Saturable absorption dynamics of a cyanovinyl-diethylaniline dye

(1)

Saturable absorption dynamics of a cyanovinyl-diethylaniline dye

A . P e n z k o f e r , A . B e i d o u n

Naturwissenschaftliche Fakultdt II - Physik, Universitdt Regensburg, D-8400 Regensburg, Federal Republic of Germany

a n d

G . W a g e n b l a s t

BASFAktiengesellschaft, D-6700 Ludwigshafen, Federal Republic of Germany

Received 16 July 1990; in final form 11 September 1990

T h e saturable absorption dynamics o f 4-tricyanovinyl-N,N-diethylaniline ( T C V A ) i n methanol is investigated with intense picosecond light pulses at 527 nm. A n excited-state absorption cross-section o f <rex= (2.5 ± 0 . 3 ) X 1 0 "1 7 c m2 and an absorption recovery time o f Ta= 3 ± 1 ps are determined. T h e fast deactivation of the Si state is due to internal conversion.

1. Introduction

Dicyanovinyl-dialkylaniline dyes [1-3] ( = d i - alkylamino-dicyanostyrenes=dialkylamino-benzyli- dene malonitriles) a n d tricyanovinyl-dialkylaniline dyes [ 1 ] (=dialkylamino-tricyanostyrenes) are applied for coloring synthetic polymer fibers [ 1 ]. They are used as sublimable dyes i n heat-transfer recording materials [4,5] a n d photoconductive recording materials [6]. They act as cytotoxic agents against tumors [7], as X - r a y protective agents [8], a n d as photostabilizers i n plastics against ultraviolet radia- tion [2].

The absorption a n d emission spectroscopic prop- erties o f 4-dicyanovinyl-N,N-diethylaniline ( D C V A )

[9-16] a n d 4-tricyanovinyl-N,N-diethylaniline ( T C V A ) [15,16] have been investigated recently.

The fluorescence quantum efficiencies q i n the vapor phase [15] a n d i n l i q u i d solutions [9-12,16] were found to be very small (q< 10 ~³) . Temperature-dependent fluorescence-quantum-efficiency measurements indicated a n intrinsic barrierless deactivation

[16]. Since the viscous drag o f the dye surroundings determines the fluorescence lifetime of these dyes [ 9 -

13,17-21 ], they are applied as micro-viscosity probes and free-volume probes i n micellar emulsions [9]

and polymers [10-13,17-21]. T h e analysis o f the

solvent-dependent spectral absorption a n d emission shifts revealed a partial intramolecular charge transfer i n the excited Si state [9,10,13,16,18,22,23].

The low fluorescence quantum efficiencies may be due to fast internal conversion caused by the rotational movement o f flexible molecular groups

[10,11,13,16,17,24]. I n this case, the fluorescence signal is emitted from the v i b r a t i o n a l ^ relaxed excited S! state w h i c h has a radiative lifetime r ^^d, a n d the fluorescence lifetime, r l¹ = qi*U> is equal to the absorption recovery time TA.

L o w fluorescence quantum efficiencies could also be the result o f fast intersystem crossing to triplet states. I n this case, the absorption recovery time r^A w o u l d become equal to the phosphorescence time Tp¹ = qrjj^d. But the investigated dyes exhibit very lit- tle triplet yield [ 13 ], so fast intersystem crossing may be excluded.

In electron-donor-acceptor systems like the disub- stituted benzenes D C V A a n d T C V A [10,13,24], an excited orthogonal zwitterionic singlet state might be formed [25-29] (twisted intramolecular charge- transfer state, T I C T state [30-34]). T h e T I C T state w o u l d have a long radiative lifetime T™f^T because its radiative transition to the electronic ground state is electric-dipole forbidden [30]. Therefore the fluorescence emission from T I C T states w o u l d result i n

(2)

low fluorescence quantum efficiencies even at slow nonradiative decay rates. The ground-state absorption recovery time T^a w o u l d become equal to the T I C T fluorescence lifetime T FICT = ^ T ™T.

