FAQs
How many MOFs are there in the CSD?
Solution
Reporting the number of Metal-Organic Frameworks (MOFs) in the CSD is a difficult question to answer because the definition of a MOF differs quite significantly depending on which researcher you ask and the intended research application. The following document (and associated references) from the MOF Commission of the International Zeolite Association (IZA) provides further information: http://www.iza-online.org/MOF/MOFforIZA.pdf
Previously the CCDC has quoted a number based on a simple search consisting of a transition metal bonded to an oxygen or nitrogen via a polymeric bond. This obtains 98,218 MOF structures (around 8% of the CSD) based on version 5.45 of the CSD.
Since the 2016 release of the CSD Portfolio (ConQuest 1.19) there have been two pre-defined hitlists of MOF structures generated via a more comprehensive series of searches. The two hitlists are:
MOF subset: The subset is deliberately intended to be as widely applicable as possible, including all metal centres (rather than based on transition metals) and 1-, 2- and 3-D frameworks that may also be termed Coordination Polymers or Networks. The subsets have been created based upon a series of CSD substructure searches that cover the bonding motifs of all common metal-organic frameworks. The subset contains 123,457 entries based on version 5.45 of the CSD (10% of the CSD).
Non-disordered MOF subset: CSD entries are normally classified as disordered if there is any non-hydrogen disorder present in the structure – including unmodelled molecules (e.g. using the SQUEEZE/PLATON or MASK/OLEX procedures). The occurrence of these unmodelled molecules is very common for MOF-like structures which often contain disordered and poorly-resolved solvent molecules within a crystallographically more well-defined framework. The Non-disordered MOF subset is intended to include structures where the framework is not disordered, but may still contain disorder relating to solvent molecules. The framework disorder definition is based on a search for multi-site or symmetry-related disorder; unmodelled disorder as part of a framework (e.g. an oxygen atom representing an ethanol molecule) will not be excluded from this subset. This subset is intended to be a starting point for virtual high-throughput screening methods where a complete framework is required.
The MOF subsets will be maintained as part of the CSD Portfolio; further details of the generation of these subsets are described in the Chemistry of Materials publication http://pubs.acs.org/doi/abs/10.1021/acs.chemmater.7b00441
The subsets can be used within ConQuest for browsing or search analysis; once the CSD Portfolio is installed, the list of identifiers is available as a .GCD file within their installation in a location like this (example given for default Windows installation):
%USERPROFILE%\CCDC\ccdc-data\csd\subsets\CSD_MOF_subsets
Should the user require more data, such as the full structure, the subsets can be viewed in ConQuest by clicking on View Databases > Subsets in CSD version 5.45 > CSD MOF subsets > Non-disordered MOF subset, then navigating to File > Export Entries as… and choosing CIF export to obtain structures in CIF format.
CSD MOF Collection: in March 2021 we launched the CSD MOF collection of over 10,000 structures free to use in academic research. Learn more and get the link to download it here.
1D MOF subset, 2D MOF subset, 3D MOF subset: The 2022.1 CSD release provides an additional three MOF subsets, with the existing Non-disordered MOF subset further divided into groups based on the dimensionality of the framework. Further information on these subsets is reported in the Chemical Science publication https://doi.org/10.1039/D0SC01297A with the algorithm available from https://github.com/ccdc-opensource/science-paper-mofs-2020