Volume 43, number 1 CHEMICAL PHYSICS LETTERS 1 October 1976
TEMPERATURE AND DEUTERATION EFFECTS ON THE FLUORESCENCE OF BENZENOI D SOLUTIONS
L.C. PEREIRA, I.C. FERREIR4 * and Marilia P.F. THOMAZ l *
Centro de Fkica e Q&mica. Univetridade do Minho. Braga. Portugal
Received 3 February 1976
Revised manusaipt received 27 April 1976
Dcutcrium isotope substitution and tcmperaturc ctTects on the fluorescence of toluenc and pxylcnc solutions have been investigated. Values of fluorescence quantum yields wcrc measured bet+ccn -80° to 70°C for p-xylcnc. gxylcnc-d~o. to- luene and tolucne+fa and also fluorescence dccly times at room temperature. The influcnoe of several cxperimcnhl factors on the fluorcsccnce intensity at various temperatures was investigated and corrections were applied to obtain correct quantum yields. The results are compatible with an absence of a normal deuteration effect, and are independent of tempratwe. for both molecules, and show the cxistencc of a non-radiative transition from the fust excited singlet state to the ground state.
I. Introduction
Deutcralion and temperature effects on photolumi- nesccnce havt provided, in recent years, valuable infor- mation for thl: understanding of radiationless transi- tions in poly:rtomic molecules.
In the Tiesent work we present and discuss results
&al+; cvith the effect of deuterium isotope substitu- tion and temperature on the fluorescence of dilute so- lutions of toluenc and p-xylene in ethanol. Several fac- tors affecting the fluorescence intensity have been in- vestigated in detail in order to obtain the correct fluo- rescence quantum yields.
2. Experimental
Toluene, p-xylene and ethanol (spectroquality) and toluenedlg and p-xylenedt,, (isotopic purity 99%) were used without further purification. No impurities were detected in any of the products by luminescence and absorption spectroscopy.
Fluorescence decay times (sM) were measured at room temperature (25°C) with a conventional pulse
l NOW at The Royal Institution, London, UK.
l * Now at Universidade de Aveiro, Portugal
fluorometer using a high pressure nitrogen lamp for ex- citation. The exciting radiation and emission were seier- ted by grating monochromators. Experimental detals of the apparatus arc described elsewhere [ 1 J . The decay curves of the deuterated and protonated solutions, after saturation with argon, were obtained immediately one after the other, under the same conditions of instrumen- tal response to the light pulse.
The values of rh, were obtained from experimental data using a computational method as described by Demas and Adamson (2 1.
Fluorescence quantum yields (@FM) were measured at room temperature in a spectrofluorimeter previousiy described [3 1. From the absorption spctra of the samples and of the standard, previously measured, a narrow bandpass of excitation at a suitable wavenumber was selected which ensures, in each case, the same optical density. The solutions (= 10e3 M) were contained in a spectrosil cell under argon atmosphere and right angle viewing was used. Solutions of toluene in methylcyclo- hexane and in hexane for which eFM = 0.14 was pre- viously determined [4] were used as standards.
Polarization effects induced by the apparatus [S) were measured as a function of wavelength. Although considerable at lower energits they.arc negIigibIe within the range of wavelengths of emission under considera- tion.
Volume 43, number 1 CHEMICAL PHYSICS LETTERS 1 October 1976
The measurements of #PM as a function of tsmpera- ture, below room temperature, was carried out by using a cryostat unit described elsewhere [4]. A specially de- signed dewar with three spectrosil windows was used to work with right-angle viewing and aIso in reflection and transmission. A continuous stream of dry nitrogen on the dcwar’s windows was used to prevent condcnsa- tion at lower temperatures. For measurements above room temperature an insulated heating block was used.
In both cases the temperature was monitored using a temperature meter connected to a recorder and the measurements were taken after convenient temperature stabilization. The solutions were contained in a cell which was sealed off after a convenient number of freeze-thaw cycles of degassing. Alternatively the so- lutions were saturated with argon and sealed via a spe- cial vacuum tap.
Variations of the optical density of the solutions with temperature were monitored by measuring the amount of light transmitted by the solvent and by the sample, at various temperatures, with the same instru- mental conditions as for fluorescence measurements.
By this method it is also possible to detect whether the solvent is transparent over ail the region of temperatures.
