Electron transfer parameters are extracted from the optical spectra of intervalence bis(hydrazine) radical cations. Compounds with 2-tert-butyl-3-phenyl-2,3-diazabicyclo[2.2.2]octyl-containing charge-bearing units that are doubly linked by 4-sigma-bond and by 6-sigma-bond saturated bridges are compared with ones having tert-butylisopropyl- and diphenyl-substituted charge bearing units and others having the aromatic units functioning as the bridge. Solvent effect studies show that the optical transition energy (E(op)) does not behave as dielectric continuum theory predicts but that solvent reorganization energy may be usefully separated from the vibrational reorganization energy by including linear terms in both the Pekar factor (gamma) and the Gutmann donor number (DN) in correlating the solvent effect. Solvation of the bridge for these compounds is too large to ignore, which makes dielectric continuum theory fail to properly predict solvent effects on either E(op) or the free energy for comproportionation.

The radical cations of properly designed bishydrazines allow comparison of observed and calculated electron transfer rate constants. These compounds have rate constants small enough to be measured by dynamic electron spin resonance spectroscopy and show charge transfer bands corresponding to vertical excitation from the energy well for the charge occurring upon one hydrazine unit to that for the electron-transferred species. Analysis of the data for all six compounds studied indicates that the shape of the adiabatic surface on which electron transfer occurs can be obtained from the charge transfer band accurately enough to successfully predict the electron transfer rate constant and that explicit tunneling corrections are not required for these compounds.

Rate constants k ESR for intramolecular electron transfer between the reduced and oxidized diazene units of dimeric 2-tert-butyl-2,3-diazabicyclo[2.2.2]octyldiazenium radical cations cations which are doubly linked through the bicyclic units by six σ-bonds, sB6σ+ and aB6σ+, were determined from their variable temperature ESR spectra in CH3CN, dimethylformamide, and CH2Cl2. These cations show solvent−sensitive charge transfer absorption bands from which the vertical electron transfer excitation energy, λ, and the electronic coupling, V J, were determined by simulation, using vibronic coupling theory. The partitioning between solvent and vibrational components of λ were made assuming that the average energy of the vibrational modes coupled to the electron transfer, hνv, is 3.15 kcal/mol (1100 cm-1). The observed rate constants interpolated to 298 K are factors of 4.7−5.8 larger than those calculated from the electron transfer parameters obtained from vibronic coupling theory analysis of the charge transfer bands, k cal, in acetonitrile and DMF, and for sB6σ+ in CH2Cl2 the factor is 2.5. The ratios k ESR(350)/k ESR(250) are 1.0−1.6 times larger than k cal(350)/k cal(250)in CH3CN and DMF and 0.9 times larger in CH2Cl2. The agreement with theory for the bis-diazeniums is far better than that obtained for doubly four σ-bond-linked bis-hydrazine radical cations (J. Am. Chem. Soc. 1997, 119, XXXX). It is suggested that the significantly smaller vibronic coupling constants S = λv/hνv for the bis-diazeniums (6.5−7.6) compared to those of the bis-hydrazines (13.6−17.5) might be principally responsible for the difference in agreement of theory with experiment.

Second-order rate constants k 12 (obsd) measured at 25°C in acetonitrile by stopped-flow for 47 electron transfer (ET) reactions among ten tetraalkylhydrazines, four ferrocene derivatives, and three p-phenylenediamine derivatives are discussed. Marcus's adiabatic cross rate formula k 12 (calcd) ) (k 11 k 22 k 12 f 12 ) 1/2 , ln f 12 ) (ln K 12 ) 2 /4 ln(k 11 k 22 /Z 2 ) works well to correlate these data. When all k 12 (obsd) values are simultaneously fitted to this relationship, best-fit self-exchange rate constants, k ii (fit), are obtained that allow remarkably accurate calculation of k 12 (obsd); k 12 (obsd)/k 12 ′(calcd) is in the range of 0.55-1.94 for all 47 reactions. The average ∆∆G ij q between observed activation free energy and that calculated using k ii (fit) is 0.13 kcal/mol. Simulations using Jortner vibronic coupling theory to calculate k 12 using parameters which produce the wide range of k ii values observed predict that Marcus's formula should be followed even when V is as low as 0.1 kcal/mol, in the weakly nonadiabatic region. Tetracyclohexylhydrazine has a higher k ii than tetraisopropylhydrazine by a factor of ca. 10. Replacing the dimethylamino groups of tetramethyl-p-phenylenediamine by 9-azabicyclo[3.3.1]nonyl groups has little effect on k ii , demonstrating that conformations which have high intermolecular aromatic ring overlap are not necessary for large ET rate constants. Replacing a γ CH 2 group of a 9-azabicyclo[3.3.1]nonyl group by a carbonyl group lowers k ii by a factor of 17 for the doubly substituted hydrazine and by considerably less for the doubly substituted p-phenylenediamine.

