Validation of crystal chemical coordination rules by automated structure data analysis

• Applicant: Professor Dr. Holger Kohlmann
• Subject Area: Solid State and Surface Chemistry, Material Synthesis
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 546112542

The composition and structure of ionic compounds are determined by a number of empirical rules, which concern, for example, coordination numbers, the structure and connectivity of polyhedra and interatomic distances. The coordination number rule states that the total number of bonds that all cations form with the anions in a compound is equal to the total number of bonds that the anions form with the cations. Recently, it was extended by H. P. Beck using the rule of simplicity, which prescribes equal environments for equal atoms. This leads to the so-called simplest numerical solution (SNS) for given cation coordination numbers. The SNS has the smallest number of different chemical environments for the anions and has the lowest possible ion segregation index Sigma, which summarizes the deviations of the real coordination numbers from the mean values. In cases where the observed crystal structure does not correspond to the SNS and therefore exhibits greater than necessary ion segregation, the reasons could be of a local electronic nature. Beck has introduced a parameter that takes into account the difference between the formal ionic charge and the actual valence sum of the neighboring cations, the so-called electrostatic imbalance Y. In order to validate Beck's coordination number rule (Bcnr) using a large data set, an algorithm was developed for the automated loading of CIF files from a database, calculations of SNS and Y and comparison with the observed crystal structure. However, it still needs to be improved and extended in order to obtain reliable results for a validation of Bcnr and this is the subject of this research project. Python codes are to be written for various parts of the processing algorithm. For example, the CIF converter will be made more reliable by introducing new functions to calculate coordination numbers and ionic charges, also for mixed occupation or high defect concentration, and the SNS calculation will be perfected. After achieving a reliability of >99% for these processes, the API retrieval tools of ICSD will be extended to PCD, COD and Materials Project and the processing of quaternary and multinary compounds will be newly introduced. Entries that have failed processing are analyzed to determine possible bias. The CIFs for all ternary halides, chalcogenides (including oxides), nitrides, phosphides, carbides, borides and hydrides available in databases will be processed and analyzed for compliance with the Bcnr in order to achieve the main objective of this research project, i.e. to decide whether the Bcnr is a valid basis for explaining the crystal structure of ionic compounds. This will lead to a deeper understanding of ionic solids and contribute to the rational design of new compounds.

• DFG Programme: Research Grants