Fig. 4: Generalization capability of Neural-Network Non-Adiabatic Molecular Dynamics (N2AMD). | Nature Communications

Fig. 4: Generalization capability of Neural-Network Non-Adiabatic Molecular Dynamics (N2AMD).

From: Advancing nonadiabatic molecular dynamics simulations in solids with E(3) equivariant deep neural hamiltonians

Fig. 4

a, c, d The use of N2AMD, trained by density functional theory (DFT) results of non-twisted bilayer MoS2, to predict c the Kohn-Sham (KS) band structure and d carrier dynamics in twisted bilayer MoS2. b, e, f The use of N2AMD, trained by DFT results of monolayer Silicon, to predict e the band structure and f carrier dynamics in the Silicon nanotube. Source data are provided as a Source Data file.

Back to article page