Rate constants (k ij ) measured by stopped flow are reported for 50 additional intermolecular electron transfer reactions between 0 and 1+ oxidation states of various compounds, enlarging our data set to 141 reactions between 45 couples in acetonitrile containing 0.1 M tetrabutylammonium perchlorate at 25°C. Hydrazines with both saturated and unsaturated substituents, ferrocene derivatives, and heteroatom-substituted aromatic compounds are included in the couples studied. Least-squares fit of all the reactions to simple Marcus cross-reaction theory provides an internally consistent set of best fit intrinsic barriers ∆G ‡ ii (fit) (for selfelectron transfer of each couple) covering a range of over 19 kcal/mol (rate constant range 2 × 10 14 ) that predicts the k ij rather accurately. All reactions have ratios of calculated to observed k ij in the range 0.3-3.3 and 95% fall in the range 0.5-2.0. These results require that the preexponential factor for a cross reaction is close to the geometric mean of those for the self-reactions, which is not expected. Changes in internal reorganization energy (λ v ) have major effects on ∆G ‡ ii (fit), and changes in electronic overlap (H ab ) have easily detectable ones, but the reactions studied are clearly not strongly nonadiabatic, even though in many cases the only electronic overlap at the transition state is between nonbonded alkyl groups. It is argued that these reactions occur in the "elbow region" between nonadiabatic and adiabatic electron transfer. IntroductionOuter-sphere single electron transfer (ET) reactions between a neutral species i 0 , and a radical cation, j + , eq 1, are the simplest cases for calculation of rate constants. Marcus introduced the

A quantitative model of mixed-valence excited-state spectroscopy is developed and applied to 2,3-diphenyl-2,3-diazabicyclo[2.2.2]octane. The lowest-energy excited state of this molecule arises from a transition from the ground state, where the charge is located on the hydrazine bridge, to an excited state where the charge is associated with one phenyl group or the other. Coupling splits the absorption band into two components with the lower-energy component being the most intense. The sign of the coupling, derived by using a neighboring orbital model, is positive. The transition dipole moments consist of parallel and antiparallel vector components, and selection rules for each are derived. Bandwidths are caused by progressions in totally symmetric modes determined from resonance Raman spectroscopic analysis. The absorption, emission, and Raman spectra are fit simultaneously with one parameter set.

trans-2,3-Dimethyl-1-tri-tert-butylsilyl-1-triisopropylsilylsilirane (1) has been prepared in 98% yield by the treatment of ( t Bu 3 Si)( i Pr 3 Si)SiBr 2 with activated magnesium in the presence of trans-2-butene. Trapping studies suggested that silylene ( t Bu 3 Si)( i Pr 3 Si)Si: (2) was extruded by thermolysis of 1. Without added trapping agents self-trapping of silylene 2 formed a stable product, 1,1-di-tert-butyl-4,4-dimethyl-2-triisopropylsilyl-1,2-disiletane (5), by intramolecular C-H insertion. Thermolysis of 1 presented an ideal reaction for a kinetic study of the thermal decomposition of a silirane. It was found that thermal decomposition of 1 is stereospecific and first-order in benzene solution. From the rates of reaction at various temperatures activation parameters were obtained. The results were consistent with a nonleast motion pathway for silylene retroaddition and hence addition.The thermally induced extrusion of silylenes from siliranes was discovered by Seyferth in 1975 1 and has often been used for thermal generation of silylenes ever since. 2 Both a concerted mechanism for silylene extrusion, the reverse of single-step singlet state silylene addition to a CdC bond, and a stepwise process proceeding via a diradical formed by Si-C homolysis were initially considered for the pyrolysis of siliranes. 1,3 Seyferth et al. found, in 1982, that transfer of Me 2 Si from hexamethylsilirane to olefins generating new siliranes was stereospecific. 4 Boudjouk et al. found stereospecific extrusion of di-tert-butylsilylene from diastereomeric siliranes. 5 In 1991, experimental results from the Gaspar group showed that the decomposition of cis-and trans-2,3-dimethyl-1,1-diadamantylsilirane to the corresponding 2-butene was stereospecific, as was the transfer to diastereomeric olefins of the diadamantylsilylene unit Ad 2 Si. 6 Stereospecific transfer of silylene units upon photolysis of their non-silirane precursors had been established by Jones et al. 7,8 and by Kumada, Ishikawa et al. 9 for Me 2 Si, 7,8 Ph 2 Si, 8 and PhMeSi. 9 The stereospecificity of silylene transfer from siliranes and to olefins was interpreted in terms of a concerted mechanism: single-step thermal extrusion of free silylenes as the inverse of single-step addition of free silylenes to olefins.Despite initial doubts regarding the formation of free silylenes upon the pyrolysis of siliranes and photolysis of oligosilanes, 10 clear evidence for the intermediacy of free silylenes has accumulated over the years. 2 This can be summarized as being comprised of the clear indication of a common reactive intermediate from quite different plausible precursors of the same silylene and the direct detection of several silylenes by electronic spectroscopy upon the photolysis of a variety of oligosilanes. 2,8 The case for concerted addition and retroaddition reactions of silylenes is more tenuous. Scheme 1 compares concerted and stepwise silylene addition and retroaddition mechanisms.The general acceptance of the concerted addition (k add )-retro...

CASSCF, CASPT2, CCSD(T), and (U)B3LYP electronic structure calculations have been performed in order to investigate the thermal fragmentation of P-phenylphosphirane (1) to phenylphosphinidene (PhP) and ethylene. The calculations show that generation of triplet PhP via a stepwise pathway is 21 kcal mol(-1) less endothermic and has a 12 kcal mol(-1) lower barrier height than concerted fragmentation of 1 to give singlet PhP. The formation of singlet PhP via a concerted pathway is predicted to be stereospecific, whereas formation of triplet PhP is predicted to occur with complete loss of stereochemistry. However, calculations on fragmentation of anti-cis-2,3-dimethyl-P-mesitylphosphirane (cis-1Me) to triplet mesitylphosphinidene (MesP) indicate that this reaction should be more stereospecific, in agreement with the experimental results of Li and Gaspar. Nevertheless, with a predicted free energy of activation of 42 kcal mol(-1), the formation of MesP from cis-1Me is not likely to have occurred in an uncatalyzed reaction at the temperatures at which this phosphirane has been pyrolyzed.