Saturable absorption studies, w h i c h measure the absorption recovery time, allow one to distinguish between internal conversion (short absorption recovery t i m e ) and T I C T state formation (slow absorption recovery t i m e ) .

In this paper the saturable absorption of T C V A dissolved i n methanol is investigated at r o o m temperature. The dye solutions are excited with second- harmonic picosecond light pulses of a mode-locked N d : glass laser. The excited-state absorption cross- section cr^{e x} at the second-harmonic laser frequency ( A^P = 527 n m ) and the absorption recovery time r^A are determined by measuring the intensity depen- dence of transmission.

pulses (wavelength A^P= 527 n m , duration At^P&4 ps) used for the saturable absorption studies are gener- ated i n a K D P crystal. F o r the determination of the intensity-dependent pump-pulse transmission through the saturable absorber, only the p u m p pulse along path 1 is passed through the sample S and the probe pulse (path 2) is covered. The input p u m p - pulse peak intensity is determined by transmission measurements through a two-photon absorbing r u - tile crystal (photodetectors P D 2 and P D 1 ) [36]. The relative error of the peak-intensity detection is estimated to be ± 20%. The pump-pulse transmission through the T C V A sample is measured w i t h the photodetectors P D 3 and P D 1 . The temporal behaviour o f the saturable absorber at a fixed pump-pulse peak intensity is studied by measuring the transmission of a weak probe pulse (path 2) of variable delay (photodetectors P D 4 and P D 1 ) . The experiments are carried out at r o o m temperature.

2. Experimental

The experimental arrangement for pump-pulse bleaching and time-delayed probe-pulse transmission measurements is displayed i n fig. 1. A passively mode-locked N d : p h o s p h a t e glass laser [35] gener- ates trains of picosecond light pulses (wavelength / lL= 1054 n m , duration A JL« 5 ps). Single pulses are selected w i t h an electro-optic shutter and are am- plified i n a Nd:phosphate glass amplifier (single- pulse energy up to 5 m J ) . The second-harmonic light

M.L LASER 1 SWITCH AMPLIFIER

3. Results

Intensity-dependent transmission measurements allow the determination of excited-state absorption cross-sections <7ÊX at high pump-pulse intensities, and the determination of absorption recovery times Tâ i f they are shorter than the pump-pulse duration ( TÂ< A /^p) by comparing experimental data w i t h n u - merical simulations [37,38]. Time-delayed probe- pulse transmission measurements allow the mea- surement of absorption recovery times i f they are longer than the pump-pulse duration ( Tâ > A*^P) [ 39 ].

- 4 —

— ^ —

^s

i

F 2- T PD2 TFA

SHG

cb

PD1

I [ = W

_ H -

^ F1 ^

—<

PD4 pathl

Fig. 1. Experimental setup. P D 1 - P D 4 , photodetectors, L, lens.

F l , F2, filters. D L , optical delay line. S H G , K D P crystal for sec- ond-harmonic generation. T P A , rutile crystal for peak-intensity detection. S, sample.

3.1. Pump-pulse transmission

The pump-pulse energy transmission T^E through T C V A i n methanol versus input peak intensity 7⁰p is shown by the circles i n fig. 2 (small signal trans- mission T⁰=0.021, sample length /= 1 m m ) .

