MOF-303: H2O Diffusion Data from Machine-Learned Potentials

Machine-learned interatomic potentials (DP-MLIP) were used to simulate H2O diffusion in the flexible metal-organic framework MOF-303. The dataset, created by Hasnain Sajid and last updated in April 2026, likely contains results from molecular dynamics and nudged elastic band calculations. It benchmarks diffusion coefficients and barriers against density functional theory and a classical force field.

Use Cases

• Benchmarking machine-learned interatomic potentials against DFT based on diffusion barrier calculations mentioned in the description
• Analyzing the dynamic behavior of water molecules in flexible frameworks based on molecular dynamics simulation data
• Comparing self-diffusion coefficients predicted by different computational methods (DP-MLIP, MACE, classical force field) as described
• Studying the sequence of adsorption-state stability in MOF-303 based on the mechanistic insights provided

Strengths

• Data is derived from highly accurate deep machine-learned interatomic potentials (DP-MLIP), as described
• Results are benchmarked against density functional theory (DFT) and a generic foundation model (MACE)
• The dataset focuses on a fully flexible framework, addressing a limitation noted in existing literature

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• The dataset is small (211.6 KB), indicating limited scope

Provenance

Source

figshare

Collection Method

Generated via molecular dynamics simulations using machine-learned interatomic potentials.

Time Range

Simulation timeframe is not specified.

Freshness

Last updated 2026-04-14 16:35:35

Geography

Not applicable; computational study.