The effect of pressure, temperature, H / D isotopes, and C isotopes on the kinetics of the OH + CO reaction are investigated using Rice-Ramsperger-Kassel-Marcus theory. Pressure effects are treated with a step-ladder plus steady-state model and tunneling effects are included. New features include a treatment of the C isotope effect and a proposed nonstatistical effect in the reaction. The latter was prompted by existing kinetic results and molecular-beam data of Simons and co-workers ͓J. Phys. Chem. A 102, 9559 ͑1998͒; J. Chem. Phys. 112, 4557 ͑2000͒; 113, 3173 ͑2000͔͒ on incomplete intramolecular energy transfer to the highest vibrational frequency mode in HOCO * . In treating the many kinetic properties two small customary vertical adjustments of the barriers of the two transition states were made. The resulting calculations show reasonable agreement with the experimental data on ͑1͒ the pressure and temperature dependence of the H / D effect, ͑2͒ the pressure-dependent 12 C/ 13 C isotope effect, ͑3͒ the strong non-Arrhenius behavior observed at low temperatures, ͑4͒ the high-temperature data, and ͑5͒ the pressure dependence of rate constants in various bath gases. The kinetic carbon isotopic effect is usually less than 10 per mil. A striking consequence of the nonstatistical assumption is the removal of a major discrepancy in a plot of the k OH+CO / k OD+CO ratio versus pressure. A prediction is made for the temperature dependence of the OD+ CO reaction in the low-pressure limit at low temperatures.

In single molecule studies of injection of an electron from a photoexcited dye into a semiconductor nanoparticle or into a film of such nanoparticles, the injection may be into the conduction band or into the band gap, depending on the system. The theory of the process and its return are discussed, in particular when a power law for the waiting time distribution may be expected and what that power might be. To this end a reaction−diffusion equation is set up and solved. When the injection is into the conduction band, a power law is predicted for the return of the electron to the dye cation but not for the injection. After a short time, the law for the waiting time distribution has a power of −1. At short times, before the slower return due to an increasing radius is recognized, the power law is −1/2. When the injection is into the band gap, a −1 power law is predicted for both the injection and the return. Available data are discussed in terms of the theory. A corollary is that single molecule studies for the injection can determine whether the injection is into the band gap or into the conduction band. The theory is tested by single molecule studies of various systems, such as comparing different dye−TiO 2 , dye−Al 2 O 3 , and dye−ZrO 2 systems and comparing specific dye−TiO 2 systems as a function of pH, and dye hole injection into p-type NiO.

A 2000, 108, 8905-8913. There were several typos and inconsistencies in the original surfaces. We made the following changes: (1) two strange points (2.6, 1.3, 0.0) and (2.6, 1.3, 180.0) on PES2.1 were removed, (2) the phases of the transition dipole moment were corrected, 1 (3) the value of the (1.6, 1.6, 0.0) point in the TDM2.1y was corrected to zero, (4) since according to symmetry the transition dipole moment from 1 1 Σ + to 1 1 ∆ in the linear structure is zero, these points on the surfaces were added or corrected, (5) due to forbidden transition in TDM2.1y when the molecule was in an isosceles triangle configuration, its transition dipole moment becomes zero when θ approaches 90°, and these points were added to the Nanbu-Johnson results, and (6) a missing point (1.6, 0.8,

The dependence of the rate of the reaction CO+ OH→ H+CO 2 on the CO-vibrational excitation is treated here theoretically. Both the Rice-Ramsperger-Kassel-Marcus ͑RRKM͒ rate constant k RRKM and a nonstatistical modification k non ͓W.-C. Chen and R. A. Marcus, J. Chem. Phys. 123, 094307 ͑2005͒.͔ are used in the analysis. The experimentally measured rate constant shows an apparent ͑large error bars͒ decrease with increasing CO-vibrational temperature T v over the range of T v 's studied, 298-1800 K. Both k RRKM ͑T v ͒ and k non ͑T v ͒ show the same trend over the T v -range studied, but the k non ͑T v ͒ vs T v plot shows a larger effect. The various trends can be understood in simple terms. The calculated rate constant k v decreases with increasing CO vibrational quantum number v, on going from v =0 to v = 1, by factors of 1.5 and 3 in the RRKM and nonstatistical calculations, respectively. It then increases when v is increased further. These results can be regarded as a prediction when v state-selected rate constants become available.