No abstract

8835reduced effective polarity of the interfacial microenvironmentZ8 is also a significant factor in this case in increasing the association of hydrogen ions with the surface carboxyl groups. These points have been made before in connection with the low conductivity above the cmc of perfluorooctanoic acid.I5The neutralization of the surface charges of the micelles of perfluoro acids and the resultant reduction of electrostatic de-(28). (a) Mukerjee, P.; Banerjee, K.stabilization of micelles are also likely to be responsible for an increased stability of the fluorocarbon acid micelles as compared to their salts. Literature datal4 on the strong acid, dodecylsulfonic acid, show that its cmc is lower by a factor of only about 1.15 than the cmc of the sodium salt over a wide temperature range. I n contrast, the cmc of perfluorooctanoic acid, 0.0096 M at 25 "C, is lower than that of the sodium salt, 0.0306 M (Table 11), by a much larger factor of 3.2. Acknowledgment.Two sulfonyloxyl radicals, CH,S(=O),O*, 2a, and 3-CF3C6H4S(=O),O', 2b, have been generated by 308-nm laser flash photolysis (LFP) of their parent symmetrical peroxides in CH3CN solution, in which they have lifetimes of 7-20 ps. Both radicals exhibit a broad, structureless absorption similar to that known for SO4'-with A, , , -450 nm. This absorption can be bleached for 2a but not for 2b by firing a second laser at 480 nm, presumably reflecting a photoinduced cleavage of the H3C-S03' bond. Radicals 2a and 2b react with the acetonitrile solvent by abstraction of a hydrogen atom, kH -1.6 X I Os M-' s-', kH/kD -2.0. Bimolecular rate constants for attack of these radicals on cyclohexane (viz., 1.9 X lo8 and 6.5 X IO8 M-l s-' for 2s and 2b, respectively) and chloroform (viz. ca., 3 X lo5 M-I S-I for both) demonstrate that they are more reactive than almost all other oxygen-centered radicals. Product studies demonstrate that both the photodecomposition and the thermal decomposition of the parent peroxides yield the corresponding sulfonyloxyl radicals, a result that contrasts with that we have previously obtained for the decomposition of [Ph,P(=O)O],, which yields radicals on photolysis but few if any radicals on thermolysis. Semiempirical AMI /PM3-UHF calculations on 2a are also reported. Transient, highly reactive, oxygen-centered radicals of the general structure Y,X(O)O*, 1, such as aryl-, ethenyl-, and (ethynylcarbonyl)oxyI radicalsz3 (ArC(O)O', RCH=CHC(O)O', and RC=CC(O)O'), (alkoxycarbony1)oxyl radicals4 (ROC-I , ( I I ) Hayon, E.; Treinin, A,; Wilf, J . J . Am. Chem. Soc. 1972, 94, 45-57.

Improved methods have been developed for generating aliphatic ketyl radical anions in solution for study by EPR spectroscopy. These methods involve either photolysis of a solution of potassium in the corresponding alcohol (which probably proceeds by initial electron transfer to the solvent), or abstraction of a hydrogen atom from the corresponding alcohol by photolytically generated tertbutoxyl radicals under strongly basic conditions. The EPR spectra of a number of new aldehyde and ketone radical ions are described, and compared with the spectra of their conjugate acids. The

The combination of laser flash photolysis (LFP) with timeresolved infrared (TRIR) detection provides a new and potentially extremely powerful tool for the measurement of absolute rate constants for fast reactions and for structural studies on reactive intermediates. However, IR bands generally have small extinction coefficients relative to UV bands, and this has restricted the application of the LFP/TRIR technique almost entirely to the study of metal carbonyl systems.4"7 The single exception in organic chemistry is Schuster and co-workers' investigation of 1,2-didehydroazepines.8It seemed important to us to widen the organic applications of this valuable new technique. The /3-scission of alkoxyl radicals in solution should be an ideal reaction to study since ketones (and aldehydes) have rather strong IR absorptions within the 1600-2000-cm"1 range of line-tunable carbon monoxide lasers. There has been only one direct, time-resolved measurement of an alkoxyl /3-scission in solution (vis.,9 C6H5CH2CH20• -C6H5CH2* + CH20). Reliable /3-scission rate data for other systems are required in order to understand why solvents would appear to have such a profound effect on fc^9'13 as well as to correlate k$ with structure.

No abstract