betaFeedback

WorksPhysics and Astronomy

A Concise Introduction to Quantum Electromagnetics

Abstract

This book introduces quantum electromagnetics to readers trained in classical electromagnetics, without assuming a systematic background in quantum mechanics. The central question is when classical descriptions suffice, and when the quantum states of matter or the field must be retained explicitly. Starting from familiar electromagnetic modes and material response, the discussion develops atomic dynamics, field quantization, and the interaction of quantized fields with quantum matter. Particular attention is given to how mode profiles, Green functions, and scattering matrices obtained from classical electromagnetic calculations enter quantum models. Atomic coherence, photon statistics, spontaneous emission, and cavity coupling are developed through derivations that identify the physical assumptions and limits of each description. Examples drawn from atomic clocks, Rydberg-atom receivers, and single-photon devices connect these concepts to excitation, propagation, and measurement in practical systems. Loss and noise are included in relating internal dynamics to measured signals and correlations. The final discussion examines how sensitivity and claims of quantum advantage depend on bandwidth, measurement resources, and the classical reference used for comparison. The aim is to help readers choose an appropriate model, carry out the calculation, and understand what its predictions mean for an electromagnetic measurement.

For researchers and engineers in electromagnetics

Description of matter

Macroscopic media

Quantum matter

Classical field

ε\varepsilon, μ\mu

Classical electromagnetics

Semiclassical theory

Quantum field

∣1⟩|1\rangle

Quantum field propagation

Quantum fields and quantum matter

Preface

In engineering electromagnetics, we often begin with a concrete question. Can an antenna be made smaller, receiver sensitivity improved by a few decibels, or insertion loss reduced a little further? We solve Maxwell’s equations, analyze field distributions and energy flow, and repeatedly adjust structures, materials, and feeds. Fabrication and measurement then test the design. Progress in wireless communications, radar sensing, high-speed interconnects, and optoelectronic devices has accumulated through such efforts. Generations of researchers and engineers have turned electromagnetics into useful technology, changing how we communicate, understand the world, and use energy.

With experience, another side becomes apparent. In some mature areas, performance gains become increasingly difficult. More complex structures, more design variables, and more accurate simulations do not always bring proportionate benefits. The question then extends beyond how to optimize further: where does the limitation come from? Is there room left in the design and fabrication, or do physical constraints already bound the performance? Resolving this question can be more valuable than obtaining another slightly improved curve.

For electrically small, linear passive antennas, the Chu limit relates size, stored energy, and radiation quality factor. The diffraction limit in conventional far-field imaging depends on wavelength, aperture, and the measurement scheme. Receivers must also account for thermal noise. Under impedance matching and conditions such as hf≪kBThf \ll k_{\mathrm{B}}T, the available thermal-noise power in bandwidth BB is approximately kBTBk_{\mathrm{B}}TB. Each result has its own conditions of validity. To understand a limit, we must first understand those conditions. When discussing a new technology, we should state which conditions change, what resources are added, and what costs follow. Quantum electromagnetics must meet the same standard.

Engineering electromagnetics increasingly emphasizes multifunctional integration, coordinated system design, and validation under realistic operating conditions. These are important and demanding tasks. Working at a university, I often ask how integration and performance improvements along established routes can build knowledge of lasting value. Leading companies have substantial strengths in fabrication, data, and system engineering. Competing only in these areas may not make the best use of a university’s strengths. We also need to study physical questions that remain poorly understood and may not yet lead to a product. New device principles and measurement methods often depend on this early exploration. Quantum electromagnetics is one direction that I believe deserves sustained effort.

The familiar interaction between field and matter provides a natural entry point. In conventional electromagnetic calculations, permittivity, permeability, conductivity, or equivalent circuit parameters summarize the material response. This lets us analyze complex structures without solving for each microscopic particle. But atomic transitions, coherence, and saturation may require more than fixed material parameters: the relevant quantum states must be evolved explicitly. Likewise, average fields and average power cannot distinguish a source that emits one photon in each trial from one that occasionally emits several. Nor do they suffice to predict coincidence counts at two detectors. These questions also require the quantum statistics of the field.

Electromagnetic design thus gains new tasks: preparing states of field and matter, controlling energy exchange between them, and extracting information through measurement. Atomic levels can provide frequency references. Rydberg-atom responses can be used to measure radio-frequency electric fields. Coupling emitters to resonant structures can control single-photon emission. Photon correlations reveal information beyond average intensity. Electromagnetic structure design is thereby connected to quantum-state preparation, evolution, and readout. Any improvement from quantum methods must be assessed for a specific task, including loss, noise, and the conditions required for implementation.

