Towards “on-demand” van der Waals epitaxy with adaptive ensemble sampling atomistic workflows

Traditional approaches to achieving targeted epitaxial growth involve exploring a vast parameter space of thermodynamic and kinetic drivers (e.g., temperature, pressure, chemical potential, etc.). This tedious and time-consuming approach becomes particularly cumbersome to accelerate synthesis and characterization of novel materials with complex dependencies on the local chemical environment, temperature and lattice strains, specifically for nanoscale heterostructures of layered 2D materials. We combine the strengths of next generation supercomputers at the extreme scale, machine learning and classical molecular dynamics simulations within an adaptive real time closed-loop virtual environment steered by Bayesian algorithms to enable asynchronous ensemble sampling of the synthesis space and apply it to the recrystallization of amorphous transition-metal dichalcogenide (TMDC) bilayers to form stacked moiré superstructures under various growth parameters. We show that such batch parallel Bayesian optimization-based online ensemble sampling frameworks for materials simulations can be promising for achieving and accelerating the on-demand epitaxy of van der Waals stacked moiré devices, paving the way towards a robust autonomous materials synthesis pipeline to enable discovery of unprecedented functionalities.