We have recently reported 1 on a new and efficient pathway to hydrated electrons (e aq •-S0002-7863(98)00040-7 CCC: $15.00

No abstract

The photoreactions of 9,10‐diphenylanthracene with two iodonium salts and a sulfonium salt in different solvents were investigated by fluorescence quenching, flash photolysis with UV/vis detection, measurements of the quantum yields of sensitizer decomposition, and photo‐CIDNP spectroscopy. Energy‐transfer sensitization is precluded by thermodynamics in these systems. In a solvent of lower polarity such as dimethoxyethane, the quenching rate constant is still about a third of the value in acetonitrile; the quantum yields of formation of radical cations and the quantum yields of sensitizer decomposition are decreased by about the same factor. In contrast, dramatic effects occur in the CIDNP spectra of the iodonium salts, which derive from two consecutive radicals pairs, the first containing the radical cation DPA± of the sensitizer and the radical On• of the onium salt, the second comprising DPA± and a phenyl radical produced by fragmentation of On•: In less polar solvents, the polarizations from the first pair are strongly reduced, and the polarizations from the second pair are larger by an order of magnitude. From a kinetic analysis, this effect was attributed to a higher rate of transformation of the first radical pair into the second in the less polar solvent. For the sulfonium salt, this transformation is faster than in the case of iodonium salts, so even in acetonitrile all polarizations stem from the second radical pair. S‐T±‐type CIDNP, which had been proposed in the literature for analogous chemical systems, could be excluded unambiguously, and all CIDNP effects could be explained by radical pair theory (S‐T0‐type CIDNP).

The photoreactions between 9,10‐anthraquinone and the donors C6H6–n (OMe)n, where n = 1 …3, in acetonitrile‐d3 have been investigated by time resolved measurements of chemically induced dynamic nuclear polarization (“flash CIDNP”). The activation parameters for the degenerate electron transfer between five of the donors and their radical cations have been determined. For these, the same isokinetic relationship as in the previously studied case of para‐substituted N,N‐dimethylanilines was found to hold. Theoretical rate constants for the self exchange processes were calculated from AM 1 data by a model based on the Marcus theory. Their agreement with the experimental values was satisfactory over the whole range of investigated temperatures (237 K … 298 K).

The sensitized photoreactions of diphenyliodonium salt and four of its technically relevant derivatives bearing lipophilic substituents were investigated by pseudo-steady-state and time-resolved CIDNP experiments. Singlet (naphthalene, diphenylanthracene, dimethylanthracene) and triplet (xanthone, thioxanthone, Michler's ketone) sensitizers were used in a variety of solvents (acetonitrile, dimethyl sulfoxide, dimethylformamide, methanol, chloroform, tetrahydrofurane, diethylene glycol dimethyl ether, dioxane, toluene). Under all conditions, the primary step leading to CIDNP was found to be electron transfer from the excited sensitizer *Sens to the onium cation On+. All spin-polarization effects could be explained consistently within the framework of radical pair theory (S−T0-type CIDNP). Pair substitution, i.e., the transformation of the primary radical pairs RP 1 into secondary pairs RP 2 ( where Ph• is the phenyl or an aryl radical) plays a key role for the CIDNP effects and even leads to a field dependence of the polarization phases for the system diphenyliodonium cation/naphthalene in dioxane. Decreasing solvent polarity causes an increase of the rate RP 1 → RP 2. The introduction of a long lipophilic side chain into the onium salt has the same effect, presumably owing to self-solvation.