The saturable absorption dynamics is simulated by transitions i n a four-level energy diagram. The level system is displayed by the inset i n fig. 2. The system of differential equations describing the level populations N^t (/= 1 to 4) and the laser propagation is given by [38]

(3)

INPUT PEAK INTENSITY I0P [W/cm2]

Fig. 2. Pump-pulse energy transmission versus input-peak inten- sity. Circles represent measured data. The solid curves are cal- culated for < 7P= 1 . 6 x l O -1 6 c m² [16], ( 7e x= 2 . 5 x 1 0 ~1 7 c m², r^{F C}= 0 . 5 p s (assumed), te x= 10~^{1 3} s (assumed), r^{o r}= 150 ps (as- sumed, no relevance here), A f^P= 4 ps, and T0=0.021. The ab- sorption recovery times are (1) TA= O O , (2) Ta= 8 ps, (3) Ta= 4 ps, (4) Ta= 2 ps, (5) Ta= 1 ps, and (6) Ta= 0 . 5 ps. The dashed curve belong to xA= 2 ps and < 7e x= 2 x l 0 ~1 7 c m2 (a) and

<TE X=3 X 10~1 7 c m2 (b). The other parameters are the same as for the solid curves. The energy-level diagram applied to the calcu- lations is inserted in the figure.

9#i(0)

to' hv^P

N2(9)+N3(6)

+

3 c rPc o s20 [ A T , ( 0 ) - J V2( 0 ) ]

(1)

toVi(e)

hv^P \_

- < X „ (N2(6)-

3CTPCOS20 [ A ^1( 0 ) - Ar 2( 0 ) ]

N2(6)

N²(6)+N³(d) N⁴{6)

)]

?A TFC/ Tor

(2)

6W3(0)

to'

N3(6)

# 4 ( 0 ) )

+

N²(e)+N³(9)

N2(0) N3(9) ^ N^d) N3{6)-N3

+

^: ⁽³⁾

9JV4(0)

to'

= ^ o^eAN²(6) + i V³( 0 ) - J V4( 0 ) ] -N4(6)

top 6z

n/2

7 = - /P | { 3 < 7Pc o s20 [ i V1( 0 ) - A r2( 6 / ) ]

+ ffex [ N2( 0 ) + N3(0) ]} sin 0 d 0 .

(4)

(5) T h e transformation t'=t—c0z/n a n d z ' = z is used

(c0 is the v a c u u m light velocity, n is the refractive i n d e x ) . The initial conditions are

Nx{0,t'=-oo, z', r)=N0,

N2(0, t'= - o o , z', r) =JV3(0, f' = - o o , z', r)

=N4(0,t'=-oo, z', r ) = 0 and

/^P( / ' , z' = 0, r) =/0 P exp[ - ( f ' / / p )²- ( r / rP)2] . f^P =Ajp/2 (In 2 )1 / 2 is half the 1 /e w i d t h o f the p u m p - pulse duration and rP is the 1 /e w i d t h o f the laser- beam radius. N0 is the total number density o f dye molecules. The orientation-averaged number densi- ties o f the level populations are

71/2

Ni = J Ni(6) sin Odd ( / = 1 , 2 , 3 ) .

The absorption anisotropy o f the electric-dipole i n -

(4)

teraction is taken into account for the ground-state absorption (Gp(6) = 3<J^PCOS²0). 6 is the angle be- tween the transition dipole moment for ground-state absorption o f the molecules and the direction o f the electric field o f the p u m p laser (linearly polarized).

r^{o r} describes the molecular reorientation o f the transition dipole moments. T h e energy transmission o f the p u m p laser is given b y

T SfrdrJ^IAt', z' = l,r) dt'

where / is the sample length.

The curves i n fig. 2 are calculated by solving n u - merically eqs. ( 1 ) - (6). T h e dye parameters used are listed i n the figure caption. The best numerical fit to the experimental points corresponds t o a n excited- state absorption cross-section o f cr^{e x}= ( 2 . 5 ± 0 . 3 ) X 1 0 "1 7 c m2 (<7ex/(7p = 0.16£0.02) a n d a ground-state absorption recovery time o f T^a= 3 ± 1 ps.

