Lehmann Representation

Understanding the behavior of complex quantum systems, such as the sea of interacting electrons in a crystal, presents a formidable challenge. Probing such a system yields an intricate response that is difficult to interpret directly. This complexity creates a knowledge gap: How can we translate the microscopic chaos of countless interacting particles into a clean, understandable, and measurable "fingerprint"? The answer lies in one of the most elegant and powerful tools in theoretical physics: the Lehmann representation. It acts as a universal dictionary, translating the messy response of any interacting quantum system into the fundamental language of its exact energy levels and the quantum leaps between them. This article unveils the power of this formalism. The first chapter explores its core principles and mechanisms, showing how it uses the completeness of quantum states to reveal the intrinsic spectrum of a system. The second chapter demonstrates its far-reaching applications, from explaining experimental spectra in materials to unifying concepts across different fields of science.

Imagine you are faced with a fabulously complex object—say, a crystal containing a sea of interacting electrons, a quantum soup teeming with a billion billion particles all pushing and pulling on each other. How could you possibly hope to understand its behavior? If you were to “poke” it—by shining light on it, or trying to inject an electron—the response would be bewilderingly complicated. It’s like striking a cathedral bell; the resulting sound is rich and intricate, a superposition of countless vibrations. You wouldn’t try to track the motion of every single atom in the bell. Instead, you would do a spectrum analysis, breaking the sound down into its fundamental frequencies, its resonant modes. These modes are the bell’s true “fingerprint.”

The Lehmann representation is the physicist’s way of performing a spectrum analysis on a quantum many-body system. It is a breathtakingly general and exact theoretical tool that acts as a universal dictionary. It translates the messy, complicated response of an interacting system into a clean, fundamental language: the language of the system’s exact energy levels and the quantum leaps between them. It tells us that no matter how complex the interactions are, the system's response is ultimately governed by its own intrinsic set of stationary states.

So, how does this magical dictionary work? The secret lies in one of the most powerful, yet beautifully simple, ideas in all of quantum mechanics: ​​completeness​​. The Schrödinger equation tells us that any quantum system has a set of special states, the ​​eigenstates​​ $|n\rangle$, in which the energy is precisely defined. These states are stationary; once in an eigenstate, a system stays there forever unless disturbed. The principle of completeness declares that any possible state of the system can be written as a superposition of these fundamental eigenstates. Mathematically, this is captured by the elegant statement that the sum of projectors onto all eigenstates gives the identity operator, $\sum_n |n\rangle \langle n| = \mathbb{I}$.

This equation is the key. In physics, when we want to understand how a disturbance propagates, we compute a mathematical object called a ​​correlation function​​, or ​​Green's function​​. It might, for instance, tell us the probability amplitude for a particle created at one point in space and time to be found at another. Let's consider a generic correlation function $\langle A(t) B(0) \rangle$, which correlates an event $B$ at time zero with an event $A$ at a later time $t$. In its raw form, this expression is a black box.

The genius of the Lehmann representation is to insert the identity operator, $\sum_m |m\rangle \langle m|$, right in the middle of our correlation function. It is like placing a quantum prism in the path of a light beam. Suddenly, the seemingly chaotic response is resolved into its pure, constituent parts:

Here, $\rho$ is the density matrix that describes the statistical state of the system (for example, the Boltzmann factor $e^{-\beta E_n}/Z$ for a system in thermal equilibrium). Using the law of quantum time evolution, $A(t) = e^{iHt} A(0) e^{-iHt}$, the expression transforms into a sum over all possible quantum leaps:

where $p_n$ is the probability of the system starting in state $|n\rangle$. Look what has happened! The complicated time dependence has been resolved into a set of simple oscillations, each with a frequency given by an energy difference $E_n - E_m$. The system is telling us its secrets. It can only absorb or emit energy in discrete packets corresponding to the energy gaps between its exact stationary states.

Physicists find it more convenient to work in the frequency domain than the time domain. By taking the Fourier transform of the correlation function, we distill all this information into a single, magnificent object: the ​​spectral function​​, often denoted $A(\omega)$. For the important case of a single-particle Green's function, which describes adding or removing a particle, the spectral function takes the following schematic form:

Let's dissect this expression, as it's the heart of the matter. (Here we use the zero-temperature ground state $\Psi_0^N$ for clarity, but the structure is identical at finite temperature, just with more starting states and thermal weights.

