{\frac{1}{2}}Y_{J+1\,M-1}(\vartheta,\varphi), [ Y J M J + 1 ( ϑ , φ ) ] + 1 = − [ Y J M J + 1 ( ϑ , φ ) ] − 1 = [ 2 ( J + 1 ) ( 2 J + 3 ) ( J − M + 1 ) ( J − M + 2 ) ] 2 1 Y J + 1 M − 1 ( ϑ , φ ) ,
[ Y J M J + 1 ( ϑ , φ ) ] 0 = [ Y J M J + 1 ( ϑ , φ ) ] 0 = − [ ( J − M + 1 ) ( J + M + 1 ) ( J + 1 ) ( 2 J + 3 ) ] 1 2 Y J + 1 M ( ϑ , φ ) , (1475) \left[\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right]^{0}=\left[\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right]_{0}=-\left[\frac{(J-M+1)(J+M+1)}{(J+1)(2J+3)}\right]^{\frac{1}{2}}Y_{J+1\,M}(\vartheta,\varphi), \tag*{(1475)} [ Y J M J + 1 ( ϑ , φ ) ] 0 = [ Y J M J + 1 ( ϑ , φ ) ] 0 = − [ ( J + 1 ) ( 2 J + 3 ) ( J − M + 1 ) ( J + M + 1 ) ] 2 1 Y J + 1 M ( ϑ , φ ) , (1475)
[ Y J M J + 1 ( ϑ , φ ) ] − 1 = − [ Y J M J + 1 ( ϑ , φ ) ] + 1 = [ ( J + M + 1 ) ( J + M + 2 ) 2 ( J + 1 ) ( 2 J + 3 ) ] 1 2 Y J + 1 M + 1 ( ϑ , φ ) , \left[\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right]^{-1}=-\left[\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right]_{+1}=\left[\frac{(J+M+1)(J+M+2)}{2(J+1)(2J+3)}\right]^{\frac{1}{2}}Y_{J+1\,M+1}(\vartheta,\varphi), [ Y J M J + 1 ( ϑ , φ ) ] − 1 = − [ Y J M J + 1 ( ϑ , φ ) ] + 1 = [ 2 ( J + 1 ) ( 2 J + 3 ) ( J + M + 1 ) ( J + M + 2 ) ] 2 1 Y J + 1 M + 1 ( ϑ , φ ) ,
[ Y J M J ( ϑ , φ ) ] + 1 = − [ Y J M J ( ϑ , φ ) ] − 1 = − [ ( J + M ) ( J − M + 1 ) 2 J ( J + 1 ) ] 1 2 Y J M − 1 ( ϑ , φ ) , \left[\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right]^{+1}=-\left[\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right]_{-1}=-\left[\frac{(J+M)(J-M+1)}{2J(J+1)}\right]^{\frac{1}{2}}Y_{J\,M-1}(\vartheta,\varphi), [ Y J M J ( ϑ , φ ) ] + 1 = − [ Y J M J ( ϑ , φ ) ] − 1 = − [ 2 J ( J + 1 ) ( J + M ) ( J − M + 1 ) ] 2 1 Y J M − 1 ( ϑ , φ ) ,
[ Y J M J ( ϑ , φ ) ] 0 = [ Y J M J ( ϑ , φ ) ] 0 = M J ( J + 1 ) Y J M ( ϑ , φ ) , (1476) \left[\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right]^{0}=\left[\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right]_{0}=\frac{M}{\sqrt{J(J+1)}}Y_{JM}(\vartheta,\varphi), \tag*{(1476)} [ Y J M J ( ϑ , φ ) ] 0 = [ Y J M J ( ϑ , φ ) ] 0 = J ( J + 1 ) M Y J M ( ϑ , φ ) , (1476)
[ Y J M J ( ϑ , φ ) ] − 1 = − [ Y J M J ( ϑ , φ ) ] + 1 = [ ( J − M ) ( J + M + 1 ) 2 J ( J + 1 ) ] 1 2 Y J M + 1 ( ϑ , φ ) , \left[\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right]^{-1}=-\left[\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right]_{+1}=\left[\frac{(J-M)(J+M+1)}{2J(J+1)}\right]^{\frac{1}{2}}Y_{J\,M+1}(\vartheta,\varphi), [ Y J M J ( ϑ , φ ) ] − 1 = − [ Y J M J ( ϑ , φ ) ] + 1 = [ 2 J ( J + 1 ) ( J − M ) ( J + M + 1 ) ] 2 1 Y J M + 1 ( ϑ , φ ) ,
[ Y J M J − 1 ( ϑ , φ ) ] + 1 = − [ Y J M J − 1 ( ϑ , φ ) ] − 1 = [ ( J + M ) ( J + M − 1 ) 2 J ( 2 J − 1 ) ] 1 2 Y J − 1 M − 1 ( ϑ , φ ) , \left[\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right]^{+1}=-\left[\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right]_{-1}=\left[\frac{(J+M)(J+M-1)}{2J(2J-1)}\right]^{\frac{1}{2}}Y_{J-1\,M-1}(\vartheta,\varphi), [ Y J M J − 1 ( ϑ , φ ) ] + 1 = − [ Y J M J − 1 ( ϑ , φ ) ] − 1 = [ 2 J ( 2 J − 1 ) ( J + M ) ( J + M − 1 ) ] 2 1 Y J − 1 M − 1 ( ϑ , φ ) ,
[ Y J M J − 1 ( ϑ , φ ) ] 0 = [ Y J M J − 1 ( ϑ , φ ) ] 0 = [ ( J − M ) ( J + M ) J ( 2 J − 1 ) ] 1 2 Y J − 1 M ( ϑ , φ ) , (1477) \left[\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right]^{0}=\left[\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right]_{0}=\left[\frac{(J-M)(J+M)}{J(2J-1)}\right]^{\frac{1}{2}}Y_{J-1\,M}(\vartheta,\varphi), \tag*{(1477)} [ Y J M J − 1 ( ϑ , φ ) ] 0 = [ Y J M J − 1 ( ϑ , φ ) ] 0 = [ J ( 2 J − 1 ) ( J − M ) ( J + M ) ] 2 1 Y J − 1 M ( ϑ , φ ) , (1477)
[ Y J M J − 1 ( ϑ , φ ) ] − 1 = − [ Y J M J − 1 ( ϑ , φ ) ] + 1 = [ ( J − M ) ( J − M − 1 ) 2 J ( 2 J − 1 ) ] 1 2 Y J − 1 M + 1 ( ϑ , φ ) . \left[\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right]^{-1}=-\left[\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right]_{+1}=\left[\frac{(J-M)(J-M-1)}{2J(2J-1)}\right]^{\frac{1}{2}}Y_{J-1\,M+1}(\vartheta,\varphi). [ Y J M J − 1 ( ϑ , φ ) ] − 1 = − [ Y J M J − 1 ( ϑ , φ ) ] + 1 = [ 2 J ( 2 J − 1 ) ( J − M ) ( J − M − 1 ) ] 2 1 Y J − 1 M + 1 ( ϑ , φ ) .
(b) An expansion of Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) in terms of the spherical covariant basis vectors e μ ( μ = ± 1 , 0 ) \mathbf{e}_{\mu}(\mu=\pm1,0) e μ ( μ = ± 1 , 0 ) has the form
Y J M ( λ ) ( ϑ , φ ) = ∑ μ [ Y J M ( λ ) ( ϑ , φ ) ] μ e μ = ∑ μ ( − 1 ) μ [ Y J M ( λ ) ( ϑ , φ ) ] − μ e μ . (1478) \mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi)=\sum_{\mu}\left[\mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi)\right]^{\mu}\mathbf{e}_{\mu}=\sum_{\mu}(-1)^{\mu}\left[\mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi)\right]_{-\mu}\mathbf{e}_{\mu}. \tag*{(1478)} Y J M ( λ ) ( ϑ , φ ) = μ ∑ [ Y J M ( λ ) ( ϑ , φ ) ] μ e μ = μ ∑ ( − 1 ) μ [ Y J M ( λ ) ( ϑ , φ ) ] − μ e μ . (1478)
The contravariant spherical components of Y J M ( λ ) ( ϑ , φ ) Y^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) may be written as
[ Y J M ( 1 ) ( ϑ , φ ) ] μ = J + 1 2 J + 1 C J − 1 M − μ 1 μ J M Y J − 1 M − μ ( ϑ , φ ) + J 2 J + 1 C J + 1 M − μ 1 μ J M Y J + 1 M − μ ( ϑ , φ ) , [ Y J M ( 0 ) ( ϑ , φ ) ] μ = C J M − μ 1 μ J M Y J M − μ ( ϑ , φ ) , [ Y J M ( − 1 ) ( ϑ , φ ) ] μ = J 2 J + 1 C J − 1 M − μ 1 μ J M Y J − 1 M − μ ( ϑ , φ ) − J + 1 2 J + 1 C J + 1 M − μ 1 μ J M Y J + 1 M − μ ( ϑ , φ ) . (1479) \begin{aligned}
\left[Y^{(1)}_{JM}(\vartheta,\varphi)\right]^\mu&= \sqrt{\frac{J+1}{2J+1}}C^{JM}_{J-1\,M-\mu\,1\,\mu}Y_{J-1\,M-\mu}(\vartheta,\varphi)+\sqrt{\frac{J}{2J+1}}C^{JM}_{J+1\,M-\mu\,1\,\mu}Y_{J+1\,M-\mu}(\vartheta,\varphi),\\
\left[Y^{(0)}_{JM}(\vartheta,\varphi)\right]^\mu&= C^{JM}_{J\,M-\mu\,1\,\mu}Y_{JM-\mu}(\vartheta,\varphi),\\
\left[Y^{(-1)}_{JM}(\vartheta,\varphi)\right]^\mu&= \sqrt{\frac{J}{2J+1}}C^{JM}_{J-1\,M-\mu\,1\,\mu}Y_{J-1\,M-\mu}(\vartheta,\varphi)-\sqrt{\frac{J+1}{2J+1}}C^{JM}_{J+1\,M-\mu\,1\,\mu}Y_{J+1\,M-\mu}(\vartheta,\varphi). \tag*{(1479)}
\end{aligned} [ Y J M ( 1 ) ( ϑ , φ ) ] μ [ Y J M ( 0 ) ( ϑ , φ ) ] μ [ Y J M ( − 1 ) ( ϑ , φ ) ] μ = 2 J + 1 J + 1 C J − 1 M − μ 1 μ J M Y J − 1 M − μ ( ϑ , φ ) + 2 J + 1 J C J + 1 M − μ 1 μ J M Y J + 1 M − μ ( ϑ , φ ) , = C J M − μ 1 μ J M Y J M − μ ( ϑ , φ ) , = 2 J + 1 J C J − 1 M − μ 1 μ J M Y J − 1 M − μ ( ϑ , φ ) − 2 J + 1 J + 1 C J + 1 M − μ 1 μ J M Y J + 1 M − μ ( ϑ , φ ) . (1479)
The covariant spherical components of Y J M ( λ ) ( ϑ , φ ) Y^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) are related to contravariant ones by
[ Y J M ( λ ) ( ϑ , φ ) ] μ = ( − 1 ) μ [ Y J M ( λ ) ( ϑ , φ ) ] − μ , (1480) \left[Y^{(\lambda)}_{JM}(\vartheta,\varphi)\right]_\mu=(-1)^\mu\left[Y^{(\lambda)}_{JM}(\vartheta,\varphi)\right]^{-\mu}, \tag*{(1480)} [ Y J M ( λ ) ( ϑ , φ ) ] μ = ( − 1 ) μ [ Y J M ( λ ) ( ϑ , φ ) ] − μ , (1480)
and may be written as
[ Y J M ( + 1 ) ( ϑ , φ ) ] + 1 = − [ Y J M ( + 1 ) ( ϑ , φ ) ] − 1 = [ ( J + M ) ( J + M − 1 ) ( J + 1 ) 2 J ( 2 J − 1 ) ( 2 J + 1 ) ] 1 / 2 Y J − 1 M − 1 ( ϑ , φ ) + [ ( J − M + 1 ) ( J − M + 2 ) J 2 ( J + 1 ) ( 2 J + 1 ) ( 2 J + 3 ) ] 1 / 2 Y J + 1 M − 1 ( ϑ , φ ) , [ Y J M ( + 1 ) ( ϑ , φ ) ] 0 = [ Y J M ( + 1 ) ( ϑ , φ ) ] 0 = [ ( J − M ) ( J + M ) ( J + 1 ) J ( 2 J − 1 ) ( 2 J + 1 ) ] 1 / 2 Y J − 1 M ( ϑ , φ ) − [ ( J − M + 1 ) ( J + M + 1 ) J ( J + 1 ) ( 2 J + 1 ) ( 2 J + 3 ) ] 1 / 2 Y J + 1 M ( ϑ , φ ) , [ Y J M ( + 1 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( + 1 ) ( ϑ , φ ) ] + 1 = [ ( J − M ) ( J − M − 1 ) ( J + 1 ) 2 J ( 2 J − 1 ) ( 2 J + 1 ) ] 1 / 2 Y J − 1 M + 1 ( ϑ , φ ) + [ ( J + M + 1 ) ( J + M + 2 ) J 2 ( J + 1 ) ( 2 J + 1 ) ( 2 J + 3 ) ] 1 / 2 Y J + 1 M + 1 ( ϑ , φ ) . (1481) \begin{aligned}
\left[Y^{(+1)}_{JM}(\vartheta,\varphi)\right]^{+1}&=-\left[Y^{(+1)}_{JM}(\vartheta,\varphi)\right]_{-1}\\
&=\left[\frac{(J+M)(J+M-1)(J+1)}{2J(2J-1)(2J+1)}\right]^{1/2}Y_{J-1\,M-1}(\vartheta,\varphi)\\
&\quad+\left[\frac{(J-M+1)(J-M+2)J}{2(J+1)(2J+1)(2J+3)}\right]^{1/2}Y_{J+1\,M-1}(\vartheta,\varphi),\\
\left[Y^{(+1)}_{JM}(\vartheta,\varphi)\right]^0&=\left[Y^{(+1)}_{JM}(\vartheta,\varphi)\right]_0\\
&=\left[\frac{(J-M)(J+M)(J+1)}{J(2J-1)(2J+1)}\right]^{1/2}Y_{J-1\,M}(\vartheta,\varphi)\\
&\quad-\left[\frac{(J-M+1)(J+M+1)J}{(J+1)(2J+1)(2J+3)}\right]^{1/2}Y_{J+1\,M}(\vartheta,\varphi),\\
\left[Y^{(+1)}_{JM}(\vartheta,\varphi)\right]^{-1}&=-\left[Y^{(+1)}_{JM}(\vartheta,\varphi)\right]_{+1}\\
&=\left[\frac{(J-M)(J-M-1)(J+1)}{2J(2J-1)(2J+1)}\right]^{1/2}Y_{J-1\,M+1}(\vartheta,\varphi)\\
&\quad+\left[\frac{(J+M+1)(J+M+2)J}{2(J+1)(2J+1)(2J+3)}\right]^{1/2}Y_{J+1\,M+1}(\vartheta,\varphi). \tag*{(1481)}
