J . Am. Chem. SOC.1988, 110, 2296-2298
2296 Scheme I
6 I
0
l a 1 -0
/
>Si-0
/
’c1
to flow CIDNP experiment^'^ where the polarization is generated in low magnetic fields but monitored in higher fields. Similar flow SLIT D N P experiments are presently under study in our laboratory.
enhancements occur for flow rates above -0.5 mL/min. In order to ensure that the measured SLIT D N P enhancements were not due to low level leakage of the spin label from the silica gel surface, ’H nuclear relaxation times were measured. The measured static ‘H nuclear relaxation times (Tho N 10.5 s ) were not altered after passage through region A containing the immobilized spin labels in comparison with pure benzene (no immobilized radical in region A).” Whereas, in the absence of a microwave field (Bls = 0) the monitored N M R magnetization did not decrease even at high flow rates (e.g., 10 mL/min), suggesting the high efficacy of these systems in reducing nuclear relaxation times.’ The immobilized nitroxide spin label systems la and l b have significantly reduced Overhauser enhancements relative to typical liquid values. In addition, the D N P spectral line shape (1 1Tt)DNP is broader for l b and ICrelative to the liquid-liquid intermolecular case (la). These results and the corresponding EPR spectra6,’ suggest reduced molecular motion for these immobilized spin labellbenzene interactions. Furthermore, the immobilized phenoxy spin label 2 has a dominant solid-state effect (enhancements a t we f w,) which is not only completely antisymmetric but also contains a minor Overhauser contribution.2 This suggests that the immobilized phenoxy spin label 2 has significant contributions from time independent nuclear-electron interactions. In conclusion, it is important to note the similarity between the present flow SLIT ‘H D N P results and data obtained previously for solid ‘H D N P studies. For example, the solid-state effect monitored by the messenger group (flowing benzene) for the immobilized phenoxy free radical 2 system clearly distinguishes the present study from previous liquid D N P studies. Since the reporter group reflects properties of the surface, theflow SLIT DNP experiment could provide a new method for monitoring liquid or gaslsurface intermolecular interactions. An important characteristic of the flow SLIT D N P experiment is the longer nuclear relaxation times (Tln0)for the flowing bolus after polarization transfer. This more readily facilitates experiments where the flowing bolus is polarized in low magnetic fields (region A), but the N M R magnetization is monitored in higher magnetic fields (region C).’* The latter experiment is similar in some respects
Dependence of the Lifetime of the Twisted Excited Singlet State of Tetraphenylethylene on Solvent Polarity
(1 1) The nuclear spin lattice relaxation times (T,’s) were measured by sequence. The T,’swere measured after using a standard (l8Oo-r-9O0-r) flowing benzene (degassed and recycled) over the immobilized radical for several minutes and then stopping the flow for the static measurement. The TIvalues were within experimental error of the values obtained for a pure static benzene sample (Le., 10.5 s). (12)This experiment was originally suggested by Richards, R. E., et al.: see ref 1b.
(1) See the following reviews and references cited therein: (a) Salem, L.; Rowland, C. Angew. Chem., Int. Ed. Engl. 1972, 11, 92. (b) Saltiel, J.; DAgostino, J.; Megarity, E. D.; Metts, L.; Neuberger, K. R.; Wrighton, M.; Zafiriou, 0. C. Org. Photochem. 1973,3, 1. (c) Salem, L.Acc. Chem. Res. 1979.12, 87. (d) Saltiel, J.; Charlton, J. L. Rearrangements in Ground and Excited States; de Mayo, P., Ed.;Academic Press: New York, ,1980;Vol. 3, p 25. (e) Hochstrasser, R. M. Pure Appl. Chem. 1980,52,2683. (0 Rettig, W. Angew. Chem., Inf. Ed. Engl. 1986,25,971. (8) BonaPiE-Koutecky,V.; Kouteckf, J.; Michl, J. Angew. Chem., Inf. Ed. Engl. 1987,26, 170.
0002-7863/88/ 1510-2296$01.50/0
Acknowledgment. We gratefully acknowledge helpful discussions with C. S.Yannoni (IBM) and R. A. Wind (CSU) during the early phases of this study. We also gratefully appreciate partial financial support for this work from DOW Chemical Company, Mobil Research Foundation, and the U.S. Army (Ft. Belvoir). (13) For CIDNP reviews, see: (a) Closs, G. L. Adu. Magn. Reson. 1974, 7, 157. (b) Kaptein, R. Advances in Free Radical Chemistry; Williams, G . H., Ed.; Elek Science: London, 1975;Vol. 5, p 319. (c) Wan, J. K. S. Adu. Photochem. 1980, 12,283. (d) Lawler, R. G. Prog. NMR Spectrosc. 1975, 9, 145.
Charles L. Schilling and Edwin F. Hilinski*
Department of Chemistry, The Florida State University Tallahassee, Florida 32306-3006 Received September 28, 1987 Studies on the cis-trans isomerization of olefins have demonstrated the existence of an intermediate (‘p*) on the energy surface of the lowest excited singlet state which corresponds to a geometry in which the p-orbitals of the original *-bond are mutually perpendicular.’ For a symmetrically substituted olefin, this twisted excited singlet state is described by theory in terms of a highly polarizable species whose electronic character is derived from the mixing of zwitterionic structures I and 11. On the excited-state surface, lp* resides approximately vertically above the transition state (‘p) on the ground-state surface for the thermal cis-trans isomerization of the alkene. The electronic nature of ‘p is described
0 1988 American Chemical Society
J . Am. Chem. Soc.. Vol. 110, No. 7, 1988 2297
Communications to the Editor R
R
R
R
R
R
I II m theoretically in terms of biradical 111. Although considerations of Ip* of alkenes and polyenes have long been a focus of theoretical studies,' a relatively small amount of experimental evidend exists for the zwitterionic nature of these electronic states. Here we report that the lifetime of 'p* of tetraphenylethylene depends upon the polarity of the solvent. This dependence is consistent with the theoretically prescribed zwitterionic description of 'p*. When tetraphenylethylene (1) in hexane was excited a t 305 nm with a 0.5-ps laser pulse, Greenej reported that absorption bands at 423 and 630 nm appeared within the time duration of the excitation laser pulse. The 630-nm band decayed with a time constant of 5 f 1 ps while the 423-nm band shifted to 417 nm on this time scale. Subsequently, the 417-nm band decayed with a time constant of 3.0 i 0.5 ns. Greene assigned the 630-nm band to an electronic transition from the vertical excited singlet state ('lv*) and the 420-nm band to a transition from the nonfluorescent twisted excited singlet state ('lp*). Greene's assignments of the two absorption bands are consistent with the results of the time-resolved emission experiments on 1 reported by Barbara, Rand, and R e n t ~ e p i s . The ~ quantum yields measured for other photoinduced phenomena associated with 1 also support these assignments. Leigh and Arnold5 reported the quantum yields of the light-induced cis-trans isomerization of several substituted tetraphenylethylenes in benzene. On the basis of their work, the degenerate cis-trans isomerization of 1 should account for >95% of the excited states generated upon electronic excitation. Each of several other deactivation processes, such as fluorescence6 from lv*, intersystem crossing,68 and photocyclization of 1 to give 4a,4bdihydr0-9,1O-diphenylphenanthrene,~ occurs with a quantum yield of
