- Email: [email protected]
Intersystem-crossing and excited-state absorption of C70 studied by picosecond pump and probe absorption and fluorescence measurements
Journal
of
Photokfmistry Photobiology ELSEVIER
Journal of Photochemistry and Photobiology A: Chemistry 115 ( 1998) 89-97
A:Chemistry
Intersystem-crossing and excited-state absorption of CT0 studied by picosecond pump and probe absorption and fluorescence measurements S. Reindl, A. Penzkofer *, H. Gratz Institut II-Experimentelle
und Angewandte Physik, Universitiit Regensburg, D-93040 Regensburg, Germany Received 22 December 1997; accepted 9 March 1998
Abstract The singlet and triplet absorption and emission dynamics of CT0in toluene is studied with a picosecondNd:glasslaser double-pulsepump and probe transmission and fluorescencemeasurementtechnique. The singlet S, excited-stateabsorption ( u,,,r = 6 X 10- l7 cm2), the triplet T, absorption (a, = 4.2 X lo- I7 cm’), and the triplet T, absorption (a,,,, = 4.3 X lo-l7 cm2) at 527 nm are determined.The singlet to triplet intersystem-crossing is studied by double pump pulse induced fluorescencesignal analysis and by double probe.pulse transmissionmeasurements. The yield of triplet formation is found to be near to unity. 0 1998Elsevier ScienceS.A. All rights reserved. Keywords: Intersystem-crossing; Excited-state absorption; C7s
1. Introduction Since the discovery [l] and laboratory synthesis [2] of fullerenes their study is an active field of research [ 31. The optical spectroscopy of Csc and CT0is reviewed in Refs.
[ 3,4]. The ground-state absorption of Cm [4-81 and CT0 [4,6-81 was analysed. Fluorescencestudies of C,, [9-121 and CT0 [9,1 l-151 have been performed. Small quantum yields of fluorescence (#+=2.2X lop4 for Cm [lo], and QjF=5.9X10W4 [14] to 8.5~10~~ [9,16] for C,,) were determined. The yield of triplet formation was found to be closeto 100% (&a: [4,17];C,,,: [4,18-21]).Triplet-triplet absorption spectra for CeOand C,,, are presented in Refs. [4,19,20,22,23] and Refs. [ 4,20,21,23,24], respectively. The excited-state absorption of CsO[ 11,25-331 [ 34-401 and CT0 [ 11,15,24,26-281 has been studied mainly with respect to optical limiting applications (reverse saturable absorption) [ 28-32,34,35,38-41]. In this paperthe singlet and triplet absorptionandemission dynamics of C,, in toluene at room temperatureis investigated. The S,-S, (n > 2) singlet-singlet, the T,-TI (12 2) and the T,-T, (m 2 3) triplet- triplet absorption cross-sections at 527 nm, and the quantum yield of triplet formation
by S,-T, intersystem-crossing (population accumulation in T, state after intersystem-crossing) are determined. * Corresponding regensburg.de
author.
E-mail:
[email protected]
lOlO-6030/98/$19.00 0 1998 Elsevier Science S.A. All rights reserved. PIISlOlO-6030(98)00264-O
Picosecond second-harmonic pulses of a mode-locked Nd:glass laser are used in the experiments.A double pumppulse and double probe-pulsetransmission [ 421 and a double pump pulse induced fluorescence measurementtechnique [ 43,441 are applied. Someabsorption cross-sectionsand the quantumyield of triplet formation are deducedby comparing the experimental results with numerical simulations of the absorption and emission dynamics. The first picosecond pump pulse transmission dependson the S,-S, excited-state absorptiondynamics (excited-stateabsorptioncross-section, u~~,~,and $-state relaxation time, T,,) . The first probe pulse transmissionis causedby the residual &,-ground-stateabsorption and the induced T,-T, triplet-triplet absorption at the time, to, of probe pulse passing. It dependson the yield of triplet formation, &, and the T, triplet absorption crosssection, 0,. The secondpump pulse transmissiondependson the singlet-singlet and triplet-triplet absorption and relaxation dynamics. It allows to extract the higher excited-state T,-T, absorption cross-section, u,,,~ The second probe pulse transmission depends on the population accumulation in the T, triplet state. The second probe pulse transmission together with the first probe pulse transmission allows the
separatedetermination of or and &. The first pump pulse induced fluorescencesignal, SF,iris determined by the pump pulse induced S,-state level population and the fluorescencequantum yield. The secondpump pulse induced fluorescencesignal, SF,2,is reducedcompared to the first pump pulse induced fluorescencesignal in the case
90
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry Il.5 (1998) 89-97
of singlet-triplet intersystem-crossing since then the S,-state level population is reduced. The ratio of S,,*/S,, allows the determination of the quantum yield of triplet formation, &
143,441. The S ,-S, excited-state absorption cross-section, ueex,=, and the T,-T, absorption cross-section are found to be larger than the Soground-state absorption cross-section, oL (reverse saturable absorption [ 401) . The quantum yield of triplet formation, a,, by S,-T, intersystem-crossing is determined to be close to 100%.
2. Experimental The fullerene CT0 was purchased from Aldrich ( [5,6]Fullerene Co) and is used without further purification. The solvent toluene is of analytical grade. The absorption crosssection spectrum, og( h) , of C,O in toluene at room temperature is displayed in Fig. 1 together with the fluorescence quantum distribution EF( h) (own measurements). The double-pulse pump and probe transmission measurements have been carried out with a picosecond Nd:glass laser system [ 4.51. The experimental arrangement is displayed in Fig. 2. The mode-locked laser generates a train of picosecond pulses. The time period of adjacent pulses is tR = 10 ns (resonator roundtrip time). Two adjacent pulses are separated from the pulse train by an electro-optic switch [47] and increased in energy by a Nd:phosphate glass laser amplifier. The pulses are frequency doubled in a CDA crystal [ 481 and the remaining fundamental light is filtered off (filter Fl ) . The wavelength of the second-harmonic pump pulses is hL = 527 nm. The pump pulse duration is AtL = 6 ps. The second harmonic pump pulses are directed to the sample, S (cell thickness Z= 1 mm). They are increased in intensity by lens Ll. The input peak energy density is determined by energy transmission measurement through rhodamine 6G in ethanol (saturable absorption [ 461) with photodetectors PD 1 and PD2. The energy transmission of the pump pulses is measured with the photodetectors PDl and PD3 (small signal transmission To = 0.0756 + 0.001) . The probe pulses are split-off from the pump pulses with a beam splitter and are attenuated with a filter (F2). They pass opposite to the pump pulse direction through the sample. Their temporal delay relative to the pump pulses is tn = 9 ns. The probe pulse diameter is adjusted by lens L2. The transmission of the probe pulses is registered with the detectors the PDl and PD4. The double-pump-pulse fluorescence measurements are carried out with the probe beam path covered. The fluorescence emitted in sideward direction is collected with lens, L3, and directed to the spectrometer, SP, with lens, IA. The fluorescence light in the spectral region from 640 to 840 nm is passed through the spectrometer and registered temporally resolved with a microchannel-plate photomultiplier (Hamamatsu type R1564U) and a fast digital oscilloscope (LeCroy type 9362, time resolution approximately 500 ps) .
400
600
500
700
Wavelength h (nm) Fig. 1. Absorption cross-section spectrum, u=(h) , and fluorescence quantum distribution, /I& A), of C,a dissolved in toluene.
IM.L.LaSert---~----(Amplifier------~---$------$ SHG
Fl
I w C.-.-.-.-.-.-.-.-.-.-~.~.~-.-.-.~.-.-.-.~~~-~.-~ A--L3 I
I
I
G-4 i
i
I
I !
I i
FRh
j
Fig. 2. Experimental setup. M.L. Laser, mode-locked N&glass laser. SW, single pulse selector. Amplifier, N&glass laser amplifier. SHG, frequency doubler. Fl, cut-off filter of fundamental laser wavelength. Ll-L4, lenses. Rb, rhodamine 6G in ethanol for energy density measurement [ 461. PDlPD4, photodetectors. S, sample. SP, spectrometer. MCP, microchannel-plate photomultiplier.
The intensity dependent transmission behaviour of the solvent toluene was measured separately with the experimental arrangement of Fig. 2 (pure toluene in sample cell, S) . The measured energy transmission vs. input pump pulse peak intensity is displayed in Fig. 3. The transmission reduction at high pump pulse intensities is thought to be due to two-photon absorption [ 491. The curves in Fig. 3 are calculated for various two-photon absorption coefficients, (Y@).The best fit is obtained for (Y(‘) = (2f0.3) X lo-” cm/W. The twophoton absorption of the solvent is included in the analysis of the experimental results.