\end{aligned} [ Y J M ( + 1 ) ( ϑ , φ ) ] + 1 [ Y J M ( + 1 ) ( ϑ , φ ) ] 0 [ Y J M ( + 1 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( + 1 ) ( ϑ , φ ) ] − 1 = [ 2 J ( 2 J − 1 ) ( 2 J + 1 ) ( J + M ) ( J + M − 1 ) ( J + 1 ) ] 1/2 Y J − 1 M − 1 ( ϑ , φ ) + [ 2 ( J + 1 ) ( 2 J + 1 ) ( 2 J + 3 ) ( J − M + 1 ) ( J − M + 2 ) J ] 1/2 Y J + 1 M − 1 ( ϑ , φ ) , = [ Y J M ( + 1 ) ( ϑ , φ ) ] 0 = [ J ( 2 J − 1 ) ( 2 J + 1 ) ( J − M ) ( J + M ) ( J + 1 ) ] 1/2 Y J − 1 M ( ϑ , φ ) − [ ( J + 1 ) ( 2 J + 1 ) ( 2 J + 3 ) ( J − M + 1 ) ( J + M + 1 ) J ] 1/2 Y J + 1 M ( ϑ , φ ) , = − [ Y J M ( + 1 ) ( ϑ , φ ) ] + 1 = [ 2 J ( 2 J − 1 ) ( 2 J + 1 ) ( J − M ) ( J − M − 1 ) ( J + 1 ) ] 1/2 Y J − 1 M + 1 ( ϑ , φ ) + [ 2 ( J + 1 ) ( 2 J + 1 ) ( 2 J + 3 ) ( J + M + 1 ) ( J + M + 2 ) J ] 1/2 Y J + 1 M + 1 ( ϑ , φ ) . (1481)
[ Y J M ( 0 ) ( ϑ , φ ) ] + 1 = − [ Y J M ( 0 ) ( ϑ , φ ) ] − 1 = − [ ( J + M ) ( J − M + 1 ) 2 J ( J + 1 ) ] 1 / 2 Y J M − 1 ( ϑ , φ ) , [ Y J M ( 0 ) ( ϑ , φ ) ] 0 = [ Y J M ( 0 ) ( ϑ , φ ) ] 0 = M J ( J + 1 ) Y J M ( ϑ , φ ) , [ Y J M ( 0 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( 0 ) ( ϑ , φ ) ] + 1 = [ ( J − M ) ( J + M + 1 ) 2 J ( J + 1 ) ] 1 / 2 Y J M + 1 ( ϑ , φ ) . (1482) \begin{aligned}
\left[Y^{(0)}_{JM}(\vartheta,\varphi)\right]^{+1}&=-\left[Y^{(0)}_{JM}(\vartheta,\varphi)\right]_{-1}=-\left[\frac{(J+M)(J-M+1)}{2J(J+1)}\right]^{1/2}Y_{JM-1}(\vartheta,\varphi),\\
\left[Y^{(0)}_{JM}(\vartheta,\varphi)\right]^0&=\left[Y^{(0)}_{JM}(\vartheta,\varphi)\right]_0=\frac{M}{\sqrt{J(J+1)}}Y_{JM}(\vartheta,\varphi),\\
\left[Y^{(0)}_{JM}(\vartheta,\varphi)\right]^{-1}&=-\left[Y^{(0)}_{JM}(\vartheta,\varphi)\right]_{+1}=\left[\frac{(J-M)(J+M+1)}{2J(J+1)}\right]^{1/2}Y_{JM+1}(\vartheta,\varphi). \tag*{(1482)}
\end{aligned} [ Y J M ( 0 ) ( ϑ , φ ) ] + 1 [ Y J M ( 0 ) ( ϑ , φ ) ] 0 [ Y J M ( 0 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( 0 ) ( ϑ , φ ) ] − 1 = − [ 2 J ( J + 1 ) ( J + M ) ( J − M + 1 ) ] 1/2 Y J M − 1 ( ϑ , φ ) , = [ Y J M ( 0 ) ( ϑ , φ ) ] 0 = J ( J + 1 ) M Y J M ( ϑ , φ ) , = − [ Y J M ( 0 ) ( ϑ , φ ) ] + 1 = [ 2 J ( J + 1 ) ( J − M ) ( J + M + 1 ) ] 1/2 Y J M + 1 ( ϑ , φ ) . (1482)
[ Y J M ( − 1 ) ( ϑ , φ ) ] + 1 = − [ Y J M ( − 1 ) ( ϑ , φ ) ] − 1 = [ ( J + M ) ( J + M − 1 ) 2 ( 2 J − 1 ) ( 2 J + 1 ) ] 1 / 2 Y J − 1 M − 1 ( ϑ , φ ) − [ ( J − M + 1 ) ( J − M + 2 ) 2 ( 2 J + 1 ) ( 2 J + 3 ) ] 1 / 2 Y J + 1 M − 1 ( ϑ , φ ) , [ Y J M ( − 1 ) ( ϑ , φ ) ] 0 = [ Y J M ( − 1 ) ( ϑ , φ ) ] 0 = [ ( J − M ) ( J + M ) ( 2 J − 1 ) ( 2 J + 1 ) ] 1 / 2 Y J − 1 M ( ϑ , φ ) + [ ( J − M + 1 ) ( J + M + 1 ) ( 2 J + 1 ) ( 2 J + 3 ) ] 1 / 2 Y J + 1 M ( ϑ , φ ) , [ Y J M ( − 1 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( − 1 ) ( ϑ , φ ) ] + 1 = [ ( J − M ) ( J − M − 1 ) 2 ( 2 J − 1 ) ( 2 J + 1 ) ] 1 / 2 Y J − 1 M + 1 ( ϑ , φ ) − [ ( J + M + 1 ) ( J + M + 2 ) 2 ( 2 J + 1 ) ( 2 J + 3 ) ] 1 / 2 Y J + 1 M + 1 ( ϑ , φ ) . (1483) \begin{aligned}
\left[Y^{(-1)}_{JM}(\vartheta,\varphi)\right]^{+1}&=-\left[Y^{(-1)}_{JM}(\vartheta,\varphi)\right]_{-1}\\
&=\left[\frac{(J+M)(J+M-1)}{2(2J-1)(2J+1)}\right]^{1/2}Y_{J-1\,M-1}(\vartheta,\varphi)\\
&\quad-\left[\frac{(J-M+1)(J-M+2)}{2(2J+1)(2J+3)}\right]^{1/2}Y_{J+1\,M-1}(\vartheta,\varphi),\\
\left[Y^{(-1)}_{JM}(\vartheta,\varphi)\right]^0&=\left[Y^{(-1)}_{JM}(\vartheta,\varphi)\right]_0\\
&=\left[\frac{(J-M)(J+M)}{(2J-1)(2J+1)}\right]^{1/2}Y_{J-1\,M}(\vartheta,\varphi)\\
&\quad+\left[\frac{(J-M+1)(J+M+1)}{(2J+1)(2J+3)}\right]^{1/2}Y_{J+1\,M}(\vartheta,\varphi),\\
\left[Y^{(-1)}_{JM}(\vartheta,\varphi)\right]^{-1}&=-\left[Y^{(-1)}_{JM}(\vartheta,\varphi)\right]_{+1}\\
&=\left[\frac{(J-M)(J-M-1)}{2(2J-1)(2J+1)}\right]^{1/2}Y_{J-1\,M+1}(\vartheta,\varphi)\\
&\quad-\left[\frac{(J+M+1)(J+M+2)}{2(2J+1)(2J+3)}\right]^{1/2}Y_{J+1\,M+1}(\vartheta,\varphi). \tag*{(1483)}
\end{aligned} [ Y J M ( − 1 ) ( ϑ , φ ) ] + 1 [ Y J M ( − 1 ) ( ϑ , φ ) ] 0 [ Y J M ( − 1 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( − 1 ) ( ϑ , φ ) ] − 1 = [ 2 ( 2 J − 1 ) ( 2 J + 1 ) ( J + M ) ( J + M − 1 ) ] 1/2 Y J − 1 M − 1 ( ϑ , φ ) − [ 2 ( 2 J + 1 ) ( 2 J + 3 ) ( J − M + 1 ) ( J − M + 2 ) ] 1/2 Y J + 1 M − 1 ( ϑ , φ ) , = [ Y J M ( − 1 ) ( ϑ , φ ) ] 0 = [ ( 2 J − 1 ) ( 2 J + 1 ) ( J − M ) ( J + M ) ] 1/2 Y J − 1 M ( ϑ , φ ) + [ ( 2 J + 1 ) ( 2 J + 3 ) ( J − M + 1 ) ( J + M + 1 ) ] 1/2 Y J + 1 M ( ϑ , φ ) , = − [ Y J M ( − 1 ) ( ϑ , φ ) ] + 1 = [ 2 ( 2 J − 1 ) ( 2 J + 1 ) ( J − M ) ( J − M − 1 ) ] 1/2 Y J − 1 M + 1 ( ϑ , φ ) − [ 2 ( 2 J + 1 ) ( 2 J + 3 ) ( J + M + 1 ) ( J + M + 2 ) ] 1/2 Y J + 1 M + 1 ( ϑ , φ ) . (1483)
(c) An expansion of Y J M ( λ ) ( ϑ , φ ) Y^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) in terms of the polar basis vectors e r \mathbf e_r e r , e ϑ \mathbf e_\vartheta e ϑ and e φ \mathbf e_\varphi e φ (§ 9.1.2 ) is given by
Y J M ( λ ) ( ϑ , φ ) = [ Y J M ( λ ) ( ϑ , φ ) ] r e r + [ Y J M ( λ ) ( ϑ , φ ) ] ϑ e ϑ + [ Y J M ( λ ) ( ϑ , φ ) ] φ e φ . (1484) Y^{(\lambda)}_{JM}(\vartheta,\varphi)=\left[Y^{(\lambda)}_{JM}(\vartheta,\varphi)\right]_r\mathbf e_r+\left[Y^{(\lambda)}_{JM}(\vartheta,\varphi)\right]_\vartheta\mathbf e_\vartheta+\left[Y^{(\lambda)}_{JM}(\vartheta,\varphi)\right]_\varphi\mathbf e_\varphi. \tag*{(1484)} Y J M ( λ ) ( ϑ , φ ) = [ Y J M ( λ ) ( ϑ , φ ) ] r e r + [ Y J M ( λ ) ( ϑ , φ ) ] ϑ e ϑ + [ Y J M ( λ ) ( ϑ , φ ) ] φ e φ . (1484)
where the polar components of Y J M ( L ) ( ϑ , φ ) \mathbf{Y}_{JM}^{(L)}(\vartheta,\varphi) Y J M ( L ) ( ϑ , φ ) have the form
[ Y J M ( 1 ) ( ϑ , φ ) ] σ = 0 , \left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]_{\sigma}=0, [ Y J M ( 1 ) ( ϑ , φ ) ] σ = 0 ,
[ Y J M ( 1 ) ( ϑ , φ ) ] ϑ = 1 J ( J + 1 ) ⋅ ∂ ∂ ϑ Y J M ( ϑ , φ ) = 1 2 ( J − M ) ( J + M + 1 ) J ( J + 1 ) e − i φ Y J , M + 1 ( ϑ , φ ) − 1 2 ( J + M ) ( J − M + 1 ) J ( J + 1 ) e i φ Y J , M − 1 ( ϑ , φ ) , \left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]_{\vartheta}
=\frac{1}{\sqrt{J(J+1)}}\cdot\frac{\partial}{\partial\vartheta}Y_{JM}(\vartheta,\varphi)
=\frac{1}{2}\sqrt{\frac{(J-M)(J+M+1)}{J(J+1)}}e^{-i\varphi}Y_{J,M+1}(\vartheta,\varphi)
-\frac{1}{2}\sqrt{\frac{(J+M)(J-M+1)}{J(J+1)}}e^{i\varphi}Y_{J,M-1}(\vartheta,\varphi), [ Y J M ( 1 ) ( ϑ , φ ) ] ϑ = J ( J + 1 ) 1 ⋅ ∂ ϑ ∂ Y J M ( ϑ , φ ) = 2 1 J ( J + 1 ) ( J − M ) ( J + M + 1 ) e − i φ Y J , M + 1 ( ϑ , φ ) − 2 1 J ( J + 1 ) ( J + M ) ( J − M + 1 ) e i φ Y J , M − 1 ( ϑ , φ ) ,
[ Y J M ( 1 ) ( ϑ , φ ) ] φ = 1 J ( J + 1 ) ⋅ 1 sin ϑ ⋅ ∂ ∂ φ Y J M ( ϑ , φ ) = i M J ( J + 1 ) ⋅ 1 sin ϑ Y J M ( ϑ , φ ) . (1485) \left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]_{\varphi}
=\frac{1}{\sqrt{J(J+1)}}\cdot\frac{1}{\sin\vartheta}\cdot\frac{\partial}{\partial\varphi}Y_{JM}(\vartheta,\varphi)
=\frac{iM}{\sqrt{J(J+1)}}\cdot\frac{1}{\sin\vartheta}Y_{JM}(\vartheta,\varphi). \tag*{(1485)} [ Y J M ( 1 ) ( ϑ , φ ) ] φ = J ( J + 1 ) 1 ⋅ sin ϑ 1 ⋅ ∂ φ ∂ Y J M ( ϑ , φ ) = J ( J + 1 ) i M ⋅ sin ϑ 1 Y J M ( ϑ , φ ) . (1485)
[ Y J M ( 0 ) ( ϑ , φ ) ] r = 0 , \left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]_{r}=0, [ Y J M ( 0 ) ( ϑ , φ ) ] r = 0 ,
[ Y J M ( 0 ) ( ϑ , φ ) ] ϑ = i J ( J + 1 ) ⋅ 1 sin ϑ ⋅ ∂ ∂ φ Y J M ( ϑ , φ ) = − M J ( J + 1 ) ⋅ 1 sin ϑ Y J M ( ϑ , φ ) , \left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]_{\vartheta}
=\frac{i}{\sqrt{J(J+1)}}\cdot\frac{1}{\sin\vartheta}\cdot\frac{\partial}{\partial\varphi}Y_{JM}(\vartheta,\varphi)
=-\frac{M}{\sqrt{J(J+1)}}\cdot\frac{1}{\sin\vartheta}Y_{JM}(\vartheta,\varphi), [ Y J M ( 0 ) ( ϑ , φ ) ] ϑ = J ( J + 1 ) i ⋅ sin ϑ 1 ⋅ ∂ φ ∂ Y J M ( ϑ , φ ) = − J ( J + 1 ) M ⋅ sin ϑ 1 Y J M ( ϑ , φ ) ,
[ Y J M ( 0 ) ( ϑ , φ ) ] φ = − i J ( J + 1 ) ⋅ ∂ ∂ ϑ Y J M ( ϑ , φ ) = − i 2 ( J − M ) ( J + M + 1 ) J ( J + 1 ) e − i φ Y J , M + 1 ( ϑ , φ ) + i 2 ( J + M ) ( J − M + 1 ) J ( J + 1 ) e i φ Y J , M − 1 ( ϑ , φ ) . (1486) \left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]_{\varphi}
=-\frac{i}{\sqrt{J(J+1)}}\cdot\frac{\partial}{\partial\vartheta}Y_{JM}(\vartheta,\varphi)
=-\frac{i}{2}\sqrt{\frac{(J-M)(J+M+1)}{J(J+1)}}e^{-i\varphi}Y_{J,M+1}(\vartheta,\varphi)
+\frac{i}{2}\sqrt{\frac{(J+M)(J-M+1)}{J(J+1)}}e^{i\varphi}Y_{J,M-1}(\vartheta,\varphi). \tag*{(1486)} [ Y J M ( 0 ) ( ϑ , φ ) ] φ = − J ( J + 1 ) i ⋅ ∂ ϑ ∂ Y J M ( ϑ , φ ) = − 2 i J ( J + 1 ) ( J − M ) ( J + M + 1 ) e − i φ Y J , M + 1 ( ϑ , φ ) + 2 i J ( J + 1 ) ( J + M ) ( J − M + 1 ) e i φ Y J , M − 1 ( ϑ , φ ) . (1486)
[ Y J M ( − 1 ) ( ϑ , φ ) ] r = Y J M ( ϑ , φ ) , \left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]_{r}=Y_{JM}(\vartheta,\varphi), [ Y J M ( − 1 ) ( ϑ , φ ) ] r = Y J M ( ϑ , φ ) ,
[ Y J M ( − 1 ) ( ϑ , φ ) ] ϑ = 0 , (1487) \left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]_{\vartheta}=0, \tag*{(1487)} [ Y J M ( − 1 ) ( ϑ , φ ) ] ϑ = 0 , (1487)
[ Y J M ( − 1 ) ( ϑ , φ ) ] φ = 0. \left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]_{\varphi}=0. [ Y J M ( − 1 ) ( ϑ , φ ) ] φ = 0.
