Supplementary data for the paper "Scalable quasi-pure MOF membranes for energy-efficient gas separations"

Supplementary Data This dataset contains the simulation input files, scripts, and computational configurations used to study gas transport and separation in mixed matrix membranes (MMMs) composed of ZIF-67 and 6FDA-DAM polymer. Four complementary computational approaches were employed: 1. CADSS — Grand Canonical Monte Carlo (GCMC) Adsorption Simulations 2. VASP — First-Principles DFT Calculations **Note**: POTCAR files are not included due to VASP copyright restrictions. 3. MACE — Machine-Learning-Accelerated Structure Optimization Due to the large number of atoms in the ZIF-67/6FDA-DAM slab model, conventional DFT is too expensive for high-throughput screening. This directory provides:- **`run_opt.py`**: Python script performing geometry optimization of the 6FDA-DAM polymer on the ZIF-67 surface using the MACE-MP-0 medium general-purpose machine learning potential. The BFGS optimizer is used with a force convergence threshold of 0.01 eV/Å. The optimization trajectory is saved for subsequent analysis. 4. LAMMPS, Classical Molecular Dynamics SimulationsMolecular dynamics simulations were performed to analyze gas diffusion in mixed matrix membranes under two different conditions.Additional workflows for interfacial pore networks and free porosity of mixed matrix membranes can be found in the Github repository as textural analysis: https://github.com/AlejandroDiazMarquez/AutoMMM