3. Experimental results The pump pulse energy transmissions, TE,L,, TE,L1, the probe pulse energy transmissions, Tp,, Tm, and the pump
S. Reindl et al. /Journal
of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
Input Peak Intensity IoL (W cm-2) Fig. 3. Energy transmission through toluene. Pump laser wavelength, AL = 527 nm. Sample length 1= 1 mm. Circles, experimental points. Curves are calculated for cy’*‘=5XlO-” cm/W (1). 1x10-” cm/W (2), 2 X lo-” cm/W (3), and 5 X lo-“cm/W (4).
pulse induced fluorescence signal heights, S,,, , SF,*,are measured as a function of the first pump pulse input peak intensity, I ,,L,,. Data are selected where the second pump pulse peak intensity, Z0L,2,is approximately equal to the first pump pulse peak intensity. The probe pulse peak intensities are kept a factor of 250 smaller than the pump pulse peak intensities. The probe beam diameters were a factor of two narrower than the pump beam diameters at the sample position. The saturation intensity of S,, ground-state depopulation, Z,, = hzy/ ( rLAtL) (h, Planck constant; %,, second harmonic laser frequency; crL absorption cross-section at c; AtL, duration of second harmonic pulses), is marked in Figs. (4)-( 8) (slow saturable absorption conditions apply since At,- -K rs, , where rs, is the S , -state lifetime [ 501) . The energy transmission, TE,L,, of the first pump pulse (A, = 527 nm, At, = 6 ps) through C,, in toluene (small signal transmission T,=O.O756) is displayed by circles in Fig. 4. The transmission decreases with rising input pump pulse intensity. The behaviour indicates that the excited-state absorption is larger than the ground-state absorption at 527 nm (reverse saturable absorption). The energy transmission, TE,Lz, of the second pump pulse (time delay tR = 10 ns) through the sample is shown by circles in Fig. 5. The decrease in transmission with rising pump pulse intensity is similar to the behaviour of the first pulse. It indicates a larger T,-triplet absorption than So-singlet absorption. The first probe pulse transmission, Tp, (time delay tD= 9 ns) is displayed by circles in Fig. 6. Its dependence is determined by the So ground-state absorption of the molecules, which have remained in the &,-state, and by the Titriplet absorption of the molecules, which have proceeded to
91
InputPeakintensityIoL,,(Wcm-*) Fig. 4. Energy transmission, TEL,. of first pump pulse through CT0in toluene vs. input pump pulse peak intensity, IaL.,. Small signal transmission, T,=O.O756. Circles are experimental data points. Curves are calculated using parameters of Table 1. Solid curves, r., = 60 fs with oeex,L= 4 X 10 - ” cmZ(1),5X10~“cmZ(2),6X10~“cm2(3),7X10~”cm*(4).Dashed curves,o.,,=6X10-“cmZwith~..=300fs(a),200fs(b),100fs(c). 0.08
I-
2
I’ , 3’
I
0.07
.-5 .-$3 j
0.06
z? iii 15
Input PeakIntensity IoL,,(W cme2) Fig. 5. Energy transmission of second pump pulse, TE,rz, vs. input peak intensity of first pump pulse (second pump pulse peak intensity is nearly the same as first pump pulse peak intensity). Circles are experimental data points. All curves are calculated for a,= 4.2 x lo-l7 cm*. Solid curves, (1). 3X lo-“cm (2), 3.5X 10-l’ rT= 2 ps and a,,,r =2.5X lo-“cm’ cm2(3),4X10~‘7cm2(4),4.5X10~‘7cm*(5),and5X10~’7cm2(6). Dashed curves, a.,,= =2.5X 10-l’ cm’ and rT= 10 ps (a), 100 ps (b). Other parameters are taken from Table 1.
the T,-state. The probe pulse transmission reduces with rising first pump pulse peak intensity, ZoL,,. But the decrease in transmission of the first probe pulse is slightly smaller than the decrease of transmission of the first pump pulse indicating
S. Reindl et al, /Journal of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
92
ka 0.05 108
10'0
109
Input Peak Intensity IoL,,(W cm-2)
Input Peak Intensity IoL,, (W cm”)
Fig. 6. Transmission of first probe pulse, TPI, vs. first pump pulse peak intensity, Z,,,. Circles are experimental results. The curves are calculated using the parameters of Table 1. Dashed curves, or=4 X 10-l’ cm’ and &=O (l), 0.25 (2), 0.5 (3), 0,75 (4), and 0.9 (5). Solid curves, &=0.999 and a,=4~10-” cm’ (6). 4.2x10-” cm2 (7), and 4.4X lo-“cm (8).
Fig. 8. Fluorescence signal ratio, .S,,/S,,, Circles are experimental data. To = 0.0756. first and second pump pulse are corrected [ZoL,,/ ( 1 + Z,,, ZZ,,,) ] / [ ZoL,a/( 1 + Z,&Zrat) parametersofTableland&=O(1),0.25 and 0.999 (6).
vs. input peak intensity, Z,,,,. Differences in peak intensity of by S,,s/S,,, = (S,,/S,,),,,.,, ] The curves are calculated for (2),0.5(3),0.75(4),0.9(5),
The fluorescence signal ratio, S,,*/S,,, , of the second pump pulse induced peak fluorescence signal height to the first pump pulse induced peak fluorescence signal height is displayed in Fig. 8 (circles) as a function of the first pump pulse peak intensity, ZoL,,. The ratio decreases with increasing ZoL,, indicating triplet level population by intersystem-crossing.
4. Theoretical description
I.
I
108
109
. ..I
J
IO'O
Input Peak Intensity IoL,, (W cmm2) Fig. 7. Transmission ratio of probe pulses Tpz/TPI. vs. input peak intensity of first pump pulse, Z,,,. Circles are experimental data. Error bars show standard deviation of the mean transmission ratios. The curves are calculated for &=O (l), 0.25 (2), 0.5 (3), 0.75 (4). 0.9 (5), 0.999 (6). Other parameters are taken from Table 1.
that the T,-triplet absorption cross-section, oTT,is smallerthan the S,-state excited-state absorption cross-section, o=+ The ratio of second probe pulse transmission to first probe pulse transmission, TP21TPI, is shown in Fig. 7. The transmission ratio is slightly less than one with a minimum around I OL, I = 7 X lo9 W/cm2. The intensity dependent behaviour of TP, and T,IT,, is used below to extract a, and &.
An extraction of the absorption cross-sections, u=~,~,a,, ueex,=,of the higher-excited-singlet-state relaxation time, T,,, and of the St-T, intersystem-crossing rate, k35, from the experimental transmission and fluorescence data requires a theoretical description of the absorption and emission dynamics and a numerical simulation of the experimental results. The absorption dynamics of the two pump pulses and the two probe pulses is analysed using the energy level system of Fig. 9a and the double pump pulse and double probe pulse scheme of Fig. 9b. All pulses are linearly polarized in the same direction. The pump pulse induced fluorescence dynamics is determined by the dynamics of the S, -level population. 4. I. Level population
and pulse propagation
model
The first pump pulse excites molecules from the So-groundstate (level 1) to an excited singlet state (level 2; groundstate absorption cross-section, oL) . From there the molecules relax quickly (Franck-Condon relaxation time constant, rFc,s) to a thermal equilibrium position in the S,-band (level
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry 11.5 (1998) 89-97
(a) 6 TL 7 Tz
5 T,
93
the phosphorescence lifetime, rp, on a microsecond to millisecond time scale [ 16,18,20,21,51,52]. The second probe pulse passes through the sample after a time delay of tL,,pz = t,, + tD relative to the first pump pulse. Its transmission is determined by the S, (level 1) and T, (level 5) population number densities at the time of passing the sample. Higher excited-state singlet + triplet intersystem-crossing [ 53-561 and triplet + singlet intersystem-crossing [ 42,5760] are not included in the absorption and emission dynamics, since the transmission and fluorescence behaviour can be fitted reasonably well with the presented model and the error bars of the experimental data allow no further model refinement. 4.2. Equation system
___._._. -.-.- .-.-.-. -.-.-.-.-.- .-.-.-,-,-.-.-.-.-. -.-.-.-.-.-.-.-.-.-.-.-.-.u Fig. 9. (a) Energy level scheme for singlet-singlet absorption dynamics, triplet-triplet absorption dynamics, and S,-T, singlet-triplet intersystemcrossing. (b) Scheme of double-pulse pump and double-pulse probe interaction in the sample.
3). From levels 2 and 3 there occurs S,-S, excited-state absorption to band 4 (excited-state absorption cross-section, geex,=).The population in level 4 relaxes quickly to level 3 (relaxation time constant, r,,) . The thermally relaxed S,state population (level 3) relaxes to the ground-state (relaxation rate constant, k3,) and to the triplet system (intersystem-crossing rate constant, ks5; yield of triplet formation, &). The S,-state lifetime, rs, (equal to the fluorescence lifetime, rF) is given by us, = (k3, + kj5) - ‘. The S , -state intersystem-crossing rate is k-,5= &TS, I. The first probe pulse, Pl, passes through the sample after a time delay of tL,,p, = tn which is slightly shorter than the temporal pump pulse separation, tR, but much longer than the S,-state lifetime, rs,. The transmission, Tp,, of the probe pulse, Pl, is given by the singlet ground-state absorption, which is determined by the Z&-ground-state population number density, N, (t’ = tn), and by the T,-triplet absorption, which is determined by the population number density of the lowest triplet state, N5( t’ = tD) (N, (t’ = to) = N,, - N, ( t’ = tD) , where No is the total molecule number density). The second pump pulse, L2, causes the same absorption and emission dynamics in the singlet system as the first pump pulse (but the initial ground-state population, N, ( t’ = tR), is smaller) and additionally causes excitation and relaxation in the triplet system. The molecules in the triplet ground-state, T, (level 5)) are excited to a higher lying triplet state, T,, (level 6, triplet absorption cross-section, Us). From there the molecules relax quickly to the lowest excited triplet state, T2 (Franck-Condon relaxation time constant, rFC,r). From the T,-state (level 7) the molecules relax to the T,-state with a time constant or. The back transfer from T, to So occurs with
The saturable absorption dynamics of the two pump pulses and the two probe pulses is handled by the following differential equation system for the level population number densities, N, to N,, and the laser pulse intensities, IL,, ZL2,Zr,, and Zp2[ 441. Immediate relaxation from level 8 to level 7 is assumed ( r,,,r = 0). +k31N3+ N,
(1)
7P
(2)
(3)
(4) (5) (6) aN7 -=---
Ns
NT
(7)
(8)
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
94
azPi -=8Z’
ULNI a.
