Polaron Formation and Effective Mass Renormalization

Recommended resources

The freely available resources this lesson will direct you to.

Hello! Welcome to our next lesson.

In our previous session, we established how electron-phonon scattering gives rise to temperature-dependent resistivity, culminating in the Bloch-Grüneisen law. Throughout that discussion, we treated the electron and the lattice vibrations (phonons) as distinct entities that collide and exchange energy and momentum.

Today, we will delve into a more intimate consequence of this interaction, prevalent in polar and ionic crystals. We will explore the scenario where the electron and the lattice distortion it creates are so strongly coupled that they move as a single, inseparable entity. This composite object is a new quasiparticle called a polaron.

Our goal is to understand how this quasiparticle forms and, crucially, to calculate how the electron-phonon interaction renormalizes the electron's effective mass. This concept of "renormalization" — where interactions modify the properties of a bare particle — is a cornerstone of condensed matter and quantum field theory.

1. The Polaron Concept: A Dressed Electron

Imagine a conduction electron moving through the crystal lattice of a polar material like NaCl or MgO. The electron's negative charge will attract the nearby positive ions and repel the negative ones. This displacement of ions creates a local polarization field, which is equivalent to a cloud of virtual LO (longitudinal optical) phonons surrounding the electron. As the electron moves, this polarization cloud moves with it.

The electron, "dressed" by its self-induced phonon cloud, is what we call a polaron. It is heavier and less mobile than the original "bare" band electron because it has to drag the lattice distortion along.

Visualizing Polaron Formation — This image illustrates the formation of a polaron. On the left, a single electron polarizes the surrounding ionic lattice. On the right, this electron and its associated polarization cloud are treated as a single quasiparticle, the polaron.

The canonical model to describe this phenomenon, for an electron interacting with LO phonons, is the Fröhlich Hamiltonian. To understand its structure and the key parameter that governs the interaction strength, please start by reading the introductory section of the following review article.

Polarons

This article, 'Polarons' by J. T. Devreese, is a classic review of the topic. The first section introduces the polaron concept and lays out the fundamental Fröhlich Hamiltonian.

Please read Section 1.1, 'The Polaron Concept' (pages 2-3). Pay close attention to the form of the Hamiltonian in Eq. (1a) and the definition of the dimensionless Fröhlich coupling constant, α, in Eq. (1c).

As you saw, the Fröhlich Hamiltonian has three parts:

H = Electron KE \frac{p ^{2}}{2 m _{b}} + Phonon Energy k \sum ℏ ω_{L O} a_{k}^{†} a_{k} + Interaction k \sum (V_{k} a_{k} e^{i k \cdot r} + h.c.)

Given your background in QFT, this structure should be familiar: it describes a fermion (the electron) interacting with a scalar boson field (the LO phonons).

The strength of this interaction is captured by the dimensionless Fröhlich coupling constant, $α$ :

α = \frac{e ^{2}}{ℏ c} \frac{m _{b} c ^{2}}{2ℏ ω _{L O}} (\frac{1}{ϵ _{\infty}} - \frac{1}{ϵ _{0}})

This constant neatly packages the relevant material properties:

The electron's band mass, $m_{b}$ .
The LO phonon frequency, $ω_{L O}$ .
The material's dielectric properties, specifically the difference between the high-frequency ( $ϵ_{\infty}$ ) and static ( $ϵ_{0}$ ) dielectric constants, which quantifies the strength of the ionic polarization.

Typically, $α < 1$ is considered weak coupling, $1 < α < 6$ is intermediate, and $α > 6$ is strong coupling.

2. Polaron Self-Energy and Mass Renormalization

Now, let's perform the central calculation of this lesson: finding the polaron's energy and effective mass. We will work in the weak-coupling limit ( $α ≪ 1$ ), where we can use second-order perturbation theory.

Ground State Energy Shift (Self-Energy)

First, consider an electron at rest at the bottom of the conduction band ( $p = 0$ ) in a crystal with no phonons. This is our unperturbed ground state, $∣ i ⟩ = ∣ p = 0; {0}_{ph} ⟩$ , with energy $E_{i} = 0$ .

