Ivanov Mikhail, Dubernet Marie-Lise, Babikov Dmitri
PSL Research University, Observatoire de Paris, Sorbonne Universités, UPMC Univ Paris 06, ENS, UCP, CNRS, UMR8112, LERMA, 5 Place Janssen, 92195 Meudon, France.
Department of Chemistry, Marquette University, Milwaukee, Wisconsin 53201-1881, USA.
J Chem Phys. 2014 Apr 7;140(13):134301. doi: 10.1063/1.4868715.
The mixed quantum/classical theory (MQCT) formulated in the space-fixed reference frame is used to compute quenching cross sections of several rotationally excited states of water molecule by impact of He atom in a broad range of collision energies, and is tested against the full-quantum calculations on the same potential energy surface. In current implementation of MQCT method, there are two major sources of errors: one affects results at energies below 10 cm(-1), while the other shows up at energies above 500 cm(-1). Namely, when the collision energy E is below the state-to-state transition energy ΔE the MQCT method becomes less accurate due to its intrinsic classical approximation, although employment of the average-velocity principle (scaling of collision energy in order to satisfy microscopic reversibility) helps dramatically. At higher energies, MQCT is expected to be accurate but in current implementation, in order to make calculations computationally affordable, we had to cut off the basis set size. This can be avoided by using a more efficient body-fixed formulation of MQCT. Overall, the errors of MQCT method are within 20% of the full-quantum results almost everywhere through four-orders-of-magnitude range of collision energies, except near resonances, where the errors are somewhat larger.
在空间固定参考系中制定的混合量子/经典理论(MQCT)用于计算在广泛的碰撞能量范围内,He原子碰撞对水分子几个转动激发态的猝灭截面,并与在相同势能面上的全量子计算进行对比测试。在MQCT方法的当前实现中,有两个主要误差来源:一个在能量低于10 cm⁻¹时影响结果,而另一个在能量高于500 cm⁻¹时出现。也就是说,当碰撞能量E低于态到态跃迁能量ΔE时,由于其固有的经典近似,MQCT方法变得不太准确,尽管采用平均速度原理(缩放碰撞能量以满足微观可逆性)有很大帮助。在更高能量下,MQCT预计是准确的,但在当前实现中,为了使计算在计算上可行,我们不得不截断基组大小。这可以通过使用更有效的体固定形式的MQCT来避免。总体而言,在碰撞能量的四个数量级范围内,除了在共振附近误差稍大外,MQCT方法的误差几乎在各处都在全量子结果的20%以内。