1
+~+TN,
(i=1,2)
ZPi
I
r
Ii(t’,l,r)dt’dr
II
(i=Ll,L2,Pl,P2)
TE,i ,” La”
(10)
r I,(t’,O,r)dt’dr 0 --
II
The transformation t' = t - nzlc, and z’ = z is used, where is the time, z is the distance along the propagation direction, co is the light velocity in vacuum, and it is the refractive index. I is the sample length. An absorption anisotropy is not considered since C,,, is nearly spherical symmetric [ 31. The initial conditions of the level populations are N, (t' = - w,z’,r) = No and Ni( t' = - m,z’,r) = 0 for i= 2,. ..,7. No is the total number density of dissolved fullerene molecules. The input pulse intensities are t
Z,, (t’,z’
=O,r)=I0L,i
ZL2(t’,z’ =O,r)=I
exp
-(3’-(3’1
(11)
oL,zexp[ -(yr-(3]
Zp, (t’,z’ =O,r)=~0P,,exp
ZP2(t’,z’ =O,r)=Z
[
(12)
[ py-y]
op,zexp[ -( “-~--fa)2-(
(13)
$J
(14)
r is the radial beam coordinate. r, and r, are the 1le beam radii of the pump and the probe pulses, respectively. to = 2 - ’ [In ( 2) ] - ’ ‘*At= is half the 1/e temporal pulse width ( AtL is the FWHM pulse duration).
5. Data analysis The equation system ( 1) -( 10) is solved numerically using relevant parameters of CT0 in toluene (see Table 1). Unknown parameters are determined by fitting the calculated energy transmissions of the pump and probe beams and the pump pulse induced fluorescence signals to the experimental values. In Fig. 4 the first pump pulse energy transmission, TE,Llr is simulated. TE,Ll depends on the S, -excited-state absorption and on the S,-state relaxation time concross-section, cr.=ex,L, stant, r,,. The solid curves are calculated for various u~~,~values ( rex = 60 fs), while the dashed curves are calculated for different &-state relaxation times, T,, (a,,,= = 6 X lo- I7 cm’). The best fit to the experimental data is obtained for u ex,L= 6 X lo- I7 cm2 and r,, = 240 fs. The double pump-pulse induced fluorescence signal behaviour is simulated in Fig. 8. The fluorescence signal ratio, S,,z/S,,, depends on the yield of triplet formation, &. S,,z/
S,,, is practically independent of the explicit triplet-triplet absorption dynamics (verified by numerical simulation, but not shown in Fig. 8). The curves are calculated for various quantum yields of triplet formation, &. A comparison of the calculated curves with the experimental data gives & = 0.9 + 0.1. The experimental accuracy and subsequently the accuracy of & is limited to 10% because the fluorescence quantum yield, (PF, of C70 in toluene is very low (& = 8.5 X 10e4, see Table 1). The first probe pulse transmission, Tp,, is simulated in Fig. 6. Tp, depends on the quantum yield of triplet formation, &, and on the cross-section, +, of T, -+ T, triplet-triplet absorption. Knowing &, the fitting of calculated T,, curves to experimental data allows the determination of a,. The dashed curves in Fig. 6 belong to various & values at a fixed value of g= = 4 X lo-l7 cm*. The solid curves are calculated for various or values while the triplet quantum yield is fixed to & = 0.999. The best fit to the experimental data is obtained for +=4.2X lo-l7 cm*. The probe pulse transmission ratio, Tp2 / Tp, is simulated in Fig. 7. The ratio depends on the yield of triplet formation, &, and the absorption cross-section ratio, arluL. If +T/ uL = 1 then the transmission ratio is Tp2/Tp, = 1 independent of &. The combined dependence of Tp, (Fig. 6) and Tp2/ allows the determination of both & and or. The TPI on IOU best fit is obtained for I#+= 1 - &F= 0.999 and CT= 4.2 X lo-l7 cm*. A high accuracy of or and & determination would be obtained, if cr, were significantly different from crL. For C70 in toluene the singlet ground-state absorption crosssection, crL= 3.5 X lo- l7 cm’, and the T, triplet-state absorption cross-section, +=4.2X lo-l7 cm*, are only slightly different. Therefore the accuracy of & determination by Tpl and T,/T,, measurement is moderate. The second pump pulse transmission behaviour is simulated in Fig. 5. TE,L2 depends on the T,-TI triplet absorption, crT, on the T,-T, triplet excited-state absorption, akx,r, and on the triplet-triplet relaxation time, rT. Since a, has been already determined by probe pulse transmission measurements, the curves in Fig. 5 are calculated for various o,,,T values ( rT= 2 ps, solid curves) and rT values ( flex,T= 2.5 X lo- l7 cm*, dashed curves) to study their dependences. For ueex,T similar to UT, the second pump pulse transmission, T E,L2,becomes practically independent of rr. Our numerical fit gives u~~,~= (4.3 k0.2) X lo-l7 cm*, while ur has been determined to be or = (4.2 + 0.1) X lo- I7 cm*. Therefore, determination of rT is not possible. In organic dyes typical values of 7T are of the order of 0.5 to 3 ps [ 42,441.
6. Discussion The determined parameters, 9, u~~,~, r,,, UT, %,TI and #r, of C70 in toluene at room temperature for AL = 527 nm are listed in Table 1 together with parameters reported in the literature.
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
95
Table 1 Spectroscopic parameters of C,, at room temperature Parameter
Value
Ts, (PS)
110 625 627 650 660 670 700 0.710 0.730 0.870 4-5 41 130 140 53000 0.00059 0.0007 0.00085 0.76 f 0.15 0.9 0.99 f 0.02 1*0.15 1 (10.85) 0.5 0.5 240 f 40 2 3.2~ 10-l’ 3.5 x lo-r7 1.25 x lo9 1.54 x lo9
[20],inbenzeneandhexane This work Assumed [ 6 11, used here Assumed, used here This work Assumed [42&l] [21],inbenzene Own measurement, used here [ 151, in benzene kfi3 = &Q-S;I, used here
3.6~ lo-r7 3.8~ lo-l7 3.9x 10-l’ (4.2+0.1)X10-” 4.75 x lo- ” (6.0f0.5) X10-l’ (4.3f0.2) X10-r’
[ 231, in benzene [21],inbenzene [ 201, in hexane This work [ 201, in benzene This work This work
TP(CLS)
%c,s (ps) %c,T (ps) 7,x (fs) TT (ps) flL (cm’)
hs (s-l) uT (cm’)
ueer,L(cm’) o;,.~ (cm’)
Comments
1511 [ 91, undegassed [ 131, argon degassed benzene solution Used here [ 91, degassed ]241 1271 [20], air saturated benzene solution [ 161, air saturated [ 181, air saturated benzene solution [ 181, argon-saturated [ 521, clusters in supersonic beam [ 161, degassed [21] in benzene [ 5 11, at 77 K in toluene/ 10% poly-( cY-methylstyrene) Cl41 [ 91, in benzene [9,16] and own measurement, used here [21], in benzene [I91
1181
Solvent toluene except stated otherwise.
The ratio of singlet excited-state absorption to singlet ground-state absorption is a,,,,/~, = 1.7. The S, state lifetime is rs, = 650 ps. For picosecond pulses (pulse duration, AtL, less than S,-state lifetime, rs,), CT0 in toluene behaves as a reverse saturable absorber since u~~,~> uL. On a nanosecond time scale (pulse duration or time of observation longer than rs,) the nonlinear transmission behaviour is determined by cr=T/g=. For C,. in toluene this absorption cross-section ratio is uJcrr= 1.2 and the reverse saturable absorption effect is rather weak. Our double-pulse fluorescence measurements and probe pulse transmission measurements agree well with a quantum yield of triplet formation of&=l-& (&,=8.5X lo-“). Triplet quantum yields close to 100% have been reported by several groups ( see Table 1) . The S, -state relaxation of CT0
(as well as of Cso [ 4,171) is dominated by intersystemcrossing. The rate of internal conversion is small compared to the rate of intersystem-crossing (because & close to 1) . The radiative relaxation time constant is long compared to the time constant of intersystem-crossing, 7isc( rise= k35- ’ = dvs, =65Ops; r,,d=rs,/&=760ns). A higher-excited-state relaxation time of r,, = (240 f 40) fs has been determined for CT,-,in toluene. In organic dyes the higher-excited-state S, + S, singlet relaxation is generally fast in the 50 to 100 fs time range [ 62,631. Only in the case of large energy gaps between S2 and S, longer relaxation times are expected [ 641 and observed [ 65,661. The symmetry of the CT0 molecule and its weak So-S, absorption cross-section may be responsible for the moderate S, + S, relaxation time constant, r,,, of CTO.
96
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
7. Conclusions The singlet and triplet absorption dynamics of CT,, in toluene at room temperature has been studied with picosecond pulses at 527 nm. A double pump pulse induced fluorescence signal technique and a double pump pulse induced probe pulse transmission technique have been employed to determine the yield of triplet formation. Despite the fact of low fluorescence quantum yield and only slight difference between singlet ground-state and lowest-triplet absorption cross-sections, the quantum yield of triplet formation could be determined by both methods (fluorescence technique and absorption technique). The S r--S, excited-state absorptioncross-section, the $-state relaxation time, as well as the T,-T, and T,-T, absorption cross-sections have been determined. The presented methods of double pump pulse induced fluorescence detection and double pump pulse induced probe pulse transmission measurement may be generally applied to the determination of quantum yields of triplet formation. The presented pump and probe pulse transmission analysis generally allows the extraction of excited-state absorption crosssection and excited-state relaxation parameters.
Acknowledgements The authors thank the Deutsche Forschungsgemeinschaft for financial support and the Rechenzentrum of the University of Regensburg for allocation of computer time.
References [II H.W. Kroto, J.R. Heath, SC. O’Brien, R.F. Carl, R.E. Smalley, Nature 318 (1985) 162.
