Photocurrent measurements of supercollision cooling in graphene

TL;DR

This study measures hot electron cooling in graphene using photocurrent as a thermometer, confirming supercollision cooling as the dominant mechanism over a wide temperature range.

Contribution

It provides direct experimental evidence supporting supercollision cooling as the primary electron cooling process in graphene, challenging previous disorder-free models.

Findings

Cooling rate follows $C dT/dt=-A(T^3-T_l^3)$

Results agree with supercollision cooling predictions

Applicable over broad temperature ranges

Abstract

The cooling of hot electrons in graphene is the critical process underlying the operation of exciting new graphene-based optoelectronic and plasmonic devices, but the nature of this cooling is controversial. We extract the hot electron cooling rate near the Fermi level by using graphene as novel photothermal thermometer that measures the electron temperature ($T(t)$) as it cools dynamically. We find the photocurrent generated from graphene $p-n$ junctions is well described by the energy dissipation rate $C dT/dt=-A(T^3-T_l^3)$, where the heat capacity is $C=\alpha T$ and $T_l$ is the base lattice temperature. These results are in disagreement with predictions of electron-phonon emission in a disorder-free graphene system, but in excellent quantitative agreement with recent predictions of a disorder-enhanced supercollision (SC) cooling mechanism. We find that the SC model provides a complete and unified picture of energy loss near the Fermi level over the wide range of electronic (15 to $\sim$3000 K) and lattice (10 to 295 K) temperatures investigated.