The knowledge that electromagnetic researchers already possess remains highly useful. The drive experienced by an atom depends on its local electric field. Emitter–cavity coupling requires knowledge of eigenmodes and field normalization. Green functions and scattering analysis determine the channels into which radiation is emitted and how it reaches a receiver. Practical devices also involve loss, packaging, and measurement. In many cases, classical electromagnetic calculations still do most of the spatial field solving, while quantum models add material-state dynamics and field-measurement statistics. Understanding this division of work provides a starting point for further study.

I wrote this book to make these connections clearer. There are many excellent texts on quantum mechanics and quantum optics, including books for engineering readers. Yet readers accustomed to starting from Maxwell’s equations, constitutive relations, and boundary conditions still encounter difficult transitions. How do familiar electromagnetic modes and complex amplitudes relate to photons, operators, and quantum states? How do classically calculated field distributions and scattering matrices enter quantum models? Why can classical results sometimes describe an average response, while fluctuations and correlations require a different description? Explaining these connections early makes many new concepts less unfamiliar.

This book is intended for researchers and engineers with a background in classical and engineering electromagnetics who have not systematically studied quantum mechanics. Field–matter interaction is the central theme. The book distinguishes classical and quantum descriptions of the field, and macroscopic parameters and explicit quantum states for matter, giving four classes of model. At each added level of quantum description, we compare the new unknowns, the information retained, and the questions that become accessible. The same device may require different models for average transmittance, atomic populations, and photon coincidence counts. The quantity being sought should determine the model.

Chapter 1 explains why we move beyond classical electromagnetics and introduces energy levels, photons, and quantum states. Chapter 2 retains classical driving fields and starts with atoms having a few relevant levels. It introduces probability amplitudes, density matrices, and dissipation, then applies them to atomic clocks and Rydberg-atom reception. Chapter 3 develops field quantization from LC circuits and electromagnetic modes. It covers photon statistics, photodetection, propagation and loss in linear media, and two-photon scattering correlations. Chapter 4 places field and matter in a joint quantum model to study energy exchange, spontaneous emission, cavity enhancement, and related photonic devices.

Chapter 5 returns to engineering analysis. It discusses model selection, the connection between classical and quantum calculations, and the assessment of sensitivity and quantum signatures. These discussions turn the preceding concepts into practical calculations and measurements.

“Concise” in the title refers chiefly to the selection of material. The main text focuses on basic concepts, essential equations, and physical meaning, spelling out derivation steps that beginners may find difficult. Where an analytical derivation aids understanding, it is developed explicitly. Where the algebra becomes too lengthy, the text explains how to formulate the equations, solve them numerically, and check the results. Material requiring more mathematical preparation is placed in Further Reading sections. Understanding a model requires knowing both how to calculate with it and what it neglects, including its conditions of validity.

Writing this book has also been an experiment in human–AI collaboration. Artificial intelligence assisted with organizing material, drafting, expanding derivations, and revising the prose. Repeated discussion and revision helped compare explanations. They also reminded me that fluent prose does not guarantee a correct concept, and a complete-looking equation may still confuse operator ordering or omit an approximation. I therefore checked the derivations, physical interpretations, and model assumptions individually, seeking to finalize material that I understood clearly myself. I take responsibility for the selection of material and for the final content.

I expect human–AI collaboration to become increasingly common. Tools make answers easier to obtain, but researchers must still judge why an answer holds and under what conditions. I hope this book helps readers develop that judgment.

During my work on quantum electromagnetics, I have benefited from the guidance, help, and support of Xianliang Wu, Weng Cho Chew, Wallace C. H. Choy, Nicolae Panoiu, and Er-Ping Li. I warmly thank them here. I also thank Professor Jin Pan, who suggested that I write a short book on quantum electromagnetics.

I hope this short book helps colleagues and students in engineering electromagnetics get started. Its main aim will be met if readers can identify when classical analysis is sufficient, when a quantum description is needed, and how to build a model whose results can be calculated, tested, and explained. I am still learning, and my knowledge and experience are limited. Errors and omissions are inevitable; I welcome readers’ corrections and comments.

