BiLayerStacker: A Universal and Global Program for Determining Ground-State Stacking Configurations of Bilayer 2D Nanomaterials

Abstract Comprehensively exploring the stacking configurations of bilayer two-dimensional (2D) nanomaterials and heterostructures is extremely challenging, as traditional manual modeling struggles with the vast slippage-induced stacking configuration space. Here, we developed a universal and global program, named BiLayerStacker, capable of systematically constructing potential bilayer stacking configurations from two arbitrary monolayer 2D nanomaterials and rapidly identifying the ground-state sliding-stacking configuration. To achieve the universality of BiLayerStacker, we employed an adaptive lattice-rectification scheme that corrects monolayers with over 5 % lattice mismatch via equivalent lattice conversion and minimal supercell expansion. Its globality is realized by the interlayer distance scanning via Bayesian optimization, the delicate design of sliding vectors determined by pre-tests of angle and distance step based on the “minimum repeat step principle”, and the maximum sliding distance. Utilizing BiLayerStacker, we explored 28 series of bilayer sliding-stacking configurations based on seven previously identified monolayer borophene, yielding over 320,000 stacking configurations. The 28 identified ground-state configurations each exhibit a binding energy that surpasses the highest value reported in the literature for their corresponding series, validating the effectiveness of BiLayerStacker. Remarkably, we identified two novel bilayer borophene stacking configurations (α7-α8 and α5-α8) that exhibit exceptional thermodynamic, dynamic stability, and photoelectric properties, highlighting their promise as transparent conductors. BiLayerStacker establishes a methodological foundation for universal multilayer material prediction, and bridges theory and experiment.