Development of a novel fragmentation approach towards applicability to metal complexes and clusters

• Applicant: Marco Bauer
• Subject Area: Theoretical Chemistry: Electronic Structure, Dynamics, Simulation
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 551708160

In this project a novel ab initio fragmentation scheme, so-called Excitonic Renormalization (XR), for the calculation of the electronic structure of molecules is pushed towards applicability of systems of interest. It truncates systems into fragments, which are calculated independently and then their interactions are calculated. Usually this requires calculating the dimer interaction at a given level of theory, which requires one to make trade-offs, as different fragments usually are best described, i.e. accurate and efficient, by different methods. However, XR is completely modular and only requires independently calculated intermediates of the monomers, as well as a cheap term, which connects these fragments and does not require making trade-offs between methods, in order to calculate their interaction on an ab initio level. It also operates in the state basis of the individual fragments, instead of the orbital basis, enabling massive truncation of the spaces for the fragments, without neglecting local correlation. These advantages are crucial for computations on e.g. complexes with more than one metal atom. This class of systems is very important in the context of activation of small and inert molecules, like nitrogen. For nitrogen this is currently done with the Haber-Bosch process, producing ammonia, which is essential for fertilizers. However, this process is responsible for 1-2 % of the global energy consumption. Splitting nitrogen under ambient conditions is a very hard task though, which is why proper theoretical modelling is so important, but this requires modelling large active spaces, since there are two metal centers. Nevertheless, with the XR approach this reduces to treating active spaces of isolated metal atoms, which is not a hard task. In the current state of XR development this is impossible though, as XR is still in its proof-of-concept phase. To be more precise, efficient algorithms are required to obtain the truncated set of states and the monomer quantities. The aims of this proposal can then be summarized by developing such algorithms and applying them to a target system from nitrogen activation under ambient conditions, in order to help designing improved catalysts.

• DFG Programme: WBP Fellowship
• International Connection: Sweden
• Host: Professor Dr. Patrick Norman