Hydroformylation of Alkenylbenzenes Catalyzed by Rhodium-Phosphine Complexes: Mechanistic Features and Kinetic Behavior Resolved Through Bayesian Analysis

The hydroformylation of alkenylbenzenes remains insufficiently defined, despite the relevance of these substrates as biomass-derived aromatic feedstocks within sustainable chemical transformations. In this work, we present an experimental (catalytic and kinetic) study of their conversion into aldehydes under rhodium-phosphine catalysis, using complexes bearing mono-, bi- and tridentate phosphine ligands, [Rh(H)(CO)2(PPh3)2], [Rh(H)(CO)(triphos)] and [Rh(H)(CO)2(dppe)], under mild reaction conditions (80 °C and 2–30 bar of syngas for eugenol; 80 °C and 20–50 bar of syngas for estragole and trans-anethole). The catalytic activity order of the complexes was Rh (PPh3) > Rh(triphos) > Rh(dppe), while the substrate reactivity followed the trend eugenol > estragole >> trans-anethole. Reaction rates were measured across a wide CO and H2 pressure range, revealing redistribution between the active monocarbonyl species and an off-cycle (acyl)dicarbonyl complex that becomes dominant at elevated p(CO). The kinetic behavior observed for eugenol hydroformylation with Rh(PPh3) was consistent with the established hydroformylation sequence involving alkene coordination, hydride migration to substrate and the CO-dependent re-coordination steps that determine catalyst speciation; the subsequent transfer of the alkyl group to the carbonyl ligand and hydrogenolysis complete the catalytic cycle; the H2 addition or the hydride transfer to the alkene was identified as the rate-determining step, depending on whether low or high p(H2) values were employed. To obtain statistically reliable kinetic parameters- often challenging in hydroformylation because of parameter covariance and restricted identifiability- we complemented conventional nonlinear regression with Bayesian inference based on the Markov Chain Monte Carlo approach. The resulting posterior distributions were well centered, exhibited realistic variance and provided parameter sets that are sufficiently robust to support mechanistic interpretation and subsequent kinetic modeling.