Wei E. I. Sha

College of Information Science and Electronic Engineering

Zhejiang University

November 2026

Email: [email protected]

Chapter 1 From Classical to Quantum Electromagnetics

Readers familiar with engineering electromagnetics already know a basic method of analysis. Given electromagnetic sources, material parameters, and boundary conditions, solve Maxwell’s equations, then calculate power, impedance, reflection, transmission, and other quantities from the fields. Antenna radiation, waveguide transmission, and cavity design usually follow this approach.

Quantum electromagnetics still needs this knowledge. Electromagnetic analysis determines where to place an atom, which modes the surrounding structure supports, and where energy enters and leaves. This book adds a description of energy changes within matter and of the quantum properties of electromagnetic fields during energy exchange and measurement.

This chapter introduces these ideas through familiar electromagnetic problems. No prior knowledge of quantum mechanics is needed at this stage.

What Does Classical Electromagnetics Leave Out?

To calculate the transmittance of a glass slab, its thickness and refractive index are usually enough. Glass contains many atoms, but we need not solve for the motion and state of each one. Their average effect on the field is already included in the material parameters. Representing complex microscopic processes with a few macroscopic parameters is one reason engineering electromagnetics can handle practical devices.

Some problems, however, concern precisely the details that these parameters leave out.

For example, why does an atomic gas absorb incident light strongly only near certain frequencies? How does the energy inside an atom change while an external field continues to act? If these changes alter the absorption of the gas, a fixed refractive index is no longer sufficient. We must describe the allowed energy states of the atom and how the external field changes them.

Another example is weak-light detection. A power meter reports the average power over a time interval. It does not specify how much light arrives in each experimental trial. A source that emits one photon every time can have the same average power as a source that sometimes emits nothing and sometimes emits several photons. Yet they need not perform equally well when used to transmit information. Besides the spatial field distribution, we must consider the possible outcomes of each measurement and their probabilities. The next section explains what a photon is.

We can also place an atom that has already acquired energy inside a cavity. After the external excitation stops, the atom may still release energy by emitting light. This process is called spontaneous emission. For an antenna, we usually prescribe a current and then calculate its radiation. For the atom, we must also explain how radiation arises and how energy passes from the atom to the field. The surrounding cavity participates in this process, so the emission rate may differ from its free-space value.

These problems have two aspects: how matter stores and exchanges energy, and how the electromagnetic field is excited and measured. Classical electromagnetics provides the basis for propagation and spatial field distributions. Quantum theory supplies the additional description needed for these processes.

Energy Levels and Photons

Why Describe Atomic Energy in Terms of Levels?

In a classical LC circuit, the capacitor voltage can vary continuously, and so can the stored energy. The bound electronic system inside an atom behaves differently: its allowed bound-state energies are discrete. These allowed values are called energy levels.

The lowest energy level corresponds to the ground state, and higher levels correspond to excited states. Figure 1.1 represents two such levels by horizontal lines. Their heights indicate energy, not the spatial positions of electrons or their orbits around the nucleus.

Two atomic energy levels and transitions between them

Figure 1.1. Two atomic energy levels and transitions between them. The upward arrow represents energy absorption; the downward arrow represents energy release. Vertical separation indicates only the energy difference.

A change from one energy level to another is called a transition. If the lower level has energy E1E_1 and the upper level has energy E2E_2, an upward transition requires energy E2−E1E_2-E_1. The transition frequency ff is related to the energy difference by

E2−E1=hf.(1)E_2-E_1=hf. \tag*{(1)}

Here, hh is Planck’s constant, which converts frequency into energy and has units of joule seconds. In terms of angular frequency ω=2πf\omega=2\pi f, the relation is often written as E2−E1=ℏωE_2-E_1=\hbar\omega, where ℏ=h/2π\hbar=h/2\pi. An incident electromagnetic wave near the transition frequency may cause appreciable absorption or stimulated emission, driving transitions between the two levels. This is the starting point for understanding atomic spectral lines, coherent control, and atomic frequency references.

A transition also depends on the electric-field direction and the properties of the atom. Matching the frequency alone does not guarantee that it occurs. Spectral lines also have finite widths. We will discuss transition strengths and linewidths when developing the equations for atom–field interaction.

Two kinds of discreteness must be distinguished here. A classical cavity supports only certain eigenfrequencies because its boundary conditions select the modes. Even after one mode has been selected, classical theory allows its amplitude and stored energy to vary continuously. Discrete atomic levels instead specify which energy states matter can occupy. These are different statements.

