🔬 materials science

First-principles theory of phonon renormalization from nonlinear electron-phonon interactions

This paper establishes a first-principles diagrammatic framework demonstrating that nonlinear one-electron-two-phonon interactions, unlike linear couplings, renormalize phonon frequencies across the entire Brillouin zone with distinct temperature dependence, a phenomenon quantified in polar semiconductors like LiF and KTaO $_3$ and applicable to materials with strong lattice fluctuations.

Original authors: Florian Kluibenschedl, Matthew Houtput, Jacques Tempere, Cesare Franchini, Mikhail Lemeshko, Ragheed Alhyder

Published 2026-09-18

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Florian Kluibenschedl, Matthew Houtput, Jacques Tempere, Cesare Franchini, Mikhail Lemeshko, Ragheed Alhyder

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Inside every solid object, from the silicon chip in a smartphone to the salt crystal on a dinner table, atoms are never truly still. They vibrate in a collective rhythm, a constant hum of motion that defines how the material conducts heat, carries electricity, or holds its shape. These vibrations, known to physicists as phonons, are the fundamental units of sound and heat in the solid world. For decades, scientists have understood that these vibrations are not isolated events; they are constantly interacting with the electrons that zip through the material. When an electron passes a vibrating atom, it tugs on it, and the atom tugs back. This interaction, called electron-phonon coupling, is a well-known force that can shift the pitch of these vibrations or shorten their lifespan, much like a heavy hand might dampen the ring of a bell.

However, for a long time, the standard way of describing this interaction assumed a simple, direct relationship: one electron pulls on one vibration. This linear view has served science well, explaining many properties of metals and semiconductors. But in materials where atoms are loosely held or where vibrations are particularly strong, this simple picture might be incomplete. Just as a single person might pull a rope, two people pulling together could create a more complex, nonlinear effect. The question that has lingered is whether these more complex, multi-particle interactions play a significant role in shaping the behavior of solids, particularly in materials that are prone to large atomic movements.

In a new study, researchers have taken a deep dive into this overlooked possibility, developing a theoretical framework to calculate how a single electron interacting with two phonons simultaneously might alter the vibrational spectrum of a material. They focused their investigation on two very different polar semiconductors: lithium fluoride, a hard, wide-gap insulator, and potassium tantalate, a softer material known for its quantum properties. By using powerful computer simulations based on the fundamental laws of quantum mechanics, the team mapped out how these nonlinear interactions change the energy and lifetime of phonons across the entire material, comparing these results against the standard linear effects.

The researchers found that the two types of interactions behave in strikingly different ways. The familiar linear interaction, where one electron couples to one phonon, acts like a spotlight. Its influence is intensely concentrated near the center of the material's momentum space, affecting only a tiny region of the vibrational spectrum. In contrast, the nonlinear interaction, where one electron couples to two phonons at once, acts like a diffuse light that illuminates the entire landscape. This process connects the incoming vibration to other vibrational branches throughout the material, effectively renormalizing, or shifting, the frequencies of phonons across the whole range of possible motions.

Crucially, this nonlinear effect is deeply tied to temperature. Because the process involves an intermediate phonon that must be thermally excited, the strength of the interaction grows as the material gets hotter and more vibrational modes become active. In the hard, rigid lithium fluoride, where only a few low-energy vibrations are active at room temperature, this nonlinear effect is incredibly small, almost negligible. However, in the softer potassium tantalate, where many more vibrational modes are thermally active, the effect is significantly larger, appearing two to three times stronger than in the fluoride. This difference highlights that the nonlinear channel is most potent in materials with "soft" lattices that support a rich variety of low-energy vibrations.

The study also revealed that these two interaction channels leave distinct fingerprints on the material's properties. While the linear effect dominates near the center of the momentum space, the nonlinear effect remains visible even in regions where the linear effect is suppressed by symmetry. This suggests that by carefully measuring how phonon frequencies shift across different directions in a crystal, scientists could experimentally isolate and identify the nonlinear contribution. Furthermore, the team calculated how these shifts affect the material's heat capacity. They found that the two interactions influence heat storage in opposite ways depending on the density of electrons in the material, offering a potential experimental signature to distinguish between them.

While the effects observed in the two materials studied were small, the researchers argue that this framework provides a vital tool for understanding a broader class of materials. In soft semiconductors, such as lead-halide perovskites used in solar cells, where lattice fluctuations are large and numerous, these nonlinear interactions are expected to be much more pronounced. By establishing a clear theoretical path to calculate these effects, the work opens the door to a more complete understanding of how electrons and vibrations dance together in the complex, soft materials that are central to next-generation electronics and energy technologies. The findings suggest that to fully grasp the behavior of these advanced materials, we must look beyond the simple one-to-one interactions and account for the more intricate, multi-part connections that shape the solid state.

Technical Summary: First-Principles Theory of Phonon Renormalization from Nonlinear Electron-Phonon Interactions

Problem Statement
Electron-phonon interactions are fundamental to the dynamical properties of solids, governing phonon frequencies and lifetimes. Standard theoretical treatments typically rely on two approximations: (1) evaluating self-energies within the lowest order of the electron-phonon interaction (linear coupling), and (2) approximating the electron-phonon Hamiltonian up to linear order in ionic displacements. While linear coupling explains phenomena such as Kohn anomalies and superconductivity in metals, the role of nonlinear electron-phonon interactions—specifically the one-electron-two-phonon vertex—remains largely unexplored regarding phonon properties. Although model studies and recent first-principles derivations have addressed the impact of quadratic coupling on electronic properties (e.g., polaron formation and electron mobility), a systematic framework to assess how these nonlinear interactions renormalize the phonon spectrum itself, particularly in polar semiconductors with charge carriers, has been lacking.