Solvents not completely transparent at low tempera- tures were found to produce an anomalous fluorescence
Variations of fluorcsccncc intensities with tcmpcra- ture were always measured during both the cooling and heating period and a good reproducibility was achieved (* 2%). Corrections of the fluorescence intensity due to variations of the refractive index with temperature and solvent were applied (61. Previous calculations of n’ at various temperatures [4) for a wavenumber of about 35 080 cm-I are in good agreement with those used by other authors [7].
Table 1
Fluorescence quantum yields and lifctimcs at roorx temperature
intensity, this being attributed to small particles of ice or 0tLer impurities acting as a diffuse reflector.
Small variations of the intensity of fluorescence due to some degree of polarization were detected at lower temperatures by passing the emission signal through a W polarizer at different orientations and corrections were applied [8]. The reflectivity of the optical system was found to be constant within the interval of tempera- tures considered.
3. Results and discussion
The results obtained for #FM and rM, at 25°C. are shown in table 1 where vaIues reported by other authors are also indicated for comparison. The results found for bFM are lower than those reported by Berlman [9] par-
ticularly in the case of p-xylcnc but arc in good agree-
ment with those of Froehhch and Morrison [ 111 and of Lumb and co-workers [3,10]. The results found for rM are also in reasonable agreement with those reported by other authors.
It is generally assumed that a decrease of fluorescence intensity with increasing temperature is due to a greater probability of the transition from the first excited singlet state (St) to the triplet manifold. This seems to be well established in molecules for which the sum of eFM and triplet yield (#TM) is unity but the situation is different for benzene and its alkyl derivatives [ 13- 151. Measure- ments of @TM for those molecules using.cither the cir+
tram isomerization technique of butene-2 [ 16-181 or The variation of #FM with temperature is shown in figs. 1 and 2, respectively, for the pairs toluene, toluene- d, and p-xylene, p-xylenedtO. As expected 4+M in- creases on lowering the temperature.
Compound +M a) TM a~bls) *Mb) 7~ b, (ns)
toluene 0.13 35.0 0.17 [9];0.12 [lo] 34 191
tolucnoda 0.14 36.0 0.21 [9] 35 I91
pxylene 0.24 33.7 0.40 [9] ; 0.22 [ 1 I] 30 191; 34.5 [12]
pxytene-& 0.22 32.9 0.26 [9] 30.3 [9]
-- a) This world. The estimated enor for both *M and rM is i,S$%.
b, Oth& watk.
Volume 43,‘numbcr 1 CHEMKAL PHYSICS LEI’TERS 1 October 1976
-20 0 20 40 60 60
Temp (T)
Fig. 1. Fluorescence quantum yields BS a function of tcmpcnturc for tolucnc (o) and toluen+ (0).
I :
-80 -60 -40 -20 0 20 10 60 60
Temp (TC)
Fk 2. Fluorescence quantum yields as a function of temperature for pxy~ene (0) and p-xyknwfro (0).
the sensitized biacetyl phosphorescence method (191 have shown the occurrence of a radiationless deactiva- tion of S, in addition to the intersystem crossing pro- cess. The origin of this third channel of deactivation is not weIl established. Calculations based on current theories of radiationless transitions discount the possi- bility of a direct internal convasion S, ‘So [20]. How- ever the recently devefoped tunnel effect model of ra- diationless transitions (211 applied to benzene predicts the existence of a r.on-radiative channel of relaxation S1 -Go with a rate constant with the right order of mag- nitude.
The measurement of h of dilute solutions of to- Iuene,p-xylene and other methylbenzenes [16,18,22),
as a function of temperature, have shown tahat the rate constants for intersystem crossing and korescenCe, respectiveIy km and k,, , are practically independent of temperature and the rate constant for the non-radia- tive transition S, -Go, (kGH) is temperature dependent
t 16,221.
The radiative rate constant for fluorescence as caku- lated from tabie 1 is 3.7 x lo6 s-l and 4.0 x iO6 sWf respectively for toiuene and’ toluenedg and 7.! x lo5 s’* and 6.7 x IO6 s-t for p-xylene and p-xylenedlo.
These results show that kFM is independent of deutera- tiori within experimental erior and confii the so-cal- led x$Gene effect whereby’ para-substituents on the beu- zene riig enhance +FN due toanincreaseof&& [ill.