121 W. Kr&xhmer, L.D. Lamb, K. Fostiropoulos, D.R. Huffman, Nature 347 (1990) 354. r31 MS. Dresselhaus, G. Dresselhaus, P.C. Eklund, Science of Fullerenes and Carbon Nanotubes, Academic Press, San Diego, 1996. [41 S. Leach, in: K. Prassides (Ed.), Physics and Chemistry of the Fullerenes, Kluwer Academic Publishers, Netherlands, 1994, pp. 117. r51 S. Leach, M. Vervloet, A. Despres, E. Breheret, J.P. Hare, T.J. Dennis, H.W. Kroto, R. Taylor, D.R.M. Walton, Chem. Phys. 160 (1992) 451. t61 H. Ajie, M.M. Alvarez, S.J. Am, R.D. Beck, F. Diedrich, K. Fostiropoulos, D.R. Huffman, W. Kmtschmer, Y. Rubin, K.E. S&river, D. Senskarma, R.L. Whetten, J. Phys. Chem. 94 ( 1990) 8630. [71 J.P. Hare, H.W. Kroto, R. Taylor, Chem. Phys. Lett. 177 (1991) 394. fsl J.B. Howard, J.T. McKinnon, Y. Makarousky, A.L. Lafleur, M.E. Johnson, Nature 352 (1991) 139. [91 D. Kim, M. Lee, Y.D. Suh, SK. Kim, J. Am. Chem. Sot. 114 ( 1992) 4429. fro1 Y.-P. Sun, P. Wang, N.B. Hamilton, J. Am. Chem. Sot. 115 (1993) 6378. r111 R. Seshadri, C.N.R. Rao, H. Pal, T. Mukherjee, J.P. Mittal, Chem. Phys. Lett. 205 ( 1993) 395.
[I21 Y.-P. Sun, B. Ma, G.E. Lawson, Chem. Phys. Lett. 233 (1995) 57. [I31 R.M. Williams, J.W. Verhoeven, Chem. Phys. Lett. 194 ( 1992) 446. r141 Y.-P. Sun, C.R. Bunker, J. Phys. Chem. 97 ( 1993) 6770. [I51 A. Watanabe, 0. Ito, M. Watanabe, H. Saito, M. Koishi, J. Phys. Chem. 100 (1996) 10518. [W J.W. Arbogast, C.S. Foote, J. Am. Chem. Sot. 113 (1991) 8886. r171 L. Biczok, H. Linschitz, RI. Walter, Chem. Phys. Lett. 195 (1992) 339. [I81 R.R. Hung, J.J. Grabowski, Chem. Phys. Lett. 192 (1992) 249. Cl91 J.W. Arbogast, A.P. Darmanyan, C.S. Foote, Y. Rubin, F.N. Diederich, M.M. Alvarez, S.J. Anz, R.L. Whetten, J. Phys. Chem. 95 (1991) 11. [201 D.K. Palit, A.V. Sapre, J.P. Mittal, C.N.R. Rao, Chem. Phys. Lett. 195 (1992) 1. r211 R.V. Bensasson, T. Hill, C. Lambert, E.J. Land, S. Leach, T.G. Truscott, Chem. Phys. Lett. 206 (1993) 197. [221 R.V. Bensasson, T. Hill, C. Lambert, E.J. Land, S. Leach, T.G. Truscott, Chem. Phys. Lett. 201 ( 1993) 326. r231 N.M. Dimitrijevic, P.V. Kamat, J. Phys. Chem. 96 ( 1992) 4811. 1241 K. Tanigaki, T.W. Ebbesen, S. Kuroshima, Chem. Phys. Lett. 185 (1991) 189. [=I G. Sauvt, N.M. Dimitrijevic, P.V. Kamat, J. Phys. Chem. 99 (1995) 1199. [261 M. Gevaert, P.V. Kamat, J. Phys. Chem. 96 (1992) 9883. 1271 M. Lee, O.K. Song, J.C. Seo, D. Kim, Y.D. Su, SM. Jin, S.K. Kim, Chem. Phys. Lett. 196 (1992) 325. [=I F. Henari, J. Callaghan, H. Stiel, W. Blau, D.J. Cardin, Chem. Phys. Lett. 199 (1992) 144. 1291 C. Li, L. Zhang, R. Wang, Y. Song, Y. Wang, J. Opt. Sot. Am. B 11 (1994) 1356. r301 M.P. Joshi, S.R. Mishra, H.S. Rawat, S.C. Mehendale, KC. Rustagi, Appl. Phys. Lett. 62 (1993) 1763. [311 B.L. Justus, Z.H. Kafali, A.L. Huston, Opt. Lett. 18 (1993) 1603. r321 A. Cost, L. Tutt, M.B. Klein, T.K. Dougherty, W.E. Elias, Opt. L&t. 18 (1993) 334. [331 R.J. Sension, C.M. Phillips, A.Z. Szarka, W.J. Romanow, A.R. McGhie, J.P. McCauley Jr., A.B. Smith III, R.M. Hochstrasser, J. Phys. Chem. 95 (1991) 6075. [341 D. McBranch, V. Klimov, L. Smilowitz, M. Grigorova, J.M. Robinson, A. Koskelo, B.R. Mattes, H. Wang, F. Wudl, SPIE 2854 (1996) 140. [351 R.C. Issac, C.V. Bwidhu, S.S. Harilal, G.K. Varier,V.P.N. Nampoori, C.P.G. Vallabhan, Mod. Phys. Lett. B 10 (1996) 61. [361 D.G. McClean, D.M. Brandelik, MC. Brant, R.L. Sutherland, L. Frock, in: R. Crane, K. Lewis, E. Van Stryland, M. Khoshnevisan (Eds.), Materials for Optical Limiting, Symposium, Mater. Res. Sot., Pittsburgh, 1995, p. 293. [371 N. Tang, H. Guan, J.P. Partanen, R.W. Hellwarth, SPIE 2143 (1994) 272. [381 J.R. Heflin, W. Wang, C. Figura, R. Yordanov, SPIE 2530 (1995) 176. t391 D. McBranch, L. Smilowitz, V. Klimov, A. Koskelo, J.M. Robinson, B.R. Mattes, J.C. Hummelen, F. Wudl, J.C. Withers, N.F. Borrelli, SPIE 2530 (1995) 196. [401 L.W. Tutt, T.F. Boggess, Prog. Quant. Electron. 17 (1993) 299. [411 Y. Song, X. Bao, F. Li, X. Yang, X. Zhang, R. Wang, C. Li, SPIE 2854 (1996) 230. 1421 S. Reindl, A. Penzkofer, Chem. Phys. 211 (1996) 431. [431 S. Reindl, A. Penzkofer, Chem. Phys. 213 (1996) 429. [Ml A. Penzkofer, A. Beidoun, M. Daiber, J. Luminesc. 51 (1992) 297. 1451 W. Scheidler, A. Penzkofer, Opt. Cornmun. 80 ( 1990) 127. [461 G. Gronninger, A. Penzkofer, Opt. Quant. Electron. 85 (1984) 473. [471 D. von der Linde, 0. Bemecker, A. Laubereau, Opt. Commun. 2 (1970) 215. [481 V.D. Dimitrev, G.G. Gurzadyan, D.N. Niiogosyan, Handbook of Nonlinear Optical Crystals, Springer-Verlag, New York, 1991.
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry II5 (1998) 89-97 [49] Y.R. Shen, The Principles of Nonlinear Optics, Wiley, New York, 1984. [50] A. Penzkofer, W. Baumler, Opt. Quant. Electron. 23 ( 1991) 727. [51] M.R. Wasie1enski.M.P. O’Neil, K.R. Lykke,M.J. Pellin,D.M.Gruen, J. Am. Chem. Sot. 113 (1991) 2774. [52] R.E. Haufler, L. Wang, L.P.F. Chibante, C. Jin, .I. Conceicao, Y. Chai, R.E. Smalley, Chem. Phys. Lett. 179 (1991) 449. [53] S. Reindl, A. Penzkofer, Chem. Phys. 230 (1998) 83. [54] A. Samanta, J. Am. Chem. Sot. 113 (1991) 7427. [55] V.V. Bryukhanov, B.F. Minaev, A.S. Kusenova, S.G. Karstina, J. Appl. Spectrosc. 56 (1992) 146. [56] M.R.V. Sahyan, D.K. Sharma, Chem. Phys. Lett. 189 (1992) 571. [57] H. Fukumura, K. Kikuchi, K. Koike, H. Kokubun, J. Photochem. Photobiol. A: Chem. 42 (1988) 283. [58] W.G. McGimpsey, J.C. Scaiano, J. Am. Chem. Sot. 111 ( 1989) 399.
