Predictive theory of roaming reaction branching fractions
In plain words
In a roaming reaction a molecule starts to break apart, but the loose fragment moves around the rest of the molecule and then removes an atom from it instead of leaving. Simulations reproduce known cases but cannot yet predict when roaming will be important.
Precise statement
For unimolecular decompositions with roaming channels ($\mathrm{H2CO} \to \mathrm{H2} + \mathrm{CO},\ \mathrm{CH3CHO} \to \mathrm{CH4} + \mathrm{CO},\ \mathrm{NO3} \to \mathrm{NO} + \mathrm{O2}$), derive a statistical or dynamical rate theory that predicts the roaming branching fraction and product state distributions versus internal energy from the potential energy surface without full trajectory ensembles, and a criterion on the potential for when roaming exceeds 10 percent of a channel. An answer is such a theory matched to quasiclassical trajectories and state-resolved imaging data on at least three systems.
What would settle it
Comparison of the theory's branching fractions with trajectory calculations and velocity-map imaging experiments on several molecules.
Status in the literature
A 2020 review (Suits, Annu. Rev. Phys. Chem., https://doi.org/10.1146/annurev-physchem-050317-020929) describes roaming as widespread but without a general predictive theory.