[Paper Review] Self-consistent analysis of electron-phonon coupling parameters of graphene

J. L. McChesney, Aaron Bostwick+5 more|ArXiv.org|Sep 23, 2008

Graphene research and applications3 citations

TL;DR

This study presents a self-consistent analysis of electron-phonon coupling in epitaxial graphene using angle-resolved photoemission spectroscopy, demonstrating that kinks in the electronic dispersion arise from many-body interactions rather than single-particle effects. It reveals electron-phonon coupling strengths 3.5–5 times larger than theoretical predictions, with strong doping-dependent coupling to multiple phonon modes.

ABSTRACT

We present a self-consistent analysis of the photoemission spectral function A(k, w) of graphene monolayers grown epitaxially on SiC(0001). New information derived from spectral intensity anomalies (in addition to linewidths and peak positions) confirms that sizeable kinks in the electronic dispersion at the Dirac energy ED and near the Fermi level EF arise from many-body interactions, not single-particle effects such as substrate bonding or extra bands. The relative electron-phonon scattering rate from phonons at different energy scales evolves with doping. The electron-phonon coupling strength is extracted and found to be much larger (~3.5-5 times) than predicted.

Motivation & Objective

• To resolve the controversy over the origin of spectral kinks in graphene’s electronic dispersion, distinguishing between many-body interactions and single-particle effects.
• To quantify electron-phonon coupling strength in epitaxial graphene with high precision, challenging existing theoretical predictions.
• To investigate the doping dependence of electron-phonon coupling across multiple energy scales using a self-consistent model.
• To validate the model by comparing not only spectral linewidths and peak positions but also absolute spectral intensity across varying carrier densities.

Proposed method

• Employed angle-resolved photoemission spectroscopy (ARPES) to measure the spectral function $ A(\mathbf{k}, \omega) $ of graphene monolayers on SiC(0001) across a range of carrier densities.
• Applied a self-consistent fitting procedure that assumes causality, approximate particle-hole symmetry, and weak momentum dependence of the self-energy, without assuming prior knowledge of bare bands.
• Used maximum entropy method (MEM) and momentum distribution curve (MDC) analysis to extract spectral intensity, linewidths, and dispersion energies from experimental data.
• Compared experimental spectral functions with model predictions using self-energies from theoretical calculations (e.g., Park et al., Calandra and Mauri) to assess discrepancies.
• Performed deconvolution of experimental broadening to isolate intrinsic many-body effects and validate the robustness of observed kinks.
• Quantified electron-phonon coupling strength $ \lambda $ via comparison of theoretical and experimental self-energies, using both direct fitting and scaling methods.

Experimental results

Research questions

• RQ1 Do the observed kinks in graphene’s electronic dispersion originate from many-body interactions or single-particle effects such as substrate bonding?
• RQ2 What is the true magnitude of electron-phonon coupling strength $ \lambda $ in epitaxial graphene, and how does it compare to theoretical predictions?
• RQ3 How does the relative strength of electron-phonon coupling to different phonon modes evolve with carrier doping?
• RQ4 Can the spectral intensity anomalies, beyond linewidths and peak shifts, provide independent evidence for many-body interactions?
• RQ5 What mechanisms could explain the observed discrepancy between measured and predicted electron-phonon coupling strengths?

Key findings

• Spectral kinks near the Dirac energy $ E_D $ and Fermi level $ E_F $ are conclusively attributed to many-body interactions, not single-particle effects such as substrate bonding or extra bands.
• The electron-phonon coupling strength $ \lambda $ is found to be 3.5 to 5 times larger than predicted by state-of-the-art theoretical calculations.
• The relative coupling strength to multiple phonon modes exhibits strong doping dependence, indicating a non-uniform response across the Fermi surface.
• The absolute spectral intensity $ \mathcal{A}(\omega) $, previously overlooked, provides critical validation of the self-consistent model and reveals discrepancies unexplained by experimental broadening.
• The discrepancy between measured and predicted self-energies persists even after deconvolution of experimental broadening, ruling out instrumental effects as the cause.
• Alternative explanations—such as coupling to magnons, substrate phonons, K-dopant vibrations, or ripples—were ruled out due to energy scale mismatches or negligible contribution.