The interaction term can annihilate a phonon and change the electron's momentum, or create a phonon and change the electron's momentum. Starting from the vacuum, it can only act via the creation term $V_{k}^{*} a_{k}^{†} e^{- i k \cdot r}$ . This process takes the system to an excited state $∣ f ⟩ = ∣ - ℏ k; {1_{k}}_{ph} ⟩$ , which contains one phonon of wavevector $k$ and an electron with momentum $- ℏ k$ (to conserve total momentum).

The energy of this final state is $E_{f} = \frac{ℏ ^{2} k ^{2}}{2 m _{b}} + ℏ ω_{L O}$ . Using second-order perturbation theory, the energy shift of the ground state is:

E_{0} = f \sum \frac{∣ ⟨ f ∣ H _{int} ∣ i ⟩ ∣ ^{2}}{E _{i} - E _{f}} = k \sum \frac{∣ V _{k} ∣ ^{2}}{0 - ( \frac{ℏ ^{2} k ^{2}}{2 m _{b}} + ℏ ω _{L O} )}

The following resource provides a clear, step-by-step evaluation of this sum.

Renormalization of electronic properties of semiconductors ...

The PhD thesis 'Renormalization of electronic properties...' by Juan Pablo Nery contains a detailed derivation of the weak-coupling polaron self-energy, which is exactly the calculation we need.

Please read Section 1.4.2, 'Zero-point renormalization of the conduction band minimum' (page 16). Follow the derivation in Eq. (1.67) to see how the sum over k-states is converted to an integral and evaluated.

The result of this calculation is remarkably simple:

E_{0} = - α ℏ ω_{L O}

This is the polaron self-energy. It tells us that the ground state energy of the polaron is lower than that of the non-interacting electron by $α ℏ ω_{L O}$ . This is the binding energy of the polaron; the energy released when the electron "dresses" itself with phonons.

Effective Mass Renormalization

To find the effective mass, we need to know how the polaron's energy depends on its momentum, $E (P)$ . The effective mass is defined by the curvature of the energy-momentum dispersion near $P = 0$ : $E (P) \approx E_{0} + P^{2} / (2 m^{*})$ .

We can calculate this by applying perturbation theory to an electron with a small initial momentum $P$ , but the algebra becomes more involved. For our purposes, we will turn back to the review by Devreese, which summarizes the key results of this calculation.

Polarons

Let's return to the Devreese review to see the final results for both the polaron energy and effective mass in the weak-coupling limit.

Please read the beginning of Section 1.2.2, 'Weak coupling' (page 5-6). Focus on the results from Fröhlich and Lee, Low, and Pines (LLP) given in Eqs. (5a) and (5b), and the more accurate expansions in Eqs. (7a) and (7d).

As you've read, the weak-coupling approximation for the polaron dispersion at small momentum $P$ is:

E (P) \approx - α ℏ ω_{L O} + \frac{P ^{2}}{2 m _{b}} (1 - \frac{α}{6})

Comparing this to the standard kinetic energy formula, $E (P) - E_{0} = P^{2} / (2 m^{*})$ , we can identify the polaron effective mass $m^{*}$ . The original Fröhlich result was written as:

m^{*} = \frac{m _{b}}{1 - α /6}

For weak coupling ( $α ≪ 1$ ), we can use the Taylor expansion $1/ (1 - x) \approx 1 + x$ , which gives the widely used approximation:

m^{*} \approx m_{b} (1 + \frac{α}{6})

This is a profound result. The interaction with the phonon field has renormalized the mass of the electron, making it heavier by a factor proportional to the coupling strength $α$ . The electron is literally harder to accelerate because it has to drag its polarization cloud along. Higher-order calculations, like the one by Feynman mentioned in the text, refine this result to $m^{*} / m_{b} = 1 + α /6 + 0.0236 α^{2} + \dots$ .

Test your understanding!

A researcher is studying two polar semiconductors, InSb and KCl. InSb has a very small Fröhlich coupling $α \approx 0.02$ , while KCl has a much larger one, $α \approx 3.5$ . In which material would you expect the polaron effective mass $m^{*}$ to be more significantly different from the bare band mass $m_{b}$ ? Based on the physics of the coupling constant, why does KCl have a much stronger coupling than InSb?