(d) An expansion of Y J M L ( ϑ , φ ) \mathbf{Y}_{JM}^{L}(\vartheta,\varphi) Y J M L ( ϑ , φ ) in terms of helicity basis vectors e ν ′ ( ϑ , φ ) \mathbf{e}_{\nu}'(\vartheta,\varphi) e ν ′ ( ϑ , φ ) (§ 9.1.4 ) may be written as
Y J M L ( ϑ , φ ) = ∑ ν [ Y J M L ( ϑ , φ ) ] ν e ν ′ ( ϑ , φ ) = ∑ ν ( − 1 ) ν [ Y J M L ( ϑ , φ ) ] − ν ′ e ν ′ ( ϑ , φ ) . (1488) \mathbf{Y}_{JM}^{L}(\vartheta,\varphi)
=\sum_{\nu}\left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]^{\nu}\mathbf{e}_{\nu}'(\vartheta,\varphi)
=\sum_{\nu}(-1)^{\nu}\left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]_{-\nu}'\mathbf{e}_{\nu}'(\vartheta,\varphi). \tag*{(1488)} Y J M L ( ϑ , φ ) = ν ∑ [ Y J M L ( ϑ , φ ) ] ν e ν ′ ( ϑ , φ ) = ν ∑ ( − 1 ) ν [ Y J M L ( ϑ , φ ) ] − ν ′ e ν ′ ( ϑ , φ ) . (1488)
where [ Y J M L ( ϑ , φ ) ] ν \left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]^{\nu} [ Y J M L ( ϑ , φ ) ] ν and [ Y J M L ( ϑ , φ ) ] ν ′ \left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]_{\nu}' [ Y J M L ( ϑ , φ ) ] ν ′ are contravariant and covariant helicity components, respectively.
∣ Y J M L ( ϑ , φ ) ∣ ν = 2 L + 1 4 π C L 01 ν J ν D ν − M J ( 0 , ϑ , φ ) . (1489) \left|\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right|^{\nu}
=\sqrt{\frac{2L+1}{4\pi}}C_{L0 1\nu}^{J\nu}D_{\nu-M}^{J}(0,\vartheta,\varphi). \tag*{(1489)} Y J M L ( ϑ , φ ) ν = 4 π 2 L + 1 C L 01 ν J ν D ν − M J ( 0 , ϑ , φ ) . (1489)
The covariant helicity components of Y J M L ( ϑ , φ ) \mathbf{Y}_{JM}^{L}(\vartheta,\varphi) Y J M L ( ϑ , φ ) are related to contravariant ones by
[ Y J M L ( ϑ , φ ) ] ν ′ = ( − 1 ) ν [ Y J M L ( ϑ , φ ) ] − ν , (1490) \left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]_{\nu}'
=(-1)^{\nu}\left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]^{-\nu}, \tag*{(1490)} [ Y J M L ( ϑ , φ ) ] ν ′ = ( − 1 ) ν [ Y J M L ( ϑ , φ ) ] − ν , (1490)
and may be written as
[ Y J M L ( ϑ , φ ) ] ν ′ = ( − 1 ) J + L + ν + 1 2 L + 1 4 π C L 01 ν J ν D ν − M J ( 0 , ϑ , φ ) . (1491) \left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]_{\nu}'
=(-1)^{J+L+\nu+1}\sqrt{\frac{2L+1}{4\pi}}C_{L0 1\nu}^{J\nu}D_{\nu-M}^{J}(0,\vartheta,\varphi). \tag*{(1491)} [ Y J M L ( ϑ , φ ) ] ν ′ = ( − 1 ) J + L + ν + 1 4 π 2 L + 1 C L 01 ν J ν D ν − M J ( 0 , ϑ , φ ) . (1491)
Explicitly we may write Eqs. (1489) –(1491) as follows:
[ Y J M J + 1 ( ϑ , φ ) ] + 1 = − [ Y J M J + 1 ( ϑ , φ ) ] − 1 ′ = J 8 π D − 1 − M J ( 0 , ϑ , φ ) , \left[\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]^{+1}
=-\left[\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]_{-1}'
=\sqrt{\frac{J}{8\pi}}D_{-1-M}^{J}(0,\vartheta,\varphi), [ Y J M J + 1 ( ϑ , φ ) ] + 1 = − [ Y J M J + 1 ( ϑ , φ ) ] − 1 ′ = 8 π J D − 1 − M J ( 0 , ϑ , φ ) ,
[ Y J M J + 1 ( ϑ , φ ) ] 0 = [ Y J M J + 1 ( ϑ , φ ) ] 0 ′ = − J + 1 4 π D 0 − M J ( 0 , ϑ , φ ) , \left[\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]^{0}
=\left[\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]_{0}'
=-\sqrt{\frac{J+1}{4\pi}}D_{0-M}^{J}(0,\vartheta,\varphi), [ Y J M J + 1 ( ϑ , φ ) ] 0 = [ Y J M J + 1 ( ϑ , φ ) ] 0 ′ = − 4 π J + 1 D 0 − M J ( 0 , ϑ , φ ) ,
[ Y J M J + 1 ( ϑ , φ ) ] − 1 = − [ Y J M J + 1 ( ϑ , φ ) ] + 1 ′ = J 8 π D 1 − M J ( 0 , ϑ , φ ) . (1492) \left[\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]^{-1}
=-\left[\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]_{+1}'
=\sqrt{\frac{J}{8\pi}}D_{1-M}^{J}(0,\vartheta,\varphi). \tag*{(1492)} [ Y J M J + 1 ( ϑ , φ ) ] − 1 = − [ Y J M J + 1 ( ϑ , φ ) ] + 1 ′ = 8 π J D 1 − M J ( 0 , ϑ , φ ) . (1492)
[ Y J M J ( ϑ , φ ) ] + 1 = − [ Y J M J ( ϑ , φ ) ] − 1 ′ = − 2 J + 1 8 π D − 1 − M J ( 0 , ϑ , φ ) , \left[\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]^{+1}
=-\left[\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]_{-1}'
=-\sqrt{\frac{2J+1}{8\pi}}D_{-1-M}^{J}(0,\vartheta,\varphi), [ Y J M J ( ϑ , φ ) ] + 1 = − [ Y J M J ( ϑ , φ ) ] − 1 ′ = − 8 π 2 J + 1 D − 1 − M J ( 0 , ϑ , φ ) ,
[ Y J M J ( ϑ , φ ) ] 0 = [ Y J M J ( ϑ , φ ) ] 0 ′ = 0 , (1493) \left[\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]^{0}
=\left[\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]_{0}'=0, \tag*{(1493)} [ Y J M J ( ϑ , φ ) ] 0 = [ Y J M J ( ϑ , φ ) ] 0 ′ = 0 , (1493)
[ Y J M J ( ϑ , φ ) ] − 1 = − [ Y J M J ( ϑ , φ ) ] + 1 ′ = 2 J + 1 8 π D 1 − M J ( 0 , ϑ , φ ) . \left[\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]^{-1}
=-\left[\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]_{+1}'
=\sqrt{\frac{2J+1}{8\pi}}D_{1-M}^{J}(0,\vartheta,\varphi). [ Y J M J ( ϑ , φ ) ] − 1 = − [ Y J M J ( ϑ , φ ) ] + 1 ′ = 8 π 2 J + 1 D 1 − M J ( 0 , ϑ , φ ) .
[ Y J M J − 1 ( ϑ , φ ) ] + 1 = − [ Y J M J − 1 ( ϑ , φ ) ] − 1 ′ = J + 1 8 π D − 1 − M J ( 0 , ϑ , φ ) , [ Y J M J − 1 ( ϑ , φ ) ] 0 = [ Y J M J − 1 ( ϑ , φ ) ] 0 ′ = J 4 π D 0 − M J ( 0 , ϑ , φ ) , [ Y J M J − 1 ( ϑ , φ ) ] − 1 = − [ Y J M J − 1 ( ϑ , φ ) ] + 1 ′ = J + 1 8 π D 1 − M J ( 0 , ϑ , φ ) . (1494) \begin{aligned}
\left[\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]^{+1} &= -\left[\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]'_{-1} = \sqrt{\frac{J+1}{8\pi}}D^{J}_{-1-M}(0,\vartheta,\varphi),\\
\left[\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]^{0} &= \left[\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]'_{0} = \sqrt{\frac{J}{4\pi}}D^{J}_{0-M}(0,\vartheta,\varphi),\\
\left[\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]^{-1} &= -\left[\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]'_{+1} = \sqrt{\frac{J+1}{8\pi}}D^{J}_{1-M}(0,\vartheta,\varphi). \tag*{(1494)}
\end{aligned} [ Y J M J − 1 ( ϑ , φ ) ] + 1 [ Y J M J − 1 ( ϑ , φ ) ] 0 [ Y J M J − 1 ( ϑ , φ ) ] − 1 = − [ Y J M J − 1 ( ϑ , φ ) ] − 1 ′ = 8 π J + 1 D − 1 − M J ( 0 , ϑ , φ ) , = [ Y J M J − 1 ( ϑ , φ ) ] 0 ′ = 4 π J D 0 − M J ( 0 , ϑ , φ ) , = − [ Y J M J − 1 ( ϑ , φ ) ] + 1 ′ = 8 π J + 1 D 1 − M J ( 0 , ϑ , φ ) . (1494)
(e) An expansion of Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}_{JM}^{(\lambda)}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) in terms of the helicity basis vectors e ν ′ ( ϑ , φ ) ( ν = ± 1 , 0 ) \mathbf{e}'_{\nu}(\vartheta,\varphi)(\nu=\pm1,0) e ν ′ ( ϑ , φ ) ( ν = ± 1 , 0 ) has the form
Y J M ( 1 ) ( ϑ , φ ) = 2 J + 1 8 π { D − 1 − M J ( 0 , ϑ , φ ) e 1 ′ ( ϑ , φ ) + D 1 − M J ( 0 , ϑ , φ ) e − 1 ′ ( ϑ , φ ) } , Y J M ( 0 ) ( ϑ , φ ) = 2 J + 1 8 π { − D − 1 − M J ( 0 , ϑ , φ ) e 1 ′ ( ϑ , φ ) + D 1 − M J ( 0 , ϑ , φ ) e − 1 ′ ( ϑ , φ ) } , Y J M ( − 1 ) ( ϑ , φ ) = 2 J + 1 4 π D 0 − M J ( 0 , ϑ , φ ) e 0 ′ ( ϑ , φ ) . (1495) \begin{aligned}
\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi) &= \sqrt{\frac{2J+1}{8\pi}}\left\{D^{J}_{-1-M}(0,\vartheta,\varphi)\mathbf{e}'_{1}(\vartheta,\varphi)+D^{J}_{1-M}(0,\vartheta,\varphi)\mathbf{e}'_{-1}(\vartheta,\varphi)\right\},\\
\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi) &= \sqrt{\frac{2J+1}{8\pi}}\left\{-D^{J}_{-1-M}(0,\vartheta,\varphi)\mathbf{e}'_{1}(\vartheta,\varphi)+D^{J}_{1-M}(0,\vartheta,\varphi)\mathbf{e}'_{-1}(\vartheta,\varphi)\right\},\\
\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi) &= \sqrt{\frac{2J+1}{4\pi}}D^{J}_{0-M}(0,\vartheta,\varphi)\mathbf{e}'_{0}(\vartheta,\varphi). \tag*{(1495)}
\end{aligned} Y J M ( 1 ) ( ϑ , φ ) Y J M ( 0 ) ( ϑ , φ ) Y J M ( − 1 ) ( ϑ , φ ) = 8 π 2 J + 1 { D − 1 − M J ( 0 , ϑ , φ ) e 1 ′ ( ϑ , φ ) + D 1 − M J ( 0 , ϑ , φ ) e − 1 ′ ( ϑ , φ ) } , = 8 π 2 J + 1 { − D − 1 − M J ( 0 , ϑ , φ ) e 1 ′ ( ϑ , φ ) + D 1 − M J ( 0 , ϑ , φ ) e − 1 ′ ( ϑ , φ ) } , = 4 π 2 J + 1 D 0 − M J ( 0 , ϑ , φ ) e 0 ′ ( ϑ , φ ) . (1495)
The contravariant and covariant helicity components, [ Y J M ( λ ) ] ν \left[\mathbf{Y}_{JM}^{(\lambda)}\right]^{\nu} [ Y J M ( λ ) ] ν and [ Y J M ( λ ) ] ν ′ \left[\mathbf{Y}_{JM}^{(\lambda)}\right]'_{\nu} [ Y J M ( λ ) ] ν ′ , are given by
[ Y J M ( 1 ) ( ϑ , φ ) ] + 1 = − [ Y J M ( 1 ) ( ϑ , φ ) ] − 1 ′ = 2 J + 1 8 π D − 1 − M J ( 0 , ϑ , φ ) , [ Y J M ( 1 ) ( ϑ , φ ) ] 0 = [ Y J M ( 1 ) ( ϑ , φ ) ] 0 ′ = 0 , [ Y J M ( 1 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( 1 ) ( ϑ , φ ) ] + 1 ′ = 2 J + 1 8 π D 1 − M J ( 0 , ϑ , φ ) . (1496) \begin{aligned}
\left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]^{+1} &= -\left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]'_{-1}=\sqrt{\frac{2J+1}{8\pi}}D^{J}_{-1-M}(0,\vartheta,\varphi),\\
\left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]^{0} &= \left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]'_{0}=0,\\
\left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]^{-1} &= -\left[\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]'_{+1}=\sqrt{\frac{2J+1}{8\pi}}D^{J}_{1-M}(0,\vartheta,\varphi). \tag*{(1496)}
