betaFeedback

WorksPhysics and Astronomy

Towards nonlinear electrodynamics without renormalization

Abstract

In this paper is considered nonlinear electrodynamics (NE) which does not satisfy the linear superposition principle (LSP). Since the presentation of the special theory of relativity, it has been commonly accepted that a famous formula E = mc^2 = m_0c^2 /(1-v^2/c^2)^(1/2) expresses the energy of a free particle only. However, while studying the experiment for the annihilation of particle and antiparticle and the production of the photon in terms of the law of energy conservation of particle and field, we obtain a conclusion that E includes not only the energy of a free particle but also the energy of its self -fields (electromagnetic field and gravitational field). Hence, a formula for the energy in the special theory of relativity comes to have a more inclusive meaning than Einstein had thought of it. Based upon such an idea and the correspondence principle, through introducing a new non-Euclidean four-dimensional space, so-called KR space in which the metric is the function of coordinates and 4-velocity, we reformulate the action function to be a function defined in the nonlinear version and from it find equation system for a charge and the field with a modified version, which is reduced to that in the Maxwell theory under the weak field condition: here, weak field condition is given by |U/m_0c^2|<<1, U; interactional energy. With the equation system, we show that the divergence problem arising from the Maxwell theory, without invoking renormalization, is very naturally solved by sensible mathematics and then demonstrate that the total energy of a free charge and its field equals mc^2 and so the energy of an electrostatic field leads to the finite. In this course are derived the modified version of Coulomb potential and of radiation reaction without renormalization which contains an interference effect (a nonlinear effect that does not obey LSP).

Introduction

Finding the equation of motion for a charged particle in the classical radiation field is a very old problem [1, 2, 3, 4]. When trying to consistently derive the effective equation of motion for a point charge, one comes up against the inevitable infinities or divergences arising due to taking a charged particle to be a point: an infinity of electrostatic field energy (self-field energy) and divergent contribution in radiation reaction. From this, the question arises whether the Maxwell theory is a consistent one in physics [5]. In order to solve this problem, different remedies have been explored [6]. The first approach is to modify the Maxwell theory as has been done by Born and Infeld [7] or Podolsky and Schwed [8]. They recently regained attention [9]. The second approach is to introduce an extended charge model as has been done by Abraham [10], Lorentz [11], and many others [12]; see [13] for a recent report. The third approach is the renormalization procedure which was first proposed by Dirac [14] and has been studied in different forms; see [15-17]. If we confine our consideration to take a charge to be a point, one of them, the first and third approaches, should be adopted. But it is not so satisfactory to conclude that the renormalization is a reasonable approach. The renormalization procedure violates the scientific requirement that there is a unique description for any given physical fact: subtracting the infinite (mass of self-field) from another infinite (bare mass) is not sensible mathematics and gives no unique mathematical value [18,19] (Dirac rejected strongly the renormalization procedure in the quantum electrodynamics though he was a founder of renormalization program in the classical electrodynamics). The version of function of bare mass is not defined unambiguously and that subtracting mass of self-field from bare mass leads to an unambiguous mass (renormalized mass) measured is only assumed but not testified. The additional comment is that, in Maxwell’s electrodynamics, by means of the renormalization procedure, the essential meaning of the basic concept (mass of a particle) defined in the first of the building of a theory is, in the middle of the evolution of the theory, basically modified and redefined. This shows that the renormalization procedure stands against a scientific rule to be naturally introduced for the building of a consistent theory according to which basic concepts of the theory should always preserve their own meaning in the whole area of a theory. This divergence problem in classical electrodynamics reappears in a more serious form at the stage of quantum electrodynamics (QED) [20]. In this context, there were some attempts to build QED without renormalization [21], but by a preponderance of the renormalization procedure, these were suppressed. In this paper, we consider the classical nonlinear electrodynamics (CNE) without renormalization to construct classical electrodynamics consistently and then, based upon it, touch the quantum nonlinear electrodynamics (QNE) without renormalization. For this purpose, we present a starting idea or starting postulate that the total energy of a free particle and its self-field equals mc2mc^2: it will be discussed in Sect.2. Until now, the general consensus would likely be that the Maxwell theory is a complete one, and for the elimination of the divergence is no alternative to introduce the renormalization procedure within a framework of the Maxwell theory as far as a particle is regarded as a point. Therefore, finding nonlinear electrodynamics (NE) without renormalization is beyond the hitherto thought in physics. But, our starting idea contradicts strongly renormalization and presents the limitation of the Maxwell theory. From such an unsatisfactory situation, we are motivated to write this paper. The importance of the establishment of NE is not purely confined only to the elimination of the divergence, while NE necessarily gives rise to nonlinear effects which cannot be described in terms of the Maxwell theory. It is anticipated that the study for it would be contributed to a new advance in electrodynamics.