PROBE PULSE DELAY TIME tD [ps]

- 8 - 6 - 4 - 2 0 2 l* 6 T—I — I — I — I — I — I — I — I — | — I — I — I — I — I—T

C N

0 I I I I I I^{I I I I} I^{I 1 I I} 1 1

0 2 U 6 8 10 12 14 16

TIME t [psl

Fig. 3. Probe-pulse energy transmission versus time. Pump-pulse input peak intensity is /^{O P}± 2 x l 0⁹ W / c m². T h e small signal transmission is T0=0.021. The structural formula o f T C V A is inserted in the figure.

23 November 1990

3.2. Probe-pulse transmission

The temporal position o f the probe pulse relative to the p u m p pulse is varied. The probe-pulse transmission versus time is displayed i n fig. 3. O n l y laser shots w i t h measured pump-pulse peak intensities between 1.5 X 1 09 and 2.5 X 1 09 W / c m2 were remained for the plot. T h e probe pulses were attenuated b y a factor o f 50 compared to the p u m p pulse.

The probe-pulse transmission curve versus delay time is symmetric relative to the transmission max- i m u m where p u m p pulse and probe pulse pass s i - multaneously through the sample (delay time £D= 0 ) . T h i s symmetric shape indicates that the absorption recovery time Ta is shorter than the laser-pulse d u - ration At^P i n agreement w i t h the computer simulations o f the pump-pulse transmission (section 3.1).

4. Discussion

In ref. [16], the fluorescence quantum efficiency of T C V A i n methanol at room temperature was found t o b e # « 7 x l 0 ~⁴ and a radiative lifetime o f the v i - brationally relaxated S! state was estimated to be Trad ^ 6 ns. T h e assumption o f S i - S⁰ internal con- version gives a fluorescence lifetime o f zp = qrfid - 4 ps. T h i s value is i n good agreement w i t h the ground- state absorption recovery time o f r^A= 3 ± 1 ps deter- m i n e d i n this paper. The excited T C V A molecules i n methanol relax to the ground state b y fast internal conversion at r o o m temperature. Rotational m o - tions o f the cyanovinyl group and the diethylamino group seem to be responsible for the fast excited-state deactivation [ 10,11,13,16 ]. The fast relaxation seems to hinder an efficient T I C T state formation.

5. Conclusions

The combined measurements o f fluorescence q u a n t u m efficiencies and absorption recovery times allow the distinction between different relaxation channels o f excited molecules. T h e same informa- t i o n could bej^btained b y the combined measurements o f fluorescence quantum efficiencies and time- resolved fluorescence decays (e.g. streak-camera measurements) [ 4 0 ] . T h e bleaching behaviour o f

(5)

T C V A i n methanol at 527 n m allows the application of the dye as a fast saturable absorber for lasers i n this wavelength region (e.g. second-harmonic pulses of N d lasers).

Acknowledgement

The authors thank the Deutsche Forschungsge- meinschaft for financial support a n d the Rechenzen- trum o f the U n i v e r s i t y for the allocation o f computer time.

References

[ 1 ] A . T . Peters and M.S. Wild, J. Soc. Dyers Colour 93 (1977) 126, 133.

[2] S.S.M. Hassan, J . M . Abdella and N . E . Nashed, Mikrochim.

Acta II (1984) 27.

[3]S.S. Beaven, J.S. Hargreave and D . Phillips, Advan.

Photochem. 11 (1979) 207.

[4] N . Taguchi, A. Imai, T . Niwa and Y . Murata, European Part.

Appl. EP163 145 A 2 , 4 December 1985.

[ 5 ] Mitsubishi Chemical Industries C o . Ltd., Japan. K o k a i T o k k y o K o h o JP89/7889A2 [84/78895], 7 August 1984.

[ 6 ] H . E . Wright and M . A . Berwick, Res. Disci. 158 (1977) 62.

[7] G . R . N . Jones, Nature 235 (1972) 257.

[8] F. Hernadi, Z . N . Nagy, P. Kovacz, Z. Dezsi and T . V . Nagy, Acta Biol. (Budapest) 19 (1968) 257.

