Stacking domain morphology in epitaxial graphene on silicon carbide

TL;DR

This study uses advanced microscopy to analyze the stacking domain morphology in epitaxial graphene on silicon carbide, revealing strain variations, moiré patterns, and the influence of substrate steps on domain orientation, which impact electronic properties.

Contribution

It provides a detailed characterization of stacking domains and strain effects in epitaxial graphene, highlighting the role of substrate steps and intrinsic disorder near a phase transition.

Findings

Moiré patterns caused by lattice mismatch reveal strain and dislocations.

Epitaxial graphene exhibits co-existing stripe and trigonal domains.

Substrate step edges influence domain orientation.

Abstract

Terrace-sized, single-orientation graphene can be grown on top of a carbon buffer layer on silicon carbide by thermal decomposition. Despite its homogeneous appearance, a surprisingly large variation in electron transport properties is observed. Here, we employ Aberration-Corrected Low-Energy Electron Microscopy (AC-LEEM) to study a possible cause of this variability. We characterize the morphology of stacking domains between the graphene and the buffer layer of high-quality samples. Similar to the case of twisted bilayer graphene, the lattice mismatch between the graphene layer and the buffer layer at the growth temperature causes a moir\'e pattern with domain boundaries between AB and BA stackings. We analyze this moir\'e pattern to characterize the relative strain and to count the number of edge dislocations. Furthermore, we show that epitaxial graphene on silicon carbide is close to a phase transition, causing intrinsic disorder in the form of co-existence of anisotropic stripe domains and isotropic trigonal domains. Using adaptive geometric phase analysis, we determine the precise relative strain variation caused by these domains. We observe that the step edges of the SiC substrate influence the orientation of the domains and we discuss which aspects of the growth process influence these effects by comparing samples from different sources.