Machine learned potential for defected single layer hexagonal boron nitride

The development of machine learned interatomic potentials (MLIPs) is critical for performing reliable simulations of materials at length and time scales that are comparable to those in the laboratory. We present here a MLIP suitable for simulations of the temperature dependent structure and dynamics of single layer hexagonal boron nitride (h-BN) with defects and grain boundaries, developed using a strictly local equivariant deep neural network, Allegro. The training dataset consisted of ∼30 000 images of h-BN with and without point defects generated with ab initio molecular dynamics simulations, based on density functional theory (DFT), at 500, 1000, and 1500 K. The developed MLIP predicts potential energies and forces with a mean absolute error of 4 meV/atom and 60 meV/Å, respectively. It also reproduces phonon dispersion curves and density of vibrational states of pristine bulk h-BN that are comparable with those obtained from DFT calculations. Using this MLIP to study the motion of the 4|8 grain boundary in h-BN we show that the initial motion of the first unit has an activation barrier of ∼2.2 eV. Moving the grain boundary units past the first shows much lower activation barriers of ∼0.42 eV, suggesting a facile motion of the grain boundary once the initial movement is stimulated. Molecular dynamics simulations of the grain boundary yield a scaled mobility of 1.739 × 10-11 m3/Js at 1500 K which, given the different setups without continuous e-beam irradiation, is not too far from the experimental value of 1.36 × 10-9 m3/Js. The ability to predict grain boundary mobility within reasonable agreement with experiment demonstrates the robustness of the MLIP and its suitability for reliable simulations of defect structures and dynamics in single layer h-BN.