An assessment of theoretical methods for the study of transition metal carbonyl complexes: [Cl2Rh(CO)2]− and [Cl2Rh(CO)]− as case studies

Abstract

A comparative theoretical study of the equilibrium geometries, vibrational frequencies, and reaction energetics of Rh–CO bond activation in the [Cl2Rh(CO)2]− and [Cl2Rh(CO)]− complexes has been performed using the B3LYP density functional method and the traditional correlated second-order Møller–Plesset (MP2), quadratic configuration interaction with single and double substitutions (QCISD) and coupled-cluster single double (triple) [CCSD(T)] methods. CCSD(T) is employed herein as a benchmark method to examine the validity of the B3LYP and MP2 methods in studies of transition metal complexes. The results show that the geometries and energies obtained with the B3LYP method agree quite well with the QCISD and CCSD(T) results. The conventional MP2 method yields poor results for all geometries, vibrational frequencies, and reaction energies. It is noteworthy that the relative reaction energies calculated at the highest correlated CCSD(T) level using the B3LYP and MP2 geometries are comparable even though the B3LYP and MP2 geometries differ significantly from each other. The absolute energies calculated with the QCISD and CCSD(T) methods at the B3LYP geometries are all lower than those obtained with the MP2 geometries, indicating that the B3LYP method could be more reliable for the study of the geometries and energetics of the catalytic reactions. Basis set effects have been examined by B3LYP calculations. The present results provide a comprehensive assessment of the widely used ab initio theoretical methods for the study of transition metal carbonyl complexes.