We apply the recently developed phase integration method (PIM) (Monteferrante et al. Mol Phys. 2011;109:3015-3027) to the calculation of infrared spectra of gas phase molecules. The PIM combines a generalised Monte Carlo sampling of the exact thermal density of the system with classical molecular dynamics to obtain approximate time quantum correlation functions. To describe the molecules, we adopt very simple analytical potentials that have, however, proved interesting, and surprisingly challenging, benchmarks for approximate quantum dynamical schemes. We show that, in contrast with two other commonly applied methods, our spectra do not exhibit spurious features or unphysical shifts depending on the temperature. Identifying the positions of the peaks requires only a few tens of trajectories, while an accurate evaluation of the relative intensities of the peaks is computationally more demanding.
Gas phase infrared spectra via the phase integration quasi-classical method / J., Beutier; Michele, Monteferrante; Bonella, Sara; Rodolphe, Vuilleumier; Ciccotti, Giovanni. - In: MOLECULAR SIMULATION. - ISSN 0892-7022. - STAMPA. - 40:1-3(2014), pp. 196-207. [10.1080/08927022.2013.843776]
Gas phase infrared spectra via the phase integration quasi-classical method
BONELLA, SARA;CICCOTTI, Giovanni
2014
Abstract
We apply the recently developed phase integration method (PIM) (Monteferrante et al. Mol Phys. 2011;109:3015-3027) to the calculation of infrared spectra of gas phase molecules. The PIM combines a generalised Monte Carlo sampling of the exact thermal density of the system with classical molecular dynamics to obtain approximate time quantum correlation functions. To describe the molecules, we adopt very simple analytical potentials that have, however, proved interesting, and surprisingly challenging, benchmarks for approximate quantum dynamical schemes. We show that, in contrast with two other commonly applied methods, our spectra do not exhibit spurious features or unphysical shifts depending on the temperature. Identifying the positions of the peaks requires only a few tens of trajectories, while an accurate evaluation of the relative intensities of the peaks is computationally more demanding.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.