Evidence of the water-cage effect on the photolysis of NO₃^- and FeOH²⁺, and its implications for the photochemistry at the air-water interface of atmospheric droplets

Experiments are conducted to determine the photolysis quantum yields of nitrate, FeOH2+, and H2O2 in the bulk and at the surface layer of water. Results show that the quantum yields of nitrate and FeOH2+ are enhanced at the surface compared to the bulk due to a reduced water-cage surrounding the photo-fragments (•OH+•NO2 and Fe2++•OH, respectively). However, no evidence is found for an enhanced quantum yield for H2O2 at the surface. The photolysis rate constant distribution within nitrate, FeOH2+, and H2O2 aerosols is calculated by combining the quantum yield data with Mie theory calculations of light intensity. Values for the photolysis rate constant of nitrate and FeOH2+ are significantly higher at the surface than in the bulk due to enhanced quantum yields at the surface. The results concerning the rates of photolysis of these photoactive species are applied to the assessment of the reaction between benzene and •OH in the presence of •OH scavengers in an atmospherically relevant scenario. For a droplet of 1μm radius, a large fraction of the total •OH-benzene reaction (15% for H2O2, 20% for nitrate, and 35% for FeOH2+) occurs in the surface layer, which accounts for just 0.15% of the droplet volume. By neglecting the surface effects on photochemistry, the rate of the important reactions could be underestimated by a considerable amount.

Allowance for Force Constant Changes in the Theory of Electron Transfer at the Metal-Redox Electrolyte Interface

The Gurney-Gerischer-Marcus (GGM) model for electron transfer1 is used to investigate the effects of force constant changes between initial and final states for electron transfer at the interface between a metal and a redox electrolyte. The effects on the symmetry factor, β, and on the determination of the redox electrolyte distribution of states are investigated and compared to the predictions of the GGM model using no change in force constant. Comparisons are also made between the Marcus2,3 and GGM models. The GGM model with non-identical parabolas for the potential energy-nuclear configuration diagrams is used to numerically calculate the distribution of states in the redox electrolyte from which data the rate constant distribution for reduction is obtained from the overlap between occupied states of the metal and the unoccupied states of the redox electrolyte. Numerical integration of the rate constant distribution gives the rate constant which is calculated as a function of the electrode potential. The calculated Tafel plots are found to be non-linear but do not go through a maximum. The Marcus and GGM models predict markedly different dependences of the symmetry factor on potential. Differentiation of the calculated rate constant with respect to potential gives the distribution of states for the redox electrolyte except for a small deviation which is due to the weak dependence on energy of the distribution of states in the metal. Anomalous results reported in the literature are shown to be qualitatively consistent with a difference in force constant between initial and final states for electron transfer.

The Temperature and Potential Dependence of the Rate Constant for Electron Transfer at the Metal Redox Electrolyte Interface

The Gurney-Gerischer-Marcus (GGM) model is used to investigate the potential and temperature dependence of the rate constant for electron transfer at the interface between a metal and a redox electrolyte. In this model electron transfer is described in terms of nuclear configuration-potential energy diagrams, electronic configuration-potential energy diagrams, state distribution functions and rate constant distribution functions. The model of identical parabolas, which leads to Gaussian electron distribution functions, g(E), for the redox electrolyte, is used for the nuclear configuration diagrams. The rate constant distribution, k(E), is obtained from the overlap between occupied and unoccupied state distribution functions of the metal and redox electrolyte. Integration of k(E) over the vertical transition (Franck-Condon) energies, E, gives the rate constant, k, which is calculated as a function of the electrode potential and temperature for various values of the reorganization energy, λ. Differentiation of k with respect to potential returns g(E) for the redox electrolyte except for a small deviation which is due to the weak dependence on energy of the distribution of states in the metal. For high λ the variation of symmetry factor with potential is small and the Tafel plots do not show a significant decrease in rate at high overpotentials. For small λ the Tafel plots are strongly curved but do not go through a maximum at high overpotential; the Tafel plots tend to a limiting value with only a small decrease in rate constant at high overpotential. This result is reflected in the temperature dependence of the rate constant and in the dependence of the Arrhenius activation energy, Ea, on potential; Ea does not increase at high overpotentials. These results are due to the weak dependence on energy of the distribution function for a metal compared to a redox electrolyte and emphasize the advantages of using distribution functions to describe the kinetics of electron transfer.