\end{aligned} [ Y J M ( 1 ) ( ϑ , φ ) ] + 1 [ Y J M ( 1 ) ( ϑ , φ ) ] 0 [ Y J M ( 1 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( 1 ) ( ϑ , φ ) ] − 1 ′ = 8 π 2 J + 1 D − 1 − M J ( 0 , ϑ , φ ) , = [ Y J M ( 1 ) ( ϑ , φ ) ] 0 ′ = 0 , = − [ Y J M ( 1 ) ( ϑ , φ ) ] + 1 ′ = 8 π 2 J + 1 D 1 − M J ( 0 , ϑ , φ ) . (1496)
[ Y J M ( 0 ) ( ϑ , φ ) ] + 1 = − [ Y J M ( 0 ) ( ϑ , φ ) ] − 1 ′ = − 2 J + 1 8 π D − 1 − M J ( 0 , ϑ , φ ) , [ Y J M ( 0 ) ( ϑ , φ ) ] 0 = [ Y J M ( 0 ) ( ϑ , φ ) ] 0 ′ = 0 , [ Y J M ( 0 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( 0 ) ( ϑ , φ ) ] + 1 ′ = 2 J + 1 8 π D 1 − M J ( 0 , ϑ , φ ) . (1497) \begin{aligned}
\left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]^{+1} &= -\left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]'_{-1}=-\sqrt{\frac{2J+1}{8\pi}}D^{J}_{-1-M}(0,\vartheta,\varphi),\\
\left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]^{0} &= \left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]'_{0}=0,\\
\left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]^{-1} &= -\left[\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]'_{+1}=\sqrt{\frac{2J+1}{8\pi}}D^{J}_{1-M}(0,\vartheta,\varphi). \tag*{(1497)}
\end{aligned} [ Y J M ( 0 ) ( ϑ , φ ) ] + 1 [ Y J M ( 0 ) ( ϑ , φ ) ] 0 [ Y J M ( 0 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( 0 ) ( ϑ , φ ) ] − 1 ′ = − 8 π 2 J + 1 D − 1 − M J ( 0 , ϑ , φ ) , = [ Y J M ( 0 ) ( ϑ , φ ) ] 0 ′ = 0 , = − [ Y J M ( 0 ) ( ϑ , φ ) ] + 1 ′ = 8 π 2 J + 1 D 1 − M J ( 0 , ϑ , φ ) . (1497)
[ Y J M ( − 1 ) ( ϑ , φ ) ] + 1 = − [ Y J M ( − 1 ) ( ϑ , φ ) ] − 1 ′ = 0 , [ Y J M ( − 1 ) ( ϑ , φ ) ] 0 = [ Y J M ( − 1 ) ( ϑ , φ ) ] 0 ′ = 2 J + 1 4 π D 0 − M J ( 0 , ϑ , φ ) , [ Y J M ( − 1 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( − 1 ) ( ϑ , φ ) ] + 1 ′ = 0. (1498) \begin{aligned}
\left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]^{+1} &= -\left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]'_{-1}=0,\\
\left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]^{0} &= \left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]'_{0}=\sqrt{\frac{2J+1}{4\pi}}D^{J}_{0-M}(0,\vartheta,\varphi),\\
\left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]^{-1} &= -\left[\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]'_{+1}=0. \tag*{(1498)}
\end{aligned} [ Y J M ( − 1 ) ( ϑ , φ ) ] + 1 [ Y J M ( − 1 ) ( ϑ , φ ) ] 0 [ Y J M ( − 1 ) ( ϑ , φ ) ] − 1 = − [ Y J M ( − 1 ) ( ϑ , φ ) ] − 1 ′ = 0 , = [ Y J M ( − 1 ) ( ϑ , φ ) ] 0 ′ = 4 π 2 J + 1 D 0 − M J ( 0 , ϑ , φ ) , = − [ Y J M ( − 1 ) ( ϑ , φ ) ] + 1 ′ = 0. (1498)
Complex Conjugation
The vector spherical harmonics Y J M L ( ϑ , φ ) \mathbf{Y}_{JM}^{L}(\vartheta,\varphi) Y J M L ( ϑ , φ ) and Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}_{JM}^{(\lambda)}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) are complex vectors. Under complex conjugation they transform as follows
Y J M L ∗ ( ϑ , φ ) = ( − 1 ) J + L + M + 1 Y J − M L ( ϑ , φ ) , Y J M ( λ ) ∗ ( ϑ , φ ) = ( − 1 ) M + λ + 1 Y J − M ( λ ) ( ϑ , φ ) . (1499) \begin{aligned}
\mathbf{Y}_{JM}^{L*}(\vartheta,\varphi) &= (-1)^{J+L+M+1}\mathbf{Y}_{J-M}^{L}(\vartheta,\varphi),\\
\mathbf{Y}_{JM}^{(\lambda)*}(\vartheta,\varphi) &= (-1)^{M+\lambda+1}\mathbf{Y}_{J-M}^{(\lambda)}(\vartheta,\varphi). \tag*{(1499)}
\end{aligned} Y J M L ∗ ( ϑ , φ ) Y J M ( λ ) ∗ ( ϑ , φ ) = ( − 1 ) J + L + M + 1 Y J − M L ( ϑ , φ ) , = ( − 1 ) M + λ + 1 Y J − M ( λ ) ( ϑ , φ ) . (1499)
The complex conjugates of the spherical components of Y J M L \mathbf{Y}_{JM}^{L} Y J M L and Y J M ( λ ) \mathbf{Y}_{JM}^{(\lambda)} Y J M ( λ ) are given by
[ Y J M L ( ϑ , φ ) ] μ ∗ = [ Y J M L ∗ ( ϑ , φ ) ] μ = ( − 1 ) J + L + M + μ + 1 [ Y J − M L ( ϑ , φ ) ] − μ = ( − 1 ) J + L + M + 1 [ Y J − M L ( ϑ , φ ) ] μ . (1500) \left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]_{\mu}^{*}
=\left[\mathbf{Y}_{JM}^{L*}(\vartheta,\varphi)\right]^{\mu}
=(-1)^{J+L+M+\mu+1}\left[\mathbf{Y}_{J-M}^{L}(\vartheta,\varphi)\right]_{-\mu}
=(-1)^{J+L+M+1}\left[\mathbf{Y}_{J-M}^{L}(\vartheta,\varphi)\right]^{\mu}. \tag*{(1500)} [ Y J M L ( ϑ , φ ) ] μ ∗ = [ Y J M L ∗ ( ϑ , φ ) ] μ = ( − 1 ) J + L + M + μ + 1 [ Y J − M L ( ϑ , φ ) ] − μ = ( − 1 ) J + L + M + 1 [ Y J − M L ( ϑ , φ ) ] μ . (1500)
[ Y J M L ( ϑ , φ ) ] μ ∗ = [ Y J M L ∗ ( ϑ , φ ) ] μ = ( − 1 ) J + L + M + μ + 1 [ Y J − M L ( ϑ , φ ) ] − μ = ( − 1 ) J + L + M + 1 [ Y J − M L ( ϑ , φ ) ] μ . (1501) \left[\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right]^{\mu*}
=\left[\mathbf{Y}_{JM}^{L*}(\vartheta,\varphi)\right]_{\mu}
=(-1)^{J+L+M+\mu+1}\left[\mathbf{Y}_{J-M}^{L}(\vartheta,\varphi)\right]^{-\mu}
=(-1)^{J+L+M+1}\left[\mathbf{Y}_{J-M}^{L}(\vartheta,\varphi)\right]_{\mu}. \tag*{(1501)} [ Y J M L ( ϑ , φ ) ] μ ∗ = [ Y J M L ∗ ( ϑ , φ ) ] μ = ( − 1 ) J + L + M + μ + 1 [ Y J − M L ( ϑ , φ ) ] − μ = ( − 1 ) J + L + M + 1 [ Y J − M L ( ϑ , φ ) ] μ . (1501)
[ Y J M ( λ ) ( ϑ , φ ) ] μ ∗ = [ Y J M ( λ ) ∗ ( ϑ , φ ) ] μ = ( − 1 ) M + μ + λ + 1 [ Y J − M ( λ ) ( ϑ , φ ) ] − μ = ( − 1 ) M + λ + 1 [ Y J − M ( λ ) ( ϑ , φ ) ] μ . (1502) \left[\mathbf{Y}_{JM}^{(\lambda)}(\vartheta,\varphi)\right]_{\mu}^{*}
=\left[\mathbf{Y}_{JM}^{(\lambda)*}(\vartheta,\varphi)\right]^{\mu}
=(-1)^{M+\mu+\lambda+1}\left[\mathbf{Y}_{J-M}^{(\lambda)}(\vartheta,\varphi)\right]_{-\mu}
=(-1)^{M+\lambda+1}\left[\mathbf{Y}_{J-M}^{(\lambda)}(\vartheta,\varphi)\right]^{\mu}. \tag*{(1502)} [ Y J M ( λ ) ( ϑ , φ ) ] μ ∗ = [ Y J M ( λ ) ∗ ( ϑ , φ ) ] μ = ( − 1 ) M + μ + λ + 1 [ Y J − M ( λ ) ( ϑ , φ ) ] − μ = ( − 1 ) M + λ + 1 [ Y J − M ( λ ) ( ϑ , φ ) ] μ . (1502)
[ Y J M ( λ ) ( ϑ , φ ) ] μ ∗ = [ Y J M ( λ ) ∗ ( ϑ , φ ) ] μ = ( − 1 ) M + μ + λ + 1 [ Y J − M ( λ ) ( ϑ , φ ) ] − μ = ( − 1 ) M + λ + 1 [ Y J − M ( λ ) ( ϑ , φ ) ] μ . (1503) \left[\mathbf{Y}_{JM}^{(\lambda)}(\vartheta,\varphi)\right]^{\mu*}
=\left[\mathbf{Y}_{JM}^{(\lambda)*}(\vartheta,\varphi)\right]_{\mu}
=(-1)^{M+\mu+\lambda+1}\left[\mathbf{Y}_{J-M}^{(\lambda)}(\vartheta,\varphi)\right]^{-\mu}
=(-1)^{M+\lambda+1}\left[\mathbf{Y}_{J-M}^{(\lambda)}(\vartheta,\varphi)\right]_{\mu}. \tag*{(1503)} [ Y J M ( λ ) ( ϑ , φ ) ] μ ∗ = [ Y J M ( λ ) ∗ ( ϑ , φ ) ] μ = ( − 1 ) M + μ + λ + 1 [ Y J − M ( λ ) ( ϑ , φ ) ] − μ = ( − 1 ) M + λ + 1 [ Y J − M ( λ ) ( ϑ , φ ) ] μ . (1503)
Transformations of Coordinate Systems
(a) Coordinate inversion
P ^ r Y J M L ( ϑ , φ ) = − Y J M L ( π − ϑ , π + φ ) = ( − 1 ) L + 1 Y J M L ( ϑ , φ ) (1504) \hat{P}_{r}Y^{L}_{JM}(\vartheta,\varphi)=-Y^{L}_{JM}(\pi-\vartheta,\pi+\varphi)=(-1)^{L+1}Y^{L}_{JM}(\vartheta,\varphi) \tag*{(1504)} P ^ r Y J M L ( ϑ , φ ) = − Y J M L ( π − ϑ , π + φ ) = ( − 1 ) L + 1 Y J M L ( ϑ , φ ) (1504)
P ^ r Y J M ( λ ) ( ϑ , φ ) = − Y J M ( λ ) ( π − ϑ , π + φ ) = ( − 1 ) J + λ + 1 Y J M ( λ ) ( ϑ , φ ) . (1505) \hat{P}_{r}Y^{(\lambda)}_{JM}(\vartheta,\varphi)=-Y^{(\lambda)}_{JM}(\pi-\vartheta,\pi+\varphi)=(-1)^{J+\lambda+1}Y^{(\lambda)}_{JM}(\vartheta,\varphi). \tag*{(1505)} P ^ r Y J M ( λ ) ( ϑ , φ ) = − Y J M ( λ ) ( π − ϑ , π + φ ) = ( − 1 ) J + λ + 1 Y J M ( λ ) ( ϑ , φ ) . (1505)
(b) Rotations of the coordinate system
Under rotations specified by the Euler angles α , β , γ \alpha,\beta,\gamma α , β , γ (§ 9.9 ) the vector spherical harmonics transform according to
Y J M ′ L ( ϑ ′ , φ ′ ) = D ^ ( α , β , γ ) Y J M ′ L ( ϑ , φ ) = ∑ M D M M ′ J ( α , β , γ ) Y J M L ( ϑ , φ ) Y^{L}_{JM'}(\vartheta',\varphi')=\hat{D}(\alpha,\beta,\gamma)Y^{L}_{JM'}(\vartheta,\varphi)=\sum_{M}D^{J}_{MM'}(\alpha,\beta,\gamma)Y^{L}_{JM}(\vartheta,\varphi) Y J M ′ L ( ϑ ′ , φ ′ ) = D ^ ( α , β , γ ) Y J M ′ L ( ϑ , φ ) = M ∑ D M M ′ J ( α , β , γ ) Y J M L ( ϑ , φ )
Y J M ′ ( λ ) ( ϑ ′ , φ ′ ) = D ^ ( α , β , γ ) Y J M ′ ( λ ) ( ϑ , φ ) = ∑ M J D M M ′ J ( α , β , γ ) Y J M ( λ ) ( ϑ , φ ) . (1506) Y^{(\lambda)}_{JM'}(\vartheta',\varphi')=\hat{D}(\alpha,\beta,\gamma)Y^{(\lambda)}_{JM'}(\vartheta,\varphi)=\sum_{M}^{J}D^{J}_{MM'}(\alpha,\beta,\gamma)Y^{(\lambda)}_{JM}(\vartheta,\varphi). \tag*{(1506)} Y J M ′ ( λ ) ( ϑ ′ , φ ′ ) = D ^ ( α , β , γ ) Y J M ′ ( λ ) ( ϑ , φ ) = M ∑ J D M M ′ J ( α , β , γ ) Y J M ( λ ) ( ϑ , φ ) . (1506)
where D M M ′ J ( α , β , γ ) D^{J}_{MM'}(\alpha,\beta,\gamma) D M M ′ J ( α , β , γ ) are the Wigner D D D -functions (Chap. 4), M ′ M' M ′ is the projection of total angular momentum on the new z ′ z' z ′ -axis; ϑ , φ \vartheta,\varphi ϑ , φ and ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ are polar angles of r r r in the initial and final coordinate systems, respectively. The relations between ϑ ′ , φ ′ \vartheta',\varphi' ϑ ′ , φ ′ and ϑ , φ \vartheta,\varphi ϑ , φ are given by eqs. (1.4.2), (1.4.3).