Now consider the electromagnetic field. For an ideal mode of frequency ff, quantum theory states that its excitation energy increases or decreases in units of hfhf. Each unit of excitation is called a photon. Adding one photon increases the mode energy by hfhf; removing one decreases it by hfhf. Even at zero photon number, irreducible quantum fluctuations remain. These are vacuum fluctuations.

A photon and a mode therefore describe different aspects of the field. A mode specifies its spatial distribution; the photon number specifies how strongly that mode is excited. In a linear cavity, increasing the photon number from one to two does not give the mode a new spatial shape.

Nor should a photon be pictured simply as a small ball following a definite trajectory. Light diffracts at a slit and interferes between two paths. These propagation properties remain present in single-photon experiments. Each detection records one localized event. The distribution of detection positions emerges only after many trials under the same conditions. The field distribution set by waveguides, lenses, and scatterers still helps determine these detection probabilities.

This does not mean that every weak-light problem requires a complicated quantum calculation. Classical fields often suffice for the average transmission of a linear device. But average power alone cannot tell us whether a pulse contains exactly one photon or whether two detectors can register signals together.

What Else Does a Quantum State Specify?

Having introduced energy levels and photons, we now need the word “state.” A quantum state describes the possible measurement outcomes of a system and the probability of each outcome. It also determines how the system evolves under an external field.

Consider an atom with two relevant energy levels. It may occupy the lower or the upper level. Suitable electromagnetic control can also prepare a coherent superposition of the two states. An energy measurement on this superposition still gives one of the allowed energies in each trial. Repeated preparation and measurement reveal the probabilities of the two outcomes.

A coherent superposition contains more information than the probabilities of occupying the two levels: it also has a relative phase. In engineering electromagnetics, changing the relative phase between two waves changes their interference even if their individual powers remain fixed. Superpositions of quantum states also involve phase, but the quantities being added are probability amplitudes for different states. A probability amplitude is a complex number whose squared magnitude gives the probability of the corresponding outcome. Probabilities alone discard the phase information.

A concrete comparison makes the distinction clear. Prepare one group of atoms with identical coherent control, giving equal probabilities for the lower and upper energy outcomes. Prepare a second group by randomly placing half the atoms in the lower level and half in the upper level. Direct energy measurements give the same statistics for both groups. If suitable coherent control is applied before the energy measurement, however, the first group responds to the relative phase established during preparation. The second group has no such phase dependence.

Recording only the fractions of atoms in each level can therefore miss information that affects their later response. These fractions are called populations; the phase relationships are described by the coherence between levels. Physically, populations describe the stored atomic energy, whereas coherence directly determines the oscillating material polarization. Chapter 2 places populations and coherence in one set of equations and explains how they determine absorption and polarization.

The field has a quantum state of its own. The same mode may contain exactly one photon in every trial, or have an uncertain photon number, as in the field from a conventional coherent source. States with the same average energy can still produce different measurement statistics. Chapter 3 develops their mathematical description. For now, remember that specifying a mode and its average energy does not specify all the quantum information in the field. Figure 1.2 compares two photon-counting distributions.

Two photon-counting distributions

Figure 1.2. Equal average energy need not imply equal photon statistics. Two sources emit pulses at the same frequency and repetition period. In (a), every pulse contains exactly one photon. In (b), half the pulses are empty and half contain two photons. The average powers are equal, but individual detection outcomes differ. The probabilities are obtained from many repeated measurements.

Choosing a Description of Field and Matter

A device need not use the same description for every component. When studying atoms in a vapor cell, for example, we usually do not solve individually for the atoms in its glass walls. A permittivity still describes the glass. The states that require detailed treatment are the atomic levels relevant to the measurement and their evolution.

Considering field and matter separately gives the four combinations in Figure 1.3. “Material parameters” means that quantities such as permittivity summarize the material response. “Quantum states of matter” means explicitly calculating the populations and coherence of the relevant levels. Likewise, the field can be treated classically or described by its quantum state, depending on the problem.

Four descriptions of field and matter, classified by the information retained in the calculation: classical electromagnetics, semiclassical theory, quantum field propagation, and quantum fields and quantum matter, with examples

Figure 1.3. Four descriptions of field and matter, classified by the information retained in the calculation. Different questions about the same device may call for different combinations.

Classical Electromagnetics

When macroscopic parameters describe matter and classical variables describe the field, we recover familiar engineering electromagnetics. Examples include calculating scattering by a dielectric particle with a given permittivity, or antenna impedance for a given conductor geometry. Such calculations do not explicitly track individual quantum states within the material.

