Many-body Green's function approaches to the doped Fr\"ohlich solid: Exact solutions and anomalous mass enhancement

TL;DR

This paper extends the Fr"ohlich model to doped systems, providing exact solutions for electron spectral functions and mass enhancement, revealing doping-induced anomalies and differences between two theoretical approaches.

Contribution

It introduces exact solutions for doped Fr"ohlich models and compares Dyson's and cumulant methods, highlighting their differences and the impact of doping on electron properties.

Findings

Both approaches show mass enhancement at finite doping.

Dyson's method yields better quasiparticle masses.

Cumulant approach captures satellite features more accurately.

Abstract

In polar semiconductors and insulators, the Fr\"ohlich interaction between electrons and long-wavelength longitudinal optical phonons induces a many-body renormalization of the carrier effective masses and the appearence of characteristic phonon sidebands in the spectral function, commonly dubbed 'polaron satellites'. The simplest model that captures these effects is the Fr\"ohlich model, whereby electrons in a parabolic band interact with a dispersionless longitudinal optical phonon. The Fr\"ohlich model has been employed in a number of seminal papers, from early perturbation-theory approaches to modern diagrammatic Monte Carlo calculations. One limitation of this model is that it focuses on undoped systems, thus ignoring carrier screening and Pauli blocking effects that are present in real experiments on doped samples. To overcome this limitation, we here extend the Fr\"ohlich model to the case of doped systems, and we provide exact solutions for the electron spectral function, mass enhancement, and polaron satellites. We perform the analysis using two approaches, namely Dyson's equation with the Fan-Migdal self-energy, and the second-order cumulant expansion. We find that these two approaches provide qualitatively different results. In particular, the Dyson's approach yields better quasiparticle masses and worse satellites, while the cumulant approach provides better satellite structures, at the price of worse quasiparticle masses. Both approaches yield an anomalous enhancement of the electron effective mass at finite doping levels, which in turn leads to a breakdown of the quasiparticle picture in a significant portion of the phase diagram.