Stacking atomically thin transition metal dichalcogenides(TMDs)into heterostructures enables exploration of exotic quantumphases,particularly through twist-angle-controlled moirésuperlattices.These structures exh...Stacking atomically thin transition metal dichalcogenides(TMDs)into heterostructures enables exploration of exotic quantumphases,particularly through twist-angle-controlled moirésuperlattices.These structures exhibit novel electronic and optical behaviors driven by atomic-scale structural reconstruction.However,studying such systems with DFT is computationally demanding due to their large unit cells and van der Waals(vdW)interactions between layers.To address this,we develop a transferable neural network potential(NNP)that includes long-range vdW corrections up to 12Åwith minimal overhead.Trained on vdW-corrected DFT data forMo-and W-basedTMDswith S,Se,and Te,the NNP accurately models monolayers,bilayers,heterostructures,and their interaction with h-BN substrates.It reproduces equilibrium structures,energy landscapes,phonon dispersions,and matches experimental atomic reconstructions in twisted WS2 and MoS2/WSe2 systems.We demonstrate that our NNP achieves DFT-level accuracy and high computational efficiency,enabling large-scale simulations of TMD-based moirésuperlattices both with and without substrates.展开更多
基金supported by the Theory of Materials program at the Lawrence Berkeley National Laboratory, funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division, under Contract No. DE-AC02-05CH11231. Computational resources were provided by the National Energy Research Scientific Computing Center, USA.
文摘Stacking atomically thin transition metal dichalcogenides(TMDs)into heterostructures enables exploration of exotic quantumphases,particularly through twist-angle-controlled moirésuperlattices.These structures exhibit novel electronic and optical behaviors driven by atomic-scale structural reconstruction.However,studying such systems with DFT is computationally demanding due to their large unit cells and van der Waals(vdW)interactions between layers.To address this,we develop a transferable neural network potential(NNP)that includes long-range vdW corrections up to 12Åwith minimal overhead.Trained on vdW-corrected DFT data forMo-and W-basedTMDswith S,Se,and Te,the NNP accurately models monolayers,bilayers,heterostructures,and their interaction with h-BN substrates.It reproduces equilibrium structures,energy landscapes,phonon dispersions,and matches experimental atomic reconstructions in twisted WS2 and MoS2/WSe2 systems.We demonstrate that our NNP achieves DFT-level accuracy and high computational efficiency,enabling large-scale simulations of TMD-based moirésuperlattices both with and without substrates.
