Data driven structure determination in amorphous molecular systems using a Reverse-Monte-Carlo approach

• Applicant: Dr. Benjamin Klee
• Subject Area: Theoretical Chemistry: Molecules, Materials, Surfaces ; Physical Chemistry of Solids and Surfaces, Material Characterisation ; Physical Chemistry of Molecules, Liquids and Interfaces, Biophysical Chemistry
• Term: from 2022 to 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 516175749

Since the discovery of X-ray radiation and the subsequent development of crystallography structure determination, i.e. the determination of relative atom positions in non-crystalline materials, was established as an important foundation of modern science. The periodicity of atom positions in crystalline materials leads to the diffraction of incident X-ray radiation and the formation of sharp reflexes at specific positions surrounding the irradiated sample. Based on the pattern of these signals the relative position of atoms within the crystal can be deduced. The investigation of crystalline samples via X-ray diffraction is a standard technique in technical, medical and scientific applications for the characterisation of materials and for linking macroscopic properties to microscopic ones. Non-crystalline materials such as liquids and glasses on the other hand have been increasingly attracting scientific interest because the composition and therefore the properties of such systems can be varied more freely to be fine tuned for specific needs. Non-crystalline systems cannot be subjected to crystallographic methods though, since the lack of periodicity in their structure prevents the formation of sharp reflexes. Still, amorphous materials can be probed for structural information by X-ray scattering and other experimental methods but in such cases the data interpretation is still challenging and is usually done using computer-based Reverse-Monte-Carlo (RMC) modeling. Using RMC a virtual set of atom coordinates can be found that reproduces all available experimental data. This model can then be explored by statistical methods to extract information on the structure, e.g. average coordination numbers or bond lengths. RMC is typically used for non-molecular systems like chalcogenide glasses since for molecule containing samples several yet unsolved problems occur. Therefore for analyzing molecular systems force field based methods like Molecular Dynamics (MD) simulations are usually preferred. The latter require system specific parameter sets though, that first need to be developed based on experimentally obtained information. I plan to extend the functionality of the RMC method to support structure determination in amorphorphous and liquid molecular systems.

• DFG Programme: WBP Fellowship
• International Connection: Hungary
• Host: Professor Dr. László Pusztai