GENERAL DISCUSSION Prof. J. Jortner (Tel-Aviv University) said : The studies of the electronic-vibrational excitations of dimers of aromatic molecules provide important information pertaining to the intermolecular interactions, which determine the exciton band structure in molecular crystals. For the electronic origin of the (AB) dimer, two excitonic com- ponents are expected whose energies are E1,2 = Acf + DA + (AD/2) If [(AD/2)2 + V2]1’2 where Aef is the free-molecular excitation energy, AD = DB - DA represents the difference between the changes of the electrostatic intermolecular energies in the excited state and the ground state, with DA and D, corresponding to the excitation of A and B, respectively. V is the intermolecular excitation transfer integral multiplied by the nuclear Franck-Condon factor.As pointed out by Levy, for the T-shaped dimer of tetrazine IADl 9 I VI and the level splitting essentially measures the difference between the two environmental shift terms. For the planar tetrazine dimer AD = 0 and the excitonic splitting is 2V but, however, the transition to only one excitonic component is allowed. Levy et al. have reported that “ symmetry breaking ” of the planar dimer by complexing with He results in two transitions separated by AE = 0.2 cm-l. It will be extremely interesting to determine whether this splitting originates from the diagonal contribution AD induced by the presence of the He atom, or rather from the off-diagonal contribution V. Proceeding to higher energies one expects to encounter the interesting phenomenon of cooperative electronic-vibrational excitations (CEVE) in dimers. When the electronic excitation is accompanied by a single vibrational excitation, two distinct types of zero-order excitations can be considered : (1) The electronic and vibrational excitation are both localized on the same molecule. (2) The electronic excitation is localized on one molecule, while the vibrational excitation is localized on the second molecule.The cooperative double excitation (2) is energetically split from the con- ventional excitation (1) by the difference of the vibrational molecular frequencies in the ground and excited states. Configurational interaction between the cooperative and the conventional excitations induced by the off-diagonal excitation transfer spreads the intensity for the transition to the two zero states (1) among four components. This phenomenon of CEVE is well documented in molecular crystals providing information concerning the exciton band structure.Studies of CEVE in dimers and in larger clusters will be interesting. E. I. Rashba, J. Expt. Theor. Phys., 1966 23, 708; J. Klafter and J. Jortner, Chem. Phys. 1980, 47, 25. Prof. D. H. Levy (University of Chicago) said: After hearing Prof. Jortner’s interesting suggestion of possible cooperative electronic-vibrational excitations, we examined some old data for evidence of such a process. The frequencies of bands produced by this effect are predictable from the known ground- and excited-state vibrational frequencies of tetrazine. Our data are stored so that a search after the fact retains the full sensitivity of the apparatus.To the limit of our sensitivity, we were unable to find any clear evidence for bands produced by cooperative excitations.1 74 GENERAL DISCUSSION Certainly such excitations must take place, but the cross-sections appear to be very small. Prof. J. M. Lisy (University of Illinois) said: One unresolved question in the molecular-beam studies of Van der Waals molecules has been the failure to observe structural isomers. Thus the observation of isomers of tetrazine dimer is quite intriguing. In particular, a natural question is which isomer is the most stable and by how much? A clue to the orientation of the tetrazine in the T-shape structure may be given in the structure of the benzene-HCl complex determined by Flygare et aZ.2 The HCl centre of mass is 2.627 A above the benzene C, axis with H atom pointing towards the benzene molecule.In the T-shaped tetrazine dimer, a similar situation may occur with the H atom of one tetrazine molecule pointing towards the centre