Ion correlations explain kinetic selectivity in diffusion-limited solid state synthesis reactions: Transport coefficients in amorphous Ba-Ti-O interphases

Data files: "BTO_Transport_Coefficients.csv": Onsager Transport coefficients fitted for amorphous Ba-Ti-O interphases from machine-learned interatomic potential based molecular dynamics simulations of randomly packed ions in units of (1/eV-cm3-sec)."ace_mpmorph_starting_structures.json": contains all amorphous structures obtained via ACE-based MPMorph density equilibration, at T=750,1000,1250,1500,1750K; with 5 replicates per composition. These are used as inputs to NVT MD simulations to compute the transport coefficients."data.pckl.gzip": DFT training data generated for training the ACE potential. ACE potentials: "output_potential.yaml" and "output_potential.asi": Final ACE potentials trained for amorphous Ba-Ti-O phases (use python-ace to use as an Ase Calculator), used for production mpmorph + NVT MD simulations."interim_potential_0.yaml": Intermediate ACE potential, without the ZBL core repulsion.ReactCA files: (use the rxn-ca package: https://github.com/mcgalcode/rxn-ca.git to run the simulations with these libraries) "reaction_lib_with_diffusion.json" : ReactCA reaction library used to run simulations with the new thermo + kinetics scorer. This was produced using the "DiffusionScorer" class in the rxn-ca package."reaction_lib_thermo_only.json" : ReactCA reaction library used to run simulations with thermo scorer. This was produced using the "TammanHuttigScorer" class in the rxn-ca package.Analysis scripts: "FittingExample.zip": contains an example notebook and code needed to fit the transport coefficients as described to NVT MD trajectories.