Mechanistic studies of Tsuji-Trost type reaction using density functional theory (DFT)

Abstract

The transition metal-catalyzed alkylation that involves π-allyl intermediate is a hot area because of its wide applications. A representative example is the Tsuji-Trost type reaction. The difference in the centre transition metal element, the ligand, the reactants, and even the solvent environment, could result in the similar or different outcomes. The theoretical DFT investigation on Tsuji-Trost type reaction would enhance the understanding of homogeneous metallic catalytic behaviour and the influence of its electronic structure, which potentially provide the guidance for tuning the properties of the existing catalyst series and the novel catalyst analogues design for the future demand. This thesis contains the DFT mechanistic study on the application of homogeneous Pd and Rh catalytic system in Tsuji-Trost type reaction, to reveal how Tsuji-Trost type reaction is influenced by varied factors (including non-covalent interactions) at the molecular level. The thesis consists of the literature review for both experimental and computational studies on this area (Chapter 1); the theoretical background of quantum chemistry and DFT (Chapter 2), the case study on Pd DAAA reaction (Chapter 3), asymmetric Rh-catalytic allylation (Chapter 4), asymmetric catalytic alkylation on sulfone controlled by Pd and Rh catalyst (Chapter 5), and the future work at those three projects (Chapter 6).
Chapter 1 reviewed the research on Tsuji-Trost type reaction from both experiment and computational method. The experimental research is extensive, and the Tsuji-Trost type reaction now is widely applied in different purpose synthetic work because its extraordinary selectivity and convenience. But the computational research is incomplete. Now what has known about the Tsuji-Trost type reaction theoretically is, the mechanism and selectivity are affect by different factors, including the steric effect, the hydrogen-bonding interaction, the dispersion force and so on. The current understanding cannot support the design or optimisation of catalyst for different synthetic requirements, for instance, the demand on tuning the diastereoselectivity. Moreover, different transition metal elements with proper ligands would behave similar to Pd. Therefore, the theoretical study of Tsuji-Trost type reaction is in urgent demand, and it would benefit the understanding of all transition-metal involved homogeneous catalytic reactions.
In Chapter 2 we discussed why theoretical chemistry is now application to the reaction mechanistic study and the methods that can be chosen. The theoretical study of chemistry problem base on the physical understanding of the particles properties including the movement. Schrödinger Equation is well-accepted as the standard description of small particles’ behaviour. But Schrödinger Equation cannot be solved when system contain more than two electrons, because the electron-electron interactions cannot be solved by current theory. So, researchers tried different approaches to estimate the system properties, especially to those properties they want to study. To solve chemistry problems, the HF method is the first milestone that was proposed to roughly estimate the properties of the system, but the HF result generally is too rough to be used. Now DFT and post-HF methods can be applied to well estimate most of the properties of the system that we study in chemistry, and the corresponding basis set, solvation model and weak interaction study tools like NCIPLOT were developed for various research purposes. The computational study on theory of chemistry problem is now more and more popular and adoptable.
In this thesis we conducted DFT study on the Pd DAAA reaction, asymmetric Rh-catalytic allylation by using sulfinamide as the reactant, and the asymmetric alkylation on sulfone allene by Pd and Rh catalysts. In chapter 3, the Pd DAAA reaction research, initially our DFT results presented the free energy difference between the TSs that forming the desired product and the competitive isomer but failed to give a convincible conclusion on what causes this difference. To understand the nature of this problem, NBO and NCIPLOT were used to investigate the weak interactions within the structures of desired TS and the diastereomer TS. The structures were re-optimised by dispersion-corrected method for the need of long-distance interaction consideration. Finally, we conclude that the diastereoselectivity of this Pd DAAA reaction origins from the bond angle stretch of sulfinamide t-Bu-S-N, which possibly related to the steric hindrance between sulfinamide and the ligand on Pd. In the future, the ligand investigation may give us more insight into this reaction.
The study of sulfinamide asymmetric Rh-catalytic allylation in chapter 4 was more complicated than the Pd DAAA reaction – it requires the consideration on regioselectivity, the reason that using linear and branched carbonate generated the same product, and the counter-ion effect, generated by Cl. As for the result, the mechanism suggests the allyl would be detached from the carbonate at the first step of the reaction, which explained why selectivities are the same when using linear and branched carbonate as the reactant. And the analysis on selectivity-related TSs structures indicates the steric hindrance at the reaction site is the key of the regio- and diastereoselectivity, and the Cl might contribute to this issue by occupying one coordinate region. The base is also important to the reaction because the bonding from Na to sulfinamide helps to expose the reaction site to Rh-allyl. Therefore, tuning the reaction selectivity would request the consideration on the steric effect of catalyst ligand and counter-ion. Factors such as the difference of the size of reactant ring, substituents on reactant ring, and the branch of carbonate are irrelevant to the reaction selectivity. The sulfinamide in this case directed the selectivity by introducing the steric difference between different side of reactant ring. More work can be done in the future on the effect by different the bases, solvents, and ligands.
The study in chapter 5 of the asymmetric catalytic alkylation on sulfone controlled by Pd and Rh catalyst is the most interesting – the syn-preferred reaction selectivity would be reversed by simply changing the transition metal element from Rh to Pd, or the removal of acid as the additive. The study was not completed on the Rh and Pd catalytic cycle with the presence of acid. Acid-free Rh catalytic route investigation was finished and well-explained experimental finding. The favour in anti-product origins from the rigid structure generated in allene hydrometallation. When acid existed in the system, the mechanism would be totally changed because the active proton is no longer the amide hydrogen but the acid hydrogen. The acid would bond on centre atom cause higher steric hindrance, which prevent us from obtaining a reasonable transition-state of the final allene amidation step. The work can be finished in the future by completing the exploration of these two catalytic cycles.

Date of Award	Jul 2023
Original language	English
Awarding Institution	University of Nottingham
Supervisor	Cheng Heng Pang (Supervisor) & Bencan Tang (Supervisor)