of the other. D. H. Levy, C . A. Hayman and D. V. Brumbaugh, Furuduy Discuss. Chem. Soc., 1982,73,137. ' W. G. Read, E, J. Campbell, G. Henderson and W. H. Flygare, J. Am. Chern. Soc., 1981,103, 7670. Dr. A. Tramer (University of Paris, Orsay) said : An important result of this work is a direct evidence for the broadening of the non-predissociated levels of the complex due to the vibrational redistribution : energy flow from internal (molecular) to external (complex) modes. I would like to consider a more general problem: whether in some systems the vibrational redistribution between discrete levels does not play the role of the primary step in vibrational predissociation ? The simplest example is that of an atom-diatom system (like I2 - Ar) where more than onevibrational quantum of 1-1 vibration are necessary for predissociation, Since all theoretical models suggest a strong propensity for Av = 1 coupling, predissociation may be considered as a sequential process: transition from the " cold " v-level to the discrete manifold corresponding to v - 1 quanta in the 1-1 mode with the energy excess in external modes and then to the (u - 2) dissociative continuum (fig.1). If it is so : (i) the homogeneous linewidth in absorption (fluorescence excitation) spectrum does not characterize the predissociation rate but that of (more rapid) vibrational re- distribution, (ii) energy disposal in products would depend rather on the intermediate than on the initially excited state.There exist some experimental data suggesting such a sequential behaviour. (i) In the infrared dissociation of the (HF), clusters of the type the homogeneous width of absorption bands is of the order of 20 cm-'. It seems highly improbable to assume that the cyclic complex dissociates with a simultaneous break of two strong hydrogen bonds on the timescale of 0.3 ps. A more plausible mechanism would be initial energy flow from the H-F stretching mode to the low- frequency ring modes followed (on a longer timescale) by the ring reclosing with elimination of the HF monomer.(ii) The fluorescence spectrum observed under the excitation of the 66l level of the tetrazine-argon complex is composed of: (a) the resonance emission from initially excited state of the complex, (b) the emission from lower lying T&i2 state of the complex with large amount of vibrational energy in external modes and (c) the emission from the (still lower) 16al level of the free tetrazine molecule. The 6&+T6ii2 redistribution and 6d1-+16a1 predissociation processes may be either parallel or sequential but highlyGENERAL DISCUSSION 175 V v - 1 v - 2 FIG. 1 .-Vibrational levels on an atom-diatom complex. selective population of the 16a' state of the disso ciation product suggests a stepwise process : 6d1+i6a2+16a1.Since all processes take place on the nanosecond time-scale, the direct evidence for the sequential (or parallel) process might be deduced from time-resolved experiments. Note added in proof: Recent 'data of Ramaekers et aL3 suggest the sequential behaviour. l M. F. Vernon, J. M. Lisy, D. J. Krajnovich, A. Tramer, H. S. Kwok, Y. R. Shen and Y. T. Lee, Faraday Discuss. Chem. SOC., 1982, 73, 000. (a) J. E. Kenny, D. V. Brumbaugh and D. H. Levy, J. Chem. Phys., 1979,71,4757; (6) D. H. Levy, personal communication. J. J. F. Ramaekers, J. Langebar and R. P. H. Rettschnick, Chem. Phys. Lett., in press. Dr. U. Even and Prof. J. Jortner (Tel-Auiv Uniuersity) said: We have recently obtained new information on electronic-vibrational excitations of the following classes of large Van der Waals molecules : (a) M-R Van der Waals molecules consisting of a rare-gas atom (R) bound to a large aromatic molecule (M), (6) M.X Van der Waals molecules consisting of medium- sized molecules (X = water, ammonia, methane, carbontetrachloride, methanol, acetonitrile and benzene) bound to a large aromatic molecule (M -- fluorene and tetracene), (c) The H2P.Ar molecule consisting of argon bound to free-base porphine (H,P), ( d ) P.X molecules consisting of porphyrins (P = free-base porphine or zinc- octaethylporphine) bound to medium-sized molecules (X = water, acetonitrile