Material parameters can incorporate quantum effects of atoms and electrons. Using them in a classical calculation means that these microscopic processes have been summarized in the material response, not that the processes are absent.

Classical Fields Driving Quantum Matter: Semiclassical Theory

Semiclassical theory applies when we need to study atomic level dynamics while the driving field can still be represented as a classical wave. For example, we can first solve for a microwave field in a cavity. We then calculate how that field changes the probabilities of occupying the atomic levels.

If the atoms appreciably affect the field, their average polarization must be fed back into Maxwell’s equations as a source, and the field recalculated. This resembles the self-consistent treatment of fields and material response in engineering electromagnetics. Here, however, the material response follows from changes in atomic states.

Many basic processes in atomic clocks and the average response of Rydberg-atom receivers can be studied at this level. Using quantum atoms does not require quantizing every driving field.

Quantum-Field Propagation in a Given Structure

Sometimes the question is how one or a few photons pass through a waveguide, beam splitter, or scatterer. We then need the quantum state of the field, while parameters such as permittivity can still describe the device materials. Classical electromagnetic calculations determine the effect of the structure on propagation; quantum theory supplies the probabilities of detection outcomes.

This treatment belongs to macroscopic quantum electrodynamics. “Macroscopic” means that the background material is described by its collective response rather than by tracking its atoms individually. If it absorbs, we must include both the associated energy exchange and the field fluctuations accompanying loss, or quantum noise. Chapter 3 starts with lossless modes and gradually develops these ideas.

Joint Evolution of Quantum Fields and Quantum Matter

To study how an atom emits a photon without external driving, or repeatedly exchanges energy with one or a few cavity photons, we must describe both atomic and field states. The field cannot always remain a prescribed drive. Nor can the atom be replaced by a fixed material parameter, or even by self-consistent feedback between its average polarization and a classical field.

This is the fully quantum framework used in this book. It concerns only the atoms and fields relevant to the problem, not every particle in the device. The surrounding substrate, package, and much of the dielectric environment can usually retain a macroscopic electromagnetic description.

Model selection should begin with a precise statement of what is to be calculated. A resonance frequency and an atomic transition probability require different information. So do average power and a single-photon detection probability. Small size, high frequency, or weak signals alone do not determine the model. A suitable model retains the physical processes that control the quantity of interest.

Applications and the Chapters Ahead

Engineering applications of quantum electromagnetics can be viewed through measurement, photonic devices, and quantum circuits.

An atomic clock uses an atomic transition as a frequency reference. The field from a local oscillator interacts with the atoms. Measurements of their state changes reveal frequency errors and provide a feedback signal. The atoms supply the reference, while the electromagnetic system controls the field frequency, amplitude, polarization, and phase. Field uniformity across the interaction region also affects the measurement.

Rydberg atoms occupy highly excited levels, and some transitions between these levels are sensitive to radio-frequency electric fields. A receiver uses the radio-frequency field to change the atomic state, converts that change into a change in probe-light transmission, and infers the electric field from the optical signal. Atomic levels explain only part of the response. Classical electromagnetic analysis is still needed to determine how the cell and surrounding structure modify the local field.

Single-photon devices concern how many photons are generated in each trial, which propagation mode they enter, and how likely they are to be detected. Making a source very weak does not ensure exactly one photon per trial. Useful output requires understanding the emitter’s levels, its interaction with the surrounding structure, and losses during propagation and absorption. These questions directly connect atomic physics with waveguide, cavity, and scattering design.

Superconducting quantum circuits use nonlinear circuit elements to create controllable, unequally spaced discrete energy levels. Two states are selected as a quantum bit, or qubit, the basic unit of quantum information. Microwave fields control these states, while resonators and transmission lines provide coupling and readout. Modes, loss, crosstalk, and package parasitics therefore remain central design problems.

These applications use quantum properties, but performance must still be assessed for the actual task. Field sensing needs both a sufficiently large response and controlled noise. A single-photon source must suppress multiphoton events while maintaining useful output efficiency. Quantum models help us calculate these processes and their limits.

The following chapters develop the central theme of field–matter interaction. Chapter 2 combines classical fields with quantum matter. It retains populations and coherence in the relevant atomic levels, treats driving fields classically, and, where needed, solves self-consistently with Maxwell’s equations through the average polarization. Chapter 3 combines quantum fields with linear media. It retains the field state, describes the background through macroscopic material response, and includes environmental fluctuations associated with loss. Chapter 4 treats quantum fields and quantum matter together, examining excitation exchange, correlations, and photon generation and detection.