[9 ] N . E . Tonks Jr., Dissertation, University of Georgia, Athens (1986) [ U n i . Microfilms Int., Order No. DA8613520].

[10] R . O . Loutfy and K . Y . Law, J. Phys. Chem. 84 (1980) 62.

[11] M.S.A. Abdel-Mottaleb, R . O . Loutfy and R. Lapouyade, J . Photochem. Photobiol. A48 (1989) 87.

[12] R.O. Loutfy, Macromolecules 16 (1983) 678.

[13] R . O . Loutfy, in: Photophysical and photochemical tools in polymer science, ed. M . A . Winnik (Reidel, Dordrecht,

1986) p. 429.

[14] J. Aihera, K. Araya and Y . Matsunaga, Bull. Chem. Soc.

Japan 54 (1981) 615.

[15] A. V. Deshpande, A . Beidoun, A . Penzkofer and G . Wagenblast, Chem. Phys. 142 (1990) 123.

[16] A . V . Deshpande, A . Beidoun, A . Penzkofer and G . Wagenblast, Chem. Phys., to be published.

[ 17] R . O . Loutfy, Pure Appl. Chem. 58 (1986) 1239.

[18] R.O. Loutfy and K . Y . Law, Macromolecules 14 (1981) 587.

[19] K . Y . Law, Chem. Phys. Letters 75 (1980) 545.

[20] K . Y . Law, Photochem. Photobiol. 33 (1981) 799.

[21] R . O . Loutfy and R.A. Arnold, J . Phys. Chem. 86 (1982) 4205.

[22] R.O. Loutfy, Macromolecules 14 (1981) 270.

[23] R . O . Loutfy, J . Polym. Sci. Polymer Phys. E d . 20 (1982) 825.

[24] A . Safarzadeh-Amiri, Can. J . Chem. 62 (1984) 1895.

[ 2 5 ] H . O . Kalinowski and H . Kessler, Topics Stereochem. 7 (1973) 295.

[26] L. Salem, C . Lefrestier, G . Segal and R. Wetmore, J . A m . Chem. Soc. 97 (1975) 479.

[27 ] G . Orlandi and W. Siebrand, Chem. Phys. Letters 30 (1975) 352.

[28] W . G . Dauben, L. Salem and N . J . Turro, Accounts Chem.

Res. 8 (1975) 41.

[29] M . Meyer, J . C . Mialocq and B. Perly, J . Phys. Chem. 94 (1990) 98.

[ 3 0 ] Z . R . Grabowski, K . Rotkiewicz, A . Siemiarczuk, D . L . Cowley and W. Baumann, Nouv. J. C h i m . 3 (1979) 443.

[31 ] W. Rettig, Angew. Chem. Intern. E d . Eng. 25 (1986) 180.

[32] E . M . Kosower and D . Huppert, A n n . Rev. Phys. Chem. 37 (1986)127.

[33] E. Lippert, W. Rettig, V. Bonacic-Koutechy, F. Heisel and J.A. Miehe, Advan. Chem. Phys. 68 (1987) 1.

[34]P.F. Barabara and W. Jarzeba, in: Advances in photochemistry, Vol. 15, eds. D . H . Volman, G.S. H a m m o n d and K . Gollnik (Wiley, New York, 1990) p. 1.

[35] A . Penzkofer and F. Graf, Opt. Quantum Electron. 17 (1985)219.

[ 36 ] A . Penzkofer and W. Falkenstein, Opt. Commun. 17 (1976) 1.

[37] A. Penzkofer and W. Blau, Opt. Quantum Electron. 15 (1983) 325.

[38] G . Gronninger and A . Penzkofer, Opt. Quantum Electron.

16 (1984) 225.

[ 39 ] A . Penzkofer and F. Ossig, Chem. Phys. Letters 154 (1989) 111.

[40] A . Penzkofer and Y . L u , Chem. Phys. 103 (1986) 399.