DFT of transition metal carbonyls: sometimes it's good, sometimes it's not…

Recent computational and experimental studies of the vibrational spectra of homoleptic transition metal carbonyls (M(CO) 6 , (M = Cr, Mo, W), Fe(CO) 5 , Fe 2 (CO) 9 , Ru 3 (CO) 12 ) are reviewed. The use of neutron vibrational spectroscopy (inelastic neutron scattering, (INS)) has enabled the vibrational assignment to be completed for the first time because INS has no selection rules, so all of the modes could be observed. An integral part of the analysis of INS spectra is a comparison of experimental spectra and those generated from the outputs of density functional theory calculations. For some of these, the agreement between the experimental spectra and those generated from DFT outputs was unexpectedly poor. It is concluded (but not proven) that the most likely reason is how the long-range interactions are included in the calculations. • DFT calculated neutron vibrational spectra are usually in agreement with experiment. • For homoleptic transition metal carbonyls this is not always the case. • Inadequate treatment of long-range forces is suggested for the disagreement.