Volume 43, number 1 CHEMICAL PHYSICS LETTERS 1 October 1976
Using the values of 0.61 obtained by Cundall and ogilvie [22) f or t np e . 1 t yields of dilute solutions of p- xylene in methylcyclohexanc, at 25”C, one obtains for this molecule at this temperature GFM + @TM = 0.85,
assuming negligible solvent effects.
At temperatures corresponding to the plateau region oT$-~, using our results and the triplet yields detcr- mined by Cundall and Ogilvic [22] a value of OFM +
@.TM = 1.06 f 10% is obtained. This shows that as in the caSe of toluene [ 16) the non-radiative process S +So becomes negligible in the low temperatures regiton.
Representing the internal quenching of St by k,, =
kTM + kGhf, one obtains the following ratios at room temperature
k&/k;h = 0.99 for toluene, and
for p-xylene, where the symbols D and H refer respec- tively to the deuteratcd and protonatcd compounds.
The ratios obtained arc, in each case, not far from unity within experimental error. Assuming that kcM is not sensitive to deuteration, since a direct internal conversion St -*So is improbable [20], our results can be interpreted in terms of an absence of a dcutcration effect.
The results obtained show that fluorescence is very little affected by dcuteration. Similar results have been described by other authors [20:23] and in some cases a reverse deuteration effect was reported 1241. Results obtained with gaseous benzene are an exception. Using fluorescence lifetimes and fluorescence and triplet yield results, Cundall et al. [ 141 have reported the values of 7.9 x lo6 s-l and 11 x I O6 s-l”, for intersystem crossing rate constants, respectively, for heavy and light benzene, under high pressure conditions and excitation at 260 nm. Similar effects were reported by other authors [2S].
The situation is different for benzene in the liquid phase.
Cundall et al. (26) have measured +TM and have found respectively the values of 0.235 and 0.250 for dilute solutions of benzened6 and benzene-it6 in methylcy- clohexane, at 26°C. Sandros [ 191 using a different t$chnique has reported respectively G,, = 0.252 for benzene-h6 and #Tht = 0.240 for benzcned6 in dilute solutions of cyclohcxane, at 20°C. Using these results and decay times reported in the literature [9,20], one obtains kk = 8.7 x lo6 s-’ and pm.= 8.9 x lo6 s-’ .
which shows no perdeuteration effect on k,, . The dif- ferences between gas and liquid phase seem to indicate that in the gas hase the intersystem crossing transition proceeds from ‘I B2u to both ?E and 3B states.
As pointed out by Sharf and’&lbey [2:! and by Birks [ 15 1 the reverse deutcration effect is consistent with a dominant S, +T4 (Q > 2) intersystem crossing.
This is expected to happen in molecules in which the transition S, +TZ is wellestablished and was confirmed in the case of anthracenc solutions for which a reverse deutcration effect, disappearing at high temperatures, was found by Lim and Bhattachajee (241.
The S, +T, energy gaps calculated from the litera- ture (201 for toluene and p-xylenc are in both cases
= 8270 cm-‘. This may be large enough to expect some dependence of the intersystem crossing rate constant upon dcutcration due to changes in the Franck-Condon overlap factor. The absence of a normal deuteration ef- fect as found in the present work is consistent with a S, +T, intersystem crossing process for dilute solutions of p-xylcne and toluene.
Assuming k,, and k,, to be practically constant with temperature [ 161, the effect of temperature on fluorescence can be analysed using the expression l/Q%, = 1 + k,,/k,, + (k&$-M) exk-&&kV.
in which k& and IV,,, are respectively the frequency factor and activation energy for the nonradiative transi- tion St *So. Using the fluorescence results obtained in this work and triplet yields determined by Cundall and co-workers [16,22] we obtain kGM = 3.15 x 10” s-l and IVGM
s-l and HI
= 0.24 eV for toluenc and k,, = 3.2 x 10’
2”
= 0.22 eV for p-xylene. The frequency factors an activation energies are reasonably compara- ble with the corresponding values reported for liquid benzene solutions (4.15) and suggest a spin-allowed character for channel 3. Further results of triplet yields for tolucneil8 and p-xylene-d10 may indicate whether perdeuteration is relevant to the nonradiative transition s1 +s,.
