J. Chem. SOC.,Faraday Trans. 2, 1984, 80, 1093-1 105 Photoisomerization Mechanism and Conformational Equilibria of Styrylnaphthalenes A Study Based on Photophysical Properties and Molecular-orbital Calculations BARTOCCI,BY GIAMPIERO FAUSTO MASETTI AND UGO MAZZUCATO* Dipartimento di Chimica, Universiti di Perugia, 1-06 100 Perugia, Italy AND GIANCARLOMARCONI Istituto di Fotochimica e Radiazioni di Alta Energia del C.N.R., 1-40126 Bologna, Italy Received 30th January, 1984 The photophysical properties of trans-1-and trans-2-styrylnaphthalene(1-StN and 2-StN) have been studied. The conformational isomers of 2-StN have been reinvestigated by measur- ing the solvent and temperature effects and by m.0. calculations (CNDO/S) on the nature of the lowest excited states.Fluorescence quantum yields and lifetimes and kinetic parameters of decay have been obtained for the two conformers in n-hexane and ethanol. The energy difference between the two conformers and the activation energy for their trans --* cis photo-isomerization have been obtained from the temperature effect in both solvents. Evidence has been obtained for two photoisomerization mechanisms, a triplet mechanism predominant below room temperature and a singlet mechanism favoured in ethanol and/or at higher temperatures. The mechanism of the trans --+cis photoisomerization of stilbene has been satisfactorily clarified over the last decade.' It is now generally accepted that the S1-+TIintersystem crossing (ISC) is a negligible process, as indicated by theoretical and experimental results.The reaction mechanism implies an activated internal rotation in the singlet manifold, from S1, reached by absorption of light, to S2, a doubly excited state with a minimum energy in the perpendicular (perp) configur- ation. From there, S2--+ So internal conversion and partitioning to the trans and cis sides takes place. A similar mechanism has recently been shown to be operative for the aza analogues of stilbene, the styrylpyridines, for which, however, ISC can become important in inert solvents at low temperatures.2 When a phenyl group of stilbene is replaced by a naphthyl group, the nature of S1 might change from (largely) ethylenic to (largely) aromatic, thus changing the radiative and photoreactive behaviour? Simple considerations based on the energy, oscillator strength and orientation of the transition moment for the lowest electronic transitions of styrene and naphthalene imply that excitation to S1 is probably localized in the naphthyl group for 2-styrylnaphthalene (2-StN), at least for one of its conformational isomers (uide infra).To confirm this, a detailed study of the photophysical properties of isomeric 1-and 2-StN has been carried out, with particular emphasis on (a) the effects of temperature and solvent on the rate of isomerization and (b) m.0. calculations. Moreover, evidence has recently been obtained for the existence, in solutions of these 1,2-diarylethylenes, of an equilibrium between rotational conformers involv- ing the quasi-single bonds between the aromatic groups and the ethylenic carbon atoms.4 Absorption and emission properties of the postulated rotamers are often 1093 1094 PHOTOPHYSICAL PROPERTIES OF STYRYLNAPHTHALENES different, leading to fluorescence behaviour which is generally dependent on both excitation (Aexc) and emission (Aem) wavelengths.Analysis of such behaviour allows the photophysical parameters of the distinct conformers to be eval~ated.