Data for: "Transfer Learning Meets Embedded Correlated Wavefunction Theory for Chemically Accurate Molecular Simulations: Application to Calcium Carbonate Ion-Pairing"

This repository provides all materials associated with the forthcoming publication, in which we develop and apply the ECW-TL framework to construct high-accuracy machine-learned interatomic potentials (MLIPs) for Ca–CO₃ ion pairing dynamics in aqueous solution. It includes input files for density functional theory (DFT), embedded correlated wavefunction (ECW) calculations, MLIP training and fine-tuning, and molecular dynamics (MD) simulations. The repository also contains the final trained machine learning models, as well as outputs from the final enhanced sampling and constrained molecular dynamics (MD) simulations. All files required to reproduce the simulations are provided, along with processed analysis data and the notebooks used to generate every figure in the manuscript.