ENERGY TRANSFER IN AROMATIC VAPOURS; THE BENZENESENSITIZED FLUORESCENCE OF ANTHRACENE VAPOUR AT 2652 A BY B. STEVENS Dept. of Chemistry, The University, Sheffield 10 Received 2nd February, 1959 The intensity of anthracene vapour fluorescence excited by the 2652 8, mercury line at 170°C is found to increase with the pressure of added benzene vapour. The lifetime of the excited anthracene molecule under these conditions is found from oxygen quenching measurements to be equal to the value found for the same molecule excited by the 3660 8, line, showing that no energy-dependent first-order deactivation of anthracene molecules takes place. Fluorescence enhancement in this case cannot therefore be due to collisional stabilization of the excited anthracene molecules, but must be due to energy transfer to anthracene from excited benzene molecules produced at this wavelength.The lifetime of the excited benzene molecule, determined from oxygen quenching measurements, together with the anthracene fluorescence intensity dependence on benzene pressure, enables a value of 7.6 8, to be calculated for the transfer distance. The transfer of electronic excitation energy plays an important role in photo- synthesis,l in the phenomenon of radiation protection 2 and in the concentration quenching of dye solutions.3 In these systems the efficiency of transfer, which may involve several hundreds of molecules, is often dependent on the nature of the medium which may control the relative dipole orientation of the transferring species or even participate in the transfer process itself.The part played by the medium will therefore be better understood if transfer processes can be studied in its absence, i.e. in the gas phase. Although it is not always possible to do this for molecules encountered in biological systems, since the temperatures required to produce appreciable vapour pressures would undoubtedly cause pyrolysis, it should be possible to investigate transfer in more thermally stable systems with similar electronic properties. The transfer of excitation energy between atoms and simpler molecules, as in the quenching of the mercury resonance line, often takes place with a greater efficiency than that anticipated on the basis of simple collision theory, especially if the molecule is unsaturated.4 Qualitative observations of transfer between more complex molecules include the naphthalene-sensitized fluorescence of acridine, acridonimine, certain phthalimide derivatives and indigo blue, and of aluminium 8-quinolinolate and Mg-phthalocyanine where direct excitation fails.5 Anti- Stokes sensitization of aniline vapour by indigo blue at 3900A and of benzene by aniline at 2800A have also been reported.6 To this author's knowledge, however, the only quantitative measurements of transfer between aromatic mole- cules in the gas phase are those of Dubois7 on the benzene-,!3-naphthylamine system at 2537 ,& which is complicated by the simultaneous collisional stabilization of excited 8-naphthylamine molecules by the sensitizing gas.This paper presents some recently accumulated data on the benzene-sensitized fluorescence of anthracene vapour excited by the Hg 2652A line at 170°C.EXPERIMENTAL MATERIALS Anthracene (m.p. 217") was purified as described elsewhere ; 8 Hopkins and Williams analar benzene was sublimed 3 times in vacuum to remove dissolved gases ; 0 2 was taken from a bulb supplied by the British Oxygen Co. sealed on to the vacuum line. 34B . STEVENS 35 APPARATUS Fig. 1 shows a plan of the optical circuit. The fluorescent vapour was contained in a rectangular quartz cell C measuring 2 x 2 x 5 cm with an absorption path of 1.8 cm, connected through a Hoke packless valve to a standard vacuum line equipped with storage bulbs, McLeod gauge and manometer for measuring gas pressures ; a lower limb of the cell containing solid anthracene rested in an aluminium block which could be electrically heated to control the anthracene pressure.The cell and the metal valve were contained in an asbestos hot-box fitted with a fan, a thermometer and 4 heating coils H. To avoid -J I FIG. 1.