A wide spectrum of metal-metal and metal-ligand interactions makes predicting properties of transition metal complexes particularly challenging. The most accurate quantum chemical methods for first principles predictions have high computational costs and can only be performed on small systems, while cost-effective, approximate methods suffer from large, uncontrolled errors. Here incremental Full Configuration Interaction (iFCI) is applied to mono- and bi-metallic third-row transition metal complexes to estimate spin gaps and electron density distributions. iFCI is a many-body expansion of FCI that results in smooth convergence to the FCI limit with polynomial scaling costs. While iFCI has been shown to attain 1 kcal/mol accuracy in small, main group polyatomics, it has never before been applied to transition metals. In this presentation, we will show that 3-body increments allow 1 kcal/mol accuracy, in comparison to experimental or existing high-level computational values. One such system, a bi-metallic vanadium complex, contains 26 heavy atoms and 142 electrons correlated in 380 active orbitals. This system would have a dimension of 10<sup>156</sup> in FCI, but instead is solved at approximately 10<sup>9</sup> dimensional costs using the iFCI strategy.

Towards full configuration interaction for metal complexes

2019 Spring National Meeting

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