and pyridine). The large Van der Waals molecules were synthesized in pulsed supersonic jets of He (stagnation pressure 1500-2000 Torr expanded through a 600 ,urn nozzle) seeded176 GENERAL DISCUSSION with the two molecules which form the complex. This preparation method allows for effective internal cooling of the Van der Waals complex.We have explored intravalence electronic excitations of these complexes. The microscopic spectral shifts, dv, for the electronic origin of the first spin-allowed S,+S, transition of M or of P in these Van der Waals complexes originates from the follow- ing additive effects: (1) Short-range repulsive interactions in S, resulting in a spectral shift towards higher energies. These small blue spectral shifts are expected to dominate Sv when the dispersive contributions are small, being exhibited in He-M complexes. (2) Dispersive interactions. The red, dispersive, spectral shifts induced in a non- polar aromatic molecule bound to rare-gas atoms Ar, Kr and Xe, or to non-polar molecules, e.g.CH, and CCI,, can be quite well rationalized by the relation Sv = Aa, where a is the polarizability of the ligand (table 1). Although this correlation is expected to hold only for small non-polar ligands, it appears to be roughly obeyed even for a large ligand such as benzene (table 1). (-3) Polar interactions. These can result in the enhancement of either attractive or repulsive electrostatic interactions in the S, state of M-X, where X is a polar molecule. TABLE RED SPECTRAL SHIFTS FOR THE ELECTRONIC ORIGIN OF TETRACENE - X VAN DER WAALS COMPLEXES X Gv/cm-' - (6v/a)/l 024 cm - Ar - 38 24 (3-34 - 68 26 NH3 - 105 48 HzO -113 75 ccl4 - 180 18 C6H6 - 350 (35) For the case of complexes containing the non-polar tetracene (T) molecule, we have observed large red spectral shifts for T-H20 and T-NH3, which are considerably higher than Sv for T-CH, (table 1).The enhancement of the red spectral shift is attributed to the stabilization of SL by dipolar interactions. In the flu0rene.X complexes (table 2), where the M molecule is polar and undergoes a change in its dipole moment upon electronic excitation, a repulsive contribution is exhibited by the polar ligand on the S, state. This repulsive contribution to Sv practically cancels out the attractive, dispersive contribution for X = H20, NH3 and CH,OH, and results in pronounced blue shift for X = CH3CN. This repulsive dipolar contribution roughly increases with the size of the dipole moment, ,u, of the polar ligand.(4) Intramolecular configurational modifications induced by Van der Waals bind- TABLE 2.-sPECTRAL SHIFTS FOR THE ELECTRONIC ORIGIN OF FLUORENE ' x VAN DER WAALS COMPLEXES CH4 - 78 0 NH3 - 17 1.47 H20 0 1.82 CH30H 15 1.7 CHjCN 157 3.6GENERAL DISCUSSION 177 ing. We have found that the spectral shift of the So+SIx transition (the Qx band) of the free-base porphine-Ar (H,P.Ar) complex is 6v = +9 cm-l to the blue. This observation is rather surprising as the spectral shift on a nn* transition of a large aromatic molecule induced by the binding of an Ar atom is expected to be dominated by red dispersive interactions. As the SO+Slx transition of H2P is characterized by a low oscillator strengthf= 0.01 ,l one expects a modest dispersive contribution to dv, so that other contributions to 6v will not be masked by the dispersive contribution and could be amenable to observation.The experimental blue shift presumably originates from medium-induced changes in the geometry of the two internal N-H bonds induced by the binding of Ar, which modifies the energy levels of H2P. It should be emphasized that the physical origins of the contributions to the spectral shifts (1)-(4) considered herein pertain to intravalence excitations, involving a minor charge expansion of M in its electronically excited state. For extravalence Rydberg- type transitions of Van der Waals complexes, the dominant contribution to the spectral shifts will originate from short-range repulsive interactions in the excited state. Regarding the vibrational excitations in the S1 state of large Van der Waals complexes, we found it interesting to study the excitations involving the intermolecular nuclear motion of R with respect to M in M-R, or of X with respect to M in M-X, while the internal vibrations of M are not excited.Three types of such intermolecular vib- rational excitations were observed by us: (A) In-plane motion of a rare-gas atom along the surface of a large aromatic molecule. The So+Sl excitation of the tetracenedAr complex at 38 cm-I above the electronic origin of the bare tetracene molecule was reinvestigated in He jets under superior conditions of internal cooling, revealing the splitting of the spectral features into two components split by 2 cm-l. The second high-energy component of this doublet is tentatively attributed to the excitation of a low-frequency in-plane motion of Ar parallel to the long axis of the tetracene molecule.(B) Out-of-plane motion of a rare-gas atom with respect to the surface of the aromatic molecule. In fluorene-R complexes we have observed vibrational excitation at 40 & 2 cm-l for fluorene-Ar, at 30 5 4 cm-l for fluorene-Kr, and at 40 & 4 cm-l for fluorene.Xe, while for tetracenesxe a vibrational excitation at 32 & 4 cm-' is exhibited. These vibrational excitations are assigned to the out-of-plane motion of R. The experimental frequencies are in accord with the results of numerical simu- lations of potential surfaces of large MR1 complexes, which result in the frequencies of 40-60 cm-I for the out-of-plane motion for a heavy R atom (Ar, Kr or Xe) in a direc- tion perpendicular to the plane of the tetracene molecule. In the fluorene-benzene complex we have observed a well characterized vibrational pro- gression which corresponds to the intermolecular nuclear motion of the two large molecules constituting the complex (fig.2). The vibrational frequency is 28 c111-l. The electronic-vibrational coupling strength is S N" 0.4, indicating a finite configur- ational change in the intermolecular fluorene-benzene equilibrium separation upon SO+Sl excitation of the fluorene molecule. L. Edwards, D. H. Dolphin, M. Gouterman and A. D. Adle, J. MoZ. Spectrosc., 1971, 38, 16. (C) Intermolecular nuclear motion of large ligands in M-X complexes. Prof. F. A. Gianturco (University ofRome) said: In relation to the results reported by Even and co-workers, it seems of some interest to further comment on their observed anomalous behaviour of large Van der Waals molecules with He and Ne atoms.These authors report, in fact, that the energetic shifts, 6v, of the electronic origin of the So+Sl excitation of a well characterized MR, Van der Waals complex relative to178 GENERAL DISCUSSION F 0-0 FB 0-0 FB (28) FB 10 2950 2960 2970 2980 2! wavelength18i 30 FIG. 2.-Fluorescence excitation spectrum of the fluorene molecule and of the fluorene-benzene com- plex in pulsed supersonic expansions of the He. Stagnation pressure p = 2000 Torr, nozzle temp- erature T = 150 "C and nozzle diameter D = 600 pm. The electronic origin of the So+S1 transition of the bare fluorene molecule is labelled F 0-0, while the electronic origin of the So+S1 transition of the complex is labelled F B 0-0.The numbers in parentheses represent the vibrational excitations of the complex (in cm-') above its electronic. the 0-0 transition of the bare molecule M are all towards lower energies (red shifts) when the molecule contains Ne, Ar, Kr and Xe atoms. On the other hand, blue spec- tral shifts are exhibited by complexes with He atoms, a fact which they say could be attributed to the increased importance of repulsive interactions as opposed to the dis- persive terms which are in turn becoming weaker for complexes containing helium atoms. We investigated the structure of simpler complexes containing one 0, molecule and an R atom (R = He, Ar) and also found that their equilibrium geometries strongly suggest the increased importance of repulsive terms when He is concerned. The latter, smaller atom in fact gets closer to the centre-of-mass of the molecule and samples more strongly the highly directional repulsive forces between the oxygen atoms.Thus the Cmh geometry (collinear configuration) shows a minimum position of ca. 7.6 au for ArO, and ca. 7.2 