The plan of the paper is as follows. Sect. 2 puts forward starting postulate I and in Sect. 3 is formulated the action function for the motion of a charge and discussed the equations for the motion of a charge. Sect. 4 studies the experimental verification for the theoretical results under the strong field condition, obtained from the equations. In Sect. 5 is considered the generalized conservation law of charge and application limit of the gauge symmetry. Sect. 6 studies the energy-momentum tensor and field equations. Sect. 7 introduces KR space (new non-Euclidean space) on which Lagrangian is formulated, and gives physical analysis and normalization rules for main functions defined on KR space. Sect. 8 discusses the main results of CNE: modified versions of Coulomb’s law and radiation reaction, and total energy of a charge and its self-field which is considered without renormalization. In Sect. 9, we briefly touch quantum nonlinear electrodynamics (QNE) in which a modified version of the Dirac equation is derived and the divergence problem coming from the S-matrix of quantum electrodynamics is naturally solved without renormalization. Then we present that consideration for the scattering of an electron by the modified version of Coulomb potential produced by a nucleus yields a new correction to be verified by experiment in the future.

Starting postulate I: basic idea for the total energy of a particle and its self-field

Historically, finding classical nonlinear electrodynamics was motivated by the credit for the ‘principle of finiteness’ by which a satisfactory theory should be the absence of divergences in physical quantities [7, 8, 9]. Unfortunately, even to the present day, there is not yet a leading principle for building a unique theory, a basic principle that enables one to satisfy the ‘principle of finiteness’ and specifies intrinsic criteria of validation or reputation for attempts at the theory.

In this section, we present a starting postulate I (providing a candidate of the leading principle) that underlies CNE, which can allow one to define the total energy of a charged point particle and its self-field as the finite: in this case a particle is taken to be a point. For this purpose, we consider the annihilation of particle-antiparticle well known in physics. The formula of energy conservation can be written as follows:

m0c21−v12/c2+m0c21−v22/c2=2ℏω(1)\frac{m_0c^2}{\sqrt{1-v_1^2/c^2}}+\frac{m_0c^2}{\sqrt{1-v_2^2/c^2}}=2\hbar\omega \tag*{(1)}

where m0c21−v12/c2\frac{m_0c^2}{\sqrt{1-v_1^2/c^2}} is the energy of a free particle and m0c21−v22/c2\frac{m_0c^2}{\sqrt{1-v_2^2/c^2}} the energy of a free antiparticle, which is the main result of the special theory of relativity (SR): v1v_1 is the velocity of the particle and v2v_2 the velocity of the antiparticle. Since Einstein presented the special theory of relativity(SR), mc2mc^2 has been considered only as the energy confined to the particle. When describing Lagrangian for the motion of a particle in the electromagnetic field, the rest mass m0m_0 that enters the particle-related part was regarded as “bare mass” devoid of field. On the other hand, while considering radiation reaction, in order to eliminate the divergence of self-field energy, “dressed mass” was redefined by the renormalization procedure [14]. But, when considering (1), from the outset, it should have been viewed that E=mc2=m0c2/1−v2/c2E=mc^2=m_0c^2/\sqrt{1-v^2/c^2} includes not only the energy of a particle but also the energy of the field (self-field energy) produced by the particle. Why should it be viewed in that way? It, in a word, is based upon the fact according to which the total energy of particles and fields created by them should be always conserved. In the case of considering mc2mc^2 to be energy confined to a particle only, the conservation formula of the total energy of particles and fields can be represented as

