DFT Functional Benchmarking on the Energy Splitting of Chromium Spin States and Mechanistic Study of Acetylene Cyclotrimerization over the Phillips Cr(II)/Silica Catalyst

Abstract

In this work, a two-state reaction mechanism for the acetylene cyclotrimerization over a cluster model for the Phillips Cr(II)/silica catalyst were systematically investigated using density functional theory (DFT). Since spin crossover phenomenon was confirmed in the catalytic cycle, an accurate prediction of the energy gap between low- and high-spin states is crucial for the description of a reaction involving a two-state reactivity. Therefore, a massive DFT functional benchmarking test has been conducted on the cluster model by taking a CASPT2 energy gap as a reference. Consequently, B3PW91* with 28% Hartree-Fock exchange energy was selected for the following mechanistic investigation. Each of the possible potential energy surface including singlet, triplet, and quintet surfaces was explored. On the quintet surface, the reaction begins with a coordination of an acetylene on the chromium center to generate a π-coordinated complex. The following oxidative coupling through further coordination with a second acetylene was predicted to be a two-step reaction to generate a chromacyclopentadiene species. This transformation was found to be energetically prohibitive by the presence of the transition state (5)TS[C-E] (ΔG(‡) = 31.1 kcal/mol). On the triplet surface, however, the coordination of an acetylene generates a chromacyclopropene species without showing any activation barrier. The second acetylene incorporation proceeding via a coordination on the chromium center followed by an insertion into a Cr-C σ-bond of the chromacyclopropene was predicted to be a facile reaction pathway (ΔG(‡) = 10.2 kcal/mol). The third acetylene was captured by the cluster model through the formation of a hydrogen bond. The later transformation on the triplet surface was found to be an intermolecular [4 + 2] cycloaddition to finish the cyclization. The lack of the aromaticity of the benzene ring in (3)L results in an uncompleted reaction pathway on a single triplet surface. Consequently, a two-state reaction pathway that is connected by two low-lying minimum-energy crossing points (MECPs) of the two surfaces is thus described. It is worthy of note that the third acetylene in the tri(acetylene)chromium complex captured by the cluster model only through the formation of a hydrogen bond rules out the [2 + 2 + 2] concerted one-step reaction pathway proposed by Zecchina et al. [Phys. Chem. Chem. Phys.2003, 5, 4414]. The singlet reaction profile is far higher in energy compared with that proceeded on the triplet and quintet surfaces.