~ The conformers of trans-2- StN, previously reported,’ have been reinvestigated. Fluorescence quantum yields, lifetimes and decay kinetic parameters have been obtained for the two species in n-hexane (or methylcyclohexane) and ethanol. The energy difference between the two conformers and the activation energy for their geometrical photoisomerization have been calculated.A m.0. description of their photophysical properties allows the experimental results to be rationalized. Evidence for two photoisomerization mechanisms is reported, a triplet mechanism pre- dominant below room temperature and a singlet mechanism favoured by polar solvents and at higher temperatures. EXPERIMENTAL The styrylnaphthalenes were synthesized by standard procedures from previous The solvents were Carlo Erba chemicals of RPE grade. The solutions were de-aerated by bubbling argon through them. The fluorescence spectra were measured with a Perkin-Elmer MPF-44 spectrophotofluorimeter equipped with an accessory for spectrum correction based on rhodamine B as quantum counter. Slit widths of 2 nm were used throughout this work.The standard used for fluorescence quantum yields was a-NPD in cyclohexane (4F= 0.58).’ For low-temperature measurements (carried out in methylcyclohexane instead of n-hexane in order to investigate a larger temperature range), the c$~value at room temperature was used as the reference, taking into account changes in absorbance and refractive index with temperature. The mean deviation of three independent experiments was ca. 4%. Fluorescence lifetimes (mean deviation of three independent experiments ca. 7%) were measured using a home-built fluorimeter (Laben electronics and Applied Photophysics source and mono-chromators) based on the single-photon time-correlation technique.The data, accumulated in the multichannel analyser, were fed into a 68000-based Cromemco CS 1D2E microcomputer and then processed. The decay curves were analysed using a single- or double-exponential deconvolution program with a non-linear least-squares fitting procedure. The low-temperature effect was measured with an Oxford Instruments CF-204 continuous flow cryostat. The uncertainty in the parameters obtained from the Arrhenius plots was ca. k0.4 kcal mol-’ for AE and k0.5 for log A. RESULTS AND DISCUSSION FLUORESCENCE ANALYSIS OF THE CONFORMER MIXTURE A fluorimetric analysis of the two conformers based on observation of the fluorescence quantum yield of the mixture (&) and of the monochromatic fluores- cence response function of the mixture F( t)= AAexp (-t/TA) +AB exp (-t/ TB) at different A,,, was previously applied to trans-2-StN in n-he~ane.~ The same analysis has been now extended to dilute solutions in a polar solvent (ethanol).The properties of the two fluorescent species are indicated by subscripts A and B, the first species being taken as that with a shorter lifetime (TA< T~).The barred symbols refer to the properties of the mixture. Note that the pre-exponential factors corres- pond to Ai =.h(Aexc) Fi(Aem)/ Ti (2) where is the fraction of excited molecules belonging to each species, Fi is the corresponding fluorescence intensity and i = A or B. Deconvolution analysis of G. BARTOCCI, F. MASETTI, U. MAZZUCATO AND G. MARCONI 1095 Table 1.Fluorescence analysis of conformers of trans-2-styrylnaphthalene in ethanol at 293 K experimental parameters derived parameters 270 0.58 0.09 0.91 272 0.57 0.13 0.87 277 0.59 0.05 0.95 282 0.56 0.17 0.83 290 0.58 0.09 0.91 300 0.59 0.05 0.95 305 0.58 0.09 0.91 3 10 0.57 0.13 0.87 315 0.57 3.7 20.8 0.9 0.14 0.86 318 0.56 0.17 0.83 325 0.56 0.17 0.83 332 0.56 0.17 0.83 335 0.52 3.5 20.3 1.76 0.23 0.77 340 0.5 1 0.34 0.66 350 0.48 0.52 0.48 355 0.47 3.3 18.7 5-4 0.49 0.51 averages 3.5 19.9 the decay curve (measured at the isoemissive ALm = 395 nm) leads to evaluation of the fluorescence lifetimes of the two species and the ratio of the pre-exponential factors, (x = AJAB =fATB/ fB7-A.The fractions of excited molecules belonging to each species J;(Aexc) = &i(Aexc)Ci/ E(Aexc)z (3) can thus be obtained. Table 1 reports the experimental parameters and those derived from the analysis. These data allow the fluorescence quantum yields and the relative absorption spectra of the two components to be ~alculated.