Theoretical Study On Ethylene Dimerization And Alkyne Cyclotrimerization Catalyzed By Cr-Based Catalysts

Phillips chromium catalyst is one of the most important industrial catalysts, which is producing more than10million tons polyethylene annually. After more than50years industrial and academic researches, the precise structure of the active site, the active oxidation state of the chromium center and the initiation mechanism for ethylene polymerization still remain unclear for the Phillips chromium catalyst. In particular, Phillips chromium catalyst could initiate ethylene polymerization without using any organometallic cocatalyst. As reported in the literature, ethylene dimerizes into1-butene on Cr(Ⅱ)OH+and Cr(Ⅲ)O+ cations, and acetylene (methylacetylene) cyclotrimerizes into benzene (trimethylbenzene) catalyzed by Phillips chromium catalyst. Both of these reactions are free of cocatalyst. The study on these two catalytic systems could possibly shed some light on the understanding of the active site and the initiation mechanism of the Phillips catalyst. In this work, theoretical calculations from first principle have been employed to investigate the mechanisms of ethylene dimerization over Cr(Ⅱ)OH+and Cr(Ⅲ)O+cations (Chapter3and Chapter4) and acetylene and methylacetylene cyclotrimerization catalyzed by the Phillips chromium catalyst (Chapter5and Chapter6).The mechanistic study of ethylene dimerization catalyzed by Cr(Ⅱ)OH+and Cr(Ⅲ)O+cations was presented in chapter3and chapter4, respectively. On the quintet potential energy surface, Cr(Ⅱ)OH+ cation could coordinate with up to five ethylene molecules which gives eight possible stable Cr(II)OH+·(C2H4)n (n=1-5) π-complexes, determined at a medium level of theory. However, the complex Cr(Ⅱ)OH+·(C2H4)5turned into Cr(Ⅱ)OH+·(C2H4)4when optimized at a higher level of theory. Interestingly, all7Cr(Ⅲ)O+·(C2H4)n (n=1-5) π-complexes with up to5coordinated ethylenes were survived at all levels. For ethylene dimerization catalyzed by Cr(Ⅱ)OH+and Cr(Ⅲ)O+cations, the reaction follows a triplet metallacycle mechanism, which rules out the proposed carbene mechanism in the literature. Thus, the spin-flipping reaction from high-spin potential energy surface to the adjacent low-spin potential energy surface is determined by presenting an minimum energy crossing point between two surfaces. The reactivity of the catalytic cycle was calculated using the recently developed energetic span model theory. Finally, a higher catalytic reactivity of the Cr(Ⅱ)OH+cation over the Cr(Ⅲ)O+cation is recognized through a comparison of the results in chapter3and chapter4.A detailed DFT functional benchmarking test has been performed on a widely used Cr-based cluster model (Cr(Ⅱ)/SiO2) for the Phillips chromium catalyst. The CAPST2singlet-quintet energy gap (1-5ΔE) of Cr(Ⅱ)/SiO2was taken as a reference for the DFT functional benchmarking test. All the210DFT functionals, including180non-hybrid DFT functionals and30hybrid DFT functionals, were employed during the benchmarking test. In general, the exchange functional in the prediction of1-5ΔE for the cluster model are in the following sequence:PW91, TPSS, mPW, PBEh, wPBEh, PBE<B<G96<PKZB<O<BRx,while the correlation functionals in the prediction of1-5ΔE for the cluster model are in the following sequence:PL<VWN<VWN5<LYP, V5LYP<B95<P86<KCIS<PKZB<VP86<PW91<TPSS<PBE. That is to say, PW91PL composed with an exchange functional PW91and a correlation functional PL tends to give the smallest1-5ΔE in all cases. However, most of the non-hybrid DFT functionals underestimated the singlet-quintet energy gap. By introducing part of the Hartree-Fock exchange energy, all the hybrid DFT functionals show a better performance for the prediction of the singlet-quintet energy gap. Eventually, a hybrid DFT functional B3PW91with28%Hartree-Fock exchange energy was recognized and thus selected for all of the following mechanistic studies in chapter5and chapter6.The mechanistic study of acetylene and methylacetylene cyclotrimerization catalyzed by Cr(Ⅱ)/SiO2cluster model was presented in chapter5and chapter6, respectively. All the reaction pathways on the quintet potential energy surface failed to show any observable catalytic reactivity for acetylene and methylacetylene cyclotrimerization catalyzed by Cr(Ⅱ)/SiO2cluster model. The spin-flipping reaction was prohibited by presenting a very large triplet-quintet energy gap of40.9kcal/mol of the Cr(Ⅱ)/SiO2cluster model. Interestingly, the triplet-quintet energy gaps of the complexes were reduced enormously through the coordination of the monomers. Therefore, a spin-flipping reaction is highly expected at the alkyne coordinated complexes. After that, the acetylene and methylacetylene cyclotrimerization on the triplet surface follows a [4+2] cycloaddition mechanism, which rules out the proposed [2+2+2] concerted one-step cycloaddition mechanism in the literature. It is worthy of note that the acetylene cyclotrimerization prefers a catalytic cycle on a single triplet surface rather than a two-state reactivity mechanism involving two spin-inversion reactions. For methylacetylene cyclotrimerization, three pathways produce1,2,4-trimethylbenzene, while one pathway generates1,3,5-trimethylbenzene. By employing a silica supported cluster model, all the four pathways showed much improved reactivities with an improved selectivity of1,3,5-trimethylbenzene in the products.