au for HeO,, with well depths of ca. 3.4 x au in the former case and ca. 0.79 x au in the latter. The C,, geometry, however (" perpendicular configuration ") exhibits equilibrium distances from the centre of mass of ca. 6.8 au for ArO, and of ca. 6.2 au for HeO,. The corresponding well depths are ca. 5.2 x in the former system and ca. 1.3 x Our present study therefore seems also to provide further evidence that the genera: trend of the present spectral shifts strongly relates to the relative and specific interplay of short-range and dispersion interactions, hence to the " size " of the charge cloud in the rare-gas partners.in the latter.GENERAL DISCUSSION 179 Dr. B. Soep (University of Paris, Orsay) said : Prof. Jortner and collaborators have reported the interesting observation of sequence vibrational transitions in the Van der Waals modes of rare-gas-tetracene molecules. The presence of vibrational transitions in the optical spectra of rare-gas molecular complexes is usually manifested by only very faint vibrational This supposedly arises from the similarity of both Van der Waals potentials in the ground and excited states.We (N. Halberstadt, C. Jouvet and myself) have found unusually long and intense vibrational progressions in glyoxal (CHO-CH0)-H,, D, complexes when excited in the lAu+lAg transition. Van der Waals complexes appear as satellite bands in the molecular spectra pro- duced in supersonic expansion. Their intensity rises as the pressure of the complexing gas is increased. Thus the band assignment is usually done from the pressure dependence of their intensities. However, bands exhibiting the same pressure effect can either be assigned to vibrational transitions or to chemical isotopes. In fig. 3, we present the fluorescence excitation spectrum of glyoxal-H, and glyoxal- 100 50 0 v1cm-l FIG. 3.-Fluorescence excitation spectra of complexes of glyoxal with Hz (A) and Dz (B) Hz and D2 complexes in the 8; spectral domain.D, molecules. Each vibrational transition of the uncomplexed molecule is accom- panied by satellite peaks (a, b, c, d and e). We attribute the (h), (c), ( d ) and (e) peaks to a vibrational progression in the Van der Waals mode (O-tO, 1,2,3), and the (a) peak to a hot band (130). For glyoxal-H, we have determined Lo = 17 cm-' and cox = 1.4180 GENERAL DISCUSSION b-a c-b d-c e-d f-e 1 2 3 4 5 n 1 2 3 4 v‘+ 1 Avlcrn-’ [= v(n + 1) - v(n), v(u’ + 1) - v(u’)] FIG. 4.-Interpretation of H2-glyoxal Van der Waals progressions. cm-l for the frequency and anharmonicity of the van der Waals mode in the elect- ronic excited state, from fig. 4. If this interpretation is correct we expect for glyoxal-D, a reduction of the frequency by a factor ql’,, where q is the ratio between the reduced masses ‘I = p (glyoxal-D,/p(glyoxal-H,) = 1.98, while the anhar- monicity should be reduced by the factor q.We expect then w = 12.2 cm-l and wx = 0.7 cm-l for glyoxal-D,. From our results presented in fig. 3 we get w = 12 cm-l and wx = 1 cm“, which agrees with the above prediction within experimental error. These results confirm the attribution of the satellite peaks of fig. 3 to a vibration pro- gression in the Van der Waals mode. The existence of such progressions must arise from a profound change in the com- plex geometry caused by an important modification of the electronic distribution in the excited molecular state. A Franck-Condon factor simulation predicts a dis- placement of the van der Waals equilibrium distance of 1.7 A.As the hydrogen is most likely attached above the molecular plane, the corresponding stretching vibration could not be so perturbed. Rotational contours of the vibrational progres- sions are under investigation. Preliminary results imply the other possibility of an hydrogen displacement parallel to the molecular plane, in the excited state, in accord- ance with the electronic distribution in excited ~ t a t e s . ~ D. H. Levy, Adu. Chem. Phys., 1981,47, 323. A. Amirav, U. Even and J. Jortner, J. Chem. Phys., 1981, 75, 2489. N. Halberstadt and B. Soep, Chem. Phys. Lett., 1982, 87, 109. C. E. Dykstra and H. F. Schaefer 111, J. Am. Chem. SOC., 