~ Fig. 1 shows the absorption spectrum of 2-StN in ethanol and the relative spectra of the two confor- mers. On the assumption that the Franck-Condon envelopes of the first absorption band systems (S, +S,) are similar in shape for the two conformers, it is to be expected from the radiative lifetime relation that (&A)OO/(&B)OO TE/Tk = 4ATB/ 4BTA- (4) From the relative absorption spectra one obtains ( EA)OOCA/ ( EB)OOCB, which combined with eqn (4) gives the relative conformer concentrations.Since one could argue that the assumption of a similar Franck-Condon envelope for the first electronic transitions of A and B could not hold if S, has a different character for the two species (vide infru), we tried to obtain the relative abundance through an independent procedure. The ratio /3 =fA/fB, obtained from lifetime measurements at different temperatures, was plotted against 1/ T (fig. 2) and led to the enthalpy differences between the two conformers reported in table 2. From the Boltzmann distribution (neglecting entropy terms) one obtains c,/ cBvalues at 293 K (in parentheses in table 2) in good agreement with those obtained from the analysis.1096 PHOTOPHYSICAL PROPERTIES OF STYRYLNAPHTHALENES 5 4 3 & 2 1 250 300 350 hlnm Fig. 1. Spectra of 2-StN in ethanol. (a) Absorption spectrum E(Aexc), (b)relative absorption spectrum of conformer B and (c)relative absorption spectrum of conformer A. (b)and (c) are derived from the analysis. A revision of the previously reported analysis in n-he~ane,~ using more experi- mental points, led to slightly different values of the photophysical parameters. These are shown in table2 for comparison with the new values in ethanol. Attempts to perform a similar analysis for 1-StN were unsuccessful, the emission properties being independent of A,,, and the fluorescence excitation spectrum being a replica of the absorption spectrum in fluid ,rolutions at room temperature. This behaviour is explained by the fact that one conformer predominates in solution, the second conformer being separated by a large energy gap because of steric interference and diminished ~oplanarity.~ Only at low temperatures, where high viscosity barriers and matrix-induced stabilizations come into play, was multiple fluorescence also observed for 1-StN.' EFFECT OF TEMPERATURE The effect of temperature on 6, and r was studied for both 1- and 2-StN in inert and polar solvents.The equations used were +Ern/4,=1 AT"^ exp (-AE/RT) (5) 1/r= l/~'~~+Aexp(-AE/l?T) (6) for the quantum yields and lifetimes, respectively. The limiting values were measured G.BARTOCCI, F. MASETI, U. MAZZUCATO AND G. MARCONI 1097 \e 1 1 I \ 3.0 4.0 5.0 lo3K/ T Fig. 2. Plot of In P(h,,,) against 1/ T for 2-StN in (---) ethanol and (-) methylcyclohexane (A,,, = 355 nm). Table 2. Photophysical parameters of 1- and 2-StN at 293 K, enthalpy differences and relative conformer concentrations (for 2-StN) in two solvents n-hexane ethanol parameters 1-StN 2-StN(A) 2-StN(B) 1-StN 2-StN(A) 2-StN(B) #F 0.60 0.83 0.50 0.3 1 0.37 0.60 T/ns 1.8 5.2 27.6 1.1 3.5 19.9 AH/cal mol-' 680 720 relative abundance (Oh) 24 (24)" 76 (76) 20 (22) 80 (78) a Values in parentheses were obtained from T =f( T)(see text). at low temperatures when the activated process (trans--+perp internal rotation in the singlet manifold) is inhibited.Table 3 reports the quantum-yield values for a large temperature range and the corresponding Arrhenius parameters, calculated in the range 300-346 K in methylcyclohexane and 280-346 K in ethanol. In the case of 2-StN, because of the concurrent temperature effect on the conformational equilibrium, the parameters obtained are in parentheses as they cannot be assigned only to the activated isomerization but are the result of two temperature-dependent processes. In fact, & changes with temperature also because the relative concentra- tions of the two conformers, which have different fluorescence efficiencies, change with temperature. Therefore, a study of r =f(T)was carried out to obtain the true barrier of the activated isomerization path for the two distinct conformers of 2-StN.Table 4 reports the corresponding lifetimes in two solvents and fig. 3 shows, as an example, the trend in r =f(T)for the two conformers of 2-StN and the corresponding