Derivation of a true (t->0+) quantum transition-state theory. II. Recovery of the exact quantum rate in the absence of recrossing

TL;DR

This paper demonstrates that ring-polymer molecular dynamics transition-state theory (RPMD-TST) accurately predicts quantum reaction rates in the absence of recrossing, validating many previous RPMD rate calculations.

Contribution

It proves RPMD-TST is exact without recrossing and clarifies its applicability limits, supporting its use in quantum reaction rate calculations.

Findings

RPMD-TST is exact if no recrossing occurs.

RPMD-TST provides good approximations at moderate temperatures.

Theoretical validation of RPMD rate calculations.

Abstract

In Part I [J. Chem. Phys. 138, 084108 (2013)] we derived a quantum transition-state theory by taking the t->0+ (short-time) limit of a new form of quantum flux-side time-correlation function containing a ring-polymer dividing surface. This t->0+ limit appears to be unique in giving positive-definite Boltzmann statistics, and is identical to ring-polymer molecular dynamics (RPMD) TST. Here, we show that quantum TST (i.e. RPMD-TST) is exact if there is no recrossing (by the real-time quantum dynamics) of the ring-polymer dividing surface, nor of any surface orthogonal to it in the space describing fluctuations in the polymer-bead positions along the reaction coordinate. In practice, this means that RPMD-TST gives a good approximation to the exact quantum rate for direct reactions, provided the temperature is not too far below the cross-over to deep tunnelling. We derive these results by comparing the long-time limit of the ring-polymer flux-side time-correlation function with that of a hybrid flux-side time-correlation function (containing a ring-polymer flux operator and a Miller-Schwarz-Tromp side function), and by representing the resulting ring-polymer momentum integrals as hypercubes. Together with Part I, the results of this article validate a large number of RPMD calculations of reaction rates.