1976,98,401. Prof. J. Jortner (Tel-Aviv University) said: Dr.Soep is correct in stating that the appearance of vibrational progressions of intermolecular vibrational modes in the electronic spectra of large complexes is quite rare, indicating that the intermolecular potential surfaces in the So and in the S, electronic states are similar. We have observed a well characterized vibrational progression of 28 cm-l, clearly exhibiting the 0-1 and 0-2 members for the fluorene-benzene complex. The relative intensity of the 0-1 transition relative to the 0-0 is 0.4, indicating a substantial change of the intermolecular equilibrium configuration.GENERAL DISCUSSION 181 Dr. P. G. Burton (University of Wollongong) said: Prof. Jortner has illustrated energetic shifts of the electronic origin of S,+S, excitations of a molecule M complexed with one or more rare-gas atoms (MR,), compared to the isolated molecule value, and raised the question about modification of the vibrational frequencies in MR, relative to those of the bare molecule.I should like to draw attention to an initial survey undertaken by Meyer et al.' of the modification of the H, potential curve by the proximity of a helium atom. Fig. 8 of their paper gives results which indicate that for collinear geometries, the internal H2 potential curvature increases near re, while it decreases for perpendicular geometries, when a series of fixed intermolecular distances, R, are sampled. In the absence of a full solution to the internal vibrational dynamics of such a complex, we might compute the internal H2 frequencies for a range of angles at each separation of the He and H2, and perform an angular average on the computed vibrational frequencies to estimate the effect of zero-point and higher frequency changes of the H2 at each value R on the isotropic HeH, potential.These internal frequencies thus depending parametrically on R could be averaged over any dimer vibrational level wavefunctions. Relying on the much greater frequency of internal vibration compared to the dimer vibration, this should provide a reasonable estimate to the influence of Van der Waals complex- ation on isolated molecule frequencies, provided of course the underlying electronic potential is sufficiently reliable, and lead to an effective vibronic dimer potential. Any difference between the rigid-rotor electronic potential and such an effective vibronic potential should be most marked for the H2 case, due to its large vibrational amplitude. We are in the process of performingjust such calculations at Wollongong, based on new potentials we have computed for He-H, and for the more complicated H2-H, case.In the latter case, we believe that modifications to the H2 zero-point energies in Van der Waals dimers is at least in part responsible for the isotropic component of our rigid-rotor H1-H2 electronic potentials lying slightly deeper in the Van der Waals minimum than semiempirical potentials. Although this discrepancy (ca. 1 cm-' for the variational PNOCI, ca. 7 cm-' for CEPA2-PNO) is a very small fraction of the computed total energy in our (superposition corrected) supermolecule calculations (1 : 100 000), it represents a significant fraction of the H2-H2 well depth.However to complete such an analysis of the influence of internal zero-point energy changes on the effective isotropic potential is a time-consuming process with the large basis sets we are using, because of the large range of internal coordinate and angular variations that must be completed at each R. W. Meyer, P. G. Hariharan and W. Kutzelnigg, J. Chem. Phys., 1980, 73, 1880. P. G. Burton and U. E. Senff, J. Chem. Phys., 1982,76, 6073. Prof. J. Jortner and Dr. U. Even (Tel-Auiu University) said: The problem of the modification of the intramolecular vibrational frequencies of M in MR, complexes relative to those of the bare molecule is of considerable interest not only in triatomic Van der Waals complexes, as alluded to by Dr.Burton, but also in large complexes. Preliminary evidence pertaining to this problem was obtained recently from the spec- troscopy of the He complexes of 1,4-dihydroxyanthraquinone [H,H] and of its deuter- ated analogue [D,D], studied in our laboratory by G. Smulevich, M. Marzocchi, U. Even and J. Jortner. The electronic origin of the S,+S, transition of the He[H,H] complex is blue-shifted by Av = 5 & 1 cm-' from the electronic origin (0-0) of the S,+S, transition of the bare [H,H]. This observation can tentatively be attributed to the effect of short-range repulsive interactions. An intriguing deuterium isotope effect on the microscopic solvent shift of He[H,H] was observed. We have found that182 GENERAL DISCUSSION Av increases from 5 1 cm-' for He[H,H] to 11 5 1 cm-' for He[D,D].This effect may manifest to modification of the zero-point energies of [H,H] and [D,D] upon binding of an He atom and can be expressed in terms of third-order differences between the zero-point energies of the He[H,H] and He[D,D] complexes relative to the zero- point energies of the bare molecule in s, and in s,. Prof. T. E. Gough, Dr. M. Keil, Mr. D. G. Knight and Prof. G. Scoles (University of Waterloo) said: We would like to give an example of the infrared spectral shift upon attaching rare-gas atoms to a " guest " molecule. These shifts are much smaller than those observed in electronic spectroscopy, but likewise clearly show an evolution from the isolated molecule to one attached to many rare-gas at0ms.l The experiment utilizes the optothermal spectroscopy technique developed by Gough et al., Upon laser-induced photodissociation of Van der Waals molecules, we observe a reduction of molecular-beam inten~ity,~ since the photofragments are scattered beyond the detector viewing angle of 1.0".Modulation of the C.W. line- tuned CO, laser provides a chopped bolometer signal corresponding to the Van der Waals absorption spectrum. The molecular beam is formed by expanding a 1 % SF6 in Ar gas mixture through a room-temperature 17 ,urn diameter nozzle. After being skimmed, the molecular beam is crossed with a 7.5 W laser beam, and after a further flight path of 149 mm, impinges on the 2 K bolometer. Rotational cooling of the SF, v3 mode-is sufficient that no absorption signal of the bare SF, molecule is seen: only absorptions due to Van der Waals molecules are observed.In fig. 5 we show the evolution of absorption features from low ( p , = 1050 Torr) to high ( p , = 6500 Torr) source pressures. For comparison, we show at the very top of fig. 5 low- and high- resolution spectra of SF6 in an Ar matrix. At low pressure, the single spectral feature is red-shifted by ca. 2 cm-' relative to the bare SF6 molecule. As the source pressure is increased, a second feature appears. This second feature is red-shifted by ca. 4 cm-l and displaces the first feature at p s = 3100 Torr. At yet higher source pressures, the peak absorption again shifts to the red, resulting in a fairly narrow peak, red-shifted by ca.10 cm-I relative to the bare SF,. This highest- pressure peak ( p , = 6500 Torr) lies at the absorption maximum of the matrix-isolated SF6 m01ecule.~*~ Other spectra taken between p s = 3900 Torr and p s = 6500 Torr (not shown) demonstrates a rather sudden appearance for the high-pressure peak at 939 cm-l. The results presented here show that the stepwise evolution from Van der Waals molecules, through large clusters, to condensed phase species, may conveniently be followed for vibrational transitions. Further work will be needed to identify the number of Ar atoms bound to the SF6 " guest " molecule. This would enable one to estimate the coordination number necessary to prepare a localized " matrix-isolated '' environment. We thank Paul Rowntree and Karen Fox for their help in gathering the experi- mental data.U. Even, A. Amirav, S. Leutwyler, M. J. Ondrechen, Z. Berkovitch-Yellin and J. Jortner, Faraday Discuss. Chem. Suc., 1982, 73, 153. * T. E. Gough, R. E. Miller and G. Scoles, AppE. Phys. Lett., 1977,30, 338. T. E. Gough, R. E. Miller and G. Scoles, J. Phys. Chem., 1981, 85, 4041. R. V. Ambartzumian, Yu. A. Gorokhov, G. N. Makarov, A. A. Puretzky and N. P. Furzikov, Proc. VKOLS Laser Spectroscopy Conf., ed. B. P. Stoicheff, A. R. W. McKellar and T. Oka, (1981), p. 439. B. I. Swanson and L. H. Jones, J. Chem. Phys., 1981, 74, 3205.GENERAL DISCUSSION 183 I I I 2600 f l 1050 I 5 940 945 9 laser frequencylcm- 0 FIG. 5.