Arrhenius plots. Similar measurements for 1-StN are less indicative because of the 1098 PHOTOPHYSICAL PROPERTIES OF STYRYLNAPHTHALENES Table3. Fluorescence quantum yields ofthe trans-styrylnaphthalenesas a function of temperature and corresponding Arrhenius parameters in two solvents (A,,, = 320 nm) me thylcyclo hexane ethanol T/ K 1-StN 2-StN 1-StN 2-5tn 346 0.34 0.089 0.33 343 0.37 337 0.39 0.12 0.39 328 0.45 0.43 0.15 0.46 313 0.48 305 0.54 0.24 0.56 300 0.54 297 0.57 0.53 0.30 0.62 29I 0.60 0.32 0.66 280 0.60 0.52 0.39 0.71 265 0.65 0.55 0.46 0.72 250 0.64 0.59 0.52 0.74 235 0.63 0.56 0.55 0.73 220 0.62 0.55 0.56 0.73 205 0.6 1 0.58 0.58 0.7 1 190 0.60 0.57 0.56 0.7 I 175 0.60 0.54 I60 0.58 0.54 AE/kcal mol-' log A 10.8 14.8 (74(12.2) 7.3 14.0 (8.7)(13.4) Table 4.Fluorescence lifetimes (ns) of the conformers of trans-2-styrylnaph- thalene as a function of temperature and corresponding Arrhenius parameters in two solvents (Aexc = 355 nm) meth ylcyclohexane ethanol T/ K A B A B 190 4.2 20.2 230 3.8 23.0 4.1 20.8 270 3.9 22.8 4.1 19.9 290 3.6 19.9 297 3.2 18.1 300 3.9 23.0 305 3.7 21.5 3 13 3.6 21.0 324 3.4 20.6 327 3.2 19.3 2.3 14.8 337 3 .O 17.5 I .6 12.9 343 2.9 16.9 347 2.8 15.8 1.5 10.3 AE/kcal mol-' 10.5 10.3 8.1 8.2 log A 14.3 13.8 13.8 12.8 G.BARTOCCI, F. MASETTT, U. MAZZUCATO AND G. MARCONI 1099 I I 1 1 I 4 .c 3.c 2 I \\2 '.20 1 \* \ \2.c .-1912: -\ \ \ .\\.I-I 218 I-C *-I \17 1 .o 2a 30 32 34 , ~o~K/T, I I I I 1 20 z k-15 --. 18 '\ .\, \ \\.\ \ \ \. \ \ 10 \ 4 2.8 1 30 32lo3K/Tl 34 I I I 150 200 2 50 300 3 50 T/K Fig. 3. Plots of fluorescence lifetimes against T for (a) 2-StN(A) and (b) 2-StN(B) in (-) methylcyclohexane and (---) ethanol. The insets show the corresponding Arrhenius plots. short lifetime (1.7 ns in decalin at 77 K) (its change with temperature was measured in the range 297-347 K when T decreases from 1.8 to 1.1 ns in methylcyclohexane).The corresponding AE (ca.8 kcal mol-') is less reliable but is in reasonable agree- ment with that obtained from quantum-yield measurements. On cooling, the confor- mer mixture of 2-StN is enriched in the hypsochromically shifted longer-lived B conformer, which is then taken as the more stable conf~rmation.~.~ The temperature effect is practically the same for the two species, and equal values were found for the activation energies of the A and B conformers while the differences observed in the frequency factors are within the experimental uncertainty. PHOTOREACTION MECHANISM The high-frequency factors obtained from the photophysical measurements in the high-temperature range, when both & and T are temperature dependent, lead to the exclusion of a spin-forbidden process as the rate-determining step.Therefore it can be concluded that a singlet mechanism is operative in this temperature range, 1100 PHOTOPHYSICAL PROPERTIES OF STYRYLNAPHTHALENES Table 5. Kinetic parameters and quantum yields for radiative and reactive deactivation of the trans-styrylnaphthalenes in two solvents at 293 K (data for stilbene are reported for comparison) n-hexane ethanol parameters St" 1-StN 2-'StN(A) 2-StN(B) 1-StN 2-StN(A) 2-StN(B) kF/lo8 s-' 5.9 3.3 1.6 0.18 2.9 1.1 0.3 k,,,/ lo8 s-l 0.4 2.1 0.3 0.17 2.7 1.3 0.15 2.2' 0.19' 2.2' 0.15' ktp/10' s-' 160 0.05 0.03 0.01 3.6 0.57 0.05 4F 0.036 0.60 0.83 0.50 0.31 0.37 0.60 4CC 0.43 O.ltid (0.16) 0.29/ (0.19)/4?IC 0.18 0.085 (0.20) 0.25 0.32 0.32 (0.22) 0.20 " From ref.(1 b) and (3). 'Calculated from low-temperature measurements of [( 1 -+,=)/TIwhen the activated process is inhibited. Data in parentheses refer to the conformer mixture. dA,,, =325 nm, from G. G. Aloisi, G. Bartocci, G. Favaro and U. Mazzucato, J. Chem. Phys., 1980,84, 2020. A,,, = 310 nm. A,,,= 325 nm, in acetonitrile +water (2/3 v/v), from G. G. Aloisi, U. Mazzucato, G. G. Aloisi and F. Marsetti, J. Photochem., 1982, 18, 211. as for stilbene at all temperatures. Below room temperature, when both photo- physical parameters remain constant up to 77 K, the activated internal rotation in S1 is no longer an important isomerization path because of its high energy barrier. A triplet mechanism can thus become the predominant pathway.This is in agreement with previous conclusions of Birks," who stated that an increase in the size of the aryl side groups with respect to stilbene reduces the S1('B*)energy relative to that of &('A*) and thus tends to increase the S, potential barrier. This inhibits the trans-perp internal rotation and the S1-+T1 ISC remains the only significant process competing with the allowed fluorescence. This is now confirmed for 1-and 2-StN, at least below room temperature. From the data in table 2 we can calculate the kinetic parameters for the deactiva- tion of S,. Since fluorescence and trans 4 cis photoisomerization take into account all quanta absorbed by the trans-StN (c#+ +24, == no internal conversion into So seems to be operative.Thus the total non-radiative rate parameter is the sum of only two contributions, namely ISC and the activated internal rotation in Slywhich can easily be calculated: (7)knr(T)= = kIsc +A exp (-AEIRT). (7)T Table 5 reports these kinetic parameters together with the quantum yields [or fluorescence and trans -+ cis photoisomerization. The corresponding values for stilbene are also reported for comparison. The kIsc values obtained using eqn (7) are in a very good agreement with those calculated from the fluorescence parameters in a rigid matrix at 77 K or in fluid solutions at low temperatures (values in parentheses), as expected for a temperature-independent rate parameter.The main effect of the presence of the naphthyl group is a larger & value compared with stilbene. The complementary photoisomerization yield is relatively small at room temperature since the activated internal rotation in the singlet pathway (k,)is characterized by higher barrier (AE=10 kcalmol-' in an inert solvent compared with ca. 3 kcal mol-I for stilbene). This means that kpskIsc at room G. BARTOCCI, F. MASETTI, U. MAZZUCATO AND G. MARCONI 1101 temperature. Above room temperature, kpincreases and the coupled fluorescence yield then decreases. Below room temperature, the isomerization rate constant becomes temperature independent and practically equal to kIsc. Our conclusions seem to be in disagreement with laser-flash-photolysis measure- ments1'-I3 which show that the triplet yield observed by direct excitation of 2-StN is practically negligible.Considering all the results we believe that this is only an apparent disagreement since those measurements were all performed at room tem- perature and at A,,,, where one species (the short-lived one) predominates. In fact, laser-flash-photolysis experiments in progress in our laboratory seem to give a direct confirmation of the relevant S1--.* T1 ISC of 1- and 2-StN at low temperatures, since the triplet state is easily shown by transient T1-+T, absorptions on cooling the solutions to below room temperature. l4 Therefore the photoreaction mechanism at low temperatures is essentially a triplet mechanism.The temperature at which the mechanism appears to become a triplet process is just below 300 K in n-hexane, whilst it is lower (ca.250 K) in ethanol. The crossing to the triplet manifold is more important in 1- and 2-StN compared with stilbene, probably because T2 is adjacent to S1, as in naphthalene.' The most remarkable result in table 5 is the great difference in kF of the two conformers of 2-StN, particularly in an inert solvent. It is much smaller (almost one order of magnitude) for conformer B, indicating a different fluorescent state (videinfra). This is accompanied by smaller values of both kIsc and kpfor conformer B, particularly in ethanol. However, since these rate constants are higher than those in n-hexane the photoisomerization quantum yield is higher in the hydroxylic solvent.The photoreaction yield of the mixture can be calculated by 4:'" = (l -a){fATACkISC,A exp (-AEA/ RT)l where (1 -a) is the fraction of twisted molecules which decay to the cis ground state. Eqn (8) simplifies in the case of 1-StN where only one species is present in solution. From data in tables 1 to 5 and assuming a =0.5,' some 4:'" values were obtained at specific A,,, which resulted in satisfactory agreement with the available experimental values reported in previous papers (see table 5). This confirms the correctness of the analysis and the reliability of the proposed mechanism. A study of the temperature dependence of the fluorescence and photoisomeriz- ation of 1- and 2-StN and related compounds has been carried out by Fischer and coworker^.'^ They report plots of quantum yields as a function of temperature which show the fluorescence and trans-+ cis photoisomerization to be complemen- tary processes in a broad temperature range, as for stilbene,' but at considerably higher temperatures. Fischer and coworkers report AE values calculated from the photoreaction [&=f(T)], but not from fluorescence.The trend of 4Fwith tem- perature is in qualitative agreement with that described in the present work. However, their absolute & values for the 1-and 2-StN are ca. 50% larger than those reported in table3. According to our results, contrary to the results for stilbene,' & does not approach unity at low temperatures as reported by Fischer and coworker~'~ because of the concurrency of ISC, which is not negligible. The AE values reported by Fischer and coworkers for & =f(T)in the range 280-360 K are 2 and 3 kcal mol-' for 2-StN and 1-StN, respectively, and decrease with tem- perature.These values are much lower than those obtained by us from the variation of +F and rF with temperature (tables 3 and 4). This is rather surprising since one expects AE measured from photoisomerization to be equal to or larger than those 1102 PHOTOPHYSICAL PROPERTIES OF STYRYLNAPHTHALENES measured from fluorescence. In fact, in the low-temperature range, they should also include viscosity barriers. A reasonable explanation of this behaviour is that the +c values contain a contribution from the non-activated [or less activated (uide infra)] triplet route which becomes predominant at lower temperatures, thus attenuating the temperature dependence.Another explanation of small changes in +F and +c in the temperature range where the activated process is negligible (below ca. 280 K) could be the shift of the conformational equilibrium between two species with different phot~reactivity.~’~ For instance, in the case of 2-StN, whose equi- librium in ethanol shifts towards the less reactive B conformer on decreasing the temperature, this could explain the decrease in +c at temperatures <0 “C. The fact that +c decreases with temperature, even in the range where +F and T~ remain practically constant, requires further investigation, which is in progress in our laboratory.At present we suggest, as a tentative explanation, that photo- isomerization in the triplet manifold (at low temperatures) is characterized by a small energy barrier, as observed for similar 1-naphthylethylene derivatives.I6 EFFECT OF SOLVENT As found previ~usly,~ the effect of solvent polarity on the kinetic parameters and on the fluorescence and photoisomerization quantum yields of these hydrocar- bons is large. The main effect observed for 1-StN on going from an inert solvent to ethanol is a decrease in AE (table 3). The +c value increases and +F becomes smaller because of an enhanced contribution of the singlet mechanism (table 5). In the case of 2-StN, AE is again smaller in ethanol for both conformers (table4). Here, however, a marked increase in kIsc for conformer A leads to a further decrease in $+(A) which becomes smaller than +F(B) (table 5).However, since species B prevails in the conformer mixture, the overall increase in & at room temperature remains relatively small. M.O. CALCULATIONS The results discussed above appear to be dependent on the electronic structure of the lowest excited states involved in the photophysical processes and therefore a molecular-orbital description of these states is expected to provide useful informa- tion about the properties examined. A calculation of the spectra of the naphthyl analogues of stilbene was reported by Wettermark et aZ.,17who used a method based on the complete separability of the (T and T electrons (Pariser-Parr-Pople method).However, the presence of two or more conformational isomers of these compounds cannot be shown without taking into account the non-bonded interactions between hydrogen atoms of the ethylenic and naphthalenic sub-units. For this reason we performed a calculation using a valence basis set with a Hamiltonian (CNDO/S) which is suitable for describing the electronic properties of stilbenic compounds.18 The results, reported in table 6, were obtained using the Mataga-Nishimoto formula for the electronic-repulsion integrals.” The geometry was assumed planar for all the species with the bond length deduced from the available X-ray crystallographic data.’ A general qualitative consideration of the results leads to the conclusion that the ethylenic and naphthalenic states of the three species reported in table6 are close enough together to justify the different behaviour of 1-and 2-StN compared with stilbene.From a more detailed inspection of table 6, one observes that while 1-StN has a well separated pattern of singlet states in the region of 3.6-4.5 eV (only G. BARTOCCI, F. MASETTI, U. MAZZUCATO AND G. MARCONI 1103 Table 6. Energy levels, oscillator strengths and compositions for the low excited states of the trans-styrylnaphthalenes, as obtained from CNDO/ S calculations (HOMO = 43 ; LUMO = 44) state EIeV fx fY f composition 1-StN, more stable species s1 3.68 0.9424 0.0788 1.0212 0.97 (43 -+ 44) s2 3.91 0.0032 0.0173 0.0205 ethylenic 0.74 (43 -+ 46) +0.55 (41 --* 44) naphthalenic s3 4.44 0.0 0.0032 0.0032 0.64 (43 -+47) +0.44 (42 -+ 47) mixed 2-StN, conformer A & s1 3.79 0.4709 0.0002 0.471 1 0.55 (43 -+ 44) +0.60 (43 -+ 45) mixed s2 3.80 0.5653 0.0083 0.5737 0.79 (43 -+ 44) +0.3 1 (43 -+45) mixed s3 4.46 0.0053 0.0001 0.0054 0.40 (40 --* 46) +0.60 (43 -+ 47) naphthalenic w2-StN, conformer B s1 3.79 0.0691 0.0030 0.0721 0.70 (43 -+ 45) +0.50 (42 -+ 44) naphthalenic s2 3.91 1.0786 0.2332 1.3112 0.95 (43 -+ 44) ethylenic s3 4.47 0.0015 0.0102 0.01 17 0.60 (43 -+ 47) +0.40 (40 --+ 44) naphthalenic the more stable form of the conformational isomers was considered), the conformers of 2-StN have closer overlap of the two lowest singlets, which, especially for the A species, are computed to be quasi-degenerate.Consequently, while the first excited singlet state of 1-StN shows ethylenic character and the B conformer of 2-StN shows naphthalenic character, the low-lying singlets of conformer A are strongly mixed, as reflected by the composition of the states, also reported in table 6. The different compositions of the emitting states of the three species explain the different photo- physical behaviour and the solvent effects observed experimentally. By applying the approximate relationship2' fv'kSFfk ---( fv2)st (9) where f and v are the calculated oscillator strength and the frequency of S, of 1-and 2-StN, respectively, and the superscript refers to stilbene, we obtain values of 2.66, 1.30 and 0.20 x IO's-' for kF of 1-StN, 2-StN(A) and 2-StN(B), respectively, 1104 PHOTOPHYSICAL PROPERTIES OF STYRYLNAPHTHALENES in good agreement with the experimental values reported in table 5.The structures of the two conformers of 2-StN (shown below), assigned on the basis of the present analysis, are also in agreement with the results of a recent study of its fluorescence detected at 4 K in a Shpolskii matrix, where the vibronic structure of 2-StN(B) could be partially resolved’ [in ref. (9) the notation is inverted, the A species being the longer-lived one]. While the present model calculation reaches no conclusions about the lack of planarity of the two conformers, the identification of 2-StN(B) with the longer-lived species in S, agrees with the present analysis.21 n 