Collision-induced absorption in Ar–Kr gas mixtures: A molecular dynamics study with new potential and dipole data
We have implemented a scheme for classical molecular dynamics simulations of collision-induced absorption. The program has been applied to a gas mixture of argon (Ar) and krypton (Kr). The simulations are compared with accurate quantum dynamical calculations. The comparisons of the absorption coeffi...
Gespeichert in:
Veröffentlicht in: | The Journal of chemical physics 2019-10, Vol.151 (14), p.144303-144303 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | We have implemented a scheme for classical molecular dynamics simulations of collision-induced absorption. The program has been applied to a gas mixture of argon (Ar) and krypton (Kr). The simulations are compared with accurate quantum dynamical calculations. The comparisons of the absorption coefficients show that classical molecular dynamics is correct within 10% for photon wave numbers up to 220 cm−1 at a temperature of 200 K for this system. At higher temperatures, the agreement is even better. Molecular dynamics accounts for many-body interactions, which, for example, give rise to continuous dimer formation and destruction in the gas. In this way, the method has an advantage compared with bimolecular classical (trajectory) treatments. The calculations are carried out with a new empirical Ar–Kr pair potential. This has been obtained through extensive analysis of experimental thermophysical and transport properties. We also present a new high level ab initio Ar–Kr potential curve for comparison, as well as ab initio interaction-induced dipole curves computed with different methods. In addition, the Ar–Kr polarizability and hyperpolarizability are reported. A comparison of the computed absorption spectra with an experiment taken at 300 K shows satisfactory agreement although a difference in absolute magnitude of 10%–15% persists. This discrepancy we attribute mainly to experimental uncertainty. |
---|---|
ISSN: | 0021-9606 1089-7690 1089-7690 |
DOI: | 10.1063/1.5099700 |