As Crystal Structure Prediction (CSP) becomes a more common tool in the study of polymorphism, identification of the putative structures in the CSP energy landscape that match the experimental targets in an accurate, consistent, and quantitative manner is necessary. A quantitative measure of structure similarity is also valuable for assessing the accuracy with which different methods generate the target structures, where minor changes in intermolecular connectivity can result in substantial changes in the bulk properties. The comparison of crystal structures can be done by either comparing simulated powder diffractograms or measuring the root-mean-squared-difference (RMSD) of atomic positions between structures, with each of these methods having different advantages and disadvantages. Here, a procedure that aims to resolve many of the disadvantages posed by the comparison of powder diffractograms has been developed. The main features of this new procedure include automated unit-cell reduction and an anisotropic volume correction prior to powder diffractogram comparison. The method has been applied to targets of the 6th CSP blind test in order to demonstrate its ability to identify the target structure within the list of putative structures submitted by contributors. The importance of using a volume correction in the quantitative comparison of crystal structures is further demonstrated by correlating results from RMSD measurements and powder diffractogram comparisons. Insights provided by the method into the accuracy of the generated structures matching the targets are also discussed.

Quantitative identification of target structures in CSP structure-energy landscapes

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.