Answers are grounded in this paper and cite their sources.
We examine how the angular potential affects entry into the inner well and roaming in a modification of Chesnavich’s model of barrierless ion–molecule dissociation. The model has two degrees of freedom. We replace V0(r)sin2 θ by B(r)sin2 θ, so that the frequency of small angular oscillations has an interior maximum. With the Chesnavich radial potential VCh(r), at E = 2 kcal/mol and with the same construction of inward initial conditions, the percentage entering the well falls from 57% to 15%. With B(r)sin2 θ, most trajectories return directly to large separation. We classify roaming by at least three crossings of a surface at a specified radius, called the classifier radius. The nonreactive roaming fractions are similar at 3.5 Å, but appreciable roaming occurs at larger sampled classifier radii with B(r)sin2 θ. Increasing V0(r) by a constant factor shows that an interior frequency maximum is not required for reduced entry or appreciable roaming at larger classifier radii. For VLJ(r) + B(r)sin2 θ, with VLJ(r) the Lennard-Jones radial potential, we locate tight, free rotor, and outer orbiting transition states as unstable periodic orbits. The directional flux through surfaces of constant radius has a local minimum defining an entropic bottleneck: angular phase space is restricted although the radial potential has no barrier. At this minimum, the flux through one of the two equivalent allowed angular intervals around θ = 0 and θ = π exceeds the dividing surface flux of one corresponding tight orbit by 0.15%. Noncaptured return time distributions are nonexponential, combining faster direct returns with slower roaming returns.
No takes yet. Share an insight, caveat, or question.
Stephen Wiggins (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: