Neural-Network Force Field Backed Nested Sampling: Study of the Silicon p-T Phase Diagram
Nested sampling is a promising method for calculating phase diagrams of materials, however, the computational cost limits its applicability if ab-initio accuracy is required. In the present work, we report on the efficient use of a neural-network force field in conjunction with the nested-sampling algorithm. We train our force fields on a recently reported database of silicon structures and demonstrate our approach on the low-pressure region of the silicon pressure-temperature phase diagram between 0 and \SI{16}{GPa}. The simulated phase diagram shows a good agreement with experimental results, closely reproducing the melting line. Furthermore, all of the experimentally stable structures within the investigated pressure range are also observed in our simulations. We point out the importance of the choice of exchange-correlation functional for the training data and show how the meta-GGA r2SCAN plays a pivotal role in achieving accurate thermodynamic behaviour using nested-sampling. We furthermore perform a detailed analysis of the exploration of the potential energy surface and highlight the critical role of a diverse training data set.