97
[59] H. Fukumura, K. Kikuchi, H. Kokubun, Chem. Phys. Lett. 92 ( 1992) 29. [601 R.W. Redmond, I.E. Kochevar, M. Krieg, G. Smith, W.G. McGimpsey, J. Phys. Chem. A 101 (1997) 2773. [611 A. Penzkofer, W. Falkenstein, W. Kaiser, Chem. Phys. Lett. 44 (1976) 82. [621 C.V. Shank, E.P. Ippen, 0. Teschke, Chem. Phys. Lett. 45 (1977) 291. [63] F. Graf, A. Penzkofer, Opt. Quant. Electron. 17 (1985) 53. [64] B.R. Henry, W. Siebrand, in: J.B. Birks (Ed.), Organic Molecular Photophysics, Vol. 1, Wiley, New York, 1973, p. 153. [65 I A. Maciejewski, A. Safarzadeh-Amiri, R.E. Verrall, R.P. Steer, Chem. Phys. 87 (1984) 295. 1661 M. Wittmann, A. Penzkofer, Appl. Phys. B 65 (1997) 761.
of
Photokfmistry Photobiology ELSEVIER
Journal of Photochemistry and Photobiology A: Chemistry 115 ( 1998) 89-97
A:Chemistry
Intersystem-crossing and excited-state absorption of CT0 studied by picosecond pump and probe absorption and fluorescence measurements S. Reindl, A. Penzkofer *, H. Gratz Institut II-Experimentelle
und Angewandte Physik, Universitiit Regensburg, D-93040 Regensburg, Germany Received 22 December 1997; accepted 9 March 1998
Abstract The singlet and triplet absorption and emission dynamics of CT0in toluene is studied with a picosecondNd:glasslaser double-pulsepump and probe transmission and fluorescencemeasurementtechnique. The singlet S, excited-stateabsorption ( u,,,r = 6 X 10- l7 cm2), the triplet T, absorption (a, = 4.2 X lo- I7 cm’), and the triplet T, absorption (a,,,, = 4.3 X lo-l7 cm2) at 527 nm are determined.The singlet to triplet intersystem-crossing is studied by double pump pulse induced fluorescencesignal analysis and by double probe.pulse transmissionmeasurements. The yield of triplet formation is found to be near to unity. 0 1998Elsevier ScienceS.A. All rights reserved. Keywords: Intersystem-crossing; Excited-state absorption; C7s
1. Introduction Since the discovery [l] and laboratory synthesis [2] of fullerenes their study is an active field of research [ 31. The optical spectroscopy of Csc and CT0is reviewed in Refs.
[ 3,4]. The ground-state absorption of Cm [4-81 and CT0 [4,6-81 was analysed. Fluorescencestudies of C,, [9-121 and CT0 [9,1 l-151 have been performed. Small quantum yields of fluorescence (#+=2.2X lop4 for Cm [lo], and QjF=5.9X10W4 [14] to 8.5~10~~ [9,16] for C,,) were determined. The yield of triplet formation was found to be closeto 100% (&a: [4,17];C,,,: [4,18-21]).Triplet-triplet absorption spectra for CeOand C,,, are presented in Refs. [4,19,20,22,23] and Refs. [ 4,20,21,23,24], respectively. The excited-state absorption of CsO[ 11,25-331 [ 34-401 and CT0 [ 11,15,24,26-281 has been studied mainly with respect to optical limiting applications (reverse saturable absorption) [ 28-32,34,35,38-41]. In this paperthe singlet and triplet absorptionandemission dynamics of C,, in toluene at room temperatureis investigated. The S,-S, (n > 2) singlet-singlet, the T,-TI (12 2) and the T,-T, (m 2 3) triplet- triplet absorption cross-sections at 527 nm, and the quantum yield of triplet formation
by S,-T, intersystem-crossing (population accumulation in T, state after intersystem-crossing) are determined. * Corresponding regensburg.de
author.
E-mail:
[email protected]
lOlO-6030/98/$19.00 0 1998 Elsevier Science S.A. All rights reserved. PIISlOlO-6030(98)00264-O
Picosecond second-harmonic pulses of a mode-locked Nd:glass laser are used in the experiments.A double pumppulse and double probe-pulsetransmission [ 421 and a double pump pulse induced fluorescence measurementtechnique [ 43,441 are applied. Someabsorption cross-sectionsand the quantumyield of triplet formation are deducedby comparing the experimental results with numerical simulations of the absorption and emission dynamics. The first picosecond pump pulse transmission dependson the S,-S, excited-state absorptiondynamics (excited-stateabsorptioncross-section, u~~,~,and $-state relaxation time, T,,) . The first probe pulse transmissionis causedby the residual &,-ground-stateabsorption and the induced T,-T, triplet-triplet absorption at the time, to, of probe pulse passing. It dependson the yield of triplet formation, &, and the T, triplet absorption crosssection, 0,. The secondpump pulse transmissiondependson the singlet-singlet and triplet-triplet absorption and relaxation dynamics. It allows to extract the higher excited-state T,-T, absorption cross-section, u,,,~ The second probe pulse transmission depends on the population accumulation in the T, triplet state. The second probe pulse transmission together with the first probe pulse transmission allows the
separatedetermination of or and &. The first pump pulse induced fluorescencesignal, SF,iris determined by the pump pulse induced S,-state level population and the fluorescencequantum yield. The secondpump pulse induced fluorescencesignal, SF,2,is reducedcompared to the first pump pulse induced fluorescencesignal in the case
90
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry Il.5 (1998) 89-97
of singlet-triplet intersystem-crossing since then the S,-state level population is reduced. The ratio of S,,*/S,, allows the determination of the quantum yield of triplet formation, &
143,441. The S ,-S, excited-state absorption cross-section, ueex,=, and the T,-T, absorption cross-section are found to be larger than the Soground-state absorption cross-section, oL (reverse saturable absorption [ 401) . The quantum yield of triplet formation, a,, by S,-T, intersystem-crossing is determined to be close to 100%.
2. Experimental The fullerene CT0 was purchased from Aldrich ( [5,6]Fullerene Co) and is used without further purification. The solvent toluene is of analytical grade. The absorption crosssection spectrum, og( h) , of C,O in toluene at room temperature is displayed in Fig. 1 together with the fluorescence quantum distribution EF( h) (own measurements). The double-pulse pump and probe transmission measurements have been carried out with a picosecond Nd:glass laser system [ 4.51. The experimental arrangement is displayed in Fig. 2. The mode-locked laser generates a train of picosecond pulses. The time period of adjacent pulses is tR = 10 ns (resonator roundtrip time). Two adjacent pulses are separated from the pulse train by an electro-optic switch [47] and increased in energy by a Nd:phosphate glass laser amplifier. The pulses are frequency doubled in a CDA crystal [ 481 and the remaining fundamental light is filtered off (filter Fl ) . The wavelength of the second-harmonic pump pulses is hL = 527 nm. The pump pulse duration is AtL = 6 ps. The second harmonic pump pulses are directed to the sample, S (cell thickness Z= 1 mm). They are increased in intensity by lens Ll. The input peak energy density is determined by energy transmission measurement through rhodamine 6G in ethanol (saturable absorption [ 461) with photodetectors PD 1 and PD2. The energy transmission of the pump pulses is measured with the photodetectors PDl and PD3 (small signal transmission To = 0.0756 + 0.001) . The probe pulses are split-off from the pump pulses with a beam splitter and are attenuated with a filter (F2). They pass opposite to the pump pulse direction through the sample. Their temporal delay relative to the pump pulses is tn = 9 ns. The probe pulse diameter is adjusted by lens L2. The transmission of the probe pulses is registered with the detectors the PDl and PD4. The double-pump-pulse fluorescence measurements are carried out with the probe beam path covered. The fluorescence emitted in sideward direction is collected with lens, L3, and directed to the spectrometer, SP, with lens, IA. The fluorescence light in the spectral region from 640 to 840 nm is passed through the spectrometer and registered temporally resolved with a microchannel-plate photomultiplier (Hamamatsu type R1564U) and a fast digital oscilloscope (LeCroy type 9362, time resolution approximately 500 ps) .
400
600
500
700
Wavelength h (nm) Fig. 1. Absorption cross-section spectrum, u=(h) , and fluorescence quantum distribution, /I& A), of C,a dissolved in toluene.
IM.L.LaSert---~----(Amplifier------~---$------$ SHG
Fl
I w C.-.-.-.-.-.-.-.-.-.-~.~.~-.-.-.~.-.-.-.~~~-~.-~ A--L3 I
I
I
G-4 i
i
I
I !
I i
FRh
j
Fig. 2. Experimental setup. M.L. Laser, mode-locked N&glass laser. SW, single pulse selector. Amplifier, N&glass laser amplifier. SHG, frequency doubler. Fl, cut-off filter of fundamental laser wavelength. Ll-L4, lenses. Rb, rhodamine 6G in ethanol for energy density measurement [ 461. PDlPD4, photodetectors. S, sample. SP, spectrometer. MCP, microchannel-plate photomultiplier.
The intensity dependent transmission behaviour of the solvent toluene was measured separately with the experimental arrangement of Fig. 2 (pure toluene in sample cell, S) . The measured energy transmission vs. input pump pulse peak intensity is displayed in Fig. 3. The transmission reduction at high pump pulse intensities is thought to be due to two-photon absorption [ 491. The curves in Fig. 3 are calculated for various two-photon absorption coefficients, (Y@).The best fit is obtained for (Y(‘) = (2f0.3) X lo-” cm/W. The twophoton absorption of the solvent is included in the analysis of the experimental results.
