Interplay of Transition Metals and Noncovalent Interactions in C–H Activation Catalysis

Abstract

Selective C–H activation is crucial for the synthesis of bioactive molecules and natural products, and plays an important role in pharmaceutical industry, medicinal chemistry, and the materials industry. While synthetic routes to activate unreactive C–H bonds require harsh conditions and usually show poor selectivity, biological systems, such as non-heme iron enzymes, carry out selective C–H activation efficiently under ambient conditions. These enzymes catalyze a variety of reactions including C–H halogenation, hydroxylation, epoxidation, and ring closures, several of which are mediated with the help of noncovalent interactions such as hydrogen bonds (HBs). Most reactions share a common catalytic pathway with the formation of a reactive ferryl intermediate which is hard to be characterized experimentally. Computational studies of these enzymes help to bridge the gap in experiments towards understanding enzyme mechanism and selectivity.

In this thesis, we study the interplay of noncovalent interactions and transition metals in C–H activation catalysis using quantum mechanical simulations. We employ density functional theory (DFT) and wavefunction theory to perform an extensive computational study of protein HB interactions and transition metal complex (TMC) active sites in non-heme iron halogenases and hydroxylases. Due to the fleeting nature of the ferryl intermediate, experimentalists tend to use vanadyl mimics in order to better understand the ferryl intermediate. However, these metals exhibit distinct electronic structure, motivating us to investigate if vanadyl mimics are indeed faithful to the native ferryl intermediates. Studying the mechanism of metalloenzymes using first principles methods could be challenging due to the larger system sizes. Thus, we also try to understand C–H activation carried out by 3d TMCs focusing on the specific case of partial oxidation of methane to methanol. While the oxidation and spin states of the metals in the enzyme active site are well defined through spectroscopic methods, that is not the case with TMC catalysts. Thus, modeling TMC catalysts is accompanied by the twin challenges of identifying the ground spin state and determining the appropriate method to identify the ground state since properties such as reaction energies and scaling relations are sensitive to the computational method used. Additionally, the ability of TMCs to exist in multiple spin states is often leveraged for practical applications, with one such example being spin crossover (SCO) complexes that exhibit a change in spin state as a function of external stimulus like temperature and are widely studied due to their increasing use in molecular switches. We curate an experimental data set of 95 Fe(II) SCO complexes and predict SCO behavior using DFT with the aim of identifying the best performing functional. This in turn sets the stage to design SCO complexes with tailed properties such as those that exhibit SCO behavior at room temperature. We expect that the insights from this work can directly guide efforts on biomimetic chemistry as well as both biological and synthetic C–H activation catalysis.