Novel experimental techniques and computational methods have provided new insight into the behavior of reactive intermediates in solution. The results of these studies show that some of the earlier ideas about how reactive intermediates ought to behave in solution were incomplete or even incorrect. This Perspective summarizes the new experimental and computational methods and draws attention to the shortcomings that their application has brought to light in previous models. Key areas needing further research are highlighted.

Classical molecular dynamics simulations are reported for the deazetisation and ring opening of meso-2,3-difluoro-2,3-dimethyldiazocyclopropane in three solvents: CHCl3, CHFClBr and CH3CH(OH)CF3 (TFIPA). The achiral reactant leads to enantiomeric allene products, and the question addressed in the study is whether either of the chiral, enantiomerically pure solvents can induce significant enantiomeric excess in the products. The direct dynamics calculations use an empirical valence bond potential for the solute, with empirical parameters optimised against M06-2X/cc-pVTZ density functional results. The results reveal that the exothermic N2 loss and ring opening promote transient strong solvent–solute interactions within the first ∼100 fs of the reaction. Because of the bifurcating reaction path, these interactions occur at time when the “decision” about which enantiomer of the product to form has yet to be made (at least for many of the trajectories). Hence, it is possible in principle that the solvent could exert a larger-than-normal influence on the course of the reaction. In fact, the results reveal no such effect for CHFClBr but do predict that TFIPA should induce 15.2 ± 2.1% enantiomeric excess. This is roughly an order of magnitude larger than solvent-induced enantiomeric excesses found experimentally in reactions where the conversion of reactant(s) to enantiomeric products occur over separate transition states.

Following previous work [ J. Chem. Phys. 2013, 139, 154108] on a simple model of a reaction with a post-transition state valley ridge inflection point, we study the chemically important example of the electrocyclic cyclopropyl radical ring-opening reaction using direct dynamics and a reduced dimensional potential energy surface. The overall reaction requires con- or disrotation of the methylenes, but the initial stage of the ring-opening involves substantial internal rotation of only one methylene. The reaction path bifurcation is then associated with the relative sense of rotation of the second methylene. Clear deviations of reactive trajectories from the disrotatory intrinsic reaction coordinate (IRC) for the ring-opening are observed and the dynamical mechanism is discussed. Several features observed in the model system are found to be preserved in the more complex and higher dimensional ring-opening reaction. Most notable is the sensitivity of the reaction mechanism to the shape of the potential manifested as a Newtonian kinetic isotope effect upon deuterium substitution of one of the methylene hydrogens. Dependence of the product yield on frictional dissipation representing external environmental effects is also presented. The dynamics of the post-transition state cyclopropyl radical ring-opening are discussed in detail, and the use of low dimensional models as tools to analyze complicated organic reaction mechanisms is assessed in the context of this reaction.

A molecular dynamics simulation reveals the occurrence of nonstatistical dynamical effects in the ring-opening and subsequent [1,5] H migration of bicyclo[2.1.0]pent-2-ene. The symptoms of the effects do not show up in the overall kinetics or product branching ratios of the reaction, which are well explained by a master-equation analysis, but in an oscillatory preference for migration of the two methylene hydrogens. It is predicted that these oscillations could have an observable effect on final product ratios in isotopically labeled analogues, and that the effect might be greater in certain solvents than in the gas phase.

The formation of spiropentane, by addition of singlet (1A1) methylene to methylenecyclopropane, and the unimolecular reactions of spiropentane have all been studied computationally. Benchmark calculations on two key biradicals were conducted by the multireference Mukherjee’s coupled-cluster (MkCC) method. Various single-reference coupled-cluster methods and multireference second-order perturbation theory were then compared for accuracy against experimental data and the MkCC results. The object of the exercise was to get the best possible description of the potential energy surface for formation and reactions of spiropentane, as a prelude to molecular dynamics simulation of the reactions. The principal conclusions of the study were that none of the unimolecular reactions of spiropentane can be classified as pericyclic processes and that the observed stereoselectivities are probably of dynamical origin. A possible resolution of a disagreement between two studies on the dynamics of cyclopropanation reactions is also offered. Of the various approximate computational models evaluated in this study, the best fit came from a composite coupled-cluster approach in which the lower-energy result was selected from a restricted coupled-cluster and a broken-symmetry, unrestricted coupled-cluster calculation on each stationary point. However, such an approach is not strictly defensible, since coupled-cluster methods are not variational, and so further evaluation of its validity would be desirable.

Classical molecular dynamics simulations are reported for the deazetisation and ring opening of meso-2,3-difluoro-2,3-dimethyldiazocyclopropane in three solvents: CHCl3, CHFClBr and CH3CH(OH)CF3 (TFIPA). The achiral reactant leads to enantiomeric allene products, and the question addressed in the study is whether either of the chiral, enantiomerically pure solvents can induce significant enantiomeric excess in the products. The direct dynamics calculations use an empirical valence bond potential for the solute, with empirical parameters optimised against M06-2X/cc-pVTZ density functional results. The results reveal that the exothermic N2 loss and ring opening promote transient strong solvent–solute interactions within the first ∼100 fs of the reaction. Because of the bifurcating reaction path, these interactions occur at time when the “decision” about which enantiomer of the product to form has yet to be made (at least for many of the trajectories). Hence, it is possible in principle that the solvent could exert a larger-than-normal influence on the course of the reaction. In fact, the results reveal no such effect for CHFClBr but do predict that TFIPA should induce 15.2 ± 2.1% enantiomeric excess. This is roughly an order of magnitude larger than solvent-induced enantiomeric excesses found experimentally in reactions where the conversion of reactant(s) to enantiomeric products occur over separate transition states.