-Photodissociation spectra of SF,.Ar,, clusters observed at the indicated source pressures (in Ton).The spectra were recorded as a laser-induced attenuation of the molecular beam intensity for various COz laser lines (data points). Matrix spectra are from ref. (4) and (5), at low- and high- resolution, respectively. The dashed vertical line corresponds to the bare SF6 absorption at 948.5 cm-'. A 17 pm room-temperature nozzle was used for expanding a gas mixture of 1 % SF6 in Ar. Dr. A. Amirav, Dr. U. Even and Prof. J. Jortner (Tel-Auiu Uniuersity) said: The interesting data of Keil and Scholes on the infrared spectra of SF6 in supersonic expansions of At at high stagnation pressure provide information on clustering of Ar atoms on SF6. For some time we have been interested in the effects of selective and gradual solvation on excited-state energetics and dynamics of large molecules in clusters in order to bridge the gap between the photophysics of isolated molecules and of the same molecules in condensed phases.We have conducted studies of electronic spectroscopy of large clusters of Ar, each containing a single aromatic molecule184 GENERAL DISCUSSION (anthracene, tetracene or pentacene), which acts as a fluorescent pr0be.l These clusters were synthesized in supersonic jets of the seeded Ar, expanded through a 150 pm nozzle and stagnation pressures p = 1500- 13 000 Torr. Fig. 6 portrays the laser- induced fluorescence (LIF) excitation spectra of the electronic origin of the " isolated " tetracene molecule (at p = 180 Torr), of the small tetracene-Ar, (n = 1-8) clusters (at p = 710 Torr), while a further increase of the stagnation pressure t o p > 3300 Torr exhibits the LIF spectra of tetracene in large clusters of Ar.The red spectral shifts of the electronic origin (fig. 7) are dominated by dispersive interactions, reaching an I I A 1 p=710 Torr 0 *- 4460 4550 4f 50 wavelength/A FIG. 6.-Fluorescence excitation spectra of tetracene (T) in a seeded supersonic beam of Ar expanded in the pressure rangep = 180-8300 Torr, expanded through a 150 pm nozzle. In all cases the lowest- energy spectral features are shown. The spectrum at 180 Torr is dominated by the electronic origin of the bare T molecule, while the weak satellite band corresponds to T-Ar. The spectra at 710 and 1200 Torr reveal well defined spectral features which could be assigned to TAr,, complexes with n = 1-8. The spectra in the high-flow supersonic expansions at 3350 and 8300 Torr correspond to the electronic origin and to the first 314 cm-' vibrational excitation of large TAr,, clusters. asymptotic value of -705 5 cm-' at p > 6000 Torr, which is comparable to spec- tral shifts in matrix-isolated large molecules. The total spectral widths (fig. 7) of the absorption bands (f.w.h.m.) exhibit a maximum in the vicinity o f p N" 2000 Torr, and are independent of the stagnation pressure at p > 3000 Torr. It appears that the low-pressure range p < 3000 Torr corresponds to the buildup of the first coordination layer, where the effects of inhomogeneous broadening, due to the distribution ofGENERAL DISCUSSION 185 - I ] ! I ; " I I 1 I 5000 10000 1500C p/Torr FIG. 7.-The spectral shifts, relative to the0-0 transition of the bare molecule, and the widths (f.w.h.m.) of the 0-0 transition of tetracene in Ar clusters synthesized at the stagnation pressures p 1=1 2000- 14 000 Torr. cluster sizes, are severe. These effects of inhomogeneous broadening are drastically manifested by the appearance of the maximum in the spectral widths. The satur- ation of the spectral shift and linewidth at p 3 6000 Torr indicates the completion of the buildup of the second coordination layer. In this high-pressure rangep >, 6000 Torr, we have applied the technique of laser fluorescence line-narrowing to provide evidence that the line-broadening of large tetracene-Ar clusters is essentially homo- geneous, presumably originating from phonon coupling effects. A. Amirav, U. Even and J. Jortner, Chem. Phys. Lett., 1980,72, 16.