3. Experimental results The pump pulse energy transmissions, TE,L,, TE,L1, the probe pulse energy transmissions, Tp,, Tm, and the pump
S. Reindl et al. /Journal
of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
Input Peak Intensity IoL (W cm-2) Fig. 3. Energy transmission through toluene. Pump laser wavelength, AL = 527 nm. Sample length 1= 1 mm. Circles, experimental points. Curves are calculated for cy’*‘=5XlO-” cm/W (1). 1x10-” cm/W (2), 2 X lo-” cm/W (3), and 5 X lo-“cm/W (4).
pulse induced fluorescence signal heights, S,,, , SF,*,are measured as a function of the first pump pulse input peak intensity, I ,,L,,. Data are selected where the second pump pulse peak intensity, Z0L,2,is approximately equal to the first pump pulse peak intensity. The probe pulse peak intensities are kept a factor of 250 smaller than the pump pulse peak intensities. The probe beam diameters were a factor of two narrower than the pump beam diameters at the sample position. The saturation intensity of S,, ground-state depopulation, Z,, = hzy/ ( rLAtL) (h, Planck constant; %,, second harmonic laser frequency; crL absorption cross-section at c; AtL, duration of second harmonic pulses), is marked in Figs. (4)-( 8) (slow saturable absorption conditions apply since At,- -K rs, , where rs, is the S , -state lifetime [ 501) . The energy transmission, TE,L,, of the first pump pulse (A, = 527 nm, At, = 6 ps) through C,, in toluene (small signal transmission T,=O.O756) is displayed by circles in Fig. 4. The transmission decreases with rising input pump pulse intensity. The behaviour indicates that the excited-state absorption is larger than the ground-state absorption at 527 nm (reverse saturable absorption). The energy transmission, TE,Lz, of the second pump pulse (time delay tR = 10 ns) through the sample is shown by circles in Fig. 5. The decrease in transmission with rising pump pulse intensity is similar to the behaviour of the first pulse. It indicates a larger T,-triplet absorption than So-singlet absorption. The first probe pulse transmission, Tp, (time delay tD= 9 ns) is displayed by circles in Fig. 6. Its dependence is determined by the So ground-state absorption of the molecules, which have remained in the &,-state, and by the Titriplet absorption of the molecules, which have proceeded to
91
InputPeakintensityIoL,,(Wcm-*) Fig. 4. Energy transmission, TEL,. of first pump pulse through CT0in toluene vs. input pump pulse peak intensity, IaL.,. Small signal transmission, T,=O.O756. Circles are experimental data points. Curves are calculated using parameters of Table 1. Solid curves, r., = 60 fs with oeex,L= 4 X 10 - ” cmZ(1),5X10~“cmZ(2),6X10~“cm2(3),7X10~”cm*(4).Dashed curves,o.,,=6X10-“cmZwith~..=300fs(a),200fs(b),100fs(c). 0.08
I-
2
I’ , 3’
I
0.07
.-5 .-$3 j
0.06
z? iii 15
Input PeakIntensity IoL,,(W cme2) Fig. 5. Energy transmission of second pump pulse, TE,rz, vs. input peak intensity of first pump pulse (second pump pulse peak intensity is nearly the same as first pump pulse peak intensity). Circles are experimental data points. All curves are calculated for a,= 4.2 x lo-l7 cm*. Solid curves, (1). 3X lo-“cm (2), 3.5X 10-l’ rT= 2 ps and a,,,r =2.5X lo-“cm’ cm2(3),4X10~‘7cm2(4),4.5X10~‘7cm*(5),and5X10~’7cm2(6). Dashed curves, a.,,= =2.5X 10-l’ cm’ and rT= 10 ps (a), 100 ps (b). Other parameters are taken from Table 1.
the T,-state. The probe pulse transmission reduces with rising first pump pulse peak intensity, ZoL,,. But the decrease in transmission of the first probe pulse is slightly smaller than the decrease of transmission of the first pump pulse indicating
S. Reindl et al, /Journal of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
92
ka 0.05 108
10'0
109
Input Peak Intensity IoL,,(W cm-2)
Input Peak Intensity IoL,, (W cm”)
Fig. 6. Transmission of first probe pulse, TPI, vs. first pump pulse peak intensity, Z,,,. Circles are experimental results. The curves are calculated using the parameters of Table 1. Dashed curves, or=4 X 10-l’ cm’ and &=O (l), 0.25 (2), 0.5 (3), 0,75 (4), and 0.9 (5). Solid curves, &=0.999 and a,=4~10-” cm’ (6). 4.2x10-” cm2 (7), and 4.4X lo-“cm (8).
Fig. 8. Fluorescence signal ratio, .S,,/S,,, Circles are experimental data. To = 0.0756. first and second pump pulse are corrected [ZoL,,/ ( 1 + Z,,, ZZ,,,) ] / [ ZoL,a/( 1 + Z,&Zrat) parametersofTableland&=O(1),0.25 and 0.999 (6).
vs. input peak intensity, Z,,,,. Differences in peak intensity of by S,,s/S,,, = (S,,/S,,),,,.,, ] The curves are calculated for (2),0.5(3),0.75(4),0.9(5),
The fluorescence signal ratio, S,,*/S,,, , of the second pump pulse induced peak fluorescence signal height to the first pump pulse induced peak fluorescence signal height is displayed in Fig. 8 (circles) as a function of the first pump pulse peak intensity, ZoL,,. The ratio decreases with increasing ZoL,, indicating triplet level population by intersystem-crossing.
4. Theoretical description
I.
I
108
109
. ..I
J
IO'O
Input Peak Intensity IoL,, (W cmm2) Fig. 7. Transmission ratio of probe pulses Tpz/TPI. vs. input peak intensity of first pump pulse, Z,,,. Circles are experimental data. Error bars show standard deviation of the mean transmission ratios. The curves are calculated for &=O (l), 0.25 (2), 0.5 (3), 0.75 (4). 0.9 (5), 0.999 (6). Other parameters are taken from Table 1.
that the T,-triplet absorption cross-section, oTT,is smallerthan the S,-state excited-state absorption cross-section, o=+ The ratio of second probe pulse transmission to first probe pulse transmission, TP21TPI, is shown in Fig. 7. The transmission ratio is slightly less than one with a minimum around I OL, I = 7 X lo9 W/cm2. The intensity dependent behaviour of TP, and T,IT,, is used below to extract a, and &.
An extraction of the absorption cross-sections, u=~,~,a,, ueex,=,of the higher-excited-singlet-state relaxation time, T,,, and of the St-T, intersystem-crossing rate, k35, from the experimental transmission and fluorescence data requires a theoretical description of the absorption and emission dynamics and a numerical simulation of the experimental results. The absorption dynamics of the two pump pulses and the two probe pulses is analysed using the energy level system of Fig. 9a and the double pump pulse and double probe pulse scheme of Fig. 9b. All pulses are linearly polarized in the same direction. The pump pulse induced fluorescence dynamics is determined by the dynamics of the S, -level population. 4. I. Level population
and pulse propagation
model
The first pump pulse excites molecules from the So-groundstate (level 1) to an excited singlet state (level 2; groundstate absorption cross-section, oL) . From there the molecules relax quickly (Franck-Condon relaxation time constant, rFc,s) to a thermal equilibrium position in the S,-band (level
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry 11.5 (1998) 89-97
(a) 6 TL 7 Tz
5 T,
93
the phosphorescence lifetime, rp, on a microsecond to millisecond time scale [ 16,18,20,21,51,52]. The second probe pulse passes through the sample after a time delay of tL,,pz = t,, + tD relative to the first pump pulse. Its transmission is determined by the S, (level 1) and T, (level 5) population number densities at the time of passing the sample. Higher excited-state singlet + triplet intersystem-crossing [ 53-561 and triplet + singlet intersystem-crossing [ 42,5760] are not included in the absorption and emission dynamics, since the transmission and fluorescence behaviour can be fitted reasonably well with the presented model and the error bars of the experimental data allow no further model refinement. 4.2. Equation system
___._._. -.-.- .-.-.-. -.-.-.-.-.- .-.-.-,-,-.-.-.-.-. -.-.-.-.-.-.-.-.-.-.-.-.-.u Fig. 9. (a) Energy level scheme for singlet-singlet absorption dynamics, triplet-triplet absorption dynamics, and S,-T, singlet-triplet intersystemcrossing. (b) Scheme of double-pulse pump and double-pulse probe interaction in the sample.
3). From levels 2 and 3 there occurs S,-S, excited-state absorption to band 4 (excited-state absorption cross-section, geex,=).The population in level 4 relaxes quickly to level 3 (relaxation time constant, r,,) . The thermally relaxed S,state population (level 3) relaxes to the ground-state (relaxation rate constant, k3,) and to the triplet system (intersystem-crossing rate constant, ks5; yield of triplet formation, &). The S,-state lifetime, rs, (equal to the fluorescence lifetime, rF) is given by us, = (k3, + kj5) - ‘. The S , -state intersystem-crossing rate is k-,5= &TS, I. The first probe pulse, Pl, passes through the sample after a time delay of tL,,p, = tn which is slightly shorter than the temporal pump pulse separation, tR, but much longer than the S,-state lifetime, rs,. The transmission, Tp,, of the probe pulse, Pl, is given by the singlet ground-state absorption, which is determined by the Z&-ground-state population number density, N, (t’ = tn), and by the T,-triplet absorption, which is determined by the population number density of the lowest triplet state, N5( t’ = tD) (N, (t’ = to) = N,, - N, ( t’ = tD) , where No is the total molecule number density). The second pump pulse, L2, causes the same absorption and emission dynamics in the singlet system as the first pump pulse (but the initial ground-state population, N, ( t’ = tR), is smaller) and additionally causes excitation and relaxation in the triplet system. The molecules in the triplet ground-state, T, (level 5)) are excited to a higher lying triplet state, T,, (level 6, triplet absorption cross-section, Us). From there the molecules relax quickly to the lowest excited triplet state, T2 (Franck-Condon relaxation time constant, rFC,r). From the T,-state (level 7) the molecules relax to the T,-state with a time constant or. The back transfer from T, to So occurs with
The saturable absorption dynamics of the two pump pulses and the two probe pulses is handled by the following differential equation system for the level population number densities, N, to N,, and the laser pulse intensities, IL,, ZL2,Zr,, and Zp2[ 441. Immediate relaxation from level 8 to level 7 is assumed ( r,,,r = 0). +k31N3+ N,
(1)
7P
(2)
(3)
(4) (5) (6) aN7 -=---
Ns
NT
(7)
(8)
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
94
azPi -=8Z’
ULNI a.