1. ​​The Delta Functions, $\delta(\dots)$​​: These are the sharp peaks of our spectrum. A delta function $\delta(x-a)$ is zero everywhere except at $x=a$. This term tells us that the system's response is zero unless the frequency $\omega$ of our probe precisely matches the energy required to transition from the initial $N$-particle state to an exact $(N+1)$- or $(N-1)$-particle state. This is a profound statement of energy conservation.

2. ​​The Matrix Elements, $|\langle \dots \rangle|^2$​​: These numbers are the intensities of the spectral peaks. The term $|\langle \Psi_m^{N+1} | c^\dagger | \Psi_0^N \rangle|^2$ is the quantum mechanical probability for a transition to occur between the initial state $\Psi_0^N$ and a specific final state $\Psi_m^{N+1}$ when we attempt to add a particle using the creation operator $c^\dagger$. If this matrix element is zero, that "quantum leap" is forbidden—it’s a selection rule. This is why the measured spectrum isn't just a uniform sea of all possible energy differences; it has a rich structure of bright and dark lines.

This anatomy has a direct, beautiful connection to experiment. Spectroscopies like ​​Angle-Resolved Photoemission Spectroscopy (ARPES)​​ work by knocking an electron out of a material. The energies of the ejected electrons map directly onto the "removal" part of the spectral function. Conversely, ​​Inverse Photoemission Spectroscopy (IPES)​​ probes the energies available when we add an electron into the material, mapping out the "addition" part of the spectrum. The Lehmann representation tells us that these experimental spectra are, quite literally, a direct measurement of the system’s quantum energy levels and transition probabilities. For a finite molecule, these levels are discrete, corresponding to ionization potentials and electron affinities. For an infinite crystal, they broaden into continuous bands, but the principle is the same.

Because the Lehmann representation is built from the bedrock axioms of quantum mechanics, it reveals several exact, non-negotiable laws that any physical system—and any valid approximation of it—must obey. These are not just mathematical curiosities; they are deep consistency checks on our understanding of nature.

The Law of Positivity

Look again at the structure of the spectral function. It is a sum of terms of the form $|\langle \dots \rangle|^2$. In quantum mechanics, the squared modulus of any complex number (like a matrix element) is always real and non-negative. This represents a probability, and probabilities can't be negative! Therefore, the spectral function itself must be ​​positive definite​​: $A(\omega) \ge 0$. A theory or an approximation that predicts a negative spectral function for any frequency $\omega$ is not just wrong, it is fundamentally unphysical. It's equivalent to predicting a negative probability for an observable event.

This law of positivity is deeply connected to ​​causality​​—the principle that an effect cannot precede its cause. It can be shown that any theoretical approximation that violates causality will also, under scrutiny, violate the positivity of the spectral function. So, a simple check—is $A(\omega)$ always positive?—becomes a powerful test for the physical soundness of a complex theory.

The Law of Sum Rules

What happens if we add up all the spectral weight over all possible frequencies? Let's take the integral of the spectral function:

This is a famous ​​sum rule​​. Its derivation relies on the fundamental commutation relations of the particle [creation and annihilation operators](@article_id:180463). Physically, it means that for a given momentum $\mathbf{k}$, the total probability of something happening—either successfully adding a particle into some available state, or successfully removing one from some occupied state—must be exactly one. The spectral weight can be shuffled around between different energies by interactions, but the total amount of it is conserved.

This is an incredibly powerful constraint. In modern physics, we often cannot solve the full interacting problem, so we resort to approximations, often using complex ​​Feynman diagrams​​. How do we know if our approximation is any good? One way is to check if it respects the exact sum rule. The most robust "conserving approximations" are specifically designed to guarantee that this sum rule is perfectly obeyed, giving us confidence that even though we are approximating, we are not violating a fundamental conservation law of the quantum world.

Perhaps the most beautiful aspect of the Lehmann representation is its universality. It reveals the deep, unified structure of quantum field theory that underlies seemingly disparate areas of physics.

First, within many-body theory itself, there are many different types of Green's functions tailored for different purposes: ​​retarded functions​​ for causal responses, ​​advanced functions​​, and ​​imaginary-time functions​​ for thermal equilibrium calculations. The Lehmann formalism shows that all of these can be constructed from the very same spectral function $A(\omega)$. It is the central object that contains all the physical information, and the different Green's functions are just different "views" of it. These views are interconnected by profound mathematical relations, such as the Kramers-Kronig relations. The general structure also holds for both ​​fermions​​ (like electrons) and ​​bosons​​ (like photons or phonons), with only a subtle sign change related to their exchange statistics.