Show answer

You would expect the polaron mass to be much more significantly renormalized in KCl. With $α = 3.5$ , $m^{*} \approx m_{b} (1 + 3.5/6) \approx 1.58 m_{b}$ . The polaron is almost 60% heavier than the bare electron. For InSb, $m^{*} \approx m_{b} (1 + 0.02/6) \approx 1.003 m_{b}$ , a negligible change.

KCl has a stronger coupling primarily because it is a much more ionic crystal than InSb. This leads to a large difference between its static and high-frequency dielectric constants ( $1/ ϵ_{\infty} - 1/ ϵ_{0}$ ), which is a key factor in the formula for $α$ . InSb is a covalently bonded semiconductor with only weak polar character.

3. Experimental Signatures and Advanced View

How do we actually observe a polaron and its renormalized mass? Your experimental experience with spectroscopy provides a direct link here.

Cyclotron Resonance: This technique measures effective mass by finding the resonance frequency $ω_{c} = e B / m^{*}$ for an electron orbiting in a magnetic field $B$ . Polaron effects were first definitively confirmed by observing that the measured cyclotron mass was heavier than the band mass predicted by theory, and that this mass difference depended on temperature and magnetic field in a way consistent with polaron theory.
Angle-Resolved Photoemission Spectroscopy (ARPES): ARPES directly measures the single-particle spectral function $A (k, ω)$ , which tells you the probability of finding an electron with momentum $k$ and energy $ω$ . For a non-interacting electron, this would be a sharp delta-function peak at the band energy $ϵ_{k}$ . For a polaron, the picture is much richer.

The polaron appears as a "quasiparticle peak" in the spectral function, but it's shifted to the lower energy $E (P)$ we calculated. Furthermore, its spectral weight $Z$ is less than 1 ( $Z \approx 1 - α /2$ in the weak limit). The "missing" weight appears in a series of phonon satellite peaks at energies $E (P) + ℏ ω_{L O}$ , $E (P) + 2ℏ ω_{L O}$ , etc. These correspond to photo-emitting the electron and also leaving one or more real LO phonons behind in the crystal.

The following resource beautifully illustrates the theoretical prediction of these features.

Renormalization of electronic properties of semiconductors ...

Let's return to Nery's thesis. The introduction to Chapter 3 provides a superb connection between the polaron problem and what is observed in a spectral function, comparing different theoretical approaches.

Please read the introduction of Chapter 3 (pages 36-40), and pay close attention to Figure 5. This figure compares the polaron ground state energy as a function of the coupling α, as predicted by different theoretical methods. It highlights how different approximations handle the problem.

Figure 5 in the reading is particularly insightful. It shows that the simple Rayleigh-Schrödinger perturbation theory we used ( $E_{0} = - α ℏ ω_{L O}$ ) gives a result remarkably close to the 'exact' Monte Carlo calculations for a wide range of $α$ . This gives us confidence in our weak-coupling approach. This topic provides a perfect bridge to the more advanced many-body techniques, such as Green's functions and self-energy, which we will tackle in Modules 6 and 7.

Conclusion

Today, we've moved beyond simple scattering to see how the electron-phonon interaction can fundamentally create a new quasiparticle, the polaron.

Key Takeaways:

A polaron is a composite quasiparticle formed by an electron and the cloud of lattice polarization (phonons) that it drags along.
The Fröhlich Hamiltonian models this interaction in polar materials, governed by the dimensionless coupling constant $α$ .
In the weak-coupling limit, second-order perturbation theory shows that the polaron's ground state energy is lowered by a self-energy of $E_{0} = - α ℏ ω_{L O}$ .
The interaction renormalizes the electron's effective mass, making it heavier: $m^{*} \approx m_{b} (1 + α /6)$ .
Polarons are observable through experiments like cyclotron resonance (as a heavier mass) and ARPES (as a shifted quasiparticle peak accompanied by phonon satellites).

Preview of the Next Lesson

We have now examined several key consequences of electron-phonon coupling, from resistivity to the formation of new quasiparticles. In our final lesson for this module, we will investigate another fascinating phenomenon: how strong electron-phonon coupling can leave a distinct signature on the phonon dispersion curve itself. We will analyze the Kohn anomaly, a softening of a particular phonon mode, which serves as an experimental fingerprint for strong coupling between electrons and the lattice.