Differential Equations
(a) The vector spherical harmonics Y J M L ( ϑ , φ ) Y^{L}_{JM}(\vartheta,\varphi) Y J M L ( ϑ , φ ) are eigenfunctions of the operator L ^ 2 \hat{L}^{2} L ^ 2 ; hence, they satisfy the equation
1 sin 2 ϑ ∂ ∂ ϑ { sin ϑ ∂ ∂ ϑ Y J M L ( ϑ , φ ) } + 1 sin 2 ϑ ∂ 2 ∂ φ 2 Y J M L ( ϑ , φ ) + L ( L + 1 ) Y J M L ( ϑ , φ ) = 0. (1507) \frac{1}{\sin^{2}\vartheta}\frac{\partial}{\partial\vartheta}\left\{\sin\vartheta\frac{\partial}{\partial\vartheta}Y^{L}_{JM}(\vartheta,\varphi)\right\}+\frac{1}{\sin^{2}\vartheta}\frac{\partial^{2}}{\partial\varphi^{2}}Y^{L}_{JM}(\vartheta,\varphi)+L(L+1)Y^{L}_{JM}(\vartheta,\varphi)=0. \tag*{(1507)} sin 2 ϑ 1 ∂ ϑ ∂ { sin ϑ ∂ ϑ ∂ Y J M L ( ϑ , φ ) } + sin 2 ϑ 1 ∂ φ 2 ∂ 2 Y J M L ( ϑ , φ ) + L ( L + 1 ) Y J M L ( ϑ , φ ) = 0. (1507)
This equation may be rewritten in a compact form as
{ Δ Ω + L ( L + 1 ) } Y J M L ( ϑ , φ ) = 0 (1508) \{\Delta_{\Omega}+L(L+1)\}Y^{L}_{JM}(\vartheta,\varphi)=0 \tag*{(1508)} { Δ Ω + L ( L + 1 )} Y J M L ( ϑ , φ ) = 0 (1508)
where Δ Ω ≡ ∇ Ω 2 \Delta_{\Omega}\equiv\nabla^{2}_{\Omega} Δ Ω ≡ ∇ Ω 2 is the angular part of the Laplace operator (see Eq. (109) ).
(b) The vector functions r L Y J M L ( ϑ , φ ) r^{L}Y^{L}_{JM}(\vartheta,\varphi) r L Y J M L ( ϑ , φ ) are solutions of the Laplace equation
Δ { r L Y J M L ( ϑ , φ ) } = 0. (1509) \Delta\{r^{L}Y^{L}_{JM}(\vartheta,\varphi)\}=0. \tag*{(1509)} Δ { r L Y J M L ( ϑ , φ )} = 0. (1509)
It follows from (7.3.48) that the components of r L Y J M L ( ϑ , φ ) r^{L}Y^{L}_{JM}(\vartheta,\varphi) r L Y J M L ( ϑ , φ ) are harmonic polynomials of degree L L L .
(c) The vectors z L ( k r ) Y J M L ( ϑ , φ ) z_{L}(kr)Y^{L}_{JM}(\vartheta,\varphi) z L ( k r ) Y J M L ( ϑ , φ ) satisfy the Helmholtz equation
( Δ + k 2 ) { z L ( k r ) Y J M L ( ϑ , φ ) } = 0 (1510) (\Delta+k^{2})\{z_{L}(kr)Y^{L}_{JM}(\vartheta,\varphi)\}=0 \tag*{(1510)} ( Δ + k 2 ) { z L ( k r ) Y J M L ( ϑ , φ )} = 0 (1510)
where
z L ( k r ) = π 2 k r Z L + 1 2 ( k r ) , (1511) z_{L}(kr)=\sqrt{\frac{\pi}{2kr}}Z_{L+\frac{1}{2}}(kr), \tag*{(1511)} z L ( k r ) = 2 k r π Z L + 2 1 ( k r ) , (1511)
and Z L + 1 2 Z_{L+\frac{1}{2}} Z L + 2 1 is any of the cylinder functions of order L + 1 2 L+\frac{1}{2} L + 2 1 .
Differential Operations
Below we present the results of the action of the ∇ \nabla ∇ operator on scalar and vector spherical harmonics; here f ( r ) f(r) f ( r ) denotes an arbitrary function of r ≡ ∣ r ∣ r\equiv\lvert r\rvert r ≡ ∣ r ∣ .
∇ { f ( r ) Y J M ( ϑ , φ ) } ≡ grad { f ( r ) Y J M ( ϑ , φ ) } = J 2 J + 1 { d d r + J + 1 r } f ( r ) Y J M J − 1 ( ϑ , φ ) − J + 1 2 J + 1 { d d r − J r } f ( r ) Y J M J + 1 ( ϑ , φ ) , (1512) \nabla\left\{f(r)\mathbf{Y}_{JM}(\vartheta,\varphi)\right\}\equiv\operatorname{grad}\left\{f(r)\mathbf{Y}_{JM}(\vartheta,\varphi)\right\}
=\sqrt{\frac{J}{2J+1}}\left\{\frac{d}{dr}+\frac{J+1}{r}\right\}f(r)\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)-\sqrt{\frac{J+1}{2J+1}}\left\{\frac{d}{dr}-\frac{J}{r}\right\}f(r)\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi), \tag*{(1512)} ∇ { f ( r ) Y J M ( ϑ , φ ) } ≡ grad { f ( r ) Y J M ( ϑ , φ ) } = 2 J + 1 J { d r d + r J + 1 } f ( r ) Y J M J − 1 ( ϑ , φ ) − 2 J + 1 J + 1 { d r d − r J } f ( r ) Y J M J + 1 ( ϑ , φ ) , (1512)
∇ { f ( r ) Y J M ( ϑ , φ ) } ≡ grad { f ( r ) Y J M ( ϑ , φ ) } = d f ( r ) d r Y J M ( − 1 ) ( ϑ , φ ) + J ( J + 1 ) 1 r f ( r ) Y J M ( 1 ) ( ϑ , φ ) , \nabla\left\{f(r)\mathbf{Y}_{JM}(\vartheta,\varphi)\right\}\equiv\operatorname{grad}\left\{f(r)\mathbf{Y}_{JM}(\vartheta,\varphi)\right\}=\frac{df(r)}{dr}\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)+\sqrt{J(J+1)}\frac{1}{r}f(r)\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi), ∇ { f ( r ) Y J M ( ϑ , φ ) } ≡ grad { f ( r ) Y J M ( ϑ , φ ) } = d r df ( r ) Y J M ( − 1 ) ( ϑ , φ ) + J ( J + 1 ) r 1 f ( r ) Y J M ( 1 ) ( ϑ , φ ) ,
( r ⋅ ∇ ) { f ( r ) Y J M L ( ϑ , φ ) } = r d d r f ( r ) Y J M L ( ϑ , φ ) , (\mathbf{r}\cdot\nabla)\left\{f(r)\mathbf{Y}_{JM}^{L}(\vartheta,\varphi)\right\}=r\frac{d}{dr}f(r)\mathbf{Y}_{JM}^{L}(\vartheta,\varphi), ( r ⋅ ∇ ) { f ( r ) Y J M L ( ϑ , φ ) } = r d r d f ( r ) Y J M L ( ϑ , φ ) ,
( r ⋅ ∇ ) { f ( r ) Y J M ( λ ) ( ϑ , φ ) } = r d d r f ( r ) Y J M ( λ ) ( ϑ , φ ) . (1513) (\mathbf{r}\cdot\nabla)\left\{f(r)\mathbf{Y}_{JM}^{(\lambda)}(\vartheta,\varphi)\right\}=r\frac{d}{dr}f(r)\mathbf{Y}_{JM}^{(\lambda)}(\vartheta,\varphi). \tag*{(1513)} ( r ⋅ ∇ ) { f ( r ) Y J M ( λ ) ( ϑ , φ ) } = r d r d f ( r ) Y J M ( λ ) ( ϑ , φ ) . (1513)
∇ ⋅ [ f ( r ) Y J M J + 1 ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M J + 1 ( ϑ , φ ) ] = − J + 1 2 J + 1 ( d d r + J + 2 r ) f ( r ) Y J M ( ϑ , φ ) , \nabla\cdot\left[f(r)\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]\equiv\operatorname{div}\left[f(r)\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]=-\sqrt{\frac{J+1}{2J+1}}\left(\frac{d}{dr}+\frac{J+2}{r}\right)f(r)Y_{JM}(\vartheta,\varphi), ∇ ⋅ [ f ( r ) Y J M J + 1 ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M J + 1 ( ϑ , φ ) ] = − 2 J + 1 J + 1 ( d r d + r J + 2 ) f ( r ) Y J M ( ϑ , φ ) ,
∇ ⋅ [ f ( r ) Y J M J ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M J ( ϑ , φ ) ] = 0 , \nabla\cdot\left[f(r)\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]\equiv\operatorname{div}\left[f(r)\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]=0, ∇ ⋅ [ f ( r ) Y J M J ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M J ( ϑ , φ ) ] = 0 ,
∇ ⋅ [ f ( r ) Y J M J − 1 ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M J − 1 ( ϑ , φ ) ] = J 2 J + 1 ( d d r − J − 1 r ) f ( r ) Y J M ( ϑ , φ ) , \nabla\cdot\left[f(r)\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]\equiv\operatorname{div}\left[f(r)\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]=\sqrt{\frac{J}{2J+1}}\left(\frac{d}{dr}-\frac{J-1}{r}\right)f(r)Y_{JM}(\vartheta,\varphi), ∇ ⋅ [ f ( r ) Y J M J − 1 ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M J − 1 ( ϑ , φ ) ] = 2 J + 1 J ( d r d − r J − 1 ) f ( r ) Y J M ( ϑ , φ ) ,
∇ ⋅ [ f ( r ) Y J M ( 1 ) ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M ( 1 ) ( ϑ , φ ) ] = − J ( J + 1 ) 1 r f ( r ) Y J M ( ϑ , φ ) , \nabla\cdot\left[f(r)\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]\equiv\operatorname{div}\left[f(r)\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]=-\sqrt{J(J+1)}\frac{1}{r}f(r)Y_{JM}(\vartheta,\varphi), ∇ ⋅ [ f ( r ) Y J M ( 1 ) ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M ( 1 ) ( ϑ , φ ) ] = − J ( J + 1 ) r 1 f ( r ) Y J M ( ϑ , φ ) ,
∇ ⋅ [ f ( r ) Y J M ( 0 ) ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M ( 0 ) ( ϑ , φ ) ] = 0 , (1514) \nabla\cdot\left[f(r)\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]\equiv\operatorname{div}\left[f(r)\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]=0, \tag*{(1514)} ∇ ⋅ [ f ( r ) Y J M ( 0 ) ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M ( 0 ) ( ϑ , φ ) ] = 0 , (1514)
∇ ⋅ [ f ( r ) Y J M ( − 1 ) ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M ( − 1 ) ( ϑ , φ ) ] = ( d d r + 2 r ) f ( r ) Y J M ( ϑ , φ ) . \nabla\cdot\left[f(r)\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]\equiv\operatorname{div}\left[f(r)\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]=\left(\frac{d}{dr}+\frac{2}{r}\right)f(r)Y_{JM}(\vartheta,\varphi). ∇ ⋅ [ f ( r ) Y J M ( − 1 ) ( ϑ , φ ) ] ≡ div [ f ( r ) Y J M ( − 1 ) ( ϑ , φ ) ] = ( d r d + r 2 ) f ( r ) Y J M ( ϑ , φ ) .
∇ × [ f ( r ) Y J M J + 1 ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M J + 1 ( ϑ , φ ) ] = i J 2 J + 1 ( d d r + J + 2 r ) f ( r ) Y J M J ( ϑ , φ ) , \nabla\times\left[f(r)\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]\equiv\operatorname{curl}\left[f(r)\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)\right]=i\sqrt{\frac{J}{2J+1}}\left(\frac{d}{dr}+\frac{J+2}{r}\right)f(r)\mathbf{Y}_{JM}^{J}(\vartheta,\varphi), ∇ × [ f ( r ) Y J M J + 1 ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M J + 1 ( ϑ , φ ) ] = i 2 J + 1 J ( d r d + r J + 2 ) f ( r ) Y J M J ( ϑ , φ ) ,
∇ × [ f ( r ) Y J M J ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M J ( ϑ , φ ) ] = i J 2 J + 1 ( d d r − J r ) f ( r ) Y J M J + 1 ( ϑ , φ ) + i J + 1 2 J + 1 ( d d r + J + 1 r ) f ( r ) Y J M J − 1 ( ϑ , φ ) , (1515) \nabla\times\left[f(r)\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]\equiv\operatorname{curl}\left[f(r)\mathbf{Y}_{JM}^{J}(\vartheta,\varphi)\right]
=i\sqrt{\frac{J}{2J+1}}\left(\frac{d}{dr}-\frac{J}{r}\right)f(r)\mathbf{Y}_{JM}^{J+1}(\vartheta,\varphi)+i\sqrt{\frac{J+1}{2J+1}}\left(\frac{d}{dr}+\frac{J+1}{r}\right)f(r)\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi), \tag*{(1515)} ∇ × [ f ( r ) Y J M J ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M J ( ϑ , φ ) ] = i 2 J + 1 J ( d r d − r J ) f ( r ) Y J M J + 1 ( ϑ , φ ) + i 2 J + 1 J + 1 ( d r d + r J + 1 ) f ( r ) Y J M J − 1 ( ϑ , φ ) , (1515)
∇ × [ f ( r ) Y J M J − 1 ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M J − 1 ( ϑ , φ ) ] = i J + 1 2 J + 1 ( d d r − J − 1 r ) f ( r ) Y J M J ( ϑ , φ ) , \nabla\times\left[f(r)\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]\equiv\operatorname{curl}\left[f(r)\mathbf{Y}_{JM}^{J-1}(\vartheta,\varphi)\right]=i\sqrt{\frac{J+1}{2J+1}}\left(\frac{d}{dr}-\frac{J-1}{r}\right)f(r)\mathbf{Y}_{JM}^{J}(\vartheta,\varphi), ∇ × [ f ( r ) Y J M J − 1 ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M J − 1 ( ϑ , φ ) ] = i 2 J + 1 J + 1 ( d r d − r J − 1 ) f ( r ) Y J M J ( ϑ , φ ) ,
∇ × [ f ( r ) Y J M ( 1 ) ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M ( 1 ) ( ϑ , φ ) ] = i ( d d r + 1 r ) f ( r ) Y J M ( 0 ) ( ϑ , φ ) , \nabla\times\left[f(r)\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]\equiv\operatorname{curl}\left[f(r)\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)\right]=i\left(\frac{d}{dr}+\frac{1}{r}\right)f(r)\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi), ∇ × [ f ( r ) Y J M ( 1 ) ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M ( 1 ) ( ϑ , φ ) ] = i ( d r d + r 1 ) f ( r ) Y J M ( 0 ) ( ϑ , φ ) ,
∇ × [ f ( r ) Y J M ( 0 ) ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M ( 0 ) ( ϑ , φ ) ] = i ( d d r + 1 r ) f ( r ) Y J M ( 1 ) ( ϑ , φ ) + i J ( J + 1 ) 1 r f ( r ) Y J M ( − 1 ) ( ϑ , φ ) , (1516) \nabla\times\left[f(r)\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]\equiv\operatorname{curl}\left[f(r)\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi)\right]
=i\left(\frac{d}{dr}+\frac{1}{r}\right)f(r)\mathbf{Y}_{JM}^{(1)}(\vartheta,\varphi)+i\sqrt{J(J+1)}\frac{1}{r}f(r)\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi), \tag*{(1516)} ∇ × [ f ( r ) Y J M ( 0 ) ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M ( 0 ) ( ϑ , φ ) ] = i ( d r d + r 1 ) f ( r ) Y J M ( 1 ) ( ϑ , φ ) + i J ( J + 1 ) r 1 f ( r ) Y J M ( − 1 ) ( ϑ , φ ) , (1516)
∇ × [ f ( r ) Y J M ( − 1 ) ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M ( − 1 ) ( ϑ , φ ) ] = − i J ( J + 1 ) 1 r f ( r ) Y J M ( 0 ) ( ϑ , φ ) . \nabla\times\left[f(r)\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]\equiv\operatorname{curl}\left[f(r)\mathbf{Y}_{JM}^{(-1)}(\vartheta,\varphi)\right]=-i\sqrt{J(J+1)}\frac{1}{r}f(r)\mathbf{Y}_{JM}^{(0)}(\vartheta,\varphi). ∇ × [ f ( r ) Y J M ( − 1 ) ( ϑ , φ ) ] ≡ curl [ f ( r ) Y J M ( − 1 ) ( ϑ , φ ) ] = − i J ( J + 1 ) r 1 f ( r ) Y J M ( 0 ) ( ϑ , φ ) .
