Abstract
Species of the form CH2X–Y (X, Y = Br, I) have been proposed and identified as recombination products of the photodissociation of the parent dihalomethanes. Second-order complete active space perturbation theory (CASPT2) calculations of the vertical absorption energies considerably overestimate the experimental transient absorption band maxima, while the computationally cheaper time-dependent density functional theory (TD-DFT) method yields results with a reasonable agreement. In this work, we try to find the reason for this unexpected performance difference. In an initial study of the I2 molecule, we establish that CASPT2 is capable of providing quantitatively accurate results and that the TD-DFT values are only valid at first sight, but are qualitatively flawed. In the CASPT2 calculations for the CH2X–Y molecules, we include relativistic corrections, spin–orbit coupling, vibrational and thermal effects, and the solvent polarisation. Unfortunately, the results do not improve appreciably compared to the experimental measurements. We conclude that the good agreement of TD-DFT results is very likely fortuitous in this case as well, and that further theoretical and experimental investigations are probably needed to resolve the current discrepancy between CASPT2 and experiments.
Acknowledgements
Roland Lindh would like to thank the Swedish Research Council (VR) for financial support.
Funding
This research is supported by the National Nature Science Foundation of China [grant number 21003143], [grant number 20873010], [grant number 20720102038]; the Major State Basic Research Development Programs [grant number 2007CB815206]; SRF for ROCS, SEM. The research at KAIST is supported by funds from the National Research Foundation [grant number 2009-0053704], [grant number R111-2007-012-03001-0]; the KISTI Supercomputing Center.