Methodology
The authors develop a diagrammatic framework to calculate the phonon self-energy arising from both long-range linear one-electron-one-phonon and nonlinear one-electron-two-phonon interactions.

Hamiltonian: The study expands the electron-phonon interaction perturbatively up to the one-electron-two-phonon vertex. The linear vertex ( $g_{mn\nu}$ ) and the nonlinear vertex ( $g_{mn\nu_1\nu_2}$ ) are derived from first principles, focusing on the long-range polar limit governed by Born effective charges and the high-frequency dielectric tensor.
Self-Energy Calculation: The phonon self-energy is computed using two Feynman diagrams:
1. Linear Contribution: The standard electron-hole polarization bubble, where a phonon decays into an electron-hole pair.
2. Nonlinear Contribution: A diagram involving the one-electron-two-phonon vertex, where a phonon scatters into an electron-hole pair and an internal phonon.
First-Principles Implementation: The authors evaluate these diagrams for two polar semiconductors: Lithium Fluoride (LiF) and Potassium Tantalate (KTaO $_3$ ). Calculations utilize the Vienna Ab Initio Simulation Package (VASP) with PBEsol functionals. The long-range vertex factors are parameterized using dynamical matrix derivatives with respect to the electric field.
Simulation Conditions: To simulate carrier populations (e.g., from doping or photoexcitation), an effective chemical potential ( $\mu$ ) is introduced into the conduction band. The study evaluates the renormalization as a function of $\mu$ and temperature ( $T$ ).

Key Contributions

Derivation of Nonlinear Self-Energy: The paper derives the analytical expression for the one-electron-two-phonon self-energy (Eq. 29), showing it depends on the convolution of the electronic polarization bubble and the phonon propagator.
Qualitative Distinction of Mechanisms: The work establishes that the linear and nonlinear channels renormalize phonons in fundamentally different ways:
- Momentum Dependence: The linear contribution is sharply localized near the Brillouin-zone center ( $\Gamma$ point) due to the $1/|q|$ divergence of the long-range vertex and the restriction of electron-hole excitations to the Fermi surface. In contrast, the nonlinear contribution couples an incoming phonon to other branches throughout the entire spectrum, resulting in renormalization across the full Brillouin zone.
- Temperature Dependence: The nonlinear process introduces a pronounced temperature dependence governed by the Bose-Einstein occupation of internal phonon branches. This contrasts with the linear process, where temperature effects are limited to the broadening of the Fermi-Dirac distribution.
Benchmarking in LiF and KTaO $_3$ : The framework is applied to LiF (a wide-gap insulator) and KTaO $_3$ (a quantum paraelectric). The study quantifies that while nonlinear renormalization is small in both, it is 2–5 times larger in KTaO $_3$ due to the greater number of thermally populated phonon branches at room temperature compared to LiF.

Results

Phonon Spectrum: In both materials, the linear self-energy causes significant frequency shifts and broadening only near the $\Gamma$ point. The nonlinear self-energy produces a uniform, albeit small, renormalization across the Brillouin zone.
Magnitude: The real part of the nonlinear self-energy is on the order of $10^{-5}$ meV in LiF and slightly larger in KTaO $_3$ . While small, the nonlinear contribution can dominate in regions where symmetry suppresses the linear vertex (e.g., along specific high-symmetry paths where the linear vertex vanishes).
Heat Capacity: The authors calculate the impact of these renormalizations on lattice heat capacity. They find that electron-phonon induced renormalization generally decreases the heat capacity. Notably, the two channels exhibit opposite dependencies on carrier density: the linear correction decreases with increasing chemical potential (due to saturation of the polarization bubble), while the nonlinear correction increases with carrier density and temperature.
Material Specificity: The enhancement of nonlinear effects in KTaO $_3$ is attributed to its "soft" nature, which allows more phonon branches to be thermally accessible, thereby opening more scattering channels for the one-electron-two-phonon process.

Significance and Claims
The paper claims to establish a general first-principles framework for assessing nonlinear electron-phonon effects on phonon properties. The primary significance lies in identifying a "clean experimental signature" for one-electron-two-phonon coupling: a temperature-dependent, momentum-uniform renormalization that persists even where linear coupling is symmetry-suppressed.

The authors modestly note that for the specific materials studied (LiF and KTaO $_3$ ), the nonlinear effects are small. However, they argue that the framework predicts substantially larger effects in materials with stronger lattice fluctuations and more low-frequency phonon modes, such as soft semiconductors (e.g., lead-halide perovskites and Bi $_2$ O $_2$ Se). The work suggests that in these softer materials, the nonlinear channel could become a dominant factor in phonon renormalization, offering a pathway to distinguish linear and nonlinear interaction channels experimentally via inelastic scattering or spectroscopy. The authors also highlight that this structure allows for indirect access to the electronic susceptibility $\chi(q, \omega)$ at finite momentum through phonon dispersion measurements.

More like this