Free Energy Profile of a Model Palladium Catalyzed Fluorination of Aryl Bromide with Cesium Fluoride

Theoretical design of new ligands to boost reaction rate and selectivity in palladium-catalyzed aromatic fluorination

The development of palladium-catalyzed fluorination with biaryl monophosphine ligands has faced two important problems that limit its application for bromoarenes: the formation of regioisomers and insufficient catalysis for heteroaryl substrates as bromothiophene derivatives. Overcoming these problems requires more ligand design. In this work, reliable theoretical calculations were used to elucidate important ligand features necessary for achieving more rate acceleration and selectivity. These features include increasing the ligand-substrate repulsion and creating a negative charge in the space around the fluoride ion bonded to the palladium. The investigated L5 ligand presents these features, and the calculations predict that this ligand completely suppresses the regioisomer formation in the difficult case of 4-bromoanisole. In addition, the free energy barriers are decreased by 2-3 kcal mol-1 in comparison with the catalysis involving the AlPhos ligand. Thus, the present study points out a direction for new developments in palladium-catalyzed fluorination.

Theoretical free energy profile and benchmarking of functionals for amino-thiourea organocatalyzed nitro-Michael addition reaction

Amino-thiourea organocatalysis is an important catalytic process for enantioselective conjugate addition reactions. The interaction of the reactants with the catalyst has a substantial effect of dispersion forces and is a challenge for a reliable description when applying density functional theory. In this report, the classical addition of acetylacetone to β-nitro-styrene catalyzed by Takemoto's catalyst in toluene was studied using the PBE functional for geometry optimization and the DLPNO-CCSD(T) benchmark method for single point energy. The complete free energy profile calculated for the reaction was able to explain all experimental observations, including the fact that the carbon-carbon bond formation step is rate-determining. The overall barrier was calculated to be 22.8 kcal mol^-1 (experimental value approximately 20 kcal mol^-1), and the enantiomeric excess was calculated to be 88% (experimental value in the range of 84 to 92%). Some functionals were tested for single point energy. The hybrid B3LYP presented a high mean absolute deviation (MAD) from the DLPNO-CCSD(T) benchmark method by approximately 20 kcal mol^-1. The inclusion of empirical dispersion correction in the B3LYP method decreased the MAD to 6 kcal mol^-1. Even the double-hybrid mPW2-PLYP and B2GP-PLYP methods had MAD values of approximately 5 kcal mol^-1. The inclusion of the dispersion correction decreased the MAD to 3.6 kcal mol^-1. M06-2X and ωB97X-D3 were the most accurate among the tested functionals, with MADs of 2.5 kcal mol^-1 and 1.8 kcal mol^-1, respectively. Additivity approximation of the correlation energy was also tested and presented a MAD of only 0.6 kcal mol^-1.