The decay of donor luminescence in a rigid solution when modified by electronic energy transfer by the exchange mechanism is treated theoretically. The rate constant for the elementary process of energy transfer is taken to be of the Dexter form, const exp(−2R/L), where R is the donor—acceptor distance and L is a positive constant. Calculations are made of the yield and decay time of the donor luminescence as functions of the acceptor concentration. The resulting relationship among the above quantities enables one to analyze experimental data in a quantitative manner, and thereby to obtain information about an intermolecular exchange interaction. As an example of such an analysis, Ermolaev's data on triplet—triplet transfer between some aromatic molecules are compared with our results, and very good agreement is found with a choice of the single parameter L.

Fluorescence spectra have been measured for a variety of diphenyl and triphenyl alkanes in cyclohexane and in p-dioxane. A class of compounds which are so structured that the phenyl groups along a main alkane chain are separated by exactly three carbon atoms, e.g., 1, 3-diphenylpropane, 1,3, 5-triphenylpentane, has been found to possess unique fluorescence characteristics. These are the appearance of a long-wavelength band in the region of 330 m}.' and a marked decrease in the fluorescence yield. The long-wavelength band is attributed to an emission from an excimer (a transient dimer) formed intramolecularly by the association of excited and unexcited phenyl groups. Formation of excimers in the specific class of compounds is discussed in relation to molecular configuration. Efficiency of excimer formation, solvent effects, and quenching by dissolved oxygen are some of the topics discussed through kinetic considerations.

The absolute fluorescence quantum yield of 9,10-diphenylanthracene in deaerated cyclohexane, the value of which has long been a matter of controversy, has been determined by use of a new method based on chemical actinometry. This technique avoids many of the intrinsic error sources of the other absolute methods. The sample's emission intensity is measured by almost completely surrounding the sample with an actinometer solution, and the excitation intensity is directly measured with the same type of actinometer. The measured sample emission intensity is corrected for the fraction of exciting light which is not absorbed within the sample solution and the fraction of fluorescence which escapes through the entrance and exit ports of the excitation beam. The absolute fluorescence yield is the ratio of the "corrected" sample intensity to the excitation intensity. The value we obtain with this method for 9,10-diphenylanthracene in cyclohexane is 0.90 to within 4%.completely with an actinometer solution, measures the (1)

Fluorescence spectra and quantum yields have been obtained for methylcyclohexane solutions of benzene, toluene, ethylbenzene, cumene, p-, m-, o-xylene, and 1,3,5-, 1,2,3-, 1,2,4-trimethylbenzene as a function of aromatic concentration over the temperature range from 25 to −100°C. At low temperatures distinct excimer emissions were observed for all compounds studied. The intrinsic emission quantum yields of monomer φm and of excimer φe have been determined by a simple technique which requires no assumptions regarding the details of the monomer–excimer kinetics. With decreasing temperature, φe is observed to decrease contrary to the behavior of φm, suggesting that the rate constant for the excimer radiative transition decreases strongly as the temperature is lowered. Such temperature dependence is explained as arising from the existence of a substantial vibronic component in the transition moment that is induced by thermal excitation of upper-state vibrational motions (e.g., torsional, tilting, etc.) of one monomer with respect to the other. From analysis of the temperature dependence of the fluorescence, lower bounds on the excimer binding energies Eb have been determined. The difference between this lower bound and Eb is approximately equal to the activation energy for radiative decay of the excimer. An estimate of this activation energy indicates that, for the case of benzene, Eb > 0.36 eV. The probability for association of excited monomer to form excimer and the probability for dissociation of excimer to an excited and unexcited monomer have been determined for benzene at 25 and −78°C from an appropriate analysis of the fluorescence quenching effect of CCl4. Additionally it has been demonstrated that the observed increase in CCl4 quenching efficiency at high benzene concentrations is predominantly due to an energy migration process. The probability per encounter for formation of excimer has been determined for benzene to be ≈1.0 and to decrease with alkyl substitution in a manner consistent with the steric requirements of sandwich-type excimer configurations.

Fluorescence spectra and fluorescence quantum yields have been determined for a wide variety of alkanes, cycloalkanes, and polycycloalkanes excited at 147 and 165 nm. Correlations between emission characteristics and molecular structure are noted and discussed.