1
+~+TN,
(i=1,2)
ZPi
I
r
Ii(t’,l,r)dt’dr
II
(i=Ll,L2,Pl,P2)
TE,i ,” La”
(10)
r I,(t’,O,r)dt’dr 0 --
II
The transformation t' = t - nzlc, and z’ = z is used, where is the time, z is the distance along the propagation direction, co is the light velocity in vacuum, and it is the refractive index. I is the sample length. An absorption anisotropy is not considered since C,,, is nearly spherical symmetric [ 31. The initial conditions of the level populations are N, (t' = - w,z’,r) = No and Ni( t' = - m,z’,r) = 0 for i= 2,. ..,7. No is the total number density of dissolved fullerene molecules. The input pulse intensities are t
Z,, (t’,z’
=O,r)=I0L,i
ZL2(t’,z’ =O,r)=I
exp
-(3’-(3’1
(11)
oL,zexp[ -(yr-(3]
Zp, (t’,z’ =O,r)=~0P,,exp
ZP2(t’,z’ =O,r)=Z
[
(12)
[ py-y]
op,zexp[ -( “-~--fa)2-(
(13)
$J
(14)
r is the radial beam coordinate. r, and r, are the 1le beam radii of the pump and the probe pulses, respectively. to = 2 - ’ [In ( 2) ] - ’ ‘*At= is half the 1/e temporal pulse width ( AtL is the FWHM pulse duration).
5. Data analysis The equation system ( 1) -( 10) is solved numerically using relevant parameters of CT0 in toluene (see Table 1). Unknown parameters are determined by fitting the calculated energy transmissions of the pump and probe beams and the pump pulse induced fluorescence signals to the experimental values. In Fig. 4 the first pump pulse energy transmission, TE,Llr is simulated. TE,Ll depends on the S, -excited-state absorption and on the S,-state relaxation time concross-section, cr.=ex,L, stant, r,,. The solid curves are calculated for various u~~,~values ( rex = 60 fs), while the dashed curves are calculated for different &-state relaxation times, T,, (a,,,= = 6 X lo- I7 cm’). The best fit to the experimental data is obtained for u ex,L= 6 X lo- I7 cm2 and r,, = 240 fs. The double pump-pulse induced fluorescence signal behaviour is simulated in Fig. 8. The fluorescence signal ratio, S,,z/S,,, depends on the yield of triplet formation, &. S,,z/
S,,, is practically independent of the explicit triplet-triplet absorption dynamics (verified by numerical simulation, but not shown in Fig. 8). The curves are calculated for various quantum yields of triplet formation, &. A comparison of the calculated curves with the experimental data gives & = 0.9 + 0.1. The experimental accuracy and subsequently the accuracy of & is limited to 10% because the fluorescence quantum yield, (PF, of C70 in toluene is very low (& = 8.5 X 10e4, see Table 1). The first probe pulse transmission, Tp,, is simulated in Fig. 6. Tp, depends on the quantum yield of triplet formation, &, and on the cross-section, +, of T, -+ T, triplet-triplet absorption. Knowing &, the fitting of calculated T,, curves to experimental data allows the determination of a,. The dashed curves in Fig. 6 belong to various & values at a fixed value of g= = 4 X lo-l7 cm*. The solid curves are calculated for various or values while the triplet quantum yield is fixed to & = 0.999. The best fit to the experimental data is obtained for +=4.2X lo-l7 cm*. The probe pulse transmission ratio, Tp2 / Tp, is simulated in Fig. 7. The ratio depends on the yield of triplet formation, &, and the absorption cross-section ratio, arluL. If +T/ uL = 1 then the transmission ratio is Tp2/Tp, = 1 independent of &. The combined dependence of Tp, (Fig. 6) and Tp2/ allows the determination of both & and or. The TPI on IOU best fit is obtained for I#+= 1 - &F= 0.999 and CT= 4.2 X lo-l7 cm*. A high accuracy of or and & determination would be obtained, if cr, were significantly different from crL. For C70 in toluene the singlet ground-state absorption crosssection, crL= 3.5 X lo- l7 cm’, and the T, triplet-state absorption cross-section, +=4.2X lo-l7 cm*, are only slightly different. Therefore the accuracy of & determination by Tpl and T,/T,, measurement is moderate. The second pump pulse transmission behaviour is simulated in Fig. 5. TE,L2 depends on the T,-TI triplet absorption, crT, on the T,-T, triplet excited-state absorption, akx,r, and on the triplet-triplet relaxation time, rT. Since a, has been already determined by probe pulse transmission measurements, the curves in Fig. 5 are calculated for various o,,,T values ( rT= 2 ps, solid curves) and rT values ( flex,T= 2.5 X lo- l7 cm*, dashed curves) to study their dependences. For ueex,T similar to UT, the second pump pulse transmission, T E,L2,becomes practically independent of rr. Our numerical fit gives u~~,~= (4.3 k0.2) X lo-l7 cm*, while ur has been determined to be or = (4.2 + 0.1) X lo- I7 cm*. Therefore, determination of rT is not possible. In organic dyes typical values of 7T are of the order of 0.5 to 3 ps [ 42,441.
6. Discussion The determined parameters, 9, u~~,~, r,,, UT, %,TI and #r, of C70 in toluene at room temperature for AL = 527 nm are listed in Table 1 together with parameters reported in the literature.
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
95
Table 1 Spectroscopic parameters of C,, at room temperature Parameter
Value
Ts, (PS)
110 625 627 650 660 670 700 0.710 0.730 0.870 4-5 41 130 140 53000 0.00059 0.0007 0.00085 0.76 f 0.15 0.9 0.99 f 0.02 1*0.15 1 (10.85) 0.5 0.5 240 f 40 2 3.2~ 10-l’ 3.5 x lo-r7 1.25 x lo9 1.54 x lo9
[20],inbenzeneandhexane This work Assumed [ 6 11, used here Assumed, used here This work Assumed [42&l] [21],inbenzene Own measurement, used here [ 151, in benzene kfi3 = &Q-S;I, used here
3.6~ lo-r7 3.8~ lo-l7 3.9x 10-l’ (4.2+0.1)X10-” 4.75 x lo- ” (6.0f0.5) X10-l’ (4.3f0.2) X10-r’
[ 231, in benzene [21],inbenzene [ 201, in hexane This work [ 201, in benzene This work This work
TP(CLS)
%c,s (ps) %c,T (ps) 7,x (fs) TT (ps) flL (cm’)
hs (s-l) uT (cm’)
ueer,L(cm’) o;,.~ (cm’)
Comments
1511 [ 91, undegassed [ 131, argon degassed benzene solution Used here [ 91, degassed ]241 1271 [20], air saturated benzene solution [ 161, air saturated [ 181, air saturated benzene solution [ 181, argon-saturated [ 521, clusters in supersonic beam [ 161, degassed [21] in benzene [ 5 11, at 77 K in toluene/ 10% poly-( cY-methylstyrene) Cl41 [ 91, in benzene [9,16] and own measurement, used here [21], in benzene [I91
1181
Solvent toluene except stated otherwise.
The ratio of singlet excited-state absorption to singlet ground-state absorption is a,,,,/~, = 1.7. The S, state lifetime is rs, = 650 ps. For picosecond pulses (pulse duration, AtL, less than S,-state lifetime, rs,), CT0 in toluene behaves as a reverse saturable absorber since u~~,~> uL. On a nanosecond time scale (pulse duration or time of observation longer than rs,) the nonlinear transmission behaviour is determined by cr=T/g=. For C,. in toluene this absorption cross-section ratio is uJcrr= 1.2 and the reverse saturable absorption effect is rather weak. Our double-pulse fluorescence measurements and probe pulse transmission measurements agree well with a quantum yield of triplet formation of&=l-& (&,=8.5X lo-“). Triplet quantum yields close to 100% have been reported by several groups ( see Table 1) . The S, -state relaxation of CT0
(as well as of Cso [ 4,171) is dominated by intersystemcrossing. The rate of internal conversion is small compared to the rate of intersystem-crossing (because & close to 1) . The radiative relaxation time constant is long compared to the time constant of intersystem-crossing, 7isc( rise= k35- ’ = dvs, =65Ops; r,,d=rs,/&=760ns). A higher-excited-state relaxation time of r,, = (240 f 40) fs has been determined for CT,-,in toluene. In organic dyes the higher-excited-state S, + S, singlet relaxation is generally fast in the 50 to 100 fs time range [ 62,631. Only in the case of large energy gaps between S2 and S, longer relaxation times are expected [ 641 and observed [ 65,661. The symmetry of the CT0 molecule and its weak So-S, absorption cross-section may be responsible for the moderate S, + S, relaxation time constant, r,,, of CTO.
96
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry 115 (1998) 89-97
7. Conclusions The singlet and triplet absorption dynamics of CT,, in toluene at room temperature has been studied with picosecond pulses at 527 nm. A double pump pulse induced fluorescence signal technique and a double pump pulse induced probe pulse transmission technique have been employed to determine the yield of triplet formation. Despite the fact of low fluorescence quantum yield and only slight difference between singlet ground-state and lowest-triplet absorption cross-sections, the quantum yield of triplet formation could be determined by both methods (fluorescence technique and absorption technique). The S r--S, excited-state absorptioncross-section, the $-state relaxation time, as well as the T,-T, and T,-T, absorption cross-sections have been determined. The presented methods of double pump pulse induced fluorescence detection and double pump pulse induced probe pulse transmission measurement may be generally applied to the determination of quantum yields of triplet formation. The presented pump and probe pulse transmission analysis generally allows the extraction of excited-state absorption crosssection and excited-state relaxation parameters.
Acknowledgements The authors thank the Deutsche Forschungsgemeinschaft for financial support and the Rechenzentrum of the University of Regensburg for allocation of computer time.
References [II H.W. Kroto, J.R. Heath, SC. O’Brien, R.F. Carl, R.E. Smalley, Nature 318 (1985) 162.