2-StN(A) 2-StN( B) The solvent effect is found to influence the radiative constant of 2-StN(B), leaving the radiative lifetimes of 1-StN and 2-StN(A) and the ISC rates of 1-StN and 2-StN(B) almost unaltered.Since the solvent stabilization of S, depends on the square of the dipole we expect that in a polar solvent the species with the largest dipole moment will be the most stabilized. In fact, the computed dipole moments for S1 [0.233, 0.245 and 0.379D for 1-StN, 2-StN(B) and 2-StN(A), respectively] indicate that conformer A is the most stabilized, as is expected from the analysis of density population of S1,which is composed mainly of charge-transfer configurations.The interaction with the solvent is also active in lowering the activation-energy barriers for internal rotation in S1,this effect being particularly pronounced when Sl has stilbenic character, as for 1-StN. Moreover, in the case of the quasi-degeneracy of S1 and S2, as computed for the isolated molecule of 2-StN(A), the relatively large interaction expected in polar solvents can also change the ordering of the two low-lying singlets. This, along with a more favourable energy gap with the coupled triplet state, can be related to the large enhancement of the ISC rate for 2-StN(A) on going from unpolar to polar solvents. This work was performed with the financial support of the ‘Progetto finalizzato del C.N.R.Chimica Fine e Secondaria’. Dr A. Spalletti is thanked for help with the fluorescence quantum-yield measurements and Prof. E. Fischer (Rehovot) is thanked for stimulating discussions. ’ For reviews see (a) J. Saltiel, J. D’Agostino, E. D. Megarity, L. Metts, K. R. Neuberger, M. Wrighton and 0. C. Zafiriou, Org. Photochem., 1973, 3, 1; (b) J. Saltiel and J. L. Charlton, in Rearrangements in Ground and Excited States, ed. P. de Mayo (Academic Press, New York, 1980), VO~.3, pp. 25-89. F. Barigelletti, G. Bartocci, S. Dellonte, F. Masetti, U. Mazzucato and G. Orlandi, J. Chem. Soc., Faraday Trans. I, 1984, 80, 1123. U. Mazzucato, Pure Appl. Chem., 1982,54, 1705. E. Haas, G. Fischer and E. Fischer, J. Phys. Chem., 1978, 82, 1638; E.Fischer, J. Photochem., 1981, 17, 331 and references therein. J. B. Birks, G. Bartocci, G. G. Aloisi, S. Dellonte and F. Barigelletti, Chem. Phys., 1980, 51, 113. P. Bortolus and G. Galiazzo, J. Photochem., 1973/74, 2, 361.’J. B. Birks, Photophysics of Aromatic Molecules (Wiley-Interscience, London, 1970). G. Bartocci, F. Masetti, U. Mazzucato and E. Fischer, unpublished results. K. A. Muszkat and T. Wismontski-Knittel, Chem. Phys. Lett., 1981, 83, 87. G. BARTOCCI, F. MASETTI, U. MAZZUCATO AND G. MARCONI 1105 In J. B. Birks, Chem. Phys. Lett., 1978, 54, 430. I' H.Garner, D. W. Heaker and J. Saltiel, J. Am. Chem. Soc., 1981, 103, 7164. l2 S. A. Krysanov and M. V.Alfimov, Chem. Phys. Lett., 1983,98, 176. 13 T. Wismontski-Knittel and P.K. Das, J. Am. Chem. Soc., submitted for publication. l4 F. Masetti, U. Mazzucato, S. M. Murgia and A. Poletti, paper presented at the XVIII National Congress of Physical Chemistry, Bari, October 1983, p. 81. T. Wismontski-Knittei, G. Fischer and E. Fischer, J. Chem. Soc., Perkin Trans. 2, 1974, 1930. 16 S. Lazare, R. Bonneau and R. Lapouyade, J. Phys. Chem., 1984,88, 18. G. Wettermark, L. Tegner and 0.Martensson, Ark. Kemi, 1968,30, 185. 18 G. Distefano, U. Mazzucato, A. Modelli, S. Pignataro and G. Orlandi, J. Chem. SOC.,Faraday Trans. 2, 1975, 71, 1583; A. R. Gregory, W. Siebrand and D. F. Williams, J. Am. Chem. Soc., 1976, 101, 1903. l9 K. Nishimoto and N. Mataga, Z. Phys. Chem. (Frankfurt), 1957, 13, 335. 2o S. P. McGlynn, T. Azumi and M. Kinoshita, Molecular Spectroscopy ofthe Triplet State (Prentice Hall, Englewood Cliffs, N.Y., 1969); G. Orlandi, G. Poggi and G. Marconi, J. Chem. Soc., Faraday Trans 2, 1980, 76, 598. 21 I. Baraldi, F. Momicchioli and G. Ponterini, J. Mof. Struct., in press. 22 N. Mataga and T. Kubota, Molecular Interactions and Electronic Spectra (Marcel Dekker, New York, 1970), chap. 8. (PAPER 4/ 165)