(d) Some applications of spherical harmonics to physical problems involve the functions
F J M ( r ) ≡ z J ( k r ) Y J M ( ϑ , φ ) , F_{JM}(r)\equiv z_J(kr)Y_{JM}(\vartheta,\varphi), F J M ( r ) ≡ z J ( k r ) Y J M ( ϑ , φ ) ,
F J M ( r ) ≡ z L ( k r ) Y J M L ( ϑ , φ ) . (1517) \mathbf{F}_{JM}(r)\equiv z_L(kr)\mathbf{Y}_{JM}^{L}(\vartheta,\varphi). \tag*{(1517)} F J M ( r ) ≡ z L ( k r ) Y J M L ( ϑ , φ ) . (1517)
where z L ( k r ) ≡ π 2 k r Z L + 1 2 ( k r ) z_L(kr)\equiv\sqrt{\frac{\pi}{2kr}}Z_{L+\frac{1}{2}}(kr) z L ( k r ) ≡ 2 k r π Z L + 2 1 ( k r ) , and Z L + 1 2 Z_{L+\frac{1}{2}} Z L + 2 1 is any cylinder functions; k k k is an arbitrary parameter. For these special functions Eqs. (1512) –(1516) are reduced to the following
1 k grad { F J M J ( r ) } = J 2 J + 1 z J − 1 ( k r ) Y J M J − 1 ( ϑ , φ ) + J + 1 2 J + 1 z J + 1 ( k r ) Y J M J + 1 ( ϑ , φ ) = J 2 J + 1 F J M J − 1 ( r ) + J + 1 2 J + 1 F J M J + 1 ( r ) , (1518) \begin{aligned}
\frac{1}{k}\operatorname{grad}\left\{\mathbf{F}^{J}_{JM}(\mathbf{r})\right\} &=
\sqrt{\frac{J}{2J+1}}z_{J-1}(kr)\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)
+\sqrt{\frac{J+1}{2J+1}}z_{J+1}(kr)\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\\
&=\sqrt{\frac{J}{2J+1}}\mathbf{F}^{J-1}_{JM}(\mathbf{r})
+\sqrt{\frac{J+1}{2J+1}}\mathbf{F}^{J+1}_{JM}(\mathbf{r}), \tag*{(1518)}
\end{aligned} k 1 grad { F J M J ( r ) } = 2 J + 1 J z J − 1 ( k r ) Y J M J − 1 ( ϑ , φ ) + 2 J + 1 J + 1 z J + 1 ( k r ) Y J M J + 1 ( ϑ , φ ) = 2 J + 1 J F J M J − 1 ( r ) + 2 J + 1 J + 1 F J M J + 1 ( r ) , (1518)
1 k div { F J M J + 1 ( r ) } = − J + 1 2 J + 1 z J ( k r ) Y J M ( ϑ , φ ) = − J + 1 2 J + 1 F J M J ( r ) , \frac{1}{k}\operatorname{div}\left\{\mathbf{F}^{J+1}_{JM}(\mathbf{r})\right\}
=-\sqrt{\frac{J+1}{2J+1}}z_J(kr)Y_{JM}(\vartheta,\varphi)
=-\sqrt{\frac{J+1}{2J+1}}F^J_{JM}(\mathbf{r}), k 1 div { F J M J + 1 ( r ) } = − 2 J + 1 J + 1 z J ( k r ) Y J M ( ϑ , φ ) = − 2 J + 1 J + 1 F J M J ( r ) ,
1 k div { F J M J ( r ) } = 0 , (1519) \frac{1}{k}\operatorname{div}\left\{\mathbf{F}^{J}_{JM}(\mathbf{r})\right\}=0, \tag*{(1519)} k 1 div { F J M J ( r ) } = 0 , (1519)
1 k div { F J M J − 1 ( r ) } = − J 2 J + 1 z J ( k r ) Y J M ( ϑ , φ ) = − J 2 J + 1 F J M J ( r ) , \frac{1}{k}\operatorname{div}\left\{\mathbf{F}^{J-1}_{JM}(\mathbf{r})\right\}
=-\sqrt{\frac{J}{2J+1}}z_J(kr)Y_{JM}(\vartheta,\varphi)
=-\sqrt{\frac{J}{2J+1}}F^J_{JM}(\mathbf{r}), k 1 div { F J M J − 1 ( r ) } = − 2 J + 1 J z J ( k r ) Y J M ( ϑ , φ ) = − 2 J + 1 J F J M J ( r ) ,
1 k curl { F J M J + 1 ( r ) } = i J 2 J + 1 z J ( k r ) Y J M J ( ϑ , φ ) = i J 2 J + 1 F J M J ( r ) , 1 k curl { F J M J ( r ) } = − i J 2 J + 1 z J + 1 ( k r ) Y J M J + 1 ( ϑ , φ ) + i J + 1 2 J + 1 z J − 1 ( k r ) Y J M J − 1 ( ϑ , φ ) = − i J 2 J + 1 F J M J + 1 ( r ) + i J + 1 2 J + 1 F J M J − 1 ( r ) , 1 k curl { F J M J − 1 ( r ) } = − i J + 1 2 J + 1 z J ( k r ) Y J M J ( ϑ , φ ) = − i J + 1 2 J + 1 F J M J ( r ) . (1520) \begin{aligned}
\frac{1}{k}\operatorname{curl}\left\{\mathbf{F}^{J+1}_{JM}(\mathbf{r})\right\}
&=i\sqrt{\frac{J}{2J+1}}z_J(kr)\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)
=i\sqrt{\frac{J}{2J+1}}\mathbf{F}^{J}_{JM}(\mathbf{r}),\\
\frac{1}{k}\operatorname{curl}\left\{\mathbf{F}^{J}_{JM}(\mathbf{r})\right\}
&=-i\sqrt{\frac{J}{2J+1}}z_{J+1}(kr)\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)
+i\sqrt{\frac{J+1}{2J+1}}z_{J-1}(kr)\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\\
&=-i\sqrt{\frac{J}{2J+1}}\mathbf{F}^{J+1}_{JM}(\mathbf{r})
+i\sqrt{\frac{J+1}{2J+1}}\mathbf{F}^{J-1}_{JM}(\mathbf{r}),\\
\frac{1}{k}\operatorname{curl}\left\{\mathbf{F}^{J-1}_{JM}(\mathbf{r})\right\}
&=-i\sqrt{\frac{J+1}{2J+1}}z_J(kr)\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)
=-i\sqrt{\frac{J+1}{2J+1}}\mathbf{F}^{J}_{JM}(\mathbf{r}). \tag*{(1520)}
\end{aligned} k 1 curl { F J M J + 1 ( r ) } k 1 curl { F J M J ( r ) } k 1 curl { F J M J − 1 ( r ) } = i 2 J + 1 J z J ( k r ) Y J M J ( ϑ , φ ) = i 2 J + 1 J F J M J ( r ) , = − i 2 J + 1 J z J + 1 ( k r ) Y J M J + 1 ( ϑ , φ ) + i 2 J + 1 J + 1 z J − 1 ( k r ) Y J M J − 1 ( ϑ , φ ) = − i 2 J + 1 J F J M J + 1 ( r ) + i 2 J + 1 J + 1 F J M J − 1 ( r ) , = − i 2 J + 1 J + 1 z J ( k r ) Y J M J ( ϑ , φ ) = − i 2 J + 1 J + 1 F J M J ( r ) . (1520)
Action of Angular Momentum Operators
The action of the angular momentum operators on vector spherical harmonics is given by Eqs. (1383) and (7.1.29) for S = 1 S=1 S = 1 . In addition, we note the following relations in which S ^ \hat{\mathbf{S}} S ^ , L ^ \hat{\mathbf{L}} L ^ and J ^ = L ^ + S ^ \hat{\mathbf{J}}=\hat{\mathbf{L}}+\hat{\mathbf{S}} J ^ = L ^ + S ^ are the spin, orbital and total angular momentum operators, respectively; n \mathbf{n} n is the unit vector defined by the polar angles ϑ , φ \vartheta,\varphi ϑ , φ .
( S ^ ⋅ n ) Y J M L ( ϑ , φ ) = − 1 2 ( J + L + 3 ) ( J + L ) ( − J + L + 2 ) ( J − L + 1 ) ( 2 L + 1 ) ( 2 L + 3 ) Y J M L + 1 ( ϑ , φ ) − 1 2 ( J + L + 2 ) ( J + L − 1 ) ( − J + L + 1 ) ( J − L + 2 ) ( 2 L − 1 ) ( 2 L + 1 ) Y J M L − 1 ( ϑ , φ ) . (1521) \begin{aligned}
(\hat{\mathbf{S}}\cdot\mathbf{n})\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)
&=-\frac{1}{2}\sqrt{\frac{(J+L+3)(J+L)(-J+L+2)(J-L+1)}{(2L+1)(2L+3)}}\mathbf{Y}^{L+1}_{JM}(\vartheta,\varphi)\\
&\quad-\frac{1}{2}\sqrt{\frac{(J+L+2)(J+L-1)(-J+L+1)(J-L+2)}{(2L-1)(2L+1)}}\mathbf{Y}^{L-1}_{JM}(\vartheta,\varphi). \tag*{(1521)}
\end{aligned} ( S ^ ⋅ n ) Y J M L ( ϑ , φ ) = − 2 1 ( 2 L + 1 ) ( 2 L + 3 ) ( J + L + 3 ) ( J + L ) ( − J + L + 2 ) ( J − L + 1 ) Y J M L + 1 ( ϑ , φ ) − 2 1 ( 2 L − 1 ) ( 2 L + 1 ) ( J + L + 2 ) ( J + L − 1 ) ( − J + L + 1 ) ( J − L + 2 ) Y J M L − 1 ( ϑ , φ ) . (1521)
In particular,
( S ^ ⋅ n ) Y J M J + 1 ( ϑ , φ ) = − J 2 J + 1 Y J M J ( ϑ , φ ) , (\hat{\mathbf{S}}\cdot\mathbf{n})\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)
=-\sqrt{\frac{J}{2J+1}}\mathbf{Y}^{J}_{JM}(\vartheta,\varphi), ( S ^ ⋅ n ) Y J M J + 1 ( ϑ , φ ) = − 2 J + 1 J Y J M J ( ϑ , φ ) ,
( S ^ ⋅ n ) Y J M J ( ϑ , φ ) = − J 2 J + 1 Y J M J + 1 ( ϑ , φ ) − J + 1 2 J + 1 Y J M J − 1 ( ϑ , φ ) , (1522) (\hat{\mathbf{S}}\cdot\mathbf{n})\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)
=-\sqrt{\frac{J}{2J+1}}\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)
-\sqrt{\frac{J+1}{2J+1}}\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi), \tag*{(1522)} ( S ^ ⋅ n ) Y J M J ( ϑ , φ ) = − 2 J + 1 J Y J M J + 1 ( ϑ , φ ) − 2 J + 1 J + 1 Y J M J − 1 ( ϑ , φ ) , (1522)
( S ^ ⋅ n ) Y J M J − 1 ( ϑ , φ ) = − J + 1 2 J + 1 Y J M J ( ϑ , φ ) , (\hat{\mathbf{S}}\cdot\mathbf{n})\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)
=-\sqrt{\frac{J+1}{2J+1}}\mathbf{Y}^{J}_{JM}(\vartheta,\varphi), ( S ^ ⋅ n ) Y J M J − 1 ( ϑ , φ ) = − 2 J + 1 J + 1 Y J M J ( ϑ , φ ) ,
( S ^ ⋅ n ) Y J M ( 1 ) ( ϑ , φ ) = − Y J M ( 1 ) ( ϑ , φ ) , (\hat{\mathbf{S}}\cdot\mathbf{n})\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)
=-\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi), ( S ^ ⋅ n ) Y J M ( 1 ) ( ϑ , φ ) = − Y J M ( 1 ) ( ϑ , φ ) ,
( S ^ ⋅ n ) Y J M ( 0 ) ( ϑ , φ ) = − Y J M ( 1 ) ( ϑ , φ ) , (1523) (\hat{\mathbf{S}}\cdot\mathbf{n})\mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi)
=-\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi), \tag*{(1523)} ( S ^ ⋅ n ) Y J M ( 0 ) ( ϑ , φ ) = − Y J M ( 1 ) ( ϑ , φ ) , (1523)
( S ^ ⋅ n ) Y J M ( − 1 ) ( ϑ , φ ) = 0. (\hat{\mathbf{S}}\cdot\mathbf{n})\mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi)=0. ( S ^ ⋅ n ) Y J M ( − 1 ) ( ϑ , φ ) = 0.
