This is the first part of a series of articles reporting critically evaluated thermochemical properties of selected free radicals. The present article contains datasheets for 11 radicals: CH, CH 2 (triplet), CH 2 (singlet), CH 3 , CH 2 OH, CH 3 O, CH 3 CO, C 2 H 5 O, C 6 H 5 CH 2 , OH, and NH 2. The thermochemical properties discussed are the enthalpy of formation, as well as the heat capacity, integrated heat capacity, and entropy of the radicals. One distinguishing feature of the present evaluation is the systematic utilization of available kinetic, spectroscopic and ion thermochemical data as well as high-level theoretical results.

The photophysical properties of the N-H and N-methyl derivatives of 1,2-, 2,3-and 1,8-naphthalimides have been studied. The shift of the fluorescence emission position as a function of the solvent polarity indicates only a weak variation of dipole moment for the excited state compared with the corresponding value in the ground state (5.7 D for 26, 2.8 D for 36 and <2 D for 46, 1 D z 3.33564 x C m, and 26, 36 and 46 are N-methyl-1,2naphthalimide, N-methyl-2,3-naphthalimide and N-methyl-1 ,&naphthalimide). However, important modifications of the photophysical properties are observed which depend on the relative position of the dicarboximide moiety on the naphthalene ring: the intersystem crossing rate constant of 46 increases dramatically by three orders of magnitude compared with that of 26 ; simultaneously, the fluorescence quantum yield decreases from 0.77 to 0.03, although the corresponding rate constant, k,, increases. This difference is found to arise from the energy gap between the lowest '(n,n*) singlet excited state and the upper 3(n,n*) triplet state, which is of the order of 9 kcal mol-' for 26 and less than 2 kcal mol-' for 46 in acetonitrile solution. Protic solvents increase the energy difference between the n,n* and n,n* states thus decreasing the mixing of the two levels; as a consequence, the lifetime of 46 is increased, i.e. from <60 ps in hexane to 2.1 ns in trifluoroethanol. A triplet-triplet annihilation process occurs with the N-methyl derivatives 36 and 46 which leads to a monomer delayed fluorescence with the former, and mainly to a delayed excimer emission with the latter.

In order to clarify mechanisms of excited state interactions in hydrogen-bonded pairs, we have studied the kinetics of dynamic quenching of singlet and triplet fluorenone by a series of alcohols, phenols, and related compounds, in which hydrogen-bonding power, redox potential, and acidity are systematically varied. In addition, effects of solvent basicity or polarity and deuteration help identify the role of hydrogen-bonding in physical or chemical quenching processes. Alcohols and weak acids, with high oxidation potentials, do not quench the triplet, but quench the singlet at rates which parallel hydrogen-bonding power. This is attributed to a physical mechanism, involving vibronic coupling to the ground state via the hydrogen bond. This is much stronger in the excited state than in the ground state, and provides efficient energy dissipation in the radiationless transition. Phenols, with hydrogen-bonding power comparable to that of the alcohols but with much lower oxidation potentials, quench both singlet and triplet by electron or H-atom transfer, depending on potentials, acidities, and solvent polarity, as shown by formation of anion or neutral fluorenone radicals from the triplet. Rates increase with both decreasing oxidation potential of the phenol and increasing acidity of the incipient cation radical. Quenching proceeds via a hydrogen-bonded complex and is facilitated by proton transfer contributions to the effective excited state redox potential.

Ab initio calculations at different levels of theory have been performed for the title H-abstraction reactions. Total energies at stationary points of the potential energy surfaces for the reaction systems were obtained at MP2 and MP4 levels and improved by using Gaussian-2 (G2) methodology. The calculated G2 heats of reaction agree well with the experimental ones for both methoxy (product resulting from hydroxyl-side attack) and hydroxymethyl (product resulting from methyl-side attack) reaction channels. Calculations of the potential energy surfaces for the reaction systems show that H-abstraction from methanol by H, CH 3 , and OH (for methoxy reaction channel) proceeds by simple metathesis. The mechanism of the hydroxymethyl channel of reaction CH 3 OH + OH appears to be more complex, and it may consist of two consecutive processes. The reaction rate is determined by the energy barrier of the first process. Differences in the heights of the calculated energy barriers explain the differences in the reactivity of H, CH 3 , and OH toward methanol. The calculated barriers indicate a significant dominance of the hydroxymethyl formation channel for the CH 3 OH + H and CH 3 OH + OH reaction systems. Rationalization of the derived energy barriers has been made in terms of the polar effect. The calculated rate constants are in very good agreement with experiment and allow a description of the kinetics of the reactions under investigation in a wide temperature range with the precision that is required by practical applications such as modeling of the chemistry of methanol combustion.