Multivariate statistical treatment of solvent effects in the photoreduction of CoIII(En)2(Br)(RC6H4NH2)2+ complexes in aqua-organic solvent medial

Ultraviolet photolysis (lambda = 254 nm) studies were carried out for a series of cobalt(III) complexes, Co-III(En)(2)(Br)(RC6H4NH2)(2+), where R = m-OMe, p-F, H, m-Me, p-Me, p-OEt, and p-OMe, in various compositions of water-methanol/1,4-dioxane mixtures (0, 5, 10, 15, 20, 25, and 30 vol% organic cosolvent) at two different temperatures (287 and 300 K). The ligand-to-metal charge-transfer excited state produced in the excitation of the complex initially generated a solvent-caged (Coll; ligand radical) pair, which eventually undergoes recombination/separation into products. The quantum yield sharply increased from a mixture containing a lower mole fraction of organic cosolvent (x(org)) to higher one. In other words, when x(org) in the mixture increases, a steady increase in the quantum yield is observed. The quantum yield of Co-III(En)(2)(Br)(RC6H4NH2)(2+) in various solvent mixtures is found to exhibit a linear (log Phi(Co(II))-1/epsilon(r)) dependence. This is consistent with solvation or solvent cage effect, which may be nonspecific, specific, or both. In order to throw light on these effects, a phenomenological model of solvent effects was applied. Therefore, the quantum yield values have been correlated statistically with some success containing different solvent parameters. The solvent parameters considered in this work are Grunwald-Winstein's Y, Krygowski-Fawcett's E-T(N) and DNN along with Kamlet-Taft's alpha, beta, pi* parameters. The regression model proposed is Y-S = Y-0 + Sigma(n)(i=1) a(i)X(i), where Y-S is the solvent-dependent property (here log Phi(Co(II))) in a given solvent; Y-0 is the statistical quantity corresponding to the value of property in the reference solvent; X-1, X-2, X-3... are explanatory variables, the solvent parameters, which can explain the various solvation effects on reactants/{Co-II; ligand radical} pair, and a(1), a(2), a(3)... etc., are the regression coefficients. The coefficient values can be quantified to measure the relative importance of solvent effects on the physicochemical quantity, that is, the photoreduction yields in the present investigation.