(572c) Mofdb: An Accessible Online Database of Computational Adsorption Data for Nanoporous Materials

MOFDB is a new freely accessible database containing simulated adsorption data from molecular simulation screening studies in nanoporous materials. The database includes about 3 million simulated adsorption data points for H₂, CH₄, CO₂, Xe, Kr, N₂, and Ar in over 160 000 MOFs and over 2000 zeolites, as well as textural properties like pore sizes and surface areas and the structure files for each material. We also include all relevant metadata such as details about Monte Carlo simulations, force field parameters, and links to original to facilitate reproducibility of the calculations. The database is easily searchable by MOF properties or adsorbate, and the adsorption data are stored in a standardized JSON format that is fully interoperable with the NIST adsorption database. Each MOF structure has a unique identifier (MOFid and MOFkey) that is searchable and amenable to cheminformatic analysis. All structure files (cifs) are provided, and the database is easily accessed using a Python API.

As an example of analysis facilitated by the database, we identified MOFs that meet high performance targets for multiple applications simultaneously, such as hydrogen and methane storage or carbon capture and Xe/Kr separation, including a MOF with predicted Xe capacity of 3.8 mol/kg and Xe/Kr selectivity of 15.6.

We have made this data publicly available in a standardized format in the hope that it will enable new machine learning analysis and lead to further discoveries from this data. We also encourage the community to adopt this JSON format as a standard way of storing and sharing isotherm data.

SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525. SAND Number: SAND2020-3912 A