These combinations answer different questions. For the same device, average absorption may be accessible with the equations of Chapter 2. Single-photon propagation or emitter–field correlations require the corresponding field-state information. Chapter 5 draws these models together, showing how to select degrees of freedom, connect electromagnetic and quantum calculations, and assess results for a specific task. Numerical methods are introduced only to the extent needed to obtain the results.

For readers trained in engineering electromagnetics, familiar fields and boundary conditions remain the starting point. As the questions extend from propagation to energy exchange and measurement, we add energy levels, quantum states, and their equations of motion. This progression makes the role of each part of the theory clear in device analysis.

Classical Fields and Quantum Matter

In engineering electromagnetics, specifying a material’s permittivity allows us to solve for field propagation and scattering. But permittivity itself has a physical origin. An applied electric field drives charges within matter, producing polarization; that polarization then acts as a source that affects the electromagnetic field. For an atomic gas, resolving the absorption, phase response, and saturation of a particular transition requires an explicit calculation of this part of the material response.

Chapter 1 explained that energy-level populations represent the energy stored in an atom, while coherence between levels corresponds to polarization that oscillates in time. This chapter expresses that statement in equations. The electromagnetic field remains classical, while the atom is described by a few relevant energy levels and their probability amplitudes. We begin with a two-level atom, then consider atomic clocks and Rydberg-atom receivers. The quantum mechanics needed for each model is introduced along the way.

From Energy Levels to State Equations

Probability Amplitudes and State Vectors

Suppose that only two atomic energy levels participate in the process of interest. Denote the lower level by gg and the upper level by ee. Set the lower-level energy to zero and the upper-level energy to ℏω0\hbar\omega_0. Here ℏ=h/(2π)\hbar= h/(2\pi) is the reduced Planck constant, and ω0=2πf0\omega_0 = 2\pi f_0 is the atomic transition angular frequency (so ℏω0=hf0\hbar\omega_0 = hf_0, consistent with Chapter 1). The choice of energy zero does not affect the energy difference between the levels.

Represent the atomic state by two complex numbers, cg(t)c_g(t) and ce(t)c_e(t): an energy measurement finds the lower level with probability ∣cg∣2|c_g|^2 and the upper level with probability ∣ce∣2|c_e|^2. These complex numbers are probability amplitudes. When only these two levels are considered, the total probability is one, so

∣cg∣2+∣ce∣2=1.(2)|c_g|^2 + |c_e|^2 = 1. \tag*{(2)}

The probability amplitudes can be assembled into a column vector. Quantum mechanics commonly uses the following notation:

∣ψ⟩=cg∣g⟩+ce∣e⟩⟷(cgce),∣g⟩=(10),∣e⟩=(01).(3)|\psi\rangle= c_g|g\rangle+ c_e|e\rangle\longleftrightarrow\begin{pmatrix} c_g \\ c_e \end{pmatrix}, \quad|g\rangle= \begin{pmatrix} 1 \\ 0 \end{pmatrix}, \quad|e\rangle= \begin{pmatrix} 0 \\ 1 \end{pmatrix}. \tag*{(3)}

The symbol ∣ψ⟩|\psi\rangle is called a state vector. For this model, it is simply a column vector containing the two probability amplitudes; ∣g⟩|g\rangle and ∣e⟩|e\rangle are the chosen basis vectors. Additional levels require only additional components.

Transposing the column vector and taking the complex conjugate of each component gives the row vector ⟨ψ∣=(cg∗,ce∗)\langle\psi| = (c_g^*, c_e^*). The asterisk denotes complex conjugation. The ordinary vector inner product can therefore be written as

⟨ψ∣ψ⟩=(cg∗,ce∗)(cgce)=∣cg∣2+∣ce∣2=1.\langle\psi|\psi\rangle= (c_g^*,c_e^*) \begin{pmatrix} c_g \\ c_e \end{pmatrix} = |c_g|^2 + |c_e|^2 = 1.

Similarly, ⟨g∣ψ⟩=cg\langle g|\psi\rangle= c_g and ⟨e∣ψ⟩=ce\langle e|\psi\rangle= c_e extract the corresponding components of the state. Most of the angle-bracket expressions below can be understood as row and column vector operations. A bra on the left and a ket on the right, as in ⟨ϕ∣ψ⟩\langle\phi|\psi\rangle, form a row-times-column inner product, whose result is a complex number. A ket on the left and a bra on the right, as in ∣ψ⟩⟨ψ∣|\psi\rangle\langle\psi| in Eq. (19) below, form a column-times-row outer product, whose result is a matrix. The rule that probability equals the squared magnitude of the corresponding amplitude is a basic probabilistic postulate of quantum mechanics; it cannot be derived solely from classical electromagnetic energy formulas.