Even more remarkably, this framework extends far beyond the realm of condensed matter. In the world of high-energy particle physics, there is an analogous formula called the ​​Källén-Lehmann representation​​. It describes the propagation of elementary particles like electrons or quarks through the vacuum. It, too, is derived by inserting a complete set of states—in this case, eigenstates of four-momentum. It also leads to a positive spectral density, which tells you how a particle's properties are "dressed" by its interaction with the quantum vacuum. The language is different—Lorentz invariance replaces crystal symmetry, and the vacuum state replaces the thermal ensemble—but the core physical and mathematical principles of completeness and positivity are identical.

This is the kind of unifying revelation that makes physics so profound. From the hum of a crystal to the propagation of a fundamental particle in the void, nature uses the same deep and elegant blueprint. The Lehmann representation gives us a key to decipher it, turning a complex, interacting mess into a beautiful and intelligible spectrum of possibilities.

Now that we have acquainted ourselves with the formal machinery of the Lehmann representation, we might be tempted to leave it as a piece of abstract art, beautiful but remote. To do so would be a tremendous mistake. This framework is not a museum piece; it is a master key, unlocking doors in nearly every corner of modern physics and chemistry. It is the language we use to translate the deep, hidden quantum choreography of particles into the observable world of experiment.

Think of a complex system of interacting electrons as a grand orchestra. Left to its own devices, its music is a cacophony of individual motions. The Lehmann representation is our prism, our spectral analyzer. It takes the total, seemingly chaotic response of the system and decomposes it into its fundamental "notes"—the elementary excitations. Each note has a pitch (its energy) and a volume (its spectral weight), and by listening to this music, we can understand the story of the orchestra: who the players are, how they are arranged, and how they talk to one another.

Let's start with the simplest possible instrument: two tiny magnets—two spins—paired up. In many materials, neighboring electrons behave this way, forming an antiferromagnetic "dimer" where their spins prefer to point in opposite directions, creating a quiet, non-magnetic ground state called a singlet. This is the simplest manifestation of magnetism. What does it take to disturb this placid state? We need to inject just the right amount of energy to flip one of the spins, creating an excited triplet state. The Lehmann representation tells us precisely what an experimentalist will see: a spectrum with a single, sharp peak. The position of this peak is not arbitrary; it sits exactly at the energy difference, $J$, between the singlet and triplet states. It is the characteristic "note" of the spin pair.

This is not just a theoretical fantasy. Techniques like inelastic neutron scattering are designed to do exactly this—to "ping" the magnetic system and listen for the characteristic energies it can absorb. The spectra they measure show these sharp peaks, revealing the exchange-energy "bonds" holding the magnetic order together.

Furthermore, the Lehmann representation reveals a deep and universal law of nature when we consider the system at a finite temperature. The spectrum shows not only a peak for absorbing energy (Stokes scattering) but also for emitting it (anti-Stokes scattering). The principle of detailed balance, a direct consequence of the Lehmann formalism, tells us that the intensity of the anti-Stokes line is suppressed by a factor of $\exp(-\hbar\omega / (k_B T))$ compared to the Stokes line. This makes perfect sense! To emit energy, the system must already be in an excited state, and the probability of that is governed by the Boltzmann factor. This beautiful connection between the spectrum of fluctuations in a material (what we measure in Raman scattering) and the energy it dissipates is an example of the famous Fluctuation-Dissipation Theorem, a cornerstone of statistical mechanics. The Lehmann representation provides the microscopic justification for it.

The two-spin dimer is a lonely soloist. What happens when we have a whole sea of electrons, as in a piece of metal? If the electrons didn't interact—if every musician in our orchestra played their own tune without listening to their neighbors—the spectrum would simply be the sum of all their individual notes. The Lehmann representation for such a non-interacting gas shows poles corresponding to the energy needed to add an electron to an empty state or create a hole in a filled one.

But electrons do interact. They are charged particles, and they repel each other with a vengeance. This is where the music gets truly interesting. Instead of a cacophony of individual tunes, the orchestra can begin to play in unison, producing powerful, collective oscillations. The most famous of these is the ​​plasmon​​. A plasmon is not the motion of any single electron, but a rhythmic, coordinated wave of the entire electron density, like a wave sweeping across the surface of the sea. The Lehmann representation for the system's density response shows us why this appears not as a jumble of notes, but as a single, powerful, sharp peak in the spectrum, standing proud and distinct from the low-energy murmur of individual electron-hole excitations. When this collective mode's energy overlaps with the continuum of individual excitations, it can decay, a process known as Landau damping. In our musical analogy, this is when the beautiful collective chord dissolves back into the noise of the crowd.