r ( S ^ ⋅ ∇ ) Y J M L ( ϑ , φ ) = L 2 ( J + L + 3 ) ( J + L ) ( − J + L + 2 ) ( J − L + 1 ) ( 2 L + 1 ) ( 2 L + 3 ) Y J M L + 1 ( ϑ , φ ) − L + 1 2 ( J + L + 2 ) ( J + L − 1 ) ( − J + L + 1 ) ( J − L + 2 ) ( 2 L − 1 ) ( 2 L + 1 ) Y J M L − 1 ( ϑ , φ ) . (1524) \begin{aligned}
r(\hat{\mathbf{S}}\cdot\nabla)\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)
&=\frac{L}{2}\sqrt{\frac{(J+L+3)(J+L)(-J+L+2)(J-L+1)}{(2L+1)(2L+3)}}\mathbf{Y}^{L+1}_{JM}(\vartheta,\varphi)\\
&\quad-\frac{L+1}{2}\sqrt{\frac{(J+L+2)(J+L-1)(-J+L+1)(J-L+2)}{(2L-1)(2L+1)}}\mathbf{Y}^{L-1}_{JM}(\vartheta,\varphi). \tag*{(1524)}
\end{aligned} r ( S ^ ⋅ ∇ ) Y J M L ( ϑ , φ ) = 2 L ( 2 L + 1 ) ( 2 L + 3 ) ( J + L + 3 ) ( J + L ) ( − J + L + 2 ) ( J − L + 1 ) Y J M L + 1 ( ϑ , φ ) − 2 L + 1 ( 2 L − 1 ) ( 2 L + 1 ) ( J + L + 2 ) ( J + L − 1 ) ( − J + L + 1 ) ( J − L + 2 ) Y J M L − 1 ( ϑ , φ ) . (1524)
In particular,
r ( S ^ ⋅ ∇ ) Y J M J + 1 ( ϑ , φ ) = − ( J + 2 ) J 2 J + 1 Y J M J ( ϑ , φ ) , r ( S ^ ⋅ ∇ ) Y J M J ( ϑ , φ ) = J J 2 J + 1 Y J M J + 1 ( ϑ , φ ) − ( J + 1 ) J + 1 2 J + 1 Y J M J − 1 ( ϑ , φ ) , r ( S ^ ⋅ ∇ ) Y J M J − 1 ( ϑ , φ ) = ( J − 1 ) J + 1 2 J + 1 Y J M J ( ϑ , φ ) (1525) r(\hat{\mathbf{S}}\cdot\nabla)Y_{JM}^{J+1}(\vartheta,\varphi)=-(J+2)\sqrt{\frac{J}{2J+1}}Y_{JM}^{J}(\vartheta,\varphi),
\\
r(\hat{\mathbf{S}}\cdot\nabla)Y_{JM}^{J}(\vartheta,\varphi)=J\sqrt{\frac{J}{2J+1}}Y_{JM}^{J+1}(\vartheta,\varphi)-(J+1)\sqrt{\frac{J+1}{2J+1}}Y_{JM}^{J-1}(\vartheta,\varphi),
\\
r(\hat{\mathbf{S}}\cdot\nabla)Y_{JM}^{J-1}(\vartheta,\varphi)=(J-1)\sqrt{\frac{J+1}{2J+1}}Y_{JM}^{J}(\vartheta,\varphi) \tag*{(1525)} r ( S ^ ⋅ ∇ ) Y J M J + 1 ( ϑ , φ ) = − ( J + 2 ) 2 J + 1 J Y J M J ( ϑ , φ ) , r ( S ^ ⋅ ∇ ) Y J M J ( ϑ , φ ) = J 2 J + 1 J Y J M J + 1 ( ϑ , φ ) − ( J + 1 ) 2 J + 1 J + 1 Y J M J − 1 ( ϑ , φ ) , r ( S ^ ⋅ ∇ ) Y J M J − 1 ( ϑ , φ ) = ( J − 1 ) 2 J + 1 J + 1 Y J M J ( ϑ , φ ) (1525)
(c)
r ( S ^ ⋅ ∇ ) Y J M ( 1 ) ( ϑ , φ ) = − Y J M ( 0 ) ( ϑ , φ ) , r ( S ^ ⋅ ∇ ) Y J M ( 0 ) ( ϑ , φ ) = − Y J M ( 1 ) ( ϑ , φ ) − J ( J + 1 ) Y J M ( − 1 ) ( ϑ , φ ) , r ( S ^ ⋅ ∇ ) Y J M ( − 1 ) ( ϑ , φ ) = J ( J + 1 ) Y J M ( 0 ) ( ϑ , φ ) . (1526) r(\hat{\mathbf{S}}\cdot\nabla)Y_{JM}^{(1)}(\vartheta,\varphi)=-Y_{JM}^{(0)}(\vartheta,\varphi),
\\
r(\hat{\mathbf{S}}\cdot\nabla)Y_{JM}^{(0)}(\vartheta,\varphi)=-Y_{JM}^{(1)}(\vartheta,\varphi)-\sqrt{J(J+1)}Y_{JM}^{(-1)}(\vartheta,\varphi),
\\
r(\hat{\mathbf{S}}\cdot\nabla)Y_{JM}^{(-1)}(\vartheta,\varphi)=\sqrt{J(J+1)}Y_{JM}^{(0)}(\vartheta,\varphi). \tag*{(1526)} r ( S ^ ⋅ ∇ ) Y J M ( 1 ) ( ϑ , φ ) = − Y J M ( 0 ) ( ϑ , φ ) , r ( S ^ ⋅ ∇ ) Y J M ( 0 ) ( ϑ , φ ) = − Y J M ( 1 ) ( ϑ , φ ) − J ( J + 1 ) Y J M ( − 1 ) ( ϑ , φ ) , r ( S ^ ⋅ ∇ ) Y J M ( − 1 ) ( ϑ , φ ) = J ( J + 1 ) Y J M ( 0 ) ( ϑ , φ ) . (1526)
( S ^ L ^ ) Y J M L ( ϑ , φ ) = 1 2 { J ( J + 1 ) − L ( L + 1 ) − 2 } Y J M L ( ϑ , φ ) . (1527) (\hat{\mathbf{S}}\hat{\mathbf{L}})Y_{JM}^{L}(\vartheta,\varphi)=\frac{1}{2}\{J(J+1)-L(L+1)-2\}Y_{JM}^{L}(\vartheta,\varphi). \tag*{(1527)} ( S ^ L ^ ) Y J M L ( ϑ , φ ) = 2 1 { J ( J + 1 ) − L ( L + 1 ) − 2 } Y J M L ( ϑ , φ ) . (1527)
In particular,
( S ^ ⋅ L ^ ) Y J M J + 1 ( ϑ , φ ) = − ( J + 2 ) Y J M J + 1 ( ϑ , φ ) , ( S ^ ⋅ L ^ ) Y J M J ( ϑ , φ ) = − Y J M J ( ϑ , φ ) , ( S ^ ⋅ L ^ ) Y J M J − 1 ( ϑ , φ ) = ( J − 1 ) Y J M J − 1 ( ϑ , φ ) (1528) (\hat{\mathbf{S}}\cdot\hat{\mathbf{L}})Y_{JM}^{J+1}(\vartheta,\varphi)=-(J+2)Y_{JM}^{J+1}(\vartheta,\varphi),
\\
(\hat{\mathbf{S}}\cdot\hat{\mathbf{L}})Y_{JM}^{J}(\vartheta,\varphi)=-Y_{JM}^{J}(\vartheta,\varphi),
\\
(\hat{\mathbf{S}}\cdot\hat{\mathbf{L}})Y_{JM}^{J-1}(\vartheta,\varphi)=(J-1)Y_{JM}^{J-1}(\vartheta,\varphi) \tag*{(1528)} ( S ^ ⋅ L ^ ) Y J M J + 1 ( ϑ , φ ) = − ( J + 2 ) Y J M J + 1 ( ϑ , φ ) , ( S ^ ⋅ L ^ ) Y J M J ( ϑ , φ ) = − Y J M J ( ϑ , φ ) , ( S ^ ⋅ L ^ ) Y J M J − 1 ( ϑ , φ ) = ( J − 1 ) Y J M J − 1 ( ϑ , φ ) (1528)
( S ^ ⋅ L ^ ) Y J M ( 1 ) ( ϑ , φ ) = − Y J M ( 1 ) ( ϑ , φ ) + J ( J + 1 ) Y J M ( − 1 ) ( ϑ , φ ) , ( S ^ ⋅ L ^ ) Y J M ( 0 ) ( ϑ , φ ) = − Y J M ( 0 ) ( ϑ , φ ) , ( S ^ ⋅ L ^ ) Y J M ( − 1 ) ( ϑ , φ ) = J ( J + 1 ) Y J M ( 1 ) ( ϑ , φ ) − 2 Y J M ( − 1 ) ( ϑ , φ ) . (1529) (\hat{\mathbf{S}}\cdot\hat{\mathbf{L}})Y_{JM}^{(1)}(\vartheta,\varphi)=-Y_{JM}^{(1)}(\vartheta,\varphi)+\sqrt{J(J+1)}Y_{JM}^{(-1)}(\vartheta,\varphi),
\\
(\hat{\mathbf{S}}\cdot\hat{\mathbf{L}})Y_{JM}^{(0)}(\vartheta,\varphi)=-Y_{JM}^{(0)}(\vartheta,\varphi),
\\
(\hat{\mathbf{S}}\cdot\hat{\mathbf{L}})Y_{JM}^{(-1)}(\vartheta,\varphi)=\sqrt{J(J+1)}Y_{JM}^{(1)}(\vartheta,\varphi)-2Y_{JM}^{(-1)}(\vartheta,\varphi). \tag*{(1529)} ( S ^ ⋅ L ^ ) Y J M ( 1 ) ( ϑ , φ ) = − Y J M ( 1 ) ( ϑ , φ ) + J ( J + 1 ) Y J M ( − 1 ) ( ϑ , φ ) , ( S ^ ⋅ L ^ ) Y J M ( 0 ) ( ϑ , φ ) = − Y J M ( 0 ) ( ϑ , φ ) , ( S ^ ⋅ L ^ ) Y J M ( − 1 ) ( ϑ , φ ) = J ( J + 1 ) Y J M ( 1 ) ( ϑ , φ ) − 2 Y J M ( − 1 ) ( ϑ , φ ) . (1529)
Algebraic Relations
In the equations given below n \mathbf{n} n is the unit vector specified by the polar angles ϑ , φ \vartheta,\varphi ϑ , φ .
(a)
n Y J M ( ϑ , φ ) = J 2 J + 1 Y J M J − 1 ( ϑ , φ ) − J + 1 2 J + 1 Y J M J + 1 ( ϑ , φ ) , n Y J M ( ϑ , φ ) = Y J M ( − 1 ) ( ϑ , φ ) . (1530) \mathbf{n}Y_{JM}(\vartheta,\varphi)=\sqrt{\frac{J}{2J+1}}Y_{JM}^{J-1}(\vartheta,\varphi)-\sqrt{\frac{J+1}{2J+1}}Y_{JM}^{J+1}(\vartheta,\varphi),
\\
\mathbf{n}Y_{JM}(\vartheta,\varphi)=Y_{JM}^{(-1)}(\vartheta,\varphi). \tag*{(1530)} n Y J M ( ϑ , φ ) = 2 J + 1 J Y J M J − 1 ( ϑ , φ ) − 2 J + 1 J + 1 Y J M J + 1 ( ϑ , φ ) , n Y J M ( ϑ , φ ) = Y J M ( − 1 ) ( ϑ , φ ) . (1530)
(b)
n ⋅ Y J M J + 1 ( ϑ , φ ) = − J + 1 2 J + 1 Y J M ( ϑ , φ ) , n ⋅ Y J M J ( ϑ , φ ) = 0 , n ⋅ Y J M J − 1 ( ϑ , φ ) = J 2 J + 1 Y J M ( ϑ , φ ) . (1531) \mathbf{n}\cdot Y_{JM}^{J+1}(\vartheta,\varphi)=-\sqrt{\frac{J+1}{2J+1}}Y_{JM}(\vartheta,\varphi),
\\
\mathbf{n}\cdot Y_{JM}^{J}(\vartheta,\varphi)=0,
\\
\mathbf{n}\cdot Y_{JM}^{J-1}(\vartheta,\varphi)=\sqrt{\frac{J}{2J+1}}Y_{JM}(\vartheta,\varphi). \tag*{(1531)} n ⋅ Y J M J + 1 ( ϑ , φ ) = − 2 J + 1 J + 1 Y J M ( ϑ , φ ) , n ⋅ Y J M J ( ϑ , φ ) = 0 , n ⋅ Y J M J − 1 ( ϑ , φ ) = 2 J + 1 J Y J M ( ϑ , φ ) . (1531)
n ⋅ Y J M ( 1 ) ( ϑ , φ ) = 0 , n ⋅ Y J M ( 0 ) ( ϑ , φ ) = 0 , n ⋅ Y J M ( − 1 ) ( ϑ , φ ) = Y J M ( ϑ , φ ) . (1532) \mathbf{n}\cdot Y_{JM}^{(1)}(\vartheta,\varphi)=0,
\\
\mathbf{n}\cdot Y_{JM}^{(0)}(\vartheta,\varphi)=0,
\\
\mathbf{n}\cdot Y_{JM}^{(-1)}(\vartheta,\varphi)=Y_{JM}(\vartheta,\varphi). \tag*{(1532)} n ⋅ Y J M ( 1 ) ( ϑ , φ ) = 0 , n ⋅ Y J M ( 0 ) ( ϑ , φ ) = 0 , n ⋅ Y J M ( − 1 ) ( ϑ , φ ) = Y J M ( ϑ , φ ) . (1532)
(c)
n × Y J M J + 1 ( ϑ , φ ) = i J 2 J + 1 Y J M J ( ϑ , φ ) , n × Y J M J ( ϑ , φ ) = i J + 1 2 J + 1 Y J M J − 1 ( ϑ , φ ) + i J 2 J + 1 Y J M J + 1 ( ϑ , φ ) , n × Y J M J − 1 ( ϑ , φ ) = i J + 1 2 J + 1 Y J M J ( ϑ , φ ) , (1533) \begin{aligned}
\mathbf{n}\times\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)&=i\sqrt{\frac{J}{2J+1}}\mathbf{Y}^{J}_{JM}(\vartheta,\varphi),\\
\mathbf{n}\times\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)&=i\sqrt{\frac{J+1}{2J+1}}\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)+i\sqrt{\frac{J}{2J+1}}\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi),\\
\mathbf{n}\times\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)&=i\sqrt{\frac{J+1}{2J+1}}\mathbf{Y}^{J}_{JM}(\vartheta,\varphi), \tag*{(1533)}
\end{aligned} n × Y J M J + 1 ( ϑ , φ ) n × Y J M J ( ϑ , φ ) n × Y J M J − 1 ( ϑ , φ ) = i 2 J + 1 J Y J M J ( ϑ , φ ) , = i 2 J + 1 J + 1 Y J M J − 1 ( ϑ , φ ) + i 2 J + 1 J Y J M J + 1 ( ϑ , φ ) , = i 2 J + 1 J + 1 Y J M J ( ϑ , φ ) , (1533)
n × Y J M ( 1 ) ( ϑ , φ ) = i Y J M ( 0 ) ( ϑ , φ ) , n × Y J M ( 0 ) ( ϑ , φ ) = i Y J M ( 1 ) ( ϑ , φ ) , n × Y J M ( − 1 ) ( ϑ , φ ) = 0. (1534) \begin{aligned}
\mathbf{n}\times\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)&=i\mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi),\\
\mathbf{n}\times\mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi)&=i\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi),\\
\mathbf{n}\times\mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi)&=0. \tag*{(1534)}
\end{aligned} n × Y J M ( 1 ) ( ϑ , φ ) n × Y J M ( 0 ) ( ϑ , φ ) n × Y J M ( − 1 ) ( ϑ , φ ) = i Y J M ( 0 ) ( ϑ , φ ) , = i Y J M ( 1 ) ( ϑ , φ ) , = 0. (1534)
Sums of Vector Spherical Harmonics
(a)
∑ μ = − 1 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M L ( ϑ , φ ) ] μ = 3 4 π n ⋅ Y J M L ( ϑ , φ ) , \sum_{\mu=-1}^{1}Y^{*}_{1\mu}(\vartheta,\varphi)\left[\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)\right]_{\mu}=\sqrt{\frac{3}{4\pi}}\mathbf{n}\cdot\mathbf{Y}^{L}_{JM}(\vartheta,\varphi), μ = − 1 ∑ 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M L ( ϑ , φ ) ] μ = 4 π 3 n ⋅ Y J M L ( ϑ , φ ) ,
∑ μ = − 1 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M ( 1 ) ( ϑ , φ ) ] μ = 3 4 π n ⋅ Y J M ( 1 ) ( ϑ , φ ) . (1535) \sum_{\mu=-1}^{1}Y^{*}_{1\mu}(\vartheta,\varphi)\left[\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)\right]_{\mu}=\sqrt{\frac{3}{4\pi}}\mathbf{n}\cdot\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi). \tag*{(1535)} μ = − 1 ∑ 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M ( 1 ) ( ϑ , φ ) ] μ = 4 π 3 n ⋅ Y J M ( 1 ) ( ϑ , φ ) . (1535)
From (7.3.75) one obtains
∑ μ = − 1 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M J + 1 ( ϑ , φ ) ] μ = − 3 ( J + 1 ) 4 π ( 2 J + 1 ) Y J M ( ϑ , φ ) , \sum_{\mu=-1}^{1}Y^{*}_{1\mu}(\vartheta,\varphi)\left[\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right]_{\mu}=-\sqrt{\frac{3(J+1)}{4\pi(2J+1)}}Y_{JM}(\vartheta,\varphi), μ = − 1 ∑ 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M J + 1 ( ϑ , φ ) ] μ = − 4 π ( 2 J + 1 ) 3 ( J + 1 ) Y J M ( ϑ , φ ) ,