-Plan of optical system. PI, P2, P3-Mazda 27M3 photomulti- H-heating coils ; Q-quartz plates ; J-Corning 9863 filters ; C-quartz cell ; G-glass plate ; S-monochromator slit. pliers ; complications arising from Hg-photosensitized reactions it was decided to use the Hg 2652 A line isolated from a 125 W H.P. Hg vapour lamp by means of a Hilger-Watts quartz monochromator D222 with an exit slit width of 0.10 mm, The ratio of fluorescence intensityf, intercepted by a Mazda 27M3 photomultiplier P2, to the intensity R of a refer- ence beam reflected from a quartz plate at 45" to the incident beam was measured as previously described.9 In the sensitization runs, a glass plate was inserted in the fluores- cence beam to remove any radiation emitted by the benzene.A third photomultiplier P3 earthed through a Tinsley spot microammeter was used to measure intensities of trans- mitted light from which fluorescence was removed by a Corning 9863 filter. RESULTS Stern-Volmer plots obtained from the 0 2 quenching of anthracene and of benzene vapour fluorescence at 2652A and 170" are shown in fig. 2. With benzene, the liquid in the lower limb was cooled to - 196" during the addition of 0 2 and allowed to warm up after the valve was closed ; since the liquid temperature, and hence the benzene pressure, could not be carefully controlled in this way, absorption measurements at each 0 2 pressure were made simultaneously and the relative quantum yields fir, determined, thus the ordi- nate of fig.2 for benzene is foZ,/fl;, where IB and 1; are absorbed intensities in the presence and absence of 0 2 respectively. Assuming a collisional quenching efficiency of unity, and collision radii of 4.0, 3.0 and 1-7A for anthracene, benzene and 0 2 , the lifetimes 7A and '78 of excited anthracene and benzene molecules obtained from the appropriate quenching constants KQ are : anthracene (p = 0.76 mm) ; K, = 950 l./mole ; 7A = 2.6 x 10-9 sec ; benzene (p = C. 55 mm) ; KQ = 6000 l./mole; T~ = 2.3 x 10-8 sec,36 ENERGY TRANSFER IN AROMATIC VAPOURS Fig. 3 shows the variation of anthracene vapour fluorescence intensity with pressure of added benzene, the results being expressed as the ratio of intensity f, in the presence of benzene to the intensityfo in the absence of benzene.The temperature of solid anthra- cene was kept at 148.0" throughout, controlling the vapour pressure 8 at 0-76 mm. 3. I 1.c fa f 2.5 2.C I .! [02] mole/l. x lo4 (benzene quenching) I 2 I 3 I 4 I 9' / 9' k , , , , [02] mole/l. x 104 (anthracene quenchhg) 0, KQ = 6OOO I./mole 5 10 I5 2 0 25 x, KQ = 950 l./mole FIG. 2.-Stern-Volmer plots of 0 2 quenching data for anthracene and benzene at 2652 A and 170". 0-benzene ; x -anthracene. [benzene] molell. x 104 FIG. 3.Variation of relative increase j J f 0 of anthracene vapour fluorescence intensity with concentration of added benzene at 2652 8, and 170".B .STEVENS 37 Absorption data for anthracene and benzene at 2652A and 170" are shown in fig. 4. A pressure-broadening effect is observable at low anthracene pressures, the extinction coefficient approaching a constant value over the pressure range used in this work. Similar effects have been noted with 8-naphthylamine 10 and with anthracene 11 at 3660 A. benzene pressure, mm Hg 10 2 0 3 0 4 0 5 0 6 0 1 \ \ %\? -0.08 I I I 1 I I I I . 0.1 0 . 2 0.3 0.4 0 5 0.6 0 . 7 0.8 0.9 I anthracene pressure, mm Hg 0, qo=l8.6 1. mole-1 cm-1 x , rlo=523 1. mole-1 an-1 FIG. 4.-Variation of loglo (transmitted intensity/incident intensity) with vapour pressure of anthracene and of benzene at 2652 8, and 170". 0-benzene ; x -anthracene.From the slopes of the curves the following values are obtained for the decadic extinction coefficients taking 1.8 mm as the cell depth : anthracene : €2652 = 523 1. mole-' cm-1 at 170", benzene : €2652 = 18-6 1. mole-1 cm-1 at 170". DLSCUSSLON The lifetime of the anthracene molecule excited at 2652A under conditions such that self-quenching is negligible, is in excellent agreement with the value of 2.5 x lO-9sec obtained from previous quenching data12 at 3660A using the same collision diameters. The independence of T on wavelength shows that an energy-dependent first-order deactivation is not operative, hence the fluorescence enhancement produced by benzene cannot be ascribed to collisional stabilization of the excited molecule as for aniline,l3 p-naphthylamine 10 and perylene.14 It also indicates that internal conversion from the second excited singlet state of anthracene excited by 2652A, to the lowest excited singlet responsible for fluor- escence emission, is much faster than the process of emission itself so that only the first excited electronic state need be considered.The fluorescence enhancement is readily accounted for in terms of a benzene sensitized anthracene fluorescence. Since benzene does not exhibit self-quenching at the pressures used,7 the following scheme is the most probable :38 ENERGY TRANSFER I N AROMATIC VAPOURS 1 . 