Interactions can do more than just organize; they can create entirely new soloists out of thin air. Consider the ​​Kondo effect​​, a celebrated puzzle in condensed matter physics. When a single magnetic atom is placed in a non-magnetic metal, something strange happens at low temperatures. The electrons in the metal conspire to screen the impurity's spin, forming a complex, highly correlated many-body state. The Lehmann representation provides the sharpest view of this phenomenon. Without interactions ($U=0$), the impurity's spectrum is a simple, broad resonance. But when the interaction is turned on, the Lehmann representation shows that a new, sharp, and dramatic resonance appears right at the Fermi energy—the "zero-energy" reference point of the metal. This is the Kondo resonance. It is a purely many-body effect, a spectral signature of a collective dance between the impurity and the entire Fermi sea, a phenomenon utterly absent in a world without interactions.

Sometimes the interactions are so strong that the orchestra simply refuses to play. This is the ​​Mott insulator​​. According to simple band theory, these materials should be metals, but the enormous energy cost $U$ of putting two electrons on the same atom brings the symphony to a grinding halt. The electrons are "jammed," and the material cannot conduct electricity. What happens if we gently "dope" the system by removing a few electrons? The Lehmann framework, via its associated sum rules, gives a staggeringly simple and beautiful answer. New states appear inside the energy gap that was created by the interaction, and the total spectral weight of these new "in-gap" states is exactly proportional to the number of holes we introduced, $2\delta$. This "spectral weight transfer" is a central theme in the physics of high-temperature superconductors, where doping a Mott insulator is the first step toward creating one of the most mysterious and promising materials known to science.

You might think that this machinery, born to describe electrons in a solid, would be confined to the messy world of materials. But the most beautiful ideas in physics have a habit of showing up everywhere.

In ​​quantum chemistry​​, a primary goal is to predict the color of a molecule, which is determined by the energies it can absorb from light. Modern computational methods like Time-Dependent Density Functional Theory (TDDFT) are the workhorses for this task. At its core, the Lehmann formalism helps us understand why these sophisticated methods work. A simple calculation (the "Kohn-Sham" spectrum) often gives the wrong colors. TDDFT corrects this by accounting for how the elementary single-particle excitations mix and talk to each other. The Lehmann representation of the true response function shows poles (the correct absorption energies) that are shifted away from the simple ones, and the spectral weight is reshuffled among them. Advanced versions of the theory even use a frequency-dependent "kernel" that, as the Lehmann structure makes clear, can introduce new poles to describe complex excitations that are crucial for photochemistry and materials design.

The representation is not just for describing dynamics and excitations. By integrating the spectral function over all occupied states, we get back a fundamental static property: the total number of particles, $N$, in the system. This might seem trivial, but it's a profound consistency check, a mathematical guarantee that our dynamic picture of excitations is perfectly compatible with the simple act of counting the particles that are present. It is a glimpse of the internal logical perfection of quantum theory.

Perhaps the most breathtaking application lies in ​​Quantum Field Theory (QFT)​​, the language of fundamental particles. The Källén-Lehmann representation applies here as well, but the players are not electrons in a solid; they are the elementary particles of the Standard Model. What we call an "electron" is not a simple, bare point charge. It is a more complex object, "dressed" in a shimmering cloud of virtual particles that it constantly emits and reabsorbs. The full propagator, or correlation function, for this dressed electron has a Lehmann representation. It shows a single-particle pole, corresponding to the stable electron we know and love, plus a continuum representing the states with the electron plus one or more photons or other particles. The residue of the single-particle pole is a number, $Z$, called the wavefunction renormalization constant. This number, which is always less than one, quantifies the overlap between the "bare" mathematical electron and the "dressed," physical electron. It tells us what fraction of the time our physical electron is actually just its bare self. The fact that the same mathematical structure describes the nature of a fundamental particle and the magnetic properties of a crystal is a stunning testament to the unity of physics.

From a simple spin-pair to the dance of plasmons, from the color of a molecule to the very fabric of an elementary particle, the Lehmann representation is our universal guide. It translates the abstract score of a system's Hamiltonian into the rich and observable music of its excitations, revealing the deep, interconnected beauty of the quantum world.