∑ μ = − 1 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M J ( ϑ , φ ) ] μ = 0 , \sum_{\mu=-1}^{1}Y^{*}_{1\mu}(\vartheta,\varphi)\left[\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right]_{\mu}=0, μ = − 1 ∑ 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M J ( ϑ , φ ) ] μ = 0 ,
∑ μ = − 1 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M J − 1 ( ϑ , φ ) ] μ = 3 J 4 π ( 2 J + 1 ) Y J M ( ϑ , φ ) , \sum_{\mu=-1}^{1}Y^{*}_{1\mu}(\vartheta,\varphi)\left[\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right]_{\mu}=\sqrt{\frac{3J}{4\pi(2J+1)}}Y_{JM}(\vartheta,\varphi), μ = − 1 ∑ 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M J − 1 ( ϑ , φ ) ] μ = 4 π ( 2 J + 1 ) 3 J Y J M ( ϑ , φ ) ,
∑ μ = − 1 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M ( 1 ) ( ϑ , φ ) ] μ = 0 , \sum_{\mu=-1}^{1}Y^{*}_{1\mu}(\vartheta,\varphi)\left[\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)\right]_{\mu}=0, μ = − 1 ∑ 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M ( 1 ) ( ϑ , φ ) ] μ = 0 ,
∑ μ = − 1 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M ( 0 ) ( ϑ , φ ) ] μ = 0 , \sum_{\mu=-1}^{1}Y^{*}_{1\mu}(\vartheta,\varphi)\left[\mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi)\right]_{\mu}=0, μ = − 1 ∑ 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M ( 0 ) ( ϑ , φ ) ] μ = 0 ,
∑ μ = − 1 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M ( − 1 ) ( ϑ , φ ) ] μ = 3 4 π Y J M ( ϑ , φ ) . (1536) \sum_{\mu=-1}^{1}Y^{*}_{1\mu}(\vartheta,\varphi)\left[\mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi)\right]_{\mu}=\sqrt{\frac{3}{4\pi}}Y_{JM}(\vartheta,\varphi). \tag*{(1536)} μ = − 1 ∑ 1 Y 1 μ ∗ ( ϑ , φ ) [ Y J M ( − 1 ) ( ϑ , φ ) ] μ = 4 π 3 Y J M ( ϑ , φ ) . (1536)
(b)
∑ M = − J J Y J M ∗ ( ϑ , φ ) Y J M L ( ϑ , φ ) = ( 2 J + 1 ) ( 2 L + 1 ) 4 π C J 0 L 0 J 0 n . (1537) \sum_{M=-J}^{J}Y^{*}_{JM}(\vartheta,\varphi)\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=\frac{\sqrt{(2J+1)(2L+1)}}{4\pi}C^{J0}_{J0L0}\mathbf{n}. \tag*{(1537)} M = − J ∑ J Y J M ∗ ( ϑ , φ ) Y J M L ( ϑ , φ ) = 4 π ( 2 J + 1 ) ( 2 L + 1 ) C J 0 L 0 J 0 n . (1537)
In more detailed form (7.3.78) reads
∑ M = − J J Y J M ∗ ( ϑ , φ ) Y J M J + 1 ( ϑ , φ ) = − ( J + 1 ) ( 2 J + 1 ) 4 π n , ∑ M = − J J Y J M ∗ ( ϑ , φ ) Y J M J ( ϑ , φ ) = 0 , ∑ M = − J J Y J M ∗ ( ϑ , φ ) Y J M J − 1 ( ϑ , φ ) = J ( 2 J + 1 ) 4 π n . (1538) \begin{aligned}
\sum_{M=-J}^{J}Y^{*}_{JM}(\vartheta,\varphi)\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)&=-\frac{\sqrt{(J+1)(2J+1)}}{4\pi}\mathbf{n},\\
\sum_{M=-J}^{J}Y^{*}_{JM}(\vartheta,\varphi)\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)&=0,\\
\sum_{M=-J}^{J}Y^{*}_{JM}(\vartheta,\varphi)\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)&=\frac{\sqrt{J(2J+1)}}{4\pi}\mathbf{n}. \tag*{(1538)}
\end{aligned} M = − J ∑ J Y J M ∗ ( ϑ , φ ) Y J M J + 1 ( ϑ , φ ) M = − J ∑ J Y J M ∗ ( ϑ , φ ) Y J M J ( ϑ , φ ) M = − J ∑ J Y J M ∗ ( ϑ , φ ) Y J M J − 1 ( ϑ , φ ) = − 4 π ( J + 1 ) ( 2 J + 1 ) n , = 0 , = 4 π J ( 2 J + 1 ) n . (1538)
Analogous relations for Y J M ( λ ) ( ϑ , φ ) \mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi) Y J M ( λ ) ( ϑ , φ ) have the form
∑ M = − J J Y J M ( 1 ) ∗ ( ϑ , φ ) Y J M ( 1 ) ( ϑ , φ ) = ∑ M = − J J Y J M ( 0 ) ∗ ( ϑ , φ ) Y J M ( 0 ) ( ϑ , φ ) = 0 , (1539) \sum_{M=-J}^{J}\mathbf{Y}^{(1)*}_{JM}(\vartheta,\varphi)\mathbf{Y}^{(1)}_{JM}(\vartheta,\varphi)=\sum_{M=-J}^{J}\mathbf{Y}^{(0)*}_{JM}(\vartheta,\varphi)\mathbf{Y}^{(0)}_{JM}(\vartheta,\varphi)=0, \tag*{(1539)} M = − J ∑ J Y J M ( 1 ) ∗ ( ϑ , φ ) Y J M ( 1 ) ( ϑ , φ ) = M = − J ∑ J Y J M ( 0 ) ∗ ( ϑ , φ ) Y J M ( 0 ) ( ϑ , φ ) = 0 , (1539)
∑ M = − J J Y J M ∗ ( ϑ , φ ) Y J M ( − 1 ) ( ϑ , φ ) = 2 J + 1 4 π n . \sum_{M=-J}^{J}\mathbf{Y}^{*}_{JM}(\vartheta,\varphi)\mathbf{Y}^{(-1)}_{JM}(\vartheta,\varphi)=\frac{2J+1}{4\pi}\mathbf{n}. M = − J ∑ J Y J M ∗ ( ϑ , φ ) Y J M ( − 1 ) ( ϑ , φ ) = 4 π 2 J + 1 n .
(c)
∑ M = − J J Y J M L ′ ∗ ( ϑ , φ ) ⋅ Y J M L ( ϑ , φ ) = 2 J + 1 4 π δ L L ′ , (1540) \sum_{M=-J}^{J}\mathbf{Y}^{L'*}_{JM}(\vartheta,\varphi)\cdot\mathbf{Y}^{L}_{JM}(\vartheta,\varphi)=\frac{2J+1}{4\pi}\delta_{LL'}, \tag*{(1540)} M = − J ∑ J Y J M L ′ ∗ ( ϑ , φ ) ⋅ Y J M L ( ϑ , φ ) = 4 π 2 J + 1 δ L L ′ , (1540)
∑ M = − J J Y J M ( λ ′ ) ∗ ( ϑ , φ ) ⋅ Y J M ( λ ) ( ϑ , φ ) = 2 J + 1 4 π δ λ λ ′ . \sum_{M=-J}^{J}\mathbf{Y}^{(\lambda')*}_{JM}(\vartheta,\varphi)\cdot\mathbf{Y}^{(\lambda)}_{JM}(\vartheta,\varphi)=\frac{2J+1}{4\pi}\delta_{\lambda\lambda'}. M = − J ∑ J Y J M ( λ ′ ) ∗ ( ϑ , φ ) ⋅ Y J M ( λ ) ( ϑ , φ ) = 4 π 2 J + 1 δ λ λ ′ .
(d) Below a \mathbf{a} a is an arbitrary complex vector, n \mathbf{n} n is the unit vector specified by the polar angles ϑ , φ \vartheta,\varphi ϑ , φ .
∑ M = − J J ∣ a ⋅ Y J M J + 1 ( ϑ , φ ) ∣ 2 = 1 8 π { J ∣ a ∣ 2 + ( J + 2 ) ∣ n ⋅ a ∣ 2 } , (1541) \sum_{M=-J}^{J}\left|\mathbf{a}\cdot\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right|^2=\frac{1}{8\pi}\left\{J|\mathbf{a}|^2+(J+2)|\mathbf{n}\cdot\mathbf{a}|^2\right\}, \tag*{(1541)} M = − J ∑ J a ⋅ Y J M J + 1 ( ϑ , φ ) 2 = 8 π 1 { J ∣ a ∣ 2 + ( J + 2 ) ∣ n ⋅ a ∣ 2 } , (1541)
∑ M = − J J ∣ a ⋅ Y J M J ( ϑ , φ ) ∣ 2 = 2 J + 1 8 π { ∣ a ∣ 2 − ∣ n ⋅ a ∣ 2 } , (1542) \sum_{M=-J}^{J}\left|\mathbf{a}\cdot\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right|^2=\frac{2J+1}{8\pi}\left\{|\mathbf{a}|^2-|\mathbf{n}\cdot\mathbf{a}|^2\right\}, \tag*{(1542)} M = − J ∑ J a ⋅ Y J M J ( ϑ , φ ) 2 = 8 π 2 J + 1 { ∣ a ∣ 2 − ∣ n ⋅ a ∣ 2 } , (1542)
∑ M = − J J ∣ a ⋅ Y J M J − 1 ( ϑ , φ ) ∣ 2 = 1 8 π { ( J + 1 ) ∣ a ∣ 2 + ( J − 1 ) ∣ n ⋅ a ∣ 2 } , (1543) \sum_{M=-J}^{J}\left|\mathbf{a}\cdot\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right|^2=\frac{1}{8\pi}\left\{(J+1)|\mathbf{a}|^2+(J-1)|\mathbf{n}\cdot\mathbf{a}|^2\right\}, \tag*{(1543)} M = − J ∑ J a ⋅ Y J M J − 1 ( ϑ , φ ) 2 = 8 π 1 { ( J + 1 ) ∣ a ∣ 2 + ( J − 1 ) ∣ n ⋅ a ∣ 2 } , (1543)
∑ M = − J J ( a ⋅ Y J M J + 1 ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J ( ϑ , φ ) ) = − i J ( 2 J + 1 ) 8 π n ⋅ [ a ∗ × a ] , (1544) \sum_{M=-J}^{J}\left(\mathbf{a}\cdot\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right)^*\left(\mathbf{a}\cdot\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right)=-i\frac{\sqrt{J(2J+1)}}{8\pi}\mathbf{n}\cdot[\mathbf{a}^*\times\mathbf{a}], \tag*{(1544)} M = − J ∑ J ( a ⋅ Y J M J + 1 ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J ( ϑ , φ ) ) = − i 8 π J ( 2 J + 1 ) n ⋅ [ a ∗ × a ] , (1544)
∑ M = − J J ( a ⋅ Y J M J + 1 ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J − 1 ( ϑ , φ ) ) = J ( J + 1 ) 8 π { ∣ a ∣ 2 − 3 ∣ n ⋅ a ∣ 2 } , (1545) \sum_{M=-J}^{J}\left(\mathbf{a}\cdot\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right)^*\left(\mathbf{a}\cdot\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right)=\frac{\sqrt{J(J+1)}}{8\pi}\left\{|\mathbf{a}|^2-3|\mathbf{n}\cdot\mathbf{a}|^2\right\}, \tag*{(1545)} M = − J ∑ J ( a ⋅ Y J M J + 1 ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J − 1 ( ϑ , φ ) ) = 8 π J ( J + 1 ) { ∣ a ∣ 2 − 3∣ n ⋅ a ∣ 2 } , (1545)
∑ M = − J J ( a ⋅ Y J M J ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J + 1 ( ϑ , φ ) ) = − i J ( 2 J + 1 ) 8 π n ⋅ [ a ∗ × a ] , (1546) \sum_{M=-J}^{J}\left(\mathbf{a}\cdot\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right)^*\left(\mathbf{a}\cdot\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right)=-i\frac{\sqrt{J(2J+1)}}{8\pi}\mathbf{n}\cdot[\mathbf{a}^*\times\mathbf{a}], \tag*{(1546)} M = − J ∑ J ( a ⋅ Y J M J ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J + 1 ( ϑ , φ ) ) = − i 8 π J ( 2 J + 1 ) n ⋅ [ a ∗ × a ] , (1546)
∑ M = − J J ( a ⋅ Y J M J ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J − 1 ( ϑ , φ ) ) = − i ( J + 1 ) ( 2 J + 1 ) 8 π n ⋅ [ a ∗ × a ] . (1547) \sum_{M=-J}^{J}\left(\mathbf{a}\cdot\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right)^*\left(\mathbf{a}\cdot\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right)=-i\frac{\sqrt{(J+1)(2J+1)}}{8\pi}\mathbf{n}\cdot[\mathbf{a}^*\times\mathbf{a}]. \tag*{(1547)} M = − J ∑ J ( a ⋅ Y J M J ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J − 1 ( ϑ , φ ) ) = − i 8 π ( J + 1 ) ( 2 J + 1 ) n ⋅ [ a ∗ × a ] . (1547)
∑ M = − J J ( a ⋅ Y J M J − 1 ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J + 1 ( ϑ , φ ) ) = J ( J + 1 ) 8 π { ∣ a ∣ 2 − 3 ∣ n ⋅ a ∣ 2 } , (1548) \sum_{M=-J}^{J}\left(\mathbf{a}\cdot\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right)^*\left(\mathbf{a}\cdot\mathbf{Y}^{J+1}_{JM}(\vartheta,\varphi)\right)=\frac{\sqrt{J(J+1)}}{8\pi}\left\{|\mathbf{a}|^2-3|\mathbf{n}\cdot\mathbf{a}|^2\right\}, \tag*{(1548)} M = − J ∑ J ( a ⋅ Y J M J − 1 ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J + 1 ( ϑ , φ ) ) = 8 π J ( J + 1 ) { ∣ a ∣ 2 − 3∣ n ⋅ a ∣ 2 } , (1548)
∑ M = − J J ( a ⋅ Y J M J − 1 ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J ( ϑ , φ ) ) = − i ( J + 1 ) ( 2 J + 1 ) 8 π n ⋅ [ a ∗ × a ] . (1549) \sum_{M=-J}^{J}\left(\mathbf{a}\cdot\mathbf{Y}^{J-1}_{JM}(\vartheta,\varphi)\right)^*\left(\mathbf{a}\cdot\mathbf{Y}^{J}_{JM}(\vartheta,\varphi)\right)=-i\frac{\sqrt{(J+1)(2J+1)}}{8\pi}\mathbf{n}\cdot[\mathbf{a}^*\times\mathbf{a}]. \tag*{(1549)} M = − J ∑ J ( a ⋅ Y J M J − 1 ( ϑ , φ ) ) ∗ ( a ⋅ Y J M J ( ϑ , φ ) ) = − i 8 π ( J + 1 ) ( 2 J + 1 ) n ⋅ [ a ∗ × a ] . (1549)