121 W. Kr&xhmer, L.D. Lamb, K. Fostiropoulos, D.R. Huffman, Nature 347 (1990) 354. r31 MS. Dresselhaus, G. Dresselhaus, P.C. Eklund, Science of Fullerenes and Carbon Nanotubes, Academic Press, San Diego, 1996. [41 S. Leach, in: K. Prassides (Ed.), Physics and Chemistry of the Fullerenes, Kluwer Academic Publishers, Netherlands, 1994, pp. 117. r51 S. Leach, M. Vervloet, A. Despres, E. Breheret, J.P. Hare, T.J. Dennis, H.W. Kroto, R. Taylor, D.R.M. Walton, Chem. Phys. 160 (1992) 451. t61 H. Ajie, M.M. Alvarez, S.J. Am, R.D. Beck, F. Diedrich, K. Fostiropoulos, D.R. Huffman, W. Kmtschmer, Y. Rubin, K.E. S&river, D. Senskarma, R.L. Whetten, J. Phys. Chem. 94 ( 1990) 8630. [71 J.P. Hare, H.W. Kroto, R. Taylor, Chem. Phys. Lett. 177 (1991) 394. fsl J.B. Howard, J.T. McKinnon, Y. Makarousky, A.L. Lafleur, M.E. Johnson, Nature 352 (1991) 139. [91 D. Kim, M. Lee, Y.D. Suh, SK. Kim, J. Am. Chem. Sot. 114 ( 1992) 4429. fro1 Y.-P. Sun, P. Wang, N.B. Hamilton, J. Am. Chem. Sot. 115 (1993) 6378. r111 R. Seshadri, C.N.R. Rao, H. Pal, T. Mukherjee, J.P. Mittal, Chem. Phys. Lett. 205 ( 1993) 395.
[I21 Y.-P. Sun, B. Ma, G.E. Lawson, Chem. Phys. Lett. 233 (1995) 57. [I31 R.M. Williams, J.W. Verhoeven, Chem. Phys. Lett. 194 ( 1992) 446. r141 Y.-P. Sun, C.R. Bunker, J. Phys. Chem. 97 ( 1993) 6770. [I51 A. Watanabe, 0. Ito, M. Watanabe, H. Saito, M. Koishi, J. Phys. Chem. 100 (1996) 10518. [W J.W. Arbogast, C.S. Foote, J. Am. Chem. Sot. 113 (1991) 8886. r171 L. Biczok, H. Linschitz, RI. Walter, Chem. Phys. Lett. 195 (1992) 339. [I81 R.R. Hung, J.J. Grabowski, Chem. Phys. Lett. 192 (1992) 249. Cl91 J.W. Arbogast, A.P. Darmanyan, C.S. Foote, Y. Rubin, F.N. Diederich, M.M. Alvarez, S.J. Anz, R.L. Whetten, J. Phys. Chem. 95 (1991) 11. [201 D.K. Palit, A.V. Sapre, J.P. Mittal, C.N.R. Rao, Chem. Phys. Lett. 195 (1992) 1. r211 R.V. Bensasson, T. Hill, C. Lambert, E.J. Land, S. Leach, T.G. Truscott, Chem. Phys. Lett. 206 (1993) 197. [221 R.V. Bensasson, T. Hill, C. Lambert, E.J. Land, S. Leach, T.G. Truscott, Chem. Phys. Lett. 201 ( 1993) 326. r231 N.M. Dimitrijevic, P.V. Kamat, J. Phys. Chem. 96 ( 1992) 4811. 1241 K. Tanigaki, T.W. Ebbesen, S. Kuroshima, Chem. Phys. Lett. 185 (1991) 189. [=I G. Sauvt, N.M. Dimitrijevic, P.V. Kamat, J. Phys. Chem. 99 (1995) 1199. [261 M. Gevaert, P.V. Kamat, J. Phys. Chem. 96 (1992) 9883. 1271 M. Lee, O.K. Song, J.C. Seo, D. Kim, Y.D. Su, SM. Jin, S.K. Kim, Chem. Phys. Lett. 196 (1992) 325. [=I F. Henari, J. Callaghan, H. Stiel, W. Blau, D.J. Cardin, Chem. Phys. Lett. 199 (1992) 144. 1291 C. Li, L. Zhang, R. Wang, Y. Song, Y. Wang, J. Opt. Sot. Am. B 11 (1994) 1356. r301 M.P. Joshi, S.R. Mishra, H.S. Rawat, S.C. Mehendale, KC. Rustagi, Appl. Phys. Lett. 62 (1993) 1763. [311 B.L. Justus, Z.H. Kafali, A.L. Huston, Opt. Lett. 18 (1993) 1603. r321 A. Cost, L. Tutt, M.B. Klein, T.K. Dougherty, W.E. Elias, Opt. L&t. 18 (1993) 334. [331 R.J. Sension, C.M. Phillips, A.Z. Szarka, W.J. Romanow, A.R. McGhie, J.P. McCauley Jr., A.B. Smith III, R.M. Hochstrasser, J. Phys. Chem. 95 (1991) 6075. [341 D. McBranch, V. Klimov, L. Smilowitz, M. Grigorova, J.M. Robinson, A. Koskelo, B.R. Mattes, H. Wang, F. Wudl, SPIE 2854 (1996) 140. [351 R.C. Issac, C.V. Bwidhu, S.S. Harilal, G.K. Varier,V.P.N. Nampoori, C.P.G. Vallabhan, Mod. Phys. Lett. B 10 (1996) 61. [361 D.G. McClean, D.M. Brandelik, MC. Brant, R.L. Sutherland, L. Frock, in: R. Crane, K. Lewis, E. Van Stryland, M. Khoshnevisan (Eds.), Materials for Optical Limiting, Symposium, Mater. Res. Sot., Pittsburgh, 1995, p. 293. [371 N. Tang, H. Guan, J.P. Partanen, R.W. Hellwarth, SPIE 2143 (1994) 272. [381 J.R. Heflin, W. Wang, C. Figura, R. Yordanov, SPIE 2530 (1995) 176. t391 D. McBranch, L. Smilowitz, V. Klimov, A. Koskelo, J.M. Robinson, B.R. Mattes, J.C. Hummelen, F. Wudl, J.C. Withers, N.F. Borrelli, SPIE 2530 (1995) 196. [401 L.W. Tutt, T.F. Boggess, Prog. Quant. Electron. 17 (1993) 299. [411 Y. Song, X. Bao, F. Li, X. Yang, X. Zhang, R. Wang, C. Li, SPIE 2854 (1996) 230. 1421 S. Reindl, A. Penzkofer, Chem. Phys. 211 (1996) 431. [431 S. Reindl, A. Penzkofer, Chem. Phys. 213 (1996) 429. [Ml A. Penzkofer, A. Beidoun, M. Daiber, J. Luminesc. 51 (1992) 297. 1451 W. Scheidler, A. Penzkofer, Opt. Cornmun. 80 ( 1990) 127. [461 G. Gronninger, A. Penzkofer, Opt. Quant. Electron. 85 (1984) 473. [471 D. von der Linde, 0. Bemecker, A. Laubereau, Opt. Commun. 2 (1970) 215. [481 V.D. Dimitrev, G.G. Gurzadyan, D.N. Niiogosyan, Handbook of Nonlinear Optical Crystals, Springer-Verlag, New York, 1991.
S. Reindl et al. /Journal of Photochemistry and Photobiology A: Chemistry II5 (1998) 89-97 [49] Y.R. Shen, The Principles of Nonlinear Optics, Wiley, New York, 1984. [50] A. Penzkofer, W. Baumler, Opt. Quant. Electron. 23 ( 1991) 727. [51] M.R. Wasie1enski.M.P. O’Neil, K.R. Lykke,M.J. Pellin,D.M.Gruen, J. Am. Chem. Sot. 113 (1991) 2774. [52] R.E. Haufler, L. Wang, L.P.F. Chibante, C. Jin, .I. Conceicao, Y. Chai, R.E. Smalley, Chem. Phys. Lett. 179 (1991) 449. [53] S. Reindl, A. Penzkofer, Chem. Phys. 230 (1998) 83. [54] A. Samanta, J. Am. Chem. Sot. 113 (1991) 7427. [55] V.V. Bryukhanov, B.F. Minaev, A.S. Kusenova, S.G. Karstina, J. Appl. Spectrosc. 56 (1992) 146. [56] M.R.V. Sahyan, D.K. Sharma, Chem. Phys. Lett. 189 (1992) 571. [57] H. Fukumura, K. Kikuchi, K. Koike, H. Kokubun, J. Photochem. Photobiol. A: Chem. 42 (1988) 283. [58] W.G. McGimpsey, J.C. Scaiano, J. Am. Chem. Sot. 111 ( 1989) 399.
97
[59] H. Fukumura, K. Kikuchi, H. Kokubun, Chem. Phys. Lett. 92 ( 1992) 29. [601 R.W. Redmond, I.E. Kochevar, M. Krieg, G. Smith, W.G. McGimpsey, J. Phys. Chem. A 101 (1997) 2773. [611 A. Penzkofer, W. Falkenstein, W. Kaiser, Chem. Phys. Lett. 44 (1976) 82. [621 C.V. Shank, E.P. Ippen, 0. Teschke, Chem. Phys. Lett. 45 (1977) 291. [63] F. Graf, A. Penzkofer, Opt. Quant. Electron. 17 (1985) 53. [64] B.R. Henry, W. Siebrand, in: J.B. Birks (Ed.), Organic Molecular Photophysics, Vol. 1, Wiley, New York, 1973, p. 153. [65 I A. Maciejewski, A. Safarzadeh-Amiri, R.E. Verrall, R.P. Steer, Chem. Phys. 87 (1984) 295. 1661 M. Wittmann, A. Penzkofer, Appl. Phys. B 65 (1997) 761.