Development and applications of multiconfigurational methods for spectroscopy of transition metal catalysts
Title: Development and applications of multiconfigurational methods for spectroscopy of transition metal catalysts
SNIC Project: SNIC 2021/5-24
Project Type: SNIC Medium Compute
Principal Investigator: Mickaël Delcey <delcey@kth.se>
Affiliation: Kungliga Tekniska högskolan
Duration: 2021-02-01 – 2022-02-01
Classification: 10407 10404
Keywords:

Abstract

Transition metals complexes form one of the main class of homogenous catalysts both in nature and in industry. They are however difficult to study, owing in particular to the variety of oxidation, spin and configuration states they can exist. Experimentally, X-ray spectroscopy is an important method to probe those metal centres thanks to its element-specificity. Yet, theory is often needed to extract information out of the spectra. Most quantum chemical methods can fail to adequately model the very covalent bonds of first row transition metals can form, and multiconfigurational methods then become the best available option. In recent years, we have successfully developed and applied multiconfigurational methods to study first row transition metal complexes. Several applications are currently on-going, including in particular for photosensitizers as well as catalysts for carbon activation and oxygen evolution. On top of these applications, we have now put our focus on new developments which would enable us to greatly extend the range of applicability of multiconfigurational methods, in particular to multinuclear complexes. Such complexes are frequently found in nature, for instance as catalysts for oxygen and hydrogen evolution as well as nitrogen fixation. A lot is left to understand in their chemistry, and accurate theoretical calculations are of invaluable help. The new development we are suggesting would make these studies possible, and the softwares we will write will strive to make the best possible use of modern computer architecture, especially massively parallel CPUs and GPUs.