In-plane thermal transport in graphene/quasi-hexagonal phase C60 heterostructure: insight from machine learning molecular dynamics

This dataset contains the training and testing configurations used to develop the NEP_Gr-qHPC₆₀ machine learning potential for predicting the structural and thermal transport properties of the graphene/quasi-hexagonal phase C₆₀ (Gr/qHPC₆₀) van der Waals heterostructure. The dataset includes atomic configurations, energies, forces, and virial stresses generated from NPT molecular dynamics simulations and DFT-MD calculations. The final dataset consists of 996 configurations, including 789 training structures and 207 testing structures. These data were used for NEP training, validation, and thermal transport simulations using GPUMD.