Acknowfedgement
The authors gratefully acknowledge partial financial support from a NATO grant and facilities provided by the Physics Department at the University of Lourenco Marques where part of this work was carried out.
Volume 43, number 1 CHEMICAL PHYSICS LETTERS 1 October 1976
References
I41 ISI I61
I71 I81 I91 I101
[Ill 1121
1131
I141
M.F. Thomas 1. Barradas and J.A. Ferreirz. J. Lumines- cence 11 (1’975) 55.
J.N. Demasand A.W. Adamson, J. Phys. Chem. 75 (1971) 2463.
M.D. Lumb, C. Lloyd Braga and L.C. Pcreira, Trans. Fa- raday Sot. 65 (1969) 1992;
M.D. Lumb. MI. Barrada, LC. Pcrcira and C. Lloyd Braga, Rev. F&. Q&n. Eng. U.LM. 1 (1969) 1.
LC. Pereim, Ph.D. Thesis, University of Nottingham (1972).
M. Almgcn, Photochcm. Photobiol. 8 (1968) 231.
M.D. Lumb, Rev. F& Q&n. Eng U.LM. 1 (1970) 1;
J.N. Dcmasand G.A. Crosby, J. Phys, Chcm. 75 (1971) 991.
P.M. Froelich and H-A. Morrison, 1. Phys Chem. 76 (1972) W.W. Mantulin and J-R. Hubcr, Photochcm. Photobiol.
17 (1973) 139.
3566.
E.D. Cchelnik, K.D. Mielenz and R.A. VcIapoldi, J. Res.
Natl. Bur. Std. US 79A (1975) 1.
LB. Berlman. Handbook of fluorescence spectra of aro- matic molecufes (Academic Press, Ns:w York, 1971).
M.D. Lumb and D.A. Weyl, J. Mol. Spectry. 23 (1962) 365.
C.D. Amata, M. Burton, W.P. Helman. P.K. Ludwig and S-A. Rodcmaycr. J. Chcm. Phyr 48 (1968) 2374.
R.B. CundalI and S. McD. Ogilvie, in: Organic molecular photophysics, VoL 2, ed. J.B. Birks (Wiley-Interscience, New York. 1974) pp. 33-93.
R.B. Cur&II, D.A. Robinson and L.C. Per&a, submitted for publication.
[lS] J.B. Birks, cd., Organic molecular photophysics. VoL 2 (Wiley-Intetscience, New York, 1974).
1161 R.B. CurtdalI, LC. Perciraand D.A. Robinson, J. Chem.
Sot. Faraday II 61(1973) 701.
(17 J RB. Cundail and D.A. Robinson, 1. Chem. Sac. Faraday II 68 (1972) 1145.1323.
[I81 R.B. CundalI and LC. Percira. Chem Phyr Lettcxs 16 (1972) 371.
[19] K. Sandros, ActaChem. Stand. 25 (1971) 3651.
[ZO] J.B. Bitks. Photophysics of aromatic molcculcs (Wiley- Interscier& New York, 1970);
- _
WI
J-B. Birks, ed., Organic molecular photophysics, Vol. 1 (Wiley-Intersciencc, New York, 1974).
S.J. Formosinho, J. Chcm. Sot. Faraday II 70 (1974) 605;
S.J. Formosinho and J. Dias da Silva. Mol. Photochcm 6 (1974) 409.
WI
1231 1241
R-B. CundalI and S. McD. Ogilvie. to be published.
B. Sharf, J. Chem. Phys 55 (1971) 320, and references therein.
[251
EC. Lim and H.B. Bhattacharjcc, Chcm. Phyr Lcttcrs 9 (1971) 249; J. Chcm. Phyr 55 (1971) S126,and rcfcrcn-
B.K. Selingcr and W.R. Ware, J. Chem. Phys 53 (1970) ccs therein.
G.M. Breucr and E.K.C. Lee, J. Chcm. Phys. 151 (1969)
3160.
3615;
I261 R.B. Cundall, L.C. Pcrcira and D.A. Robinson, Chcm. Phys.
Letters 13 (1972) 253;
R.B. Cur&B, S. McD. Ogilvie and D.A. Robinson, J. Photo chcm. 1 (1972/73) 417.
I271 B. Sharf and R. Silbey, Chem. Phys. Letters 5 (1970) 314.