In engineering electromagnetics, the complex amplitudes of several modes have both magnitudes and phases. Complex information must also be retained here, but the squared magnitudes of cgc_g and cec_e give measurement probabilities and cannot be identified directly with electric-field energy. Multiplying both probability amplitudes by the same phase factor does not change measurement results; their relative phase, however, affects subsequent driven evolution.

How the Hamiltonian Matrix Determines Evolution

Neglecting dissipation for now, the probability amplitudes obey the Schrödinger equation:

iℏddt∣ψ⟩=H∣ψ⟩,iℏddt(cgce)=(HggHgeHegHee)(cgce).(4)\mathrm{i}\hbar\frac{\mathrm{d}}{\mathrm{d}t}|\psi\rangle= H|\psi\rangle,\qquad \mathrm{i}\hbar\frac{\mathrm{d}}{\mathrm{d}t} \begin{pmatrix} c_g \\ c_e \end{pmatrix} = \begin{pmatrix} H_{gg} & H_{ge} \\ H_{eg} & H_{ee} \end{pmatrix} \begin{pmatrix} c_g \\ c_e \end{pmatrix}. \tag*{(4)}

Here i\mathrm{i} is the imaginary unit, and HH is the Hamiltonian matrix; every matrix element has units of energy. In the chosen energy-level basis, its diagonal elements in the absence of an applied field give the level energies, while off-diagonal elements introduced by the field describe coupling between states. Equation (4) is the fundamental law of quantum evolution used in this chapter. It cannot be derived from Maxwell’s equations alone; once it is given, the few-level equations below follow by algebra.

Expanding the matrix product gives

iℏc˙g=Hggcg+Hgece,iℏc˙e=Hegcg+Heece.\mathrm{i}\hbar\dot{c}_g = H_{gg}c_g + H_{ge}c_e,\qquad\mathrm{i}\hbar\dot{c}_e = H_{eg}c_g + H_{ee}c_e.

For example, HegcgH_{eg}c_g in the second equation is the contribution of the lower-level component to the change in the upper-level probability amplitude. For evolution without dissipation, HH equals its conjugate transpose, H=H†H = H^\dagger; thus its diagonal elements are real and Hge=Heg∗H_{ge} = H_{eg}^{*}. Substituting the state equation ∣ψ⟩˙=−(i/ℏ)H∣ψ⟩\dot{|\psi\rangle} = -(\mathrm{i}/\hbar)H|\psi\rangle and its conjugate transpose ⟨ψ∣˙=(i/ℏ)⟨ψ∣H†\dot{\langle\psi|} = (\mathrm{i}/\hbar)\langle\psi|H^\dagger (complex conjugation reverses the sign of the imaginary unit, and transposition reverses the order of a matrix product) into the time derivative of the inner product gives

ddt⟨ψ∣ψ⟩=iℏ⟨ψ∣(H†−H)∣ψ⟩=0.\frac{\mathrm{d}}{\mathrm{d}t}\langle\psi|\psi\rangle= \frac{\mathrm{i}}{\hbar}\langle\psi|(H^\dagger-H)|\psi\rangle= 0.

Thus population can be exchanged between the levels while the total probability remains one.

Without an applied field, H=diag⁡(0,ℏω0)H = \operatorname{diag}(0,\hbar\omega_0). Substitution into Eq. (4) gives cg(t)=cg(0)c_g(t) = c_g(0) and ce(t)=ce(0)e−iω0tc_e(t) = c_e(0)e^{-\mathrm{i}\omega_0t}. The populations remain unchanged, but the relative phase of the upper- and lower-level probability amplitudes evolves in time. Equations for populations alone would omit this information.

An applied field generally introduces off-diagonal elements into HH, coupling the two probability amplitudes. The solution procedure resembles that for coupled-mode equations: given the initial state and coupling coefficients, integrate a system of ordinary differential equations with complex coefficients. What distinguishes the quantum model is that these coefficients correspond to energy levels and transitions, and the resulting amplitudes determine measurement probabilities.

