Insights into catalytic activities of noble transition metal complexes from quantum-chemical calculations

Abstract

Homogeneous catalysis at transition metal centres is an essential and ubiquitous tool for the regioselective direct synthesis of fine chemicals from abundant resources. Quantum chemical models are used to gain insights into mechanisms of homogeneous catalysis with organo-transition metal complexes. The resulting overall kinetic barriers and barriers at branching leading to different products can help in steric and electronic tuning of ligands at the metal centre to control chemo- and regioselectivities. This thesis presents computational investigations on detailed mechanisms of alkene and alkyne alkoxycarbonylation at palladium catalysts and ruthenium–catalysed selective reduction of cardanol derivatives via transfer hydrogenation using state-of-the-art Density Functional Theory (DFT).

We have explored catalytic methoxycarbonylation of ethene with a bidentate tertiary phosphine (DTBPX) and palladium. Out of three pathways, (i) carbomethoxy, (ii) ketene, and (iii) hydride-hydroxyalkylpalladium, the latter is the most plausible pathway with a computed selectivity of >99% towards the formation of methyl propanoate (MePro) and a reasonably low overall kinetic barrier of 17.8 kcal mol⁻¹. Consistent with experimental data, for a less bulky bidentate phosphine, the overall barrier increases to 30.1 kcal mol⁻¹.

We have revisited in situ base mechanism of alkyne alkoxycarbonylation via a Pd catalyst with hemilabile P,N-ligands (PyPPh₂, Py = 2-pyridyl). Newly characterised acryloyl and η³-propen-1-oyl intermediates readily undergo methanolysis. The new mechanism is consistent with experimental data in terms of ligand effects on the reaction rates and selectivities, where (6-Cl-Py)PPh₂ has been shown to improve both selectivity and turnover. We further have explored the formation of a highly stable π-allyl intermediate as the reason for catalyst poisoning due to the presence of propadiene in technical propyne. Predicted regioselectivities suggest that at least 11 % of propadiene should yield this π-allyl intermediate, which requires an insurmountable barrier of 25.8 kcal mol⁻¹ to proceed to the product via methanolysis. A new ligand, (6-Cl-3-Me-Py)PPh₂, is proposed, which is more tolerant to propadiene poisoning (as only 0.2% of propadiene gets trapped at the π-allyl complex) and tremendously active (as decreases the overall barrier to 9.1 kcal mol⁻¹).

Finally, we have calculated the complete pathway leading to the monoene cardanol at ruthenium–catalysed selective reduction of cardanol derivatives via transfer hydrogenation. The overall barrier of 29.2 kcal mol⁻¹ associated with the transfer hydrogenation leading to the monoene cardanol is surmountable under the elevated temperatures of the experiments (refluxing iso-propanol). The computed barrier of 46.6 kcal mol⁻¹ leading to a fully saturated cardanol product explains the 100% selectivity towards monoene cardanol.