Studies of chemical reactivity in the condensed phase. I. The dynamics of iodine photodissociation and recombination on a picosecond time scale and comparison to theories for chemical reactions in solution

Abstract

Picosecond transient absorption measurements from 1000-295 nm are used to monitor the recombination dynamics of iodine after photodissociation in a variety of inert solvents. The high time resolution and signal-to-noise ratio of these measurements permits the development of a detailed model of this reaction, which should resolve disagreements over the time scales of geminate recombination and vibrational relaxation and over the role of excited electronic state trapping. Most of the atoms which undergo geminate recombination do so in < 15 ps, in agreement with the predictions of existing molecular dynamics simulations. The subsequent vibrational and electronic energy relaxation of the recombined molecule is relatively slow and accounts for most of the transient absorption dynamics. The relaxing X-state vibrational population distribution is extracted with an approximate method using calculated spectra of the excited vibrational levels and is compared to recent models. Vibrational relaxation times vary from -15 ps near the middle of the ground state well to -150 ps for complete relaxation to v = O.The vibrational relaxation rates do not provide support for the predicted role of resonant vibration-to-vibration energy transfer to chlorinated methane solvents, but some evidence for this mechanism is found in alkane solvents. B-state predissociation times of 10--15 ps and A '-state lifetimes of 65-2700 ps are found depending on the solvent. Current theory is not able to satisfactorily explain the large variation of the A '-state lifetime in various solvents. 788

A classical molecular dynamics study of recombination reactions in a microporous solid

“…This is in agreement with previous simulations on similar systems, [21][22][23][24] as well as with our results. Moreover, the agreement with experimental results 29,30 is connected to the observation that the curve-crossing process ͑i.e., the jump from a repulsive unbound surface or an excited bound to the ground state, preceding the recombination͒ is very fast in the system considered here; for this reason, MD simulations could adequately predict recombination times without explicitly including surface-hopping models, as in the present study. The inclusion of all electronically excited states in the simulation could probably slow down the recombination: the occasional trapping of the molecule in an excited bound electronic state may slow the recombination rate down by a factor of five.…”

“…However, usually MD simulations effectively reproduce energy transfer and solvent caging effects, and also with a simplified model, reasonable predictions on the time scales for geminate recombination should arise. In fact, the calculated recombination times are in good agreement with experimental and theoretical data on the iodine recombination; [20][21][22][23][24][25][26][27][28][29][30][31] therefore, in the following, we will refer to the reacting species as ''iodine,'' in spite of the abovementioned limitations.…”

“…In particular, V-T transfer seems to dominate when the excited molecule moves in the upper half of the Morse well ͑i.e., during the more crucial step of the overall recombination process͒. Harris and coworkers 29 did not find evidence of resonant energy transfer to solvent vibrational modes in polyatomic chlorinated solvents. The present study of the recombination process in the two environments, with the choice of a structureless model of CCl 4 , entails a direct comparison of the effectiveness of the two energy-transfer mechanisms: in each environment a different relaxation mechanism is active.…”

Molecules in Motion: Chemical Reaction and Allied Dynamics in Solution and Elsewhere

“…But lower in the I 2 potential, I 2 would come into near resonance with a CCl 4 vibration; we predicted that the additional, associated vibration-vibration transfer route would speed up the relaxation into the required timeframe. Our basic picture was later confirmed by Charles Harris and colleagues (26) and others, although it was refined and enriched in assorted ways (e.g., by the involvement of several electronic states). I think it fair to say that after this, few people automatically assumed that vibrational relaxation in solution was very slow.…”

Visualizing Photochemical Dynamics in Solution through Picosecond X-Ray Scattering

“…Within a few picoseconds, atoms trapped within this cage recombine geminately (i.e., with their original partner) onto either the A, A 0 , or X potential. Those recombining on A or A 0 undergo solvent induced curve crossing to the X potential prior to vibrational relaxation back to the ground state, and the time scale for this is strongly solvent dependent [19,22]. Atoms which break out of the cage diffuse through the solvent and recombine nongeminately on a slower time scale [25].…”