(m0c21−v12/c2+ε1)+(m0c21−v22/c2+ε2)=2ℏω+ε,(2)\left(\frac{m_0c^2}{\sqrt{1-v_1^2/c^2}}+\varepsilon_1\right)+\left(\frac{m_0c^2}{\sqrt{1-v_2^2/c^2}}+\varepsilon_2\right)=2\hbar\omega+\varepsilon, \tag*{(2)}

where ε1\varepsilon_1 is the energy of electromagnetic field and gravitational field created by the particle, and ε2\varepsilon_2 the energy of two fields created by the antiparticle: in a general sense, we assume that particles produce the electromagnetic field, as well as the gravitational field. From (1) and (2), we get

ε1+ε2=ε,(3)\varepsilon_1+\varepsilon_2=\varepsilon, \tag*{(3)}

where ε\varepsilon is the energy of another matter which appears newly except photon after the annihilation of a system of particle-antiparticle. However, until now has not yet been found any experimental data which, except photon, another matter occurred. Therefore, if there is something except photons after the annihilation of particle-antiparticle, there is no alternative to conclude that only the fields remain as it is, as an unmeasured invariant. On the other hand, as the particle-antiparticle is annihilated, the mass m0m_0 and charge ee also vanish, and thus the static gravitational field and electrostatic field created by the rest mass and charge also should disappear. Therefore, the result is

ε1+ε2≠ε,(4)\varepsilon_1+\varepsilon_2\ne\varepsilon, \tag*{(4)}

with

{ε=0ε1+ε2>0(5)\begin{cases} \varepsilon=0 \\ \varepsilon_1+\varepsilon_2>0 \tag*{(5)} \end{cases}

Consequently, when considering mc2mc^2 only as the energy confined to a particle, with the occurrence of photons, we lead to the conclusion that the energy of electromagnetic field and gravitational field should vanish, which stands against the conservation law of energy. In this context, we define starting postulate I as follows: for a system consisting of a free particle and its field, mc2mc^2 (mm is the measured mass; m=m0m=m_0 in the rest state, m=m0/1−β2m=m_0/\sqrt{1-\beta^2} in the motion state) is the same as the total energy of a particle and all fields produced by it (electromagnetic field plus gravitational field) but not the energy of the particle only: here β=v/c\beta=v/c (vv; velocity of a particle) and mm is the total mass of a particle and its self-field. Throughout the paper, β\beta denotes v/cv/c.

In this paper, the fact that the total energy of a free particle and fields created by it is just equal to mc2mc^2 is regarded as a new starting postulate or starting point for developing CNE. This shows that the formula for the energy of a free particle in SR involves more inclusive and universal physical meaning than Einstein had thought of it: an inclusive concept of the total energy of a system that consists of a particle and its self-fields. From the starting postulate I, we can obtain the following conclusions. The starting postulate I implies that in the paper, EE (the total energy of a free particle and its field) obtained from energy-momentum tensor should lead to

{E=m0c21−β2+U0=mc2U0=0(6)\begin{cases} E=\dfrac{m_0c^2}{\sqrt{1-\beta^2}}+U_0=mc^2 \\ U_0=0 \tag*{(6)} \end{cases}

where U0U_0 is the energy of the self-field of a particle, which gives no divergence, being zero according to the starting postulate I: it is because mc2mc^2 contains not only the energy of a particle but also the energy of its self-field. In this case, it should be emphasized that the system consisting of a particle and its field should be considered as integrity and so a question about what share of mc2mc^2 is divided into the energy of matter and self-field, respectively and whether the bare mass and the mass of self-field are finite or infinite are, in principle, invalid. Therefore, the paper does not need the definitions of bare mass and mass of self-field, and thus, the introduction of the renormalization procedure.

