Moire patterns and inversion boundaries in graphene/hexagonal boron nitride bilayers

In this paper a systematic examination of graphene/hexagonal boron nitride (g/hBN) bilayers is presented, through a recently developed two-dimensional phase field crystal model that incorporates out-of-plane deformations. The system parameters are determined by closely matching the stacking energies and heights of g/hBN bilayers to those obtained from existing quantum-mechanical density functional theory calculations. Out-of-plane deformations are shown to reduce the energies of inversion domain boundaries in hBN, and the coupling between graphene and hBN layers leads to a bilayer defect configuration consisting of an inversion boundary in hBN and a domain wall in graphene. Simulations of twisted bilayers reveal the structure, energy, and elastic properties of the corresponding moire patterns and show a crossover as the misorientation angle between the layers increases from a well-defined hexagonal network of domain boundaries and junctions to smeared-out patterns. The transition occurs when the thickness of domain walls approaches the size of the moire patterns and coincides with the peaks in the average von Mises and volumetric stresses of the bilayer.

Automated summary

This paper presents a systematic examination of graphene/hexagonal boron nitride (g/hBN) bilayers using a recently developed two-dimensional phase field crystal model that considers out-of-plane deformations. The system parameters are determined by matching the stacking energies and heights of g/hBN bilayers to those obtained from quantum-mechanical density functional theory calculations. The study reveals that out-of-plane deformations reduce the energies of inversion domain boundaries in hBN, and the coupling between graphene and hBN layers results in a bilayer defect configuration consisting of an inversion boundary in hBN and a domain wall in graphene. Simulations of twisted bilayers demonstrate the structure, energy, and elastic properties of moire patterns, and show a transition from well-defined hexagonal network of domain boundaries and junctions to smeared-out patterns as the misorientation angle between the layers increases.