The computation, starting from basic principles, of chemical reaction rates in realistic systems (with three or more degrees of freedom) has been a longstanding goal of the chemistry community. Our current work, which merges tube dynamics with Monte Carlo methods provides some key theoretical and computational tools for achieving this goal. We use basic tools of dynamical systems theory, merging the ideas of Koon et al. [W.S. Koon, M.W. Lo, J.E. Marsden, S.D. Ross, Heteroclinic connections between periodic orbits and resonance transitions in celestial mechanics, Chaos 10 (2000) 427-469.] and De Leon et al. [N. De Leon, M.A. Mehta, R.Q. Topper, Cylindrical manifolds in phase space as mediators of chemical reaction dynamics and kinetics. I. Theory, J. Chem. Phys. 94 (1991) 8310-8328.], particularly the use of invariant manifold tubes that mediate the reaction, into a tool for the computation of lifetime distributions and rates of chemical reactions and scattering phenomena, even in systems that exhibit non-statistical behavior. Previously, the main problem with the application of tube dynamics has been with the computation of volumes in phase spaces of high dimension. The present work provides a starting point for overcoming this hurdle with some new ideas and implements them numerically. Specifically, an algorithm that uses tube dynamics to provide the initial bounding box for a Monte Carlo volume determination is used. The combination of a fine scale method for determining the phase space structure (invariant manifold theory) with statistical methods for volume computations (Monte Carlo) is the main contribution of this paper. The methodology is applied here to a three degree of freedom model problem and may be useful for higher degree of freedom systems as well.

Keywords: Dynamical systems; Transport rates; Normal forms; Invariant manifolds; Transition state theory; Ionization; Atoms in crossed fields