Where Energy Levels and Coupling Coefficients Come From

The two-level model has not yet described the atom’s internal spatial structure. Starting from the motion of charged particles requires a wavefunction ψ(r,t)\psi(\mathbf{r},t). The quantity ∣ψ(r,t)∣2d3r|\psi(\mathbf{r},t)|^2\mathrm{d}^3r is the probability of finding the particle in a small volume near position r\mathbf{r}. The wavefunction is neither the actual electromagnetic field nor the particle’s trajectory.

Consider a nonrelativistic particle of mass mm, confined by a potential energy V(r)V(\mathbf{r}). In the absence of an applied field, its stationary states satisfy

[−ℏ22m∇2+V(r)]φn(r)=Enφn(r).(5)\left[-\frac{\hbar^2}{2m}\nabla^2+V(\mathbf{r})\right]\varphi_n(\mathbf{r})=E_n\varphi_n(\mathbf{r}). \tag*{(5)}

This is an eigenvalue problem. The first term on the left corresponds to the classical kinetic energy p2/(2m)p^2/(2m), obtained by squaring the momentum operator p^=−iℏ∇\hat{\mathbf{p}}=-i\hbar\nabla introduced below; the second term is the confining potential energy. Its form is analogous to the Helmholtz equation used to find cavity eigenmodes in electromagnetics. Here EnE_n is an allowed energy and φn\varphi_n the corresponding spatial wavefunction. The wavefunctions can be chosen normalized and mutually orthogonal, so that ∫φm∗φnd3r=δmn\int\varphi_m^*\varphi_n\mathrm{d}^3r=\delta_{mn}, where m=nm=n gives δmn=1\delta_{mn}=1 and otherwise the result is zero.

An operator represents a linear operation on a state. For example, the momentum operator p^=−iℏ∇\hat{\mathbf{p}}=-i\hbar\nabla differentiates the wavefunction. Once a basis is chosen, an operator can be represented by a matrix. Below, a hat emphasizes an operator, whereas HH directly denotes its matrix in the chosen energy-level basis.

To obtain equations for a finite set of probability amplitudes, expand the wavefunction as ψ(r,t)=∑ncn(t)φn(r)\psi(\mathbf{r},t)=\sum_n c_n(t)\varphi_n(\mathbf{r}). Denote the field-free energy operator on the left of Eq. (5) by H^0\hat{H}_0, and the applied-field interaction by H^int\hat{H}_{\mathrm{int}}. Substitute the expansion into the spatial form of the time-dependent Schrödinger equation iℏ∂tψ=(H^0+H^int)ψi\hbar\partial_t\psi=(\hat{H}_0+\hat{H}_{\mathrm{int}})\psi, multiply from the left by φm∗\varphi_m^*, and integrate over space. Using orthogonality gives

References

  1. [1]Jaynes E T, Cummings F W. Comparison of quantum and semiclassical radiation theories with application to the beam maser. Proceedings of the IEEE, 1963, 51: 89–109. doi:10.1109/PROC.1963.1664.DOI
  2. [2]Dung H T, Knöll L, Welsch D G. Spontaneous decay in the presence of dispersing and absorbing bodies: General theory and application to a spherical cavity. Physical Review A, 2000, 62: 053804. doi:10.1103/PhysRevA.62.053804. https://arxiv.org/abs/quant-ph/0003138.
  3. [3]Reiserer A, Rempe G. Cavity-based quantum networks with single atoms and optical photons. Reviews of Modern Physics, 2015, 87: 1379–1418. doi:10.1103/RevModPhys.87.1379. https://arxiv.org/abs/1412.2889.
  4. [4]QuTiP documentation. Two-time correlation functions. https://qutip.readthedocs.io/en/stable/guide/guide-correlation.html. Accessed November 11, 2026.
  5. [5]Birnbaum K M, Boca A, Miller R, et al. Photon blockade in an optical cavity with one trapped atom. Nature, 2005, 436: 87–90. doi:10.1038/nature03804.DOI
  6. [6]Blais A, Grimsmo A L, Girvin S M, et al. Circuit quantum electrodynamics. Reviews of Modern Physics, 2021, 93: 025005. doi:10.1103/RevModPhys.93.025005. https://arxiv.org/abs/2005.12667.
  7. [7]Ficek Z, Tanaś R. Entangled states and collective nonclassical effects in two-atom systems. Physics Reports, 2002, 372: 369–443. doi:10.1016/S0370-1573(02)00368-X. https://arxiv.org/abs/quant-ph/0302082.

Paper details

vlnc:W_nfb2vd7e
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 8

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