When one considers a system of many particles, the total energy of nn particles and their fields (self-field plus interactional field), EnE_n, should arrive at

En=∑i=1nmic2+∑i=1nUiUi=∑j=1(i≠j)n−1Uij(∼eiej/Rij),Uii=0(7)\begin{aligned} E_n &= \sum_{i=1}^{n} m_i c^2 + \sum_{i=1}^{n} U_i \\ U_i &= \sum_{\substack{j=1 \\ (i\ne j)}}^{n-1} U_{ij}(\sim e_i e_j/R_{ij}), \qquad U_{ii}=0 \tag*{(7)} \end{aligned}

where mic2m_i c^2 is the energy of the iith particle and its self-field, UiU_i the interaction energy of the iith particle with the whole fields produced by other particles, UijU_{ij} the interaction energy of different two particles, RijR_{ij} the distance between arbitrary two particles, and the self-field energy UiiU_{ii} should vanish. Hence, CNE should be built such that the infinite terms arising from self-field vanish naturally and only the terms of interactional energy dependent on the mutual distribution of particles remain. This explicitly shows that starting postulate I is a leading principle that enable one to satisfy the ‘principle of finiteness’. However, unfortunately from the Maxwell theory, we cannot find the conclusion obtained above. If one tries to get such a conclusion, it is no alternative to introduce the renormalization procedure. But as shown above, the starting postulate I does not need the renormalization procedure and stands strongly against it. This implies a drawback of the Maxwell theory. In the view of the correspondence principle, a new theory should include the former theory as an approximate version. the Maxwell theory is a linear theory that obeys LSP. In a general sense, a nonlinear theory comprises a linear theory as an approximate version as if the non-Euclidian geometry includes the Euclidian geometry, that is, linear geometry as an approximate version (within an infinitesimal area of space) Therefore, from starting postulate I and correspondence principle, we can conclude that true classical electrodynamics must be a nonlinear theory that includes the Maxwell theory as an approximate version: staring postulate II is discussed in Subsect.7.2.

Equation of motion of a charge in CNE

In this section, we formulate the action for the motion of a charge and derive the equation of motion of a charge in the electromagnetic field. In CNE, the action for the motion of a particle should satisfy the following requirement. First, the action function should be invariant with respect to the Lorentz transformation from Einstein’s postulates in SR. Second, the action must constitute a nonlinear function which, according to the principle of correspondence, includes the linear Lagrangian in the Maxwell theory as an approximate version. Third, mc2mc^2 in the action function for a free particle should be the total energy of a particle and its self-field: this is subject to the starting postulate I argued in Sect. 2 and so from the actions for a free particle and field produced by it should follow that the total energy of a particle and its self-field is the same as mc2mc^2, which is derived from the energy-momentum tensor obtained from the action.

From the above requirement we define the action for the motion of a charge as follows:

S=∫L ds=−m0c∫ds(gμνuμuν)12=−m0c∫[δμν(1+2αAλuλ)uμuν]12ds(8)S=\int L\,\mathrm{d}s=-m_{0}c\int\mathrm{d}s\left(g_{\mu\nu}u^{\mu}u^{\nu}\right)^{\frac{1}{2}}=-m_{0}c\int\left[\delta_{\mu\nu}\left(1+2\alpha A_{\lambda}u^{\lambda}\right)u^{\mu}u^{\nu}\right]^{\frac{1}{2}}\mathrm{d}s \tag*{(8)}

