Training sets for Halide Diffusion in Mixed-Halide Perovskites and Heterojunctions

The training sets for the machine learning potentials used in '10.1021/acs.chemmater.6c00258' are provided. These training sets, in combination with the training parameters provided in the publication, can be used for generating the force fields. They are divided as such: For CsPbBr3, the ML_AB files, which is the VASP MLFF training set format, are provided: ML_AB_I_Br^-: Negatively charged bromide interstitial ML_AB_I_Br^0: Neutral bromide interstitial ML_AB_I_Br^+: Positively charged bromide interstitial ML_AB_V_Br^-: Negatively charged bromide vacancy ML_AB_V_Br^0: Neutral bromide vacancy ML_AB_V_Br^+: Positively charged bromide vacancy ML_AB_PT: Phase transition model For CsPb(IxBr1-x)3, the training sets are provided in the .extxyz format, with the stress tensors and total energy written in the header, and forces written in the last three columns for each structure. In the publication, these sets are used for training Allegro models, but in principle, this data can be used to train any interatomic potential that reads .extxyz as a valid input training format.