2. 3. 4. 5. 6. 7. A + /IV -+ A* A* -+ A 4- hv’ A * + D B* 4 B + hv” B * + D B * + A - + B + A * . B + h v + B* A and B refer to anthracene and benzene molecules respectively, the asterisk denotes an excited electronic state, and D is the product of a radiationless transition.Under photostationary conditions processes 1-7 lead to eqn. (1) for the measured quantity : where k, = rate constant of process iz, I, = intensity of light absorbed by anthracene in the presence of benzene IB = intensity of light absorbed by benzene in the presence of anthracene Z i = intensity of light absorbed by anthracene in the absence of benzene l o = incident intensity, d = cell depth. = 10[1 - eXp (- E,[A]d), Under the experimental conditions the optical density ranges from 0.02 in the absence of benzene to 0.07 in the presence of benzene at the highest concentration (1.72 x 10-3 mole/l.) in which case the exponentials can be expanded and higher terms than the first neglected to give I* - 1 0 4 4 1 4 I B - IoEBIBld, 1; - Io€,[A]d.Eqn. (1) now becomes i.e., fS/’o varies linearly with benzene concentration as is experimentally observed. The slope of the curve (fig. 3) provides the value whence with [A] = 2.77 x 10-5 mole/l. and extinction coefficients already given equal to the ratio of decadic k7/(ks + k6) = k7Tg = KS = 10 ‘5 X 103 I./mOk. Here K,, the sensitizing constant, is equal to the quenching constant for the quenching of benzene by anthracene, and TB is the lifetime of the excited benzene molecule in the absence of anthracene, which is obtained above from 0 2 quenching.B . STEVENS 39 If MA and MB are the respective molecular weights of anthracene and benzene and rAB is the quenching diameter for benzene quenching, or the sensitizing diameter for anthracene sensitization, then Ks = T~(N/lOoo)r&[8 nRT(kf~ $- MB)/kfAMB]* l./moky which with gives which is close to the sum (7.0 A) of the collision radii assumed for these molecules. It is concluded, therefore, that transfer in this system occurs with unit efficiency in collisions of the second kind.The energies of the 'Lb state of benzene and 1Bb state of anthracene are some 38,000 cm-1 and 39,000 cm-1 above the respective ground states,lS on which grounds efficient transfer would be expected ; presum- ably long range transfer is inefficient owing to the lack of suitable orientation at these distances which, however, is achieved in a collision between these planar molecules. TB = 2.3 X 10-8 sec, rAB = 7.6 A, APPENDIX The radiative lifetime ~ ; 2 of the benzene molecule may be estimated from the integrated absorption coefficient which is related to the oscillator strengthf.For this transition in the region 2200-2700 A Almsay and Laemme116 obtain the value f = 0.002 for the vapour at 170"C, whence from the relationship T: = 5.7 x 10-7 sec, v = 3-81 x 104 cm-1. with This is consistent with a quantum yield of benzene fluorescence of The author is grateful to Mr. E. Hutton for calculating this value of T& and to the Royal Society for a grant towards purchase of equipment. 1 e.g. see Rabinowitch, J. Physic. Chem., 1957, 61, 870. 2 e.g. see M. Lefort, Ann. Rev. Physic. Chem., 1958, 9, 123. 3 Levschin and Baranova, Izvest. Akad. Nauk, S.S.S.R. ser. fiz., 1956, 20, 424. 4 Norrish and Smith, Proc. Roy. SOC. A , 1940, 176, 295. 5 Terenin and Karyakin, Izvest. Akad. Nauk. S.S.S.R. ser.fiz., 1951,15, 550 ; Doklady 6 Prileshajewa, Acta physicochim., 1934, 1, 785. Prileshajewa and Klimova, Acta 7 Dubois, J. Physic. Chcm., 1959, 63, 638. 8 Stevens, J. Chem. Soc., 1953, 2973. 9 Bowen and Metcalf, Proc. Roy. SOC. A, 1951, 206,437. 10 Neporent, Zhur. Fiz. Khim., 1950, 24, 1219. 11 McCartin, unpublished data. 12 Stevens, Trans. Faraday Soc., 1955, 51, 610. 13 Neporent, Zhur. Fit. Khim., 1939, 13, 965. 14 Bowen and Veljkovic, Proc. Roy. SOC. A , 1956, 236, 1. 15 In the notation of Platt, J. Chem. Physic., 1949, 17, 484. 16 Almsay and Laemmel, Helv. chim. Acta, 1951, 34,462. Lavorel, J . Physic. Chem., 1957, 61, 864. Akad. Nauk. S.S.S.R., 1954, 96, 269. physicochim., 1937, 7, 163.