with L=−m0c(gμνuμuν)12L=-m_{0}c\left(g_{\mu\nu}u^{\mu}u^{\nu}\right)^{\frac{1}{2}} and gμν=δμν(1+2αAλuλ)g_{\mu\nu}=\delta_{\mu\nu}\left(1+2\alpha A_{\lambda}u^{\lambda}\right), where δμν\delta_{\mu\nu} is the Minkowski metric tensor given by δ=diag⁡(1,−1,−1,−1)\delta=\operatorname{diag}(1,-1,-1,-1), AλA_{\lambda} the 4-dimensional electromagnetic field vector, uλu^{\lambda} the 4-dimensional velocity, Aλuλ=δλσAλuσA_{\lambda}u^{\lambda}=\delta_{\lambda\sigma}A^{\lambda}u^{\sigma}, α\alpha a constant defined from the approximate condition α=e/m0c2\alpha=e/m_{0}c^{2}, Aλ=Aex+AinA_{\lambda}=A_{\mathrm{ex}}+A_{\mathrm{in}} (AexA_{\mathrm{ex}}; external field and AinA_{\mathrm{in}}; self-field of a particle), and gμνg_{\mu\nu} is a non-Euclidean metric tensor (a metric tensor of KR space dependent on the velocity 4-vector and coordinates of space-time: see 7.1 for KR space and as will be shown in Sect. 8, Ain=0A_{\mathrm{in}}=0 holds for a free particle and AinA_{\mathrm{in}} is the small quantity of order e2/m0c3e^{2}/m_{0}c^{3} in the presence of radiation reaction. Hence, from the outset in the action (8), the interactions of a particle with an external field as well as self-field are treated in a unified framework, while the Lagrangian in the Maxwell theory includes only the external field due to the divergence of the self-field potential and then, with the aid of a very artificial approach of renormalization, the interaction of a particle with its self-field is, additionally from the outside of a logical system of the theory, introduced into the equation of motion, which finally leads to the Lorentz-Abraham-Dirac equation (LAD equation) [6]. As will be seen in Sect. 8, describing AλA_{\lambda} in a unified framework of AexA_{\mathrm{ex}} and AinA_{\mathrm{in}} is of paramount importance for a consistent solution to the divergence problem.

The action function (8) is conformed to the requirement mentioned above. First, (8) is invariant with respect to Lorentz transformation: as can be easily demonstrated, because in the metric tensor gμνg_{\mu\nu}, δμν\delta_{\mu\nu} is the Minkowski metric tensor and AλuλA_{\lambda}u^{\lambda} the scalar product of the field 4-vector and the velocity 4-vector, (8) is invariant with respect to Lorentz transformation. Second, as αAλuλ\alpha A_{\lambda}u^{\lambda} is in the square root, the term characterizing interaction becomes a nonlinear version and then gives birth to the interference effect of interaction, and so, the linear superposition principle (LSP), as well as gauge symmetry is broken. The gauge symmetry is a basic principle that underlies present classical and quantum electrodynamics. In this relation, serious problems may arise why gauge symmetry necessarily should be broken and if so, what becomes of the conservation law of charge, what is the associated conservation law: this is intensively argued in Sect. 5. Unfortunately, with one or several phrases can never be given a persudadible description of the problem. Throughout the paper, we have confirmed that gauge symmetry and LSP are the fundamental factors giving rise to divergence problems in classical and quantum linear electrodynamics, and only classical and quantum nonlinear electrodynamics described by logical and sensible mathematics ensures the finiteness of physical quantities. It is obvious that while considering CNE with starting postulate I and correspondence principle, the breaking of gauge symmetry and LSP are inevitable because the two principles come from the linearity of interaction Lagrangian. Now, allowing for the approximation condition r≫r0r\gg r_{0} (rr; the interaction radius, r0r_{0}; around electron radius, ∼10−13\sim10^{-13} cm), the action (8) is transformed into the action for the motion of a charge in Maxwell’s electrodynamics. The action (8) can be rewritten as

References

  1. [1]F. Rohrlich, Classical Charged Particles, World Scientific (2007).
  2. [2]S. Lyle, Self-Force and Inertia; Old Light on New Idea, Lecture Notes in Physics; Vol 796, New York (2010).
  3. [3]R. Penros and N.D. Mermin, The Emperor’s New Mind; Concerning Computers. Mind and the Laws of Physics, New York (1990).
  4. [4]H. Spohn, Dynamics of Charged Particle and their Radiation Field, Cambridge (2004).
  5. [5]M. Frisch, Inconsistency, Asymmetry, and Non-Locality: A Philosophical Investigation of Classical Electrodynamics, Oxford, 23-46 (2005).DOI
  6. [6]C. Bild and D.A. Deckert, H. Ruhl, Phys. Rev. D 99, 096001 (2019).
  7. [7]M. Born and L. Infeld, Proc. Roy. Soc. A 144, 425 (1934).
  8. [8]B. Podolsky and P. Schwed, Rev. Mod. Phys. 20, 40 (1948).
  9. [9]M.K.H. Kiessling, J. Stat. Phys. 116, 1057 (2004).

Paper details

vlnc:W_4ugpesr5
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents