ATOP dyes. Optimization of a multifunctional merocyanine chromophore for high refractive index modulation in photorefractive materials

Article abstract of DOI:10.1021/ja002321g

This paper reports synthesis, characterization and structural optimization of amino-thienyl-dioxocyano-pyridine (ATOP) chromophores toward a multifunctional amorphous material with unprecedented photorefractive performance. The structural (dynamic NMR, XRD) and electronic (UV/vis, electrooptical absorption, Kerr effect measurements) characterization of the ATOP chromophore revealed a cyanine-type π-conjugated system with an intense and narrow absorption band (ε_max = 140 000 L mol^-1 cm^-1), high polarizability anisotropy (δα₀ = 55 × 10^-40 C V^-1 m²), and a large dipole moment (13 D). This combination of molecular electronic properties is a prerequisite for strong electrooptical response in photorefractive materials with low glass-transition temperature (T_g). Other important materials-related properties such as compatibility with the photoconducting poly(N-vinylcarbazole) (PVK) host matrix, low melting point, low T_g, and film-forming capabilities were optimized by variation of four different alkyl substituents attached to the ATOP core. A morphologically stable PVK-based composite containing 40 wt % of ATOP-3 showed an excellent photorefractive response characterized by a refractive index modulation of Δn ≈ 0.007 and a gain coefficient of Γ ≈ 180 cm^-1 at a moderate electrical field strength of E = 35 V μm^-1. Even larger effects were observed with thin amorphous films consisting of the pure glass-forming dye ATOP-4 (T_g = 16 °C) and 1 wt % of the photosensitizer 2,4,7-trinitro-9-fluorenylidene-malononitrile (TNFM). This material showed complete internal diffraction at a field strength of only E = 10 V μm^-1 and Δn reached 0.01 at only E = 22 V μm^-1 without addition of any specific photoconductor.

Full text of DOI:10.1021/ja002321g

2810
J. Am. Chem. Soc. 2001, 123, 2810-2824
ATOP Dyes. Optimization of a Multifunctional Merocyanine
Chromophore for High Refractive Index Modulation in
Photorefractive Materials
,
†
†
‡
§
Frank W u1 rthner,* Sheng Yao, Joachim Schilling, R u1 diger Wortmann,
§
⊥
⊥
⊥
Mesfin Redi-Abshiro, Erwin Mecher, Francisco Gallego-Gomez, and Klaus Meerholz
Contribution from the Department of Organic Chemistry II, UniVersity of Ulm, Albert-Einstein-Allee 11,
D-89081 Ulm, Germany, Department of Inorganic Chemistry, UniVersity of Ulm, Albert-Einstein-Allee 11,
D-89081 Ulm, Germany, Department of Physical Chemistry, UniVersity of Kaiserslautern, Erwin
Schr o¨ dinger-Strasse, D-67663 Kaiserslautern, Germany, and Department of Chemistry, Physical
Chemistry, UniVersity of Munich, Butenandtstrasse 11, D-81377 M u¨ nchen, Germany
ReceiVed June 27, 2000. ReVised Manuscript ReceiVed January 5, 2001
Abstract: This paper reports synthesis, characterization and structural optimization of amino-thienyl-
dioxocyano-pyridine (ATOP) chromophores toward a multifunctional amorphous material with unprecedented
photorefractive performance. The structural (dynamic NMR, XRD) and electronic (UV/vis, electrooptical
absorption, Kerr effect measurements) characterization of the ATOP chromophore revealed a cyanine-type
-
1
-1
π-conjugated system with an intense and narrow absorption band (ꢀmax ) 140 000 L mol cm ), high
-
40
-1
2
polarizability anisotropy (δR0 ) 55 × 10
C V m ), and a large dipole moment (13 D). This combination
of molecular electronic properties is a prerequisite for strong electrooptical response in photorefractive materials
with low glass-transition temperature (Tg). Other important materials-related properties such as compatibility
with the photoconducting poly(N-vinylcarbazole) (PVK) host matrix, low melting point, low Tg, and film-
forming capabilities were optimized by variation of four different alkyl substituents attached to the ATOP
core. A morphologically stable PVK-based composite containing 40 wt % of ATOP-3 showed an excellent
photorefractive response characterized by a refractive index modulation of ∆n ≈ 0.007 and a gain coefficient
-
1
-1
of Γ ≈ 180 cm at a moderate electrical field strength of E ) 35 V µm . Even larger effects were observed
with thin amorphous films consisting of the pure glass-forming dye ATOP-4 (Tg ) 16 °C) and 1 wt % of the
photosensitizer 2,4,7-trinitro-9-fluorenylidene-malononitrile (TNFM). This material showed complete internal
-
1
-1
diffraction at a field strength of only E ) 10 V µm and ∆n reached 0.01 at only E ) 22 V µm without
addition of any specific photoconductor.
Introduction
coefficients. Promising applications of organic PR materials in
optical data storage, real-time image processing, and phase
conjugation have been demonstrated. Widespread utilization
of these materials in holography, however, requires a number
of additional qualities such as larger refractive index modulation,
faster holographic build-up time, increased hologram lifetime,
There has been considerable progress in the field of photo-
refractive (PR) organic materials during the last years. Being
observed for the first time in 1990 in an organic crystal and in
1
,4
1
2
3
1
991 in polymers, PR effects have meanwhile been realized
3
-7
in a vast number of organics, including polymer composites,
8
9,10
fully functionalized polymers, low-molar-mass glasses,
(4) (a) Meerholz, K.; Volodin, B. L.; Sandalphon; Kippelen, B.;
Peyghambarian, N. Nature 1994, 371, 497-500. (b) Zobel, O.; Eckl, M.;
Strohriegl, P.; Haarer, D. AdV. Mater. 1995, 7, 911-914. (c) Cox, A. M.;
Blackburn, D.; West, D. P.; King, T. A.; Wade, F. A.; Leigh, D. A. Appl.
Phys. Lett. 1996, 68, 2801-2803. (d) Meerholz, K.; Bittner, R.; De Nardin,
Y.; Br a¨ uchle, C.; Hendrickx, E.; Volodin, B. L.; Kippelen, B.; Peygham-
barian, N. AdV. Mater. 1997, 9, 1043-1046.
liquid crystals,11 and polymer-dispersed liquid crystals.12 Many
of these materials are already superior to well-known inorganic
PR crystals with respect to refractive index modulation and gain
*
To whom correspondence should be addressed.
†
Department of Organic Chemistry II, University of Ulm.
Department of Inorganic Chemistry, University of Ulm.
Department of Physical Chemistry, University of Kaiserslautern.
Department of Chemistry, University of M u¨ nchen.
(5) Grunnet-Jepsen, A.; Thompson, C. L.; Twieg, R. J.; Moerner, W. E.
Appl. Phys. Lett. 1997, 70, 1515-1517.
‡
§
⊥
(6) (a) Kippelen, B.; Marder, S. R.; Hendrickx, E.; Maldano, J. L.;
Guillernet, G.; Volodin, B. L.; Steele, D. D.; Enami, Y.; Sandalphon; Yao,
Y. J.; Wang, J. F.; R o¨ ckel, H.; Erskine, L.; Peyghambarian, N. Science
1998, 279, 54-58. (b) Hendrickx, E.; Herlocker, J.; Maldonado, J. L.;
Marder, S. R.; Kippelen, B.; Persoons, A.; Peyghambarian, N. Appl. Phys.
Lett. 1998, 72, 1679-1681.
(
1) (a) Moerner, W. E.; Silence, S. M. Chem. ReV. 1994, 94, 127-155.
(
1
b) Zhang, Y.; Burzynski, R.; Ghosal, S.; Casstevens, M. K. AdV. Mater.
996, 8, 111-125. (c) Moerner, W. E.; Grunnet-Jepsen, A.; Thompson, C.
L. Annu. ReV. Mater. Sci. 1997, 27, 585-623. (d) Meerholz, K. Angew.
Chem. 1997, 109, 981-985; Angew. Chem., Int. Ed. Engl. 1997, 36, 945-
(7) (a) W u¨ rthner, F.; Wortmann, R.; Matschiner, R.; Lukaszuk, K.;
Meerholz, K.; DeNardin, Y.; Bittner, R.; Br a¨ uchle, C.; Sens, R. Angew.
Chem. 1997, 109, 2933-2936; Angew. Chem., Int. Ed. Engl. 1997, 36,
2765-2768. (b) Meerholz, K.; DeNardin, Y.; Bittner, R.; Wortmann, R.;
W u¨ rthner, F. Appl. Phys. Lett. 1998, 73, 4-6. (c) W u¨ rthner, F.; Legrand,
C.; Mecher, E.; Meerholz, K.; Wortmann, R. Polym. Mater. Sci. Eng. 1999,
80, 252-254.
9
48. (e) Meerholz, K.; Kippelen, B.; Peyghambarian, N. In Electrical and
Optical Polymer Systems; Wise, D. L., Wnek, G. E., Trantolo, D. J., Gresser,
J. D., Cooper, T. M., Eds.; World Scientific: London, 1998; pp 571-632.
(
2) Sutter, K., G u¨ nter, P. J. Opt. Soc. Am. B 1990, 7, 2274-22.
(3) Ducharme, S.; Scott, J. C.; Twieg, R. J.; Moerner, W. E. Phys. ReV.
Lett. 1991, 66, 1846-1849.
1
0.1021/ja002321g CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/21/2001

ATOP Dyes
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2811
and improved morphological stability of the materials. In this
paper, we concentrate on two aspects, refractive index modula-
tion and morphological stability; we also discuss the tradeoff
between these material properties and describe systematic
strategies for their optimization.
low glass-transition temperatures (<25 °C), which facilitate
device fabrication and allow for “orientational enhancement”
(reorientation of dyes in the internal space-charge field).13
Composites will probably also be the natural choice for industrial
applications because they allow simple fine-tuning for appli-
cations with different demands. Organic CTAs are already
well-established and successfully used in photocopiers, laser
The PR effect comprises a series of events taking place when
two coherent laser beams interfere in a material that exhibits
both photoconductiVity and electrooptical (EO) response. In a
first step charge carriers are generated in the bright regions of
the interference pattern by photoexcitation of spectral sensitizers
and charge transfer to the charge-transport agent (CTA).
Subsequently, the mobile charges are transferred by the CTA
from the bright to the dark regions by diffusion or, more
effectively, by an electric-field-assisted drift process. The
charges then get trapped and create a periodic space-charge
distribution between bright and dark regions of the material. In
a final step reorientation of dipolar EO chromophores occurs
in the total electric field (external plus space-charge field), and
a refractive index grating is produced. This multistep hologram
formation process in photorefractive materials exhibits several
unique features which are not achieved with other photo-
addressable materials: (1) the refractive index gratings are out
of phase with respect to the original light intensity pattern; (2)
the writing process is fully reversible; and (3) only weak laser
intensities are required. The phase shift between intensity and
refractive index grating gives rise to energy exchange between
the two writing beams (“two-beam coupling”), an effect
characteristic of PR materials and of considerable interest for
optical data processing.
14,15
16
printers,
and, more recently, organic light-emitting devices.
EO chromophores, on the other hand, have been studied rather
1
7
extensively in the field of nonlinear optics, but hitherto
no final candidates for technological applications have been
identified. With regard to morphological properties, the com-
posite should have the capability of forming amorphous films
of high optical quality. Any microphase separation must be
avoided because it will cause light scattering.
Eventually in 1996 the so-called “orientational enhancement
1
3
effect” in low-Tg polymeric PR materials was be consistently
(3)
interpreted in terms of the Kerr susceptibility ø (-ω;ω,0,0),
(2)
in contrast to the Pockels susceptibility ø (-ω;ω,0) in crystalline
1
8
materials. This allowed the derivation of a suitable figure-of-
Kerr
merit F
for the electronic properties of PR chromophores,
Kerr
1
M
2
F
)
[9µ‚â + 2µ ‚δR/(k T)]
(1)
B
where µ is the dipole moment, δR the anisotropy of the linear
polarizability, â the second-order polarizability, k the Boltz-
B
18
mann constant, T the temperature, and M the molar mass.
Kerr
Usually, the figure-of-merit is reported as F0 , where the
subscript “0” indicates dispersion-free values at infinite wave-
length (λ f ∞).
Multicomponent organic materials seem to be well-suited to
optimize the different optical and electrical functionalities due
to their compositional flexibility. Currently, the most advanced
organic PR materials are composites of two major components
providing charge transport and EO response, small amounts of
a photosensitizer for the given laser wavelength, and signifi-
cant amounts of plastisizers. The latter are added to achieve
On the basis of eq 1, clear guidelines for the most favorable
chromophores could be developed. In a detailed study on
two series of merocyanine chromophores covering the whole
range of electronic structures from polyene-like to betaine-
like, we could unambiguously confirm eq 1 and elucidate the
1
9,20
various contributions.
It was shown that in contrast to
(8) (a) Tamura, K.; Padias, A. B.; Hall, H. K., Jr.; Peyghambarian, N.
permanently poled (e.g. crystalline or high-Tg polymeric) PR
materials, where the nonlinearity only depends on the µ‚â
Pockels term, PR effects in low-Tg materials are dominated by
Appl. Phys. Lett. 1992, 60, 1803-1805. (b) Kippelen, B.; Tamura, K.;
Peyghambarian, N.; Padias, A. B.; Hall, H. K., Jr. J. Appl. Phys. 1993, 74,
3
617-3619. (c) Kippelen, B.; Tamura, K.; Peyghambarian, N.; Padias, A.
B.; Hall, H. K., Jr. Phys. ReV. B 1993, 48, 10710-10718. (d) Yu, L.; Chen,
Y. M., Chan, W. K. J. Phys. Chem. 1995, 99, 2797-2802. (e) Yu, L.;
Chan, W. K.; Peng, Z.; Gharavi, A. Acc. Chem. Res. 1996, 29, 13-21. (f)
Peng, Z.; Gharavi, A. R.; Yu, L. J. Am. Chem. Soc. 1997, 119, 4622-
(13) Moerner, W. E.; Silence, S. M.; Hache, F.; Bjorklund, G. C. J. Opt.
Soc. Am. B 1994, 11, 320-330.
(14) Law, K.-Y. Chem. ReV. 1993, 93, 449-486.
(15) (a) Wiedemann, W. Chem.-Z. 1982, 106, 275-287. (b) Gibson, H.
W. Polymer 1984, 25, 3-27. (b) Strohriegl, P.; Grazulevicius, J. V. In
Handbook of Organic ConductiVe Molecules and Polymers; Nalwa, H. S.,
Ed.; Wiley: Chichester, 1997; Vol. 1, pp 553-620.
(16) (a) Salbeck, J. Ber. Bunsen-Ges. Phys. Chem. 1996, 100, 1667-
1677. (b) Friend, R. H.; Gymer, R. W.; Holmes, A. B.; Burroughes, J. H.;
Marks, R. N.; Taliani, C.; Bradley, D. D. C.; Dos Santos, D. A.; Br e´ das, J.
L.; L o¨ gdlund, M.; Salaneck, W. R. Nature 1999, 397, 121-128. (c) Heeger,
A. J. Solid State Commun. 1998, 107, 673-679.
(17) (a) Wolff, J. J.; Wortmann, R. AdV. Phys. Org. Chem. 1999, 32,
121-217. (b) Verbiest, T.; Houbrechts, S.; Kauranen, M.; Clays, K.;
Persoons, A. J. Mater. Chem. 1997, 7, 2175-2189. (c) Kanis, D. R.; Ratner,
M. A.; Marks, T. J. Chem. ReV. 1994, 94, 195-242. (d) Burland, D. M.;
Miller, R. D.; Walsh, C. A. Chem. ReV. 1994, 94, 31-75.
(18) (a) Wortmann, R.; Poga, C.; Twieg, R. J.; Geletneky, C.; Moylan,
C. R.; Lundquist, P. M.; DeVoe, R. G.; Cotts, P. M.; Horn, H.; Rice, J.;
Burland, D. M. J. Chem. Phys. 1996, 105, 10637-10647. (b) Kippelen,
B.; Meyers, F.; Peyghambarian, N.; Marder, S. J. Am. Chem. Soc. 1997,
119, 4559-4560.
(19) Wortmann, R.; W u¨ rthner, F.; Sautter, A.; Lukaszuk, K.; Matschiner,
R.; Meerholz, K. Proc. SPIE 1998, 3471, 41-49.
4
2
9
232. (g) Wang, L.; Wang, Q.; Yu, L. Appl. Phys. Lett. 1998, 73, 2546-
548. (h) Li, W.; Gharavi, A.; Wang, Q.; Yu, L. AdV. Mater. 1998, 10,
27-931. (i) D o¨ bler, M.; Weder, C.; Neuenschwander, P.; Suter, U. W.;
Follonier, S.; Bosshard, C.; G u¨ nter, P. Macromolecules 1998, 31, 6184-
6
189. (j) Schloter, S.; Hofmann, U.; Hoechstetter, K.; B a¨ uml, G.; Haarer,
D. J. Opt. Soc. Am. B 1998, 15, 2560-2565. (k) Bratcher, M. S.; DeClue,
M. S.; Grunnet-Jepsen, A.; Wright, D.; Smith, B. R.; Moerner, W. E.; Siegel,
J. S. J. Am Chem. Soc. 1998, 120, 9680-9681. (l) Hattemer, E.; Zentel,
R.; Mecher, E.; Meerholz, K. Macromolecules 2000, 33, 1972-1977.
(
9) Lundquist, P. M.; Wortmann, R.; Geletneky, C.; Twieg, R. J.;
Jurich, M.; Lee, V. Y.; Moylan, C. R.; Burland, D. M. Science 1996, 274,
1
182-1185.
(
10) (a) Li, W.; Gharavi, A.; Wang, Q.; Yu, L. AdV. Mater. 1998, 10,
927-931. (b) Wang, L.; Zhang, Y.; Wada, T.; Sasabe, H. Appl. Phys. Lett.
996, 69, 728-730. (c) Wang, L.; Zhang, Y.; Wada, T.; Sasabe, H. Appl.
1
Phys. Lett. 1997, 70, 2949-2951. (d) Grasruck, M.; Schreiber, A.; Hofmann,
U.; Zilker, S.; Leopold, A.; Schloter, S.; Hohle, C.; Strohriegl, P.; Haarer,
D. Phys. ReV. B 1999, 60, 16543-16548. (e) Hohle, C.; Hofmann, U.;
Schloter, S.; Thelakkat, M.; Strohriegl, P.; Haarer, D.; Zilker, S. J. J. Mater.
Chem. 1999, 9, 2205-2210.
(
11) (a) Khoo, I. C.; Li, H.; Liang, Y. Opt. Lett. 1994, 19, 1723-1725.
b) Wiederrecht, G. P.; Yoon, B. A.; Wasielewski, M. R. Science 1996,
70, 1794-1797. (c) Wiederrecht, G. P.; Yoon, B. A.; Svec, W. A.;
Wasielewski, M. R. J. Am. Chem. Soc. 1997, 119, 3358-2264.
12) (a) Ono, H.; Kawatsuki, N. Opt. Lett. 1997, 22, 1144-1146. (b)
Golemme, A.; Volodin, B. L.; Kippelen, B.; Peyghambarian, N. Opt. Lett.
997, 22, 1226-1228. (c) Golemme, A.; Kippelen, B.; Peyghambarian, N.
Appl. Phys. Lett. 1998, 73, 2408-2410.
(
(20) (a) W u¨ rthner, F.; Thalacker, C.; Matschiner, R.; Lukaszuk, K.;
Wortmann, R. Chem. Commun. 1998, 1739-1740. (b) Beckmann, S.;
Etzbach, K.-H.; Kr a¨ mer, P.; Lukaszuk, K.; Matschiner, R.; Schmidt, A. J.;
Schuhmacher, P.; Sens, R.; Seybold, G.; Wortmann, R.; W u¨ rthner, F. AdV.
Mater. 1999, 11, 536-541. (c) Wortmann, R.; Glania, C.; Kr a¨ mer, P.;
Lukaszuk, K.; Matschiner, R.; Twieg, R. J.; You, F. Chem. Phys. 1999,
245, 107-120.
2
(
1

2812 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
W u¨ rthner et al.
Chart 1
Scheme 1
2
orientational birefringence related to the µ ‚δR term. In view
2
of the importance of the µ ‚δR term the new dyes 2BNCM
9
6
7
(
(
3), DHADC-MPN (4), and ATOP-1 (5) were synthesized
Chart 1). These dyes are all characterized by high δR but
low â values and were successfully incorporated into highly
promising PR materials. At least for DHADC-MPN and
ATOP-1 based PR materials complete internal diffraction was
achieved at lower dye concentrations than ever possible before
Table 1. ATOP Dyes Synthesized According to Scheme 1
starting
materials
M
(g/mol)
1
2
3
4
dyes
R
R
R
R
ATOP-1
ATOP-2
ATOP-3
ATOP-4
ATOP-5
ATOP-6
ATOP-7
ATOP-8
ATOP-9
ATOP-10
8c, 11b
8a, 11b
8b, 11d
8c, 11d
8a, 11a
8d, 11b
8e, 11b
8e, 11c
8e, 11e
8c, 11f
Bu
Bu
Me
Me
Me
Me
Me
Me
Me
Me
Pr
Bu
Me
Et
Bu
Meb
Pip
Et
Et
Et
Bu
Bu
Me
Et
Bu
Me
427
343
427
484
301
383
400
427
427
371
with traditional EO chromophores such as F-DEANST (1) or
Kerr
a
DMNPAA (2). The highest F0
value was reported for the
2-Ethex
2-Ethexa
Me
highly dipolar merocyanine dye 6, which exhibits a betainic
character in the ground state.19
Bu
Bu
Hex
Bu
H
Although merocyanines in the cyanine limit (like ATOP-1,
Bu
Bu
Bu
Bu
5) with equal contributions of the apolar (neutral) and the
zwitterionic resonance structures exhibit the highest polar-
Me
izability anisotropy δR, the more betainic dyes such as 6
Kerr
have the highest overall F0
values because of their larger
a
-Ethylhexyl. b R , R ) Piperidine.
3
4
2
dipole moments. However, the beneficial effect of large dipole
moments for dye orientation in electric fields is only given
under ideal conditions in diluted solutions, while high dye
concentrations are required to achieve the desired modulation
Results and Discussion
Synthesis. ATOP dyes were synthesized in high yields
57-87%) by condensation reactions of aldehydes 8a-e and
(
efficiencies. Under these conditions serious problems due to
dipolar aggregation and crystallization were noted.21 Therefore,
hydroxypyridones 11a-f in acetic acid anhydride according to
Scheme 1. The required aldehydes 8a-e were obtained in yields
we decided to focus our further attempts to optimize the
materials on the dyes with the highest δR values, i.e., the ATOP
chromophore, which exhibit better solubility than dye 6 and
reduced tendency for dipolar aggregation.
2
2
of 50-70% from 2-bromo-5-formylthiophene 7 and dialkyl-
amines in nucleophilic displacement reactions and were typically
2
3
used as crude materials (purity ≈ 90%). Hydroxypyridines
1
1a-f were obtained by condensation reactions of acyl acetates
In the first part of this paper, we will characterize the
structural and functional properties of ATOP rendering this
chromophore superior to any other previously investigated dye
within the field of organic PR materials. In the second part, we
will show the important impact of alkyl substituents on the
morphological and functional properties of the dyes and the PR
composites. Finally, we will report the electrooptical and
holographic characterization of a series of PVK-based PR com-
posites with varying ATOP content (10-50 wt %). In one case
2
4
9
a,b and cyanacetic acid amides 10a-e. Table 1 lists the
compounds synthesized via this route along with their molar
mass M. Note that according to eq 1, M influences the respective
Kerr
F0
value.
Structural Properties of the ATOP Chromophore. Promis-
ing molecular structures for maximized PR response can be
designed on the basis of eq 1. In this section we will characterize
(22) Weston, A. W.; Michaels, R. J., Jr. J. Am. Chem. Soc. 1950, 72,
(ATOP-4) PR devices were obtained from the bulk chromophore
1422-1423.
(
23) (a) Nazarova, Z. N.; Pustarov, V. S. Khim. Geterotsikl. Soedin. 1969,
alone. The optimized materials display the strongest refractive
index modulation ever reported in nonliquid-crystalline organic
PR materials at low electrical field.
5
, 586. A recently published method gave significantly lower yields: (b)
Prim, D.; Kirsch, G.; Nicoud, J.-F. Synlett 1998, 383-384.
(24) (a) Guareschi, J. Ber. Dtsch. Chem. Ges. Board 4 (Referate, Patente,
Nekrologe), 1896, 29, 654-656. (b) Bello, K. A. Dyes Pigm. 1995, 28,
83-90. (c) Katritzky, A. R.; Rachwal, S.; Smith, T. J. Heterocycl. Chem.
1995, 32, 1007-1010.
(
21) W u¨ rthner, F.; Yao, S. Angew. Chem., Int. Ed. 2000, 39, 1978-
1
981.

ATOP Dyes
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2813
the structural properties of ATOP and show that this chromo-
phore is indeed an excellent choice for this goal. Moreover,
the structural investigation will reveal several geometrical
features which distinguish ATOP from other merocyanine dyes
(
such as DHADC-MPN 4) and which are considered to be
important for high dye loading in PR composite materials.
a) UV/Vis Absorption and Electrooptical Absorption
(
Spectroscopy. Donor-acceptor substituted π-systems such as
ATOP or the nonlinear optical (NLO) prototype molecule
p-nitroaniline (pNA) usually exhibit a single intense charge-
transfer transition (CT) in the UV/vis region which may be
characterized by electrooptical absorption (EOA) spectros-
copy2
5,26
and analyzed according to eq 2
ꢀE(φ, ν˜ ) ) ꢀ( ν˜ )[1 + L(φ, ν˜ )E + ...]
2
(2)
where ꢀΕ is the absorption coefficient in the presence of the
electric field E and L is a measure of the relative change of ꢀ
induced by E. L depends on the wavenumber ν˜ ) 1/λ as well
as on the angle φ between the polarization vector of the incident
light and the applied field. It has been shown that L may be
interpreted in terms of the imaginary part of a third-order
Figure 1. Optical (right scale; s) and electrooptical (left scale;
φ ) 0°, ‚‚‚; φ ) 90°, - - -) absorption spectra of ATOP-1 in dioxane at
T ) 298 K.
externally applied field increases while it decreases for the
orthogonal polarization (φ ) 90°). This confirms that the
ground-state dipole moment µg and the transition dipole moment
µag are essentially parallel and oriented along the conjugated
chain. The second effect is the band shift effect (Stark effect)
of the CT band in the electric field. It only occurs if the optical
excitation is accompanied by a change of the dipole moment
(3)
26d
susceptibility, Im{ø (-ω; ω, 0,0)}.
Usually the EOA spectrum is recorded for two polarizations
φ ) 0° and 90°. A band shape analysis in terms of the optical
absorption spectrum and its first derivative yields the ground-
state dipole moment µ as well as the dipole difference between
the ground state (g) and the excited state (a) ∆µ. Furthermore,
knowledge of ∆µ obtained from EOA allows one to estimate
the static second-order polarizability â0 and the anisotropy of
∆
µ. The EOA spectrum of ATOP-1 shows almost no shift of
the spectrum, indicating a very small dipole difference ∆µ. The
the first-order polarizability δR0 of the dye within a two-level
-
30
detailed band shape analysis yields µ ) 44 × 10
Cm (13
Cm (9.3
model:18a,26b,27
-
30
-30
D), ∆µ ) 5.6 × 10
Cm (1.7 D), µag ) 31 × 10
2
D), and λag ) 536 nm, respectively (in dioxane at T ) 298 K).
In addition, UV/vis spectra were recorded in solvents of
different polarity. As expected from the EOA result ∆µ ≈ 0,
δR ) 2µ λag/(hc)
(3)
(4)
0
ag
2
2
2
â0 ) 6µ ∆µλag /(hc)
28
ag
solvatochromism of ATOP-1 was negligibly small, i.e., λag
was found to be almost independent of the solvent even for
very different polarities: 531 (hexane), 532 (methanol), 535
where the wavelength of maximum absorption λag and the
transition dipole moment µag may be derived from the UV/vis
spectrum.
(
(
ether), 536 (dioxane), 536 (DMF), 538 (CCl4), and 541 nm
CH2Cl2). Also the molar decadic absorption coefficient at the
Figure 1 shows the EOA spectrum of ATOP-1 taken in
absorption maximum (ꢀmax) and the transition dipole moment
µag showed only little dependence on the solvent and ranged
-
5
dioxane at a concentration of about 10 M. No aggregation is
observed at this concentration (vide infra). The EOA spectrum
of ATOP-1 is governed by two effects. The first is the
electrodichroic effect caused by orientation of the chromophores
in the externally applied electric field. For ATOP-1 there is
strong positiVe electrodichroism, i.e., the absorption for parallel
polarization (φ ) 0°) of the incident light field relative to the
-1
-1
between 135 000 L mol cm in methanol and 150 000 L
-1
-1
mol cm in CH2Cl2. For all solvents studied, the narrow (half
-1
width ∆ ν˜ 1/2 ≈ 900 cm ) and intense cyanine-like transition is
preserved. This is advantageous for PR applications because
laser wavelengths close to λag may be used with such chromo-
phores without absorption problems often encountered with the
broad bands of traditional NLO dyes (cp. Figure 5 below). In
this way one can take advantage of the considerable dispersion
enhancement near the absorption edge.
(25) (a) Liptay, W. In Excited States, Vol. 1. Dipole Moments and
Polarizabilities of Molecules in Excited Electronic States; Lim, E. C., Ed.;
Academic Press: New York, 1974; pp 129-229. (b) Wortmann, R.; Elich,
K.; Lebus, S.; Liptay, W.; Borowicz, P.; Grabowska, A. J. Phys. Chem.
1
992, 96, 9724-9730. (c) Bublitz, G. U.; Boxer, S. G. Annu. ReV. Phys.
With the experimental results for λag, µag, µ, and ∆µ it
Chem. 1997, 48, 213-242. (d) Wolff, J. J.; Wortmann, R. AdV. Phys. Org.
Chem. 1999, 32, 121-217.
is possible to estimate the linear and nonlinear polarizabilities
Kerr
as well as the figure-of-merit F0
of ATOP-1 according
(26) (a) W u¨ rthner, F.; Effenberger, F.; Wortmann, R.; Kr a¨ mer, P. Chem.
-
40
-1
2
to eqs 1-4. The results are δR0 ) 52 × 10 C V m , â0 )
Phys. 1993, 173, 305-314. (b) Wortmann, R.; Kr a¨ mer, P.; Glania,C.; Lebus,
S.; Detzer, N. Chem. Phys. 1993, 173, 99-108. (c) Blanchard-Desce, M.;
Wortmann, R.; Lebus, S.; Lehn, J.-M.; Kr a¨ mer, P. Chem. Phys. Lett. 1995,
-
50
-2
3
Kerr
-74
2
-2
4
23 × 10 C V m , and F0 ) 1.40 × 10 C V
m
2
43, 526-532. (d) Blanchard-Desce, M.; Alain, V.; Midrier, L.; Wortmann,
(28) (a) Liptay, W. Angew. Chem. 1969, 81, 195-206; Angew. Chem.,
Int. Ed. Engl. 1969, 8, 177. (b) Liptay, W. Z. Naturforsch. 1965, 20a,
1441-1471.
(29) The structural and dipolar properties of the ATOP chromophore
are reproduced fairly well by AM1 calculations which are in accordance
with the crystal structure of ATOP-10 and measured dipole moments.
(30) (a) Jackman, L. M.; Cotton, F. A., Eds. Dynamic Nuclear Magnetic
Resonance Spectroscopy; Academic Press: New York, 1975. (b) Kessler,
H. Angew. Chem. 1970, 82, 237-253. (c) Stewart, W. E.; Sidall, T. H., III
Chem. ReV. 1970, 70, 517-551.
R.; Lebus, S.; Glania, C.; Kr a¨ mer, P.; Fort, A.; Muller, J.; Barzoukas, M.
J. Photochem. Photobiol. 1997, 105, 115-121. (e) Blanchard-Desce, M.;
Alain, V.; Bedworth, P. V.; Marder, S. R.; Fort, A.; Runser, C.; Barzoukas,
M.; Lebus, S.; Wortmann, R. Chem. Eur. J. 1997, 3, 1091-1104. (f) Steybe,
F.; Effenberger, F.; Beckmann, S.; Kr a¨ mer, P.; Glania, C.; Wortmann, R.
Chem. Phys. 1997, 219, 317-331. (g) Bublitz, G. U.; Ortiz, R.; Runser,
C.; Fort, A.; Barzoukas, M.; Marder, S. R.; Boxer, S. G. J. Am. Chem. Soc.
1
997, 119, 2311-2312.
(27) Oudar, J. L.; Chemla, D. S. J. Chem. Phys. 1977, 66, 2664-2668.

2814 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
W u¨ rthner et al.
-
1
7
mol kg , respectively. The value of the anisotropy of the linear
polarizability was additionally confirmed by an independent Kerr
-
40
-1
effect measurement to be δR(-ω; ω) ) 316 × 10
C V
2
m at the wavelength λ ) 589 nm. Application of the two-
-
40
-1
level dispersion correction yields δR0 ) 55 × 10 C V for
the static value in excellent agreement with the EOA result. In
view of the similar molecular structure and the almost identical
UV/vis absorption observed for all ATOP dyes we may assume
equal values of µ, δR0, and â0 for the dyes of this study, ATOP-1
Kerr
to ATOP-10. However, differences in the figure-of-merit F0
-
74
2
-2
4
-1
ranging from 1.24 × 10
C V m mol kg (ATOP-4) to
-
74
2
-2
4
-1
1
.99 × 10
C V m mol kg (ATOP-5) result from the
different molar masses (Table 1).
The electric properties of the dipolar ATOP dyes may be
analyzed in terms of a simple CT model, which describes the
ground and excited state as linear combinations of the neutral
and zwitterionic resonance structures based on a resonance
2
18a,20
parameter c .
Within this crude model donor-acceptor
substituted π-conjugated chain molecules may be classified from
2
2
polyene-type (c ≈ 0) over neutrocyanines (“cyanine limit”, c
2
≈
0.5) to betaine-type systems (c ≈ 1). A second parameter
of this model is the maximal hypothetical dipole change ∆µmax,
i.e., the difference of the dipole moments of the neutral and the
zwitterionic resonance structure.1
8a,20
Since this difference is
proportional to the charge separation distance, ∆µmax can be
regarded as a measure of the effective CT length of the donor-
2
acceptor system. The model parameters c and ∆µmax can be
determined from the values of the transition dipole moment and
the dipole difference according to
Figure 2. HOMO, LUMO, difference, and transition density for (A)
a model cyanine structure with a thiophene core and (B) ATOP dyes
calculated with the CNDO/S method for an AM1 optimized geometry
2
1
2
2 -1/2
c ) / [1 - ∆µ(4µ + ∆µ )
]
(5)
(6)
2
ag
(alkyl groups replaced by hydrogen for simplicity).
2
∆
µmax ) ∆µ/(1 - 2 c )
Scheme 2
2
For ATOP-1 the resonance parameter is found to be c ) 0.46
in dioxane solution, which indicates a structure close to the
cyanine limit. The maximum dipole difference is found to be
-
30
∆
µmax ) 62 × 10
Cm (19 D), which corresponds to an
effective CT length of 0.39 nm (∼36% of the length of the
π-conjugated system).
We further analyzed the HOMO and LUMO as well as the
transition and dipole difference densities involved in the low-
lying CT band of ATOP-1 (Figure 2b). These molecular
properties were calculated with the CNDO/S method for AM1
optimized geometry.29 The MO calculations reveal a unique
electronic structure of the HOMO and LUMO where the number
of nodes is minimized in the transition densities but maximized
in the difference density. As a consequence, the magnitude of
the transition dipole is almost maximized for the given topology,
while the dipole difference is close to zero. This general tradeoff
between µag and ∆µ has been discussed recently for a two-
center CT system. The comparison with a similar calculation
of the cyanine model chromophore shown in Scheme 2 reveals
a close correspondence of the HOMO/LUMO structure with
ATOP and supports the classification of ATOP as being close
to the cyanine limit (Figure 2a).
system. Here, the magnitude of the rotational barriers will give
valuable information about the orders of the π bonds which
have to be broken in the transition state of the internal rotation.
Dynamic NMR spectroscopy is a powerful method for getting
insight into the double bond character of C-C and C-N bonds
26d
30
of conjugated compounds. As a prerequisite, the rotational
barrier has to be in the energy range accessible by the NMR
-
1
method (≈20-100 kJ mol ), and the investigated conforma-
tions have to be populated to a detectable amount. In the case
of ATOP dyes with identical substituents at the amine, the
rotation around the C-N bond is characterized by equal amounts
of the two exchanging sites. For the other two possible rotations
the situation is more complicated because one conformation is
favored over the others (Scheme 3). As shown in Figure 3 (left),
cooling of an ATOP-1 solution in CDCl3 causes significant line-
broadening of the methylene protons of the amino group (and
also the other protons of the alkyl substituents). The coalescence
temperature (Tc) is about 255 K. At even lower temperature,
separated signals appear which imply a frozen rotation around
(b) NMR Spectroscopy. The results on the cyanine-like
electronic structure of the ATOP chromophore suggest mean
bond orders of 1.5 for each carbon-carbon bond of the
conjugated chain. However, deviations from this mean value
might be expected for the individual bonds because the
conjugated chain is partly embedded into heterocyclic units, i.e.,
the thiophene and the pyridone rings. Nevertheless, a high
contribution of zwitterionic structures should result in restricted
rotation also around those bonds, which are not fixed in a ring

ATOP Dyes
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2815
q
Table 2. Gibbs Free Activation Energies ∆G of ATOP-1 and
2a-c3 in CDCl
2a
at the Coalescence Temperature T
1
3
c
q
∆
G
T
c
(K)
(kJ mol-1)
a
ATOP-1
255
388
306
253
52.2
81.7
61.6
51.5
b
1
1
1
2a
2b
2c
a
The value at 298 K is calculated to be 49.0 kJ mol-1. b For DMF
-
1
values between 80 and 92 kJ mol are reported by several authors
using different solvents and different evaluation methods.
3
0a-c
Scheme 4
methine 12c whose C-N π bond order was calculated to be
3
3
0
.63, close to the cyanine limit of 0.5. Since identical C-N
rotational barriers suggest similar CT properties, this result can
be viewed as further evidence for the cyanine-like character of
ATOP. Obviously, the higher acceptor strength of the pyridone
acceptor in ATOP as compared to the aldehyde acceptor unit
in 12c compensates for the counteracting resonance energy of
the heterocyclic thiophene core.
3
Figure 3. Measured (left, in CDCl at the given temperature) and
calculated (right, rate constant given) 500 MHz H NMR spectra of
the AA′BB′ spin system of the (RCH ) N group of ATOP-1.
2 2
1
Scheme 3
Whereas rotation 1 in Scheme 3 could be easily studied by
1
temperature-dependent H NMR based on identical populations
of the exchanging sites, distinct conformational preferences
prevented a comparable study for the other two possible
rotations. Strong nuclear Overhauser (NOE) effects between
the methine hydrogen and the pyridone CH3 group suggest
the predominance of the structure depicted in Scheme 3, which
is also the structure of lowest energy according to AM1
2
9
calculations (gas phase). These calculations reveal that steric
repulsions between the pyridone CH3 group and the thiophene
core prevent the formation of another planar conformer by
rotation 3. On the other hand, considerable line broadening of
the C-N bond. Total line shape analyses30,31 were carried out
for each temperature to give the respective first-order rate
constants and the calculated spectra are shown on the right side
of Figure 3. A quantitative evaluation may be done on the basis
of Eyring’s equation,
1
the H resonance of the thiophene proton H3 at higher
temperatures (the temperature of maximal line broadening is
ca. 315 K in CDCl3) demonstrates a dynamic influence of a
slightly populated (<2%) second conformer and allows the
k T
B
*
*
k )
exp[-(∆H - T∆S )/RT]
(7)
h
*
estimation of the free energy of activation ∆G of rotation 2
-
1
to be at least 60 kJ mol . The existence of such slightly
populated conformations in solution as well as the bent shape
of the ATOP core may contribute to the high solubility of these
dyes (in particular ATOP-4, see Table 4) and also to the
excellent morphological stability of PR composites discussed
below.
where k is the rate constant, kB Boltzmann’s constant, T
the temperature, h Planck’s constant, and R the molar gas
constant. The activation parameters were determined by linear
*
-1
regression of log(k/T) vs 1/T to be ∆G (255 K) ) 52.2 kJ mol ,
*
-1
*
∆
H (255 K) ) 72.2 kJ mol , and ∆S (255 K) ) 78.5
J K mol . It is generally accepted that ∆S and ∆H values
-
1.
-1
*
*
(c) X-ray Crystallography. Final proof for the cyanine-type
obtained by this procedure are susceptible to considerable errors
*
highly conjugated system and the lowest energy conformation
was disclosed by the precise determination of the geometry of
ATOP-10 in a single crystal. The ORTEP view shown in Figure
while the free energies of activation ∆G are very reliable. Thus,
*
we compare only the ∆G value for the C-N rotational barrier
3
2
of ATOP-1 to those of other conjugated amines (Table 2).
4
reveals an essentially coplanar π-conjugated system with a
According to Table 2 the rotational barrier of the C-N bond
of ATOP-1 is very similar to the respective bond of penta-
twisting angle of only 7° between the mean planes of the
thiophene and the pyridone rings. This high degree of planarity
is accompanied by an expansion of the bond angle of the central
(
31) (a) Binsch, G.; Kessler, H. Angew. Chem. 1980, 92, 445-463;
Angew. Chem., Int. Ed. Engl. 1980, 19, 411. (b) Feigel, M. J. Phys. Chem.
983, 87, 3054-3058.
32) (a) Radeglia, R. Z. Phys. Chem. (Leipzig) 1967, 235, 335-339. (b)
Blanchard, M. L.; Chevallier, A.; Martin, G. J. Tetrahedron Lett. 1967, 50,
2
methine sp carbon to 137° and a reduction of the S1-O2
1
(
distance to 2.58 Å, which is significantly less than the sum of
the van der Waals radii (3.25 Å). The optical and electronic
properties of the ATOP chromophore are mainly reflected by
the bond distances along the conjugated path N1-C1-C2-
5
057-5060. (c) Filleux-Blanchard, M. L.; Clesse, F.; Bignebat, J.; Martin,
G. J. Tetrahedron Lett. 1969, 981-988. (d) Kramer, H. E. A.; Gompper,
R. Z. Phys. Chem. Neue Folge 1964, 43, 292-303. (e) Wiberg, K. B.;
Rablen, P. R.; Rush, D. J.; Keith, T. A. J. Am. Chem. Soc. 1995, 117,
4
261-4270.
(33) Leupold, D.; D a¨ hne, S. Theor. Chim. Acta 1965, 3, 1.

2816 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
W u¨ rthner et al.
With regard to the morphological properties discussed in the
second part of the paper some comments on the packing of
ATOP-10 in the crystalline state are of interest. Thus the dimeric
unit depicted in Figure 4 contains two ATOP dyes linked
together by two hydrogen bonds between the imide groups at a
distance of 2.87 Å for O2‚‚‚N2′ and N2‚‚‚O2′, respectively. This
arrangement leads to a densely packed centrosymmetric unit,
which exhibits an antiparallel orientation of the dipole moments
of the two chromophores. Thus, in addition to hydrogen bonding
also the dipolar interaction energies are considerably reduced.
The space surrounding this dimer consists of a “cocoon” of alkyl
groups made up of the dibutylamino substituents of the depicted
and the neighboring molecules without any π-π interactions
between the conjugated systems of the individual dye molecules.
It is clear that this hydrogen bond-directed arrangement is a
unique feature of ATOP-10. If no hydrogen bonding is possible
as in the other ATOP dyes, the packing in the solid state will
arise from the remaining dipolar and dispersion interactions.
Despite considerable efforts none of the other dyes bearing alkyl
substituents at the imide group could be grown in crystals of
sufficient size suited for X-ray crystallography. Possible reasons
for the resistance toward well-ordered dye packing might be
given by a high conformational disorder in solution (see the
NMR studies above) or the formation of antiparallel dimer
2
1
aggregates which cannot pack easily in a three-dimensional
lattice. Nevertheless, we assume that all ATOP dyes of this
study exhibit a perfectly flat π-conjugated core as found for
ATOP-10.
Figure 4. ORTEP drawings for ATOP-10 in the crystal showing the
dimeric unit of two ATOP-10 molecules connected by hydrogen
bonding in an antiparallel fashion (top view and side view).
To summarize our structural studies, characterization of the
molecular properties of the ATOP chromophore by optical,
electrooptical, NMR, and crystallographic methods as well
as semiempirical calculations revealed a highly delocalized
cyanine-type electronic system. In solvents of low polarity (such
as dioxane used for the EOA investigation) almost identical
Table 3. Selected Bond Distances for ATOP-10 in the Crystal
a
(Å)
atom 1
atom 2
L
atom 1
atom 2
L
C13
C17
N1
S1
N1
N1
C1
C1
C4
C2
C3
C4
C5
C6
1.469 (2)
1.470 (2)
1.334 (3)
1.744 (2)
1.769 (2)
1.417 (3)
1.356 (3)
1.420 (3)
1.375 (3)
1.417 (3)
C6
C7
C8
C8
C10
C9
N2
C11
C6
O2
C7
C8
C9
C10
N3
O1
C9
N2
C11
C11
1.427 (3)
1.380 (3)
1.447 (3)
1.428 (3)
1.150 (3)
1.229 (2)
1.380 (2)
1.377 (2)
1.442 (3)
1.243 (2)
contributions of the neutral and the zwitterionic canonical
resonance structures (Scheme 2) were observed, yielding a c2
S1
value of 0.46 characteristic of an almost perfectly balanced
polymethine chain. However, more polar environments such as
the solid state or the photoconducting PVK (see vide infra) may
further polarize the conjugated system and lead to the more
zwitterionic PR optimum suggested by eq 1.
C1
C2
C3
C4
C5
Materials Related Properties of ATOP Dyes. The ideal EO
chromophore for PR applications has to combine the highest
possible electrooptic response with other important properties
such as thermal, chemical, and photochemical stability as well
as electrical and structural compatibility with the photocon-
ducting polymer. The characterization and optimization of these
properties will be discussed in the following sections.
C3-C4-C5-C6-C7-C8-C9/C10 (Table 3), which reveal
almost equal bond lengths as expected for a cyanine-type
chromophore in accordance with the canonical structures given
in Scheme 2. For the thiophene ring we even note a higher
contribution of the quinonoid34 compared to the aromatic35
structure, i.e., the shorter carbon-carbon distance is observed
for the Câ-Câ bond and the longer distances are given for the
CR-Câ bonds (Table 3). Furthermore, the N1-C1 bond is
among the shortest bonds known between an amine and an
aromatic carbon and the methine bridge is almost symmetrical
with the shorter bond between the C4 and C5 carbon atoms.
This indicates that highly polar surroundings such as the bulk
solid-state polarize the ATOP chromophore toward a more
zwitterionic conjugated system.
(a) Thermal Behavior of Pure ATOP Dyes. The thermal
properties of ATOP dyes were investigated by thermal gravi-
metric analysis (TGA) and differential scanning calorimetry
(DSC). TGA gives information on the thermal stability of a
material, whereas DSC provides additional information on
melting enthalpies as well as glass formation capabilities.36 We
obtained decomposition temperatures (Td) of about 220 °C for
all ATOP dyes and melting temperatures, Tm, and melting
enthalpies, ∆Hm, as detailed in Table 4. A crude correlation
between the percentage of the aliphatic part and Tm is observed.
The parent chromophore ATOP-5 (all substituents ) methyl)
exhibits the highest melting point, while ATOP-4 with the
highest alkyl content of almost 50 wt % has the lowest Tm.
(
34) For crystallographic data on quinonoid dihydrothiophene-2,5-
diylidenes see: (a) Takahashi, K.; Suzuki, T.; Akiyama, K.; Ikegami, Y.;
Fukazawa, Y. J. Am. Chem. Soc. 1991, 113, 4576-4583. (b) Hieber, G.;
Hanack, M.; Wurst, K.; Str a¨ hle, J. Chem. Ber. 1991, 124, 1597-1605.
(35) For crystallographic data of oligothiophenes see: (a) Bak, B.;
Christensen, D.; Hansen-Nygaard, L.; Rastrup-Andersen, J. J. Mol.
Spectrosc. 1961, 7, 58-63. (b) Visser, G. J.; Heeres, G. J.; Wolters, J.;
Vos, A. Acta Crystallogr. Sect. B 1968, 24, 467-473. (c) Van Bolhuis, F.;
Wynberg, H.; Havinga, E. E.; Meijer, E. W.; Staring, E. G. J. Synth. Met.
(36) (a) W u¨ rthner, F.; Sens, R.; Etzbach, K.-H.; Seybold, G. Angew.
Chem. 1999, 111, 1753-1757; Angew. Chem., Int. Ed. Engl. 1999, 38,
1649-1652. (b) W u¨ rthner, F.; Yao, S.; Wortmann, R. J. Inf. Rec. 2000,
25, 69-86.
1
989, 30, 381-389.

ATOP Dyes
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2817
Table 4. Alkyl Content, Thermal Behavior during a First and a
Second Heating/Cooling Cycle in DSC Experiments, and
Solubilities in N,N-Dimethylaniline for ATOP Dyes
To probe the polarity of the photoconducting matrix polymer
we applied the solvatochromic method, which allows one to
determine the polarity of a microenvironment based on a probe
dye whose absorption maximum λag strongly depends on the
polarity of the surrounding matrix. A good choice for our
purpose is the dye 5-(dimethylamino)-5′-nitro-2,2′-bithiophene
39
M
alk/Mgesa
T
m
∆H
m
b
g
T (2nd heating) solubility
c
-
-1
(%)
(°C) (J g 1)
(°C)
(g L )
ATOP-5
ATOP-2
ATOP-10
ATOP-6
ATOP-7
ATOP-1
ATOP-3
ATOP-8
ATOP-9
ATOP-4
15
25
31
33
36
40
40
40
40
47
310
249
272
235
182
198
177
155
182
100
n.s.
n.s.
n.s.
n.s.
88
84
81
73
79
n.s.
n.s.
n.s.
n.s.
0.06
0.4
0.02
1.7
8
12
37
13
(
13), which shows a band shift from 466 nm (yellow) in unpolar
41
solvents up to 600 nm (blue) for solvents of high polarity.
This band shift was shown to depend mainly on the dielectric
properties (dipolarity and polarizability) of the medium. Thus,
dye 13 could be used to derive empirical solvent polarity
d
35
e
-
d
36
4
2
29
27
parameters (π* values) which by definition range from
d
41,42
34
cyclohexane (π* ) 0.0) to DMSO (π* ) 1.0).
Incorporation
57
16
>300
of a small amount of 13 into pure PVK resulted in λag ) 542
nm, corresponding to a π*-polarity value of 0.90 that indicates
a rather high micropolarity of the polymer PVK comparable to
a
Percentage of the alkyl part of the molecule to the total molecular
b
mass. n.s. ) not studied because of decomposition shortly after the
melting temperature. After cooling with a rate of 10 K min ; heating
rate 10 K min
of 10 K min recrystallization takes place during the cooling cycle.
Recrystallization at any cooling rate.
c
-1
43
the solvent DMF (π* ) 0.88). As expected, the polarity of
-
1 d
. Only observed after fast cooling. At a cooling rate
the PVK-based composites increases significantly with polar
additives as the azo dye DMNPAA (2, Chart 1). For 25 wt %
DMNPAA the absorption maximum was shifted to λmax ) 555
nm corresponding to a π* value of 1.03. Addition of ATOP
dyes should similarly increase the polarity. Unfortunately, the
strong absorption of ATOP at 500-550 nm obscured the
absorption band of the probe dye 13, preventing a detailed
analysis of the polarity of ATOP composites.
-
1
e
Remarkably, the latter was even difficult to isolate as a solid
because of its high solubility and vitrification tendency. The
significantly lower melting enthalpy of this dye (Table 4)
suggests a partly amorphous solid, which could be isolated from
an oil only after being cooled for several weeks in the
refrigerator. Therefore, it is not surprising that this dye did not
recrystallize from the melt in the DSC experiment and that a
glass-transition temperature of 16 °C in the second heating cycle
indicates the preferred formation of an amorphous material. A
similar DSC behavior was observed for ATOP-8 with a higher
The knowledge about the microscopic polarity could now be
used to test the dyes’ solubility properties in a suitable model
solvent. For the purposes of this work the natural choice was
N,N-dimethylaniline (π* ) 0.90), which shows similar structural
features as well as the same π*-polarity value as the PVK
polymer. Table 4 summarizes the observed solubilities of the
various ATOP dyes in this solvent. As already noted for the
melting behavior again a clear correlation between the mass
fraction of alkyl groups and the solubility is observed. The
solubilities vary over more than 4 orders of magnitude within
the given series. A closer inspection of the data in Table 4
reveals that branched alkyl chains at the imide group and a
nonsymmetrically substituted dialkylamino unit further improve
the solubility. These observations are rationalized by entropic
arguments because statistically distributed conformations have
to be ordered during the crystallization process. It is noteworthy
that ATOP-4 is soluble to such an extent in N,N-dimethylaniline
that highly viscous oils with more than 30 wt % of the dye are
stable at room temperature.
-
1
Tg of 29 °C. At faster cooling rates of >20 K min ATOP-3,
7, and -9 also gave glassy solids. However, in contrast to
-
ATOP-4, all other dyes recrystallized in a second heating cycle
indicating lower kinetic stability of their amorphous states. The
capability of forming amorphous glassy materials from the
melt recommend these dyes as promising candidates for PR
composites with long-term morphological stability.
(b) Solubility Studies. The stability of amorphous PR
composites consisting of ATOP dyes and the photoconductor
PVK against crystallization or phase separation is generally
controlled by kinetic and thermodynamic factors. In the ideal
case phase separation would be not only slowed for kinetic
3-7
reasons as in existing materials,
but thermodynamically
prevented by an unlimited miscibility of the components.
According to London’s theory dispersion interactions between
different molecules are maximized if they exhibit the same
(c) Electrochemical Properties. In addition to efficient
electrooptical response to the space-charge grating, high-
performance PR materials should have high charge carrier
mobilities. Typically this property is provided by photo-
conducting systems consisting of a photosensitizer and a
CTA, which upon illumination form mobile charges localized
on radical cationic species. Investigation of the redox behavior
of ATOP-1 by cyclic voltammetry in dichloromethane revealed
the reversible formation of radical cations at E ) + 0.50 V (vs
3
7
refractive indices. On a similar basis Hildebrand’s solubility
3
8,39
parameters
were derived. Both points of view suggest
that the highest miscibility as well as solubility is achieved
when the components exhibit similar dipole moments and
polarizabilities. Large differences in these properties should lead
to phase separation as observed for the majority of polymer
composites. In our system the ATOP dye has the higher values
for both dipolarity as well as polarizability. It seemed therefore
reasonable to reduce these values (for the given volume of
dye molecule) to get a closer match to PVK by introducing
appropriate amounts of less polar and less polarizable alkyl
chains.40
+
Fc/Fc ) and no reduction in the accessible potential range (up
+
to - 2 V vs Fc/Fc ). For comparison PVK or 9-ethylcarbazole
+
(
ECZ) are oxidized at about 0.77 V (vs Fc/Fc ) in a chemically
(
40) The refractive index of saturated alkanes is roughly 1.4 and thus
much smaller than the refractive index of PVK based organic PR materials
1.7).
(41) (a) Effenberger, F.; W u¨ rthner, F. Angew. Chem. 1993, 105,
(
(37) Israelachvili, J. N. Intermolecular and Surface Forces, 2nd ed.;
Academic Press: London, 1991; p 102.
742-744; Angew. Chem., Int. Ed. Engl. 1993, 32, 719-721. (b) Effenberger,
F.; W u¨ rthner, F.; Steybe, F. J. Org. Chem. 1995, 60, 2082-2091.
(42) Kamlet, M. J.; Abboud, J.-L. M.; Abraham, M. H.; Taft, R. W.
J. Org. Chem. 1983, 48, 2877-2887.
(43) Mecher, E.; Br a¨ uchle, C.; H o¨ rhold, H. H.; Hummelen, J. C.;
Meerholz, K. Phys. Chem. Chem. Phys. 1999, 2, 1749-1756.
(
38) (a) Small, P. A. J. Appl. Chem. 1953, 3, 71-80. (b) Kumar, R.;
Prausnitz, J. M. Solutions and Solubilities, Part 1; Dack, M. R. J., Ed.;
Wiley-Interscience: New York, 1975; pp 259-326.
(39) Reichardt, C. SolVents and SolVent Effects in Organic Chemistry,
2
nd ed.; VCH: Weinheim, 1990; pp 191-194.

2818 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
W u¨ rthner et al.
Table 5. Compositions of the Investigated Composites
a
F
DYE
PVK
ECZ
21
19
16
DPP
21
TNFM
10
20
30
40
50
47
41
37
31
25
1
1
1
1
1
19
16
14 (0 )
b
b
14 (20 )
b
b
12 (9 )
12 (9 )
a
b
Content of ATOP dye in wt % For ATOP-4 composites.
irreversible process. Thus, the ATOP chromophore is more
easily oxidized than the designated carbazole photoconductor,
suggesting that the ATOP molecules will act as hole traps in
the carbazole photoconducting manifold or might even perform
charge transport themselves (see below).
Electrooptical and Photorefractive Properties. (a) Pre-
sentation of the Studied Composites. In the following, we will
investigate PR composites with the chromophores ATOP-1,
ATOP-3, and ATOP-4 as electrooptically active components.
The latter were clearly identified as the most promising
chromophores on the basis of the solubility/compatibility study
presented in the preceding sections (Table 4). The other
components in the composites were the photoconducting
polymer poly(N-vinylcarbazole) (PVK), the plasticizers N-
ethylcarbazole (ECZ) and diphenylphthalate (DPP), and the
photosensitizer 2,4,7-trinitro-9-fluorenylidene-malononitrile
Figure 5. Absorption spectra of PR composites ATOP-4:PVK:ECZ:
DPP:TNFM containing different chromophore concentrations FDYE (20,
3
0, 40, 99, and 100 wt %). The main figure shows spectra determined
from thin films of about 1 µm thickness (the spectra of FDYE ) 99 and
00 wt % are identical within the experimental resolution); the inset is
1
a more detailed view for the NIR obtained from 37 µm thick PR films.
The dashed line indicates the laser wavelength used in the holographic
experiments.
(
TNFM). TNFM is commonly used for operating wavelengths
40 wt % dye. In films consisting of pure or 99 wt % ATOP-4
slightly more than 50% of the dye molecules are present as
antiparallel dimers.
6,7
in the near-infrared.
During the preparation of the various composites we discov-
ered that some (those with higher ATOP contents FDYE in
particular) showed a decrease in PR performance when they
were heated above 170 °C in air. By contrast, heating sealed
devices even to 190 °C did not lead to degradation. These
findings indicate that ATOP dyes decompose by oxidation at
elevated temperature in the presence of oxygen. To accomplish
processing under ambient conditions at lower temperatures
The charge-carrier generation efficiency of the PR composite
depends on the absorption coefficient at the given laser
frequency and the fraction of absorbed photons which are
converted into free charges that migrate and form a space-charge
field. Composites consisting exclusively of ATOP, PVK, ECZ,
and DPP show only weak absorbance at the operating wave-
length of the holographic experiments (λL ) 790 nm). Addition
of a small amount of the sensitizer TNFM, however, leads to
substantial absorption at λL. In all cases a higher absorption
coefficient was observed in the presence of ATOP dyes (R790
(<160 °C), the plasticizer content was chosen somewhat higher
7
than in previous work. The ATOP content (FDYE) was varied
between 10 and 50 wt % (Table 5); composites with higher
loading (e.g. >75 wt %) could not be prepared due to phase
separation of the components. The same composition was chosen
for all ATOP dyes except for ATOP-4 where lower amounts of
plasticizer had to be added to obtain similar Tg values than for
the other dyes. Nevertheless, for a given chromophore content,
the resulting Tg’s exhibit some variations, e.g. 14 °C for ATOP-1
and ATOP-3 composites and 8 °C for ATOP-4 composites
containing 40 wt % of the dye (for the Tg’s of the other
composites see the Supporting Information).
-1
44
)
62-87 cm for the 40 wt % composites) than for
-1
carbazole/TNFM (R790 ) 14 cm ), indicating the formation
of charge-transfer (CT) complexes between ATOP dyes and
TNFM similar to those reported for 2BNCM (3)/TNF and
DHADC-MPN (4)/TNFM. For the material ATOP-4:TNFM
9
6
-1
)
99:1 wt % an even larger value of R790 (111 cm ) and a
broad CT band are observed (Figure 5). This CT band also could
be observed in a UV-vis-NIR titration study where increasing
amounts of ATOP-4 were added to a 0.005 M solution of TNFM
in dioxane (see the Supporting Information). In view of this
intense absorption in the range of operation for common laser
diodes and Ti-sapphire lasers (around 800 nm), this complex
seems to be well-suited for photosensitization.
(b) Absorption Spectra. The absorption spectra were very
similar for all composites: a maximum absorption peak was
observed at ca. 550 nm, and a shoulder (for low ATOP loading)
or a second smaller peak (for high ATOP loading) appeared at
about 515 nm (Figure 5). In addition, as reported earlier for
ATOP-1,7b an increase of FDYE led to a sublinear increase of
the total absorbance. Both effects could be interpreted in terms
of dye aggregation driven by strong dipolar interactions between
the dye molecules,7b and dimer aggregates consisting of two
dye molecules in an antiparallel orientation could be character-
ized in low-polarity solvents21 (for spectra and determination
of dimerization free energies see the Supporting Information).
However, in the rather polar environment of the PVK polymer
dimer formation occurs only at rather high concentrations (i.e.,
in the percent range). From the peak ratios of the monomer
and the dimer bands of Figure 5 the amount of aggregated
ATOP-4 was estimated to raise from 40% for the composite
containing 20 wt % to about 45% for the composite containing
(c) Ellipsometry. The efficiency of the ATOP chromophores
to transform the molecular FOM into macroscopic refractive
index changes ∆n was studied by a DC ellipsometric (ELP)
45
technique, which yields information about the field-induced
Kerr nonlinearity without complication by the PR space-charge
field. A typical data set is displayed in Figure 6. Due to the
field-induced changes of the refractive index ∆n′ ) nP - nS2,
the transmission TELP shows an oscillatory behavior (sin
dependency; see eq 8 in the Experimental Section). For
(44) The R790 values for all the other composites are given in the
Supporting Information.
(
45) (a) Bittner, R.; Meerholz, K.; Br a¨ uchle, C. Appl. Opt. 1998, 37,
843-2851. (b) Bittner, R.; D a¨ ubler, T.; Neher, D.; Meerholz, K. AdV.
Mater. 1999, 11, 123-128.
2

ATOP Dyes
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2819
Figure 7. (A) Index change ∆n′ ) n
state transmission in ELP experiments and (B) index modulation for
p-polarized readout ∆n calculated from the steady-state diffraction
P
S
- n calculated from the steady-
P
efficiency in DFWM experiments on PR composites based on the
chromophores ATOP-1 (squares), ATOP-3 (circles), and ATOP-4
Figure 6. Typical field-dependent data obtained in an ELP
experiment: (A) transmission TELP and (B) field-induced index change
(triangles) for various chromophore concentrations FDYE. Inset: Data
for ATOP-4 only. The crossed area qualitatively symbolizes the contents
where ATOP-4 is nonmiscible with the PVK matrix. All data apply
∆
n′ calculated with eq 8 for the material containing 40 wt % ATOP-4.
The dashed line in part B corresponds to E ) 28.6 V µm at which
-1
-
1
for Eext ) 28.6 V µm
.
the ∆n′ data in Figure 7A were determined.
chromophore contents typically show a sublinear dependence
comparison of the various materials the field-induced index
of the performance (e.g. of the EO coefficient r33) on FDYE or
change ∆n′ was determined for a constant external electric field
even a performance optimum at intermediate dye contents.47
A
-1
of E ) 28.6 V µm (Figure 7A). It was found to be very similar
for all chromophores and to increase monotonically with FDYE.
This finding is in contrast to an earlier study where distinct
differences occurred between ATOP-1, ATOP-3, and ATOP-
possible explanation for our surprising results (despite dipolar
aggregation of ATOP, see above) is that the increase of the
matrix polarity at high dye loading FDYE reduces the relative
amount of aggregated dyes and furthermore polarizes the ATOP
chromophore beyond the cyanine limit, which leads to a
4
, and an optimum performance for some specific FDYE was
7c
observed. Both discrepancies are due to the elevated processing
temperatures (temporally >180 °C) used in our earlier study
which led to partial decomposition of the chromophore (in
particular at higher FDYE) and, thus, to a reduced nonlinearity
Kerr
significant increase of the figure-of-merit F0 . Indications for
such a matrix-assisted enhancement of the EO response were
already suggested by the bond-length alternations observed in
the crystalline state for ATOP-10, which pointed toward higher
(see also above). In the given study the processing temperature
2
c values in more polar surroundings. Nevertheless, the observed
was kept sufficiently low (<160 °C) to avoid such undesired
chemical reactions.
-
1
∆
n′ of about 0.0015 at E ) 25 V µm (Figure 6B) is still less
18a
than expected from the theoretical model based on independ-
Composites based on ATOP-4 yielded the largest ∆n′ for a
given FDYE possibly due to a ∼10 K lower Tg and thus higher
orientational mobility in the PVK matrix. The most striking
result of the ELP investigations however was that the depend-
ence of ∆n′ on FDYE is even slightly superlinear (see the inset
in Figure 7A for ATOP-4/TNFM). The so-called “oriented-gas
model” (OGM), which is well-known from second-order
nonlinear optical applications, would predict a linear dependency
ent chromophores which predict ∆n′ ≈ 0.005 for this field
(46) (a) Singer, K. D.; Sohn, J. E. Appl. Phys. Lett. 1986, 49, 248-250.
b) Singer, K. D.; Kuzyk, M. G.; Sohn, J. E. J. Opt. Soc. Am. B 1987, 4,
(
9
68-975. (c) Wu J. W. J. Opt. Soc. Am. B 1991, 8, 142-152.
(
47) (a) Dalton, L. R.; Harper, A. W.; Robinson, B. H. Proc. Natl. Acad.
Sci. U.S.A. 1997, 94, 4842-4847. (b) Robinson, B. H.; Dalton, L. R.; Harper,
A. W.; Ren, A.; Wang, F.; Zhang, C.; Todorova, G.; Lee, M.; Aniszfeld,
R.; Garner, S.; Chen, A.; Steier, W. H.; Houbrecht, S.; Persoons, A.; Ledoux,
I.; Zyss, J.; Jen, A. K. Y. Chem. Phys. 1999, 245, 35-50. (c) Yitzchaik,
S.; Di Bella, S.; Lundquist, P. M.; Wong, G. K.; Marks, T. J. J. Am. Chem.
Soc. 1997, 119, 2995-3002.
4
6,18a
of ∆n′ on FDYE assuming independent chromophores.
However, due to dipolar aggregation real systems with large

2820 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
W u¨ rthner et al.
index modulation amplitude ∆nP are displayed as a function of
the applied electric field E. Similarly to the transmission in an
ELP experiment, the diffraction efficiency ηP (defined in eq
1
0, Experimental Section) shows an oscillatory behavior in
agreement with eq 11, which allows calculating ∆nP from η.
For comparison of the various composites, ∆nP was determined
-1
for a constant external electric field of E ) 28.6 V µm (Figure
B). Unlike in ELP there are reproducible differences between
7
the different ATOP derivatives, in particular at high dye
concentrations. At the moment the origin of these differences
remains an open question because several factors are of
importance like variations in the Tg values, different aggregation,
and dye packing properties, as well as the presence of different
5
,49
numbers of trap sites.
Finally, a detailed characterization of the composite contain-
ing 40 wt % ATOP-3 was carried out (Figure 8). This material
proved to be morphologically stable and exhibits only slightly
reduced PR performance compared to the samples containing
higher ATOP contents of 50 wt %. In TBC experiments,
depending on the polarization of the writing beams, gain
-
1
-1
-1
coefficients of ΓP ) 50 cm (95 cm ) and ΓP ) 180 cm at
-
1
electric fields of |E| ) 30 (40) V µm were observed. For
-
1
comparison, the absorption coefficient was R ) 62 cm , i.e.,
-
1
net gain was observed for E > 33 (18) V µm for s
p)-polarized light, respectively. The leveling of ΓP observed
(
-
1
50
in Figure 8 for |E| > 30 V µm is related to beam fanning.
In the DFWM experiment the maximum of the diffraction
efficiency is reached at an electric field as low as E ) 23 V
-
1
µm for 105 µm thick films. The highest index modulation
-
1
-3
observed at E ) 40 V µm is ∆nP ) 7×10 . This performance
Figure 8. Field-dependent PR performance of the composite containing
0 wt % ATOP-3. (A) Gain coefficients Γ for p-polarized (open squares,
E < 0) and s-polarized write beams (solid squares, E > 0) calculated
according to eq 9 (see Experimental Section for details). Note that the
external field direction was opposite for the two polarizations. (B)
Internal diffraction efficiency for the p-polarized readout of a grating
written by s-polarized writing beams calculated according to eq 10 (solid
compares with PR composites containing the dye DHADC-MPN
4
(
4), which for identical FDYE and field strength exhibited a
-
3
dynamic range of ∆nP ≈ 4.4 × 10 ; however, at λL ) 633 nm
6b
large absorption losses occurred. At a wavelength more similar
to the one we used throughout our studies (λL ) 830 nm), the
index modulation amplitude of DHADC-MPN composites
-
3
6b
circles) and refractive index modulation ∆n
P
calculated according to
dropped to only ∆nP ≈ 1.5 × 10 (for FDYE ) 0.25).
eq 11 (open triangles). The dashed line indicates the field E ) 28.6 V
The response time of the grating buildup in DFWM was
found to increase with increasing ATOP content. For the lowest
-
1
P
µm for which the ∆n data in Figure 7B were determined.
-1
FDYE (10 wt %) we obtained τ50 ) 0.75 s for E ) 66.7 V µm ,
which is slightly slower than the response time of composites
containing 50 wt % of the dye DMNPAA (2) measured under
identical conditions. In all experiments the dynamics of the
reorientation of the chromophores as determined by ELP was
faster than the holographic response, suggesting that the PR
response was not limited by the speed of dye orientation.
Therefore, we attribute the observed time constant to the grating
formation time, which is slow due to the high polarity of
ATOP, its trapping character, and the possible presence of
additional deep trapping sites from dimeric species. Since
the focus of this work was to improve the dynamic range of
the index modulation by optimization of the chromophore,
no further studies were performed on the kinetic aspects. Of
course, for applications both aspects have to be considered
and thus in future work we plan to incorporate ATOP dyes
into photoconducting matrices with lower oxidation potentials
strength (estimated on the basis of the dipole moment and
polarizability anisotropy of ATOP and using Lorentz local field
factors). Accordingly, the observed experimental values are still
in the range of our expectations based on the molecular figure-
of-merit F0 and the given aggregation of ATOP dyes at higher
concentrations.
Kerr
(
d) Two-Beam Coupling and Degenerate Four-Wave
Mixing of Composites. The holographic characterization of
PR composites typically involves two-beam coupling (TBC)
1
experiments to prove the PR origin of the recorded gratings
and four-wave-mixing (FWM) experiments. The diffraction
efficiency η in the FWM experiment depends, besides geo-
metrical factors, mostly on the amplitude ∆n of the recorded
index grating. By contrast, the energy transfer coefficient Γ
(“gain”, eq 9, see the Experimental Section) in a TBC experi-
ment is determined by ∆n and the phase shift ú between the
4
8
light intensity pattern and the recorded index grating. The
holographic experiments were performed at a laser wavelength
of λL ) 790 nm, and the readout in the FWM experiment was
performed degenerately at the same wavelength (DFWM; see
the Experimental Section for details).
(e.g. based on the triarylamine photoconducting system). For
those materials it is expected that ATOP and ATOP dimers
will not act as traps, allowing considerably faster grating
formation times.
A typical data set is shown in Figure 8 where the gain
coefficients ΓP and ΓS, the diffraction efficiency ηP, and the
(49) Grunnet-Jepsen, A.; Wright, D.; Smith, B.; Bratcher, M. S.;
DeClue, M. S.; Siegel, J. S.; Moerner, W. E. Chem. Phys. Lett. 1998, 291,
5
53-561.
(48) The exact gain values at high fields depend on the samples’ history.
(50) Meerholz, K.; Bittner, R.; De Nardin, Y. Opt. Commun. 1998, 150,
Mecher, E.; Meerholz, K. To be submitted for publication.
205-209.

ATOP Dyes
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2821
another unambiguous proof for the formation of CT complexes
between ATOP and TNFM, leading to strongly enhanced
-1
absorption at 790 nm (increase from 8 to 100 cm after addition
of TNFM, see Figure 5). Most importantly, this CT complex
enables efficient charge generation at the operating wavelength
for the ATOP-4:TNFM 99:1 composite. The diffraction maxi-
mum in 105 µm thick devices (ηint ≈ 0.85, ηext ≈ 0.32) occurred
-
1
at only 10.5 V µm , which is the lowest value ever observed
for nonliquid-crystalline organic PR materials up to now. The
respective index modulation amplitude and gain coefficient at
-
1
-3
-1
E ) 28 V µm were ∆nP ) 10.7 × 10 and ΓP ) 130 cm .
For comparison, the first multifunctional amorphous low-molar-
mass compound 2BNCM (3 in Chart 1) gave more than 1 order
9
of magnitude smaller index modulation at the same electric field,
Kerr
in agreement with expectations based on the F0
2
()0.30 for
2
0c
BNCM ). The only organic PR materials with a comparable
dynamic range are polymer-dispersed liquid crystals (PDLC).
-
1
For PDLCs overmodulation was reported at about 8 V µm ,
-
3
and an index modulation amplitude of ∆nP ) 3.2 × 10 was
-
1
observed at E ) 22 V µm in devices of identical thickness.
51
The diffraction efficiency, however, was limited to about 25%
due to scattering of the PDLC droplets.
1
2c
Thus, amorphous films from ATOP-4 exhibit about a factor
three improvement of the index modulation amplitude over any
previous organic PR material at these low electric field values.
Comparison of our ELP and DFWM experiments indicates that
the ATOP-4 material has even more potential, since ∆nP/∆n′
(i.e., the ratio of the PR index modulation in DFWM compared
to field-induced nonlinearity in ELP) is much smaller for FDYE
)
0.99 than for FDYE ) 0.50. As a drawback, amorphous films
of ATOP-4 exhibit a rather low resistance to dielectric break-
down, limiting the range of electric fields at this point to E <
-
1
2
8 V µm . As a result of these low fields the response time
was rather slow (several tens of seconds). In addition, some
crystallization occurred within a period of several days, indicat-
ing that further chemical modification or addition of small
amounts of a binder is needed to obtain a material that is stable
against phase separation.
Conclusion
Beginning with the design of a chromophore with an
optimized polarizability anisotropy and ending with the
characterization of organic thin films of unprecedented PR
performance, this paper has described the complete process of
the development of a multifunctional material for holographic
optical applications. Various experimental techniques were
applied during the different stages. Optical and electronic
properties on the molecular level combined with stability
and morphological issues on the materials level were recognized
to be equally important in realizing high-performance bulk
functionality.
Figure 9. Field-dependence of the steady-state performance of
devices containing pure ATOP-4 (open symbols) or ATOP-4/TNFM
9
2
9:1 wt % (solid symbols), respectively: (A) gain coefficients in
BC experiments for s-polarized write beams (squares), (B) internal
diffraction efficiency for p-polarized readout of a grating written by
s-polarized writing beams calculated according to eq 10 (circles), and
P
(C) refractive index modulation ∆n in DFWM experiments calculated
according to eq 11 (up triangles), and refractive index change ∆n′
in ELP experiments calculated according to eq 8 (down triangles).
On the molecular level, a merocyanine dye was designed
that exhibits cyanine-type features, i.e., a well-conjugated chain
with equal bond orders of 1.5 along the push-pull system,
(e) TBC and DFWM of ATOP-4 Glasses. The glass-
intense and narrow absorption bands, a high polarizability
forming properties and the low Tg of ATOP-4 allowed us to
prepare PR devices using only the dye without additional
sensitizer or hole-conductor. These devices exhibited the PR
effect, but only with very limited performance (Figure 9, open
symbols). Nevertheless, this experiment demonstrated that
ATOP-4 may act as a trifunctional component (EO chromo-
phore, charge generator, and photoconductor).
Kerr
anisotropy, and most importantly a large figure-of-merit F0
.
On the materials level, it was demonstrated that morphological
properties such as compatibility, stability, and glass-transition
temperature can be controlled by a suitable choice of the
functional core which allows derivatization with plasticizing
alkyl groups at several positions. This approach bears the distinct
The steady-state and dynamic PR performance of devices
based on ATOP-4 could be strongly enhanced by adding TNFM
as a sensitizer (Figure 9, solid symbols). This finding constitutes
(
51) Figure 1 of ref 12c shows an optical loss of 50% at about 8 V µm-1.
-1
This corresponds to a loss coefficient of about 130 cm , yielding a
diffraction efficiency of 0.25 in 105 µm thick devices as used in this study.

2822 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
W u¨ rthner et al.
advantages of higher volume densities of the functional species
and the absence of any phase-separation problems compared
to the commonly used approach of incorporating the functional
dyes in a matrix of plasticizer and polymer. With regard to
the realization of amorphous bulk materials, the comparatively
high solubility (considering the strongly dipolar nature of
ATOP) and the appreciable thermal stability (for a methine
chromophore) are noteworthy properties which seem to be
related to the bent, but perfectly planar structure of ATOP.
1-Butyl-4-propyl-2,6-dioxo-1,2,3,6-tetrahydro-pyridine-3-carbo-
nitrile (11e): Ethyl cyanoacetate (5.66 g, 0.05 mol) was dropped into
o
n-butylamine (4.40 g, 0.06 mol) at 0 C within 15 min and the mixture
was allowed to warm slowly to room temperature and stirred for 12 h
to give amide 10b. Ethyl butyryl acetate (7.91 g, 0.05 mol) and
piperidine (5 mL) were added dropwise and the mixture was heated at
1
00 °C for 20 h. After the mixture was cooled to room temperature
the solvent was evaporated, the pH of the residue was adjusted to 1
with 32% aqueous HCl, and the precipitated product was filtered off,
washed with water and ether, and dried in vacuo to give 4.61 g of 11e
(
purity 98% from NMR, yield 39%). Recrystallization from ethyl acetate
Owing to the presence of four alkyl substituents given at the
ATOP core we could design dyes with high solubilities of more
1
gave an analytically pure sample: mp 208 °C; H NMR (200 MHz,
[D ]DMSO) δ 9.31 (br, 1H), 5.64 (s, 1H), 3.91 (t, J ) 7.4 Hz, 2H),
.49 (t, J ) 7.6 Hz, 2H), 1.56 (m, 4H), 1.28 (m, 2H), 0.91(m, 3H).
Anal. Calcd for C13 (234.3): C, 66.64, H, 7.74, N, 11.96.
Found: C, 66.76, H, 7.82, N, 11.96.
-Alkyl-5-dialkylamino-thiophen-2-ylmethylene-4-alkyl-2,6-dioxo-
,2,5,6-tetrahydro-pyridine-3-carbonitriles (ATOP dyes, general
-
1
6
than 300 g L in the PVK-like model solvent N,N-dimethyl-
aniline and a remarkable propensity for the formation of
amorphous solid state materials from the melt or casting from
solution. For optimized members of this dye family, i.e.,
ATOP-3 and ATOP-4, long-term morphologically stable PR
composites could be prepared which contained up to 50 wt %
of the electrooptically active dye in a photoconducting matrix
of PVK/ECZ/DPP. Photosensitization at the laser wavelength
of 790 nm was accomplished by addition of 1 wt % of the
electron acceptor TNFM, which was shown to form a CT
complex with PVK as well as with ATOP dyes. The composites
were studied by ellipsometric, degenerate four-wave-mixing, and
two-beam-coupling techniques. For all investigated ATOP
composites a continuous increase of the field-induced bire-
fringence (ELP) and the PR index modulation could be observed
with increasing dye content. However, while all composites of
a given chromophore content exhibited almost identical per-
formance in ELP, distinct differences were observed in DFWM.
2
18 2 2
H N O
1
1
procedure): Equimolar amounts of a dialkylaminothiophenaldehyde
8a-e and a hydroxypyridone 11a-f were heated in acetic anhydride
(0.5 mL per mmol of 8a-e) at 90-100 °C for 15-30 min. The deep
violet oil was allowed to cool to room temperature with stirring. The
slurry formed was diluted with 5-10 mL of 2-propanol and filtered
by suction to give a violet solid. After purification all dyes were dried
in vacuo at 40-60 °C.
ATOP-1 was synthesized from 3.32 g (12.5 mmol, 90% pure) of
c and 2.5 g (12.5 mmol) of 11b. The crude product was washed with
8
two portions of 5 mL of 2-propanol, purified by chromatography (silica
gel, eluent: toluene/AcOEt 7:3), and recrystallized from toluene: 4.27
g (80%) of a red solid, mp 198 °C; H NMR (200 MHz, CDCl ) δ
1
3
7.53 (s, 1H), 7.51 (d, J ) 5 Hz, 1H), 6.35 (d, J ) 5 Hz, 1H), 3.98 (t,
J ) 7 Hz, 2H), 3.53 (t, J ) 8 Hz, 4H), 2.48 (s, 3H), 1.69 (m
.40 (m
CH Cl
427.6): C, 67.41; H, 7.78; N, 9.83; S, 7.50. Found: C, 67.54; H, 7.67;
c
, 6H),
, 6H), 0.99 (t, J ) 7 Hz, 6H), 0.93 (t, J ) 7 Hz, 3H); UV/vis
) λmax nm (ꢀ) 538 (150 000). Anal. Calcd for C24
1
(
(
c
The composite material (ATOP-3:PVK:ECZ:DPP:TNFM 40:
33 3 2
H N O S
-
3
2
2
3
1:14:14:1) exhibited an index modulation of ∆nP ≈ 7 × 10
-
1
and a gain coefficient of ΓP ) 180 cm at electric fields of
only 30 V µm . Even higher index modulation efficiencies
N, 9.97; S, 7.27.
-
1
ATOP-3 was synthesized from 1.2 g (6 mmol, 95% pure) of 8b
and 1.6 g (6 mmol) of 11d in 5 mL of acetic anhydride. After the
product was washed with three 5 mL portions of 2-propanol, 1.8 g of
a violet solid was collected and purified by chromatography (toluene/
ethyl acetate from 9:1 to 6:4) to give 1.68 g (66%) of green crystals:
mp 177 °C, H NMR (200 MHz, CDCl
)
3
6
-
3
-1
(
∆nP ) 10.7 × 10 at 28 V µm ) were observed for films of
the pure low-Tg dye ATOP-4 sensitized with 1 wt % TNFM.
These experiments proved that ATOP dyes are truly multi-
functional π-conjugated systems combining electrooptical,
photoconducting, and glass-forming properties. Optimized multi-
functional PR systems based on ATOP dyes were demonstrated
to exhibit superior index modulation than any other organic PR
material including PDLC materials.
1
3 c
) δ 7.52 (m , 2H), 6.37 (d, J
5 Hz, 1H), 3.92 (d, J ) 7 Hz, 2H), 3.62 (q, J ) 7 Hz, 4H), 2.48 (s,
H), 1.88 (m , 1H), 1.37 (t, J ) 7 Hz, 6H), 1.28 (m , 8H), 0.89 (m
H). Anal. Calcd for C24 S (427.6): C, 67.41; H, 7.78; N, 9.83;
c
c
c
,
33 3 2
H N O
S, 7.50. Found: C, 67.19; H, 7.65; N, 9.72; S, 7.56.
ATOP-4 was synthesized from 2.4 g (10 mmol) of 8c and 2.6 g (10
mmol) of 11d in 6 mL of acetic anhydride. After the product was cooled
to room temperature a viscous oil was obtained that solidified during
stirring on an ice bath but redissolved on addition of 5 mL of
Experimental Section
Materials. 2-Brom-5-formylthiophene 7 was prepared according
2
2
to the literature. 1-Methyl- (11a), 1-butyl- (11b), 1-hexyl- (11c),
2
-propanol. The mixture was again concentrated to give an oil that
was purified by chromatography on silica with toluene/ethyl acetate
9:1 to 7:3) as eluent. After evaporation of the solvent an oil was
and 1-(2-ethylhexyl)-4-methyl-2,6-dioxo-1,2,3,6-tetrahydro-pyridine-
3-carbonitrile (11d) as well as 3-cyano-2,6-dihydroxy-4-methylpyridine
(
(11f) were donated by BASF. The characterization of all new
obtained that started to solidify during stirring on an ice bath. After
standing for several days in the refrigerator the product was collected
on a B u¨ chner funnel and washed with a little ether and several portions
1
compounds was accomplished by H NMR spectroscopy (Bruker AM
00 and AM 500), UV/vis spectroscopy (Perkin-Elmer Lambda 40P),
2
and elemental analysis. Melting points were measured with a B u¨ chi
SMP-20 and are uncorrected. All dyes used for PR experiments were
purified by chromatography and special care was taken to completely
remove a violet impurity of lower R value (silica; toluene/AcOEt) that
F
was considered to increase the absorption losses of the laser light.
of ether/pentane to give after drying 3.4 g (70%) of a red powder: mp
1
1
01 °C; H NMR (200 MHz, CDCl
3
) δ 7.51 (m
, 2H), 3.54 (t, J ) 8 Hz, 4H), 2.48 (s, 3H), 1.88 (m
H), 1.65-1.82 (m, 4H), 1.28-1.52 (m, 12H), 0.99 (t, J ) 7 Hz, 6H),
.90 (t, J ) 7 Hz, 6H); UV/vis (hexane) λmax nm (ꢀ) 531 (115 000);
) λmax nm (ꢀ)
539 (130 000); UV/vis (MeOH) λmax nm (ꢀ) 532 (135 000); UV/vis
(CH Cl ) λmax nm (ꢀ) 538 (150 000). Anal. Calcd for C28
c
, 2H), 6.37 (d, J ) 5
Hz, 1H), 3.91 (m
c
c
,
1
0
Synthesis. Dialkylaminoformylthiophenes 8a-e (General Pro-
UV/vis (ether) λmax nm (ꢀ) 535 (140 000); UV/vis (CCl
4
2
3
22
cedure): 2-Brom-5-formylthiophene 7 (0.1 mol, 19.1 g), the
respective amine (0.3 mol), and p-toluenesulfonic acid (0.5 g) were
heated at 100 °C for 24 h. The mixture was cooled and 150 mL of
water was added. After the mixture was stirred for an additional half
2
2
41 3 2
H N O S
(483.7): C, 69.53; H, 8.54; N, 8.69; S, 6.63. Found: C, 69.26; H, 8.54;
N, 8.64; S, 6.89.
hour the product was extracted with CH
2
Cl
2
. Drying with Na
2
SO
4
and
Single-Crystal X-ray Structure Determination of ATOP-10. Red
prisms of ATOP-10 were obtained from a hot saturated solution in
acetic acid by slow cooling to room temperature in a Dewar vessel.
The crystals belong to the monoclinic system with cell dimensions
a ) 10.9561(9) Å, b ) 8.4306(5) Å, c ) 21.382(2) Å, â ) 94.836(11)°,
evaporation of the solvent gave a dark oil that was chromatographed
on silica gel by elution with a 9:1 mixture of toluene/AcOEt to give
the dialkylaminothiophenes 8a-e in 50-70% yield at a purity of >90%
which was used for further reactions.

ATOP Dyes
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2823
3
and V ) 1967.9(3) Å . The space group is P2
1
/c (IT. No. 14) and Z )
Thermal gravimetric analyses (TGA) were carried out with a
Perkin-Elmer TGS 2 instrument at a heating rate of 10 K min in the
temperature range 300-700 K.
-
1
4
. The empirical formula is C20 S, the molar mass is 371.49,
H
25
N
3
O
2
-3
and the calculated density is 1.254 g cm . The three-dimensional X-ray
data were collected at 220 K by the use of graphite-monochromated
Mo KR radiation (λ ) 71.073 pm) on a Stoe IPDS diffractometer from
Solubility Studies. Concentrated dye solutions in N,N-dimethyl-
aniline were prepared by suspending an excess of the respective ATOP
dye in 1 mL of N,N-dimethylaniline to obtain a saturated solution and
subsequent filtering through 0.45 µm filters. The solutions were then
diluted and investigated by UV/vis spectroscopy. On the basis of the
2
θ ) 3.82-51.94°. The intensity data of 16401 reflections were
o o
collected and 3816 independent reflexes (all with |F | > 4σ|F |) were
used in the present X-ray analysis. The position of the sulfur atom
was located by using the Patterson method of SHELX97 and all non-
hydrogen atoms were located and refined by full-matrix least squares
-
1
-1
molar absorption coefficient of 132 000 L mol cm (determined for
three different ATOP dyes for the given solvent), the saturation
concentrations given in Table 4 were obtained.
2
52
against F with use of SHELX97. No absorption correction was
performed. Hydrogen atoms were included in the refinement by using
a riding model (1.2 times Uiso of the attached atom). Block-diagonal
least-squares refinements with anisotropic non-hydrogens and isotropic
hydrogens have converged to a conventional R-factor of 0.0405 and
Preparation of Photorefractive Devices. To obtain homogeneous
samples all components were taken up in CH Cl , the solvent was
2 2
evaporated, and the mixture was softened at temperatures below 160
°C. Devices were prepared by melt-pressing of the composites between
two ITO-coated glass substrates with use of glass spacer beads to adjust
the film thickness d.
UV/Vis Investigations of the Composites. The absorption co-
efficients in the near-infrared (NIR) were measured on 37 µm thick
glass/composite/glass devices with a Varian Cary 50 spectrometer with
a nonabsorbing reference cell. Full spectra were recorded from thin
films spin-coated from toluene:cyclohexanone 4:1 solution (d ≈ 1 µm).
The individual thicknesses were determined with a step profilometer
(Dectac 4).
2
.
2
2
a weighted R-factor of 0.0975 (w ) 1/[σ (F ) + (0.0642P) )], where
o
2
2
P ) (F
NMR spectra were recorded on Bruker AM 200 and AM 500
magnetic resonance spectrometers in CDCl or [D ]DMSO, using
TMS (0.0 ppm) and [D ]DMSO (2.49 ppm) as internal standards,
respectively. Dynamic NMR investigations were carried out at
00.13 MHz from 215 to 315 K and the temperature was calibrated
based on the separation of the highest and the third highest field peaks
o
+ 2F
c
)/3.
3
6
5
5
1
3
of the C carbons of neat 2-chlorobutane. For the evaluation of the
rotational barrier of the N-C bond a series of 7 experiments at
temperatures between 235 and 265 K separated by 5 K intervals were
used (see Figure 3) and the spectrum at 215 K was taken to define the
static parameters of the AA′BB′ spin system. For each temperature the
Ellipsometry (ELP). To study the poling process independently from
the PR grating formation, we determined the changes of the bulk
refractive index due to the induced birefringence and the electrooptic
45
effect by a DC ellipsometric (ELP) technique derived from the original
-
1
54
mean lifetime, τ ) k , was obtained by using the WIN-DYNAMICS
software package of Bruker-Franzen, version 1.0, where the computer-
generated line shapes are fitted to the experimental line shapes by
work of Teng/Man and Schildkraut. A laser beam (780 nm, 80 mW
-
2
cm ) was incident under an external angle of θext ) 60° with respect
to the sample normal. Its polarization was set to +45°, and the
transmitted light was probed through a -45° polarizer. To compensate
the phase shift not induced by the polymer, a Soleil-Babinet compensa-
tor was placed between the sample and the second polarizer. At zero
field the PR material did not affect the polarization state of the probe
beam (random orientation of the chromophores) and thus the second
polarizer blocked all the transmitted light. With the electric field applied
the transmission T through the second polarizer is given by
*
*
varying the rate constant k. The activation parameters ∆G , ∆H , and
*
∆
S
were obtained by linear regression analysis of a plot of log(k/T)
against 1/T according to the Eyring eq 7 with the transmission
coefficient set to 1.
UV/Vis, EOA, and Kerr Effect Studies. UV/vis absorption spectra
were recorded on a Perkin-Elmer Lambda 40P instrument at 293 K in
Merck Uvasol or Aldrich spectroscopic grade solvents at low concen-
trations (<10 mol L ) with 1 cm cells to determine the absorption
maxima λag and the molar decadic absorption coefficient ꢀ. EOA and
Kerr effect measurements were performed with devices described in
refs 25, 26, and 53, respectively. The experiments were carried out in
the solvent dioxane, which was purified as described in ref 25b and
carefully dried by distillation from Na/K in an argon atmosphere prior
to use.
Electrochemistry. Cyclic voltammetry was performed with a
computer-controlled EG & G PAR 273 potentiostat using a platinum
disk (1 mm diameter) working electrode, a platinum wire counter
electrode, and a Ag/AgCl wire as the reference electrode that was
internally calibrated vs the chemical redox couple of ferrocene/
-
5
-1
2
TELP ) sin (CELP‚∆n′)
(8)
where ∆n′ ) n
indices for p- and s-polarization of the incident light, respectively, and
ELP ) [2πd cos(θint)]/λ is a geometrical constant, d the sample
thickness, λ the laser wavelength, and θint the internal angle of the
P
S
- n is the difference between the bulk refractive
C
L
L
beam relative to the device normal (it was assumed that this angle is
identical for s- and p-polarized beams). ∆n′ was calculated from TELP
by using eq 8. The ELP experiments were performed at ambient
temperature of 21 ( 2 °C.
Holography. The holographic experiments were carried out at
ambient temperature in the typical tilted geometry by overlapping two
equally polarized writing beams “1” and “2” from a diode laser (Melles
+
ferrocenium (Fc/Fc ). The experiments were performed in dichloro-
methane, which was purified and dried by distillation from CaH prior
2
to use. Tetrabutylammonium hexafluorophosphate (0.1 M) was used
as the supporting electrolyte; it was recrystallized twice from methanol/
ethanol and dried in vacuo at 100 °C for several days. The dye
concentration was in the range of 1-2 × 10 mol L . After the
measurement of the sample a small amount of ferrocene was added
and an additional cycle was run for internal calibration.
Differential Scanning Calorimetry (DSC). DSC measurements
were performed on a Perkin-Elmer DSC 2 or a Mettler Toledo, model
21e. The glass-transition temperatures T
-3 mg samples by first heating the materials above their melting
temperatures (for pure dyes) or 180 °C (for composites) and then
cooling the liquid to -50 °C to equilibrate the materials' thermal history.
g
Then another heating-cooling cycle was carried out to determine T .
The heating rates are given in the text.
Griot, λ
L
) 790 nm, 28 mW) in devices with thickness d ) 105 µm.
-2
The external angles (intensities) were R1,ext ) 50° (I1, ext ) 4.5 W cm
and R2,ext ) 70° (I2, ext ) 5.4 W cm ). Assuming a refractive index of
)
-
2
-
3
-1
n
0
) 1.7 of the composites yields internal angles of R1,int ) 26.8° and
2,int ) 33.6°. For positive electric fields the positive electrode was
R
facing the writing beams. The field direction is important because of
“
beam fanning”55 resulting from energy transfer between a laser beam
and scattered light. In thin PR samples, this phenomenon leads to
coupling of fanned light into the polymer layer and ultimately to a
8
2
g
were determined with
loss of light (I
result unphysical gain values may be obtained.
1
+ I * constant) by wave-guiding phenomena. As a
2
50,56,57
To avoid this
(
54) Teng, C. C.; Man, H. T. Appl. Phys. Lett. 1990, 56, 1734-1736.
(b) Schildkraut, J. S. Appl. Opt. 1990, 29, 2839-2941.
55) (a) G u¨ nter, P.; Huignard, J. P. PhotorefractiVe Materials and their
(
(
52) (a) Sheldrick, G. M.; SHELXS-97, Program for the Solution of
Crystal Structures, G o¨ ttingen, 1997. (b) Sheldrick, G. M.; SHELXL-97,
Program for the Refinement of Crystal Structures, G o¨ ttingen, 1997.
Applications; Springer: Berlin, 1988 and 1989; Vols. 1 and 2. (b) Yeh, P.
Introduction to PhotorefractiVe Nonlinear Optics; Wiley: New York, 1993.
(56) Fischer, B.; Segev, M. Appl. Phys. Lett. 1989, 54, 684-686.
(57) Kokron, D.; Evanko, S.; Hayden, L. M. Opt. Lett. 1995, 20,
2297-2299.
(
45.
53) Briegleb, G.; Kuball, H. G. Z. Phys. Chem. N. F. 1965, 45, 339-
3

2824 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
W u¨ rthner et al.
problem, positive fields (E > 0) were used with s-polarized writing
beams and negative fields (E < 0) with p-polarized beams.50
The two-beam coupling (TBC) gain coefficient ΓS(P) for s- and
p-polarized write beams was calculated from the transmitted write beam
For determination of the response time in DFWM preexisting
gratings were erased by illuminating the sample uniformly by write
beam “1” for 10 min at zero field. The devices were then poled for 10
min by applying an electric field. During that time, the sample was
illuminated by beam “1” only. Finally, write beam “2” was switched
on by a magnetic shutter (opening time ≈ 1 ms). Data collection started
intensities I
1
2
and I according to:
1
0 ms prior to the switching process. The temporal resolution of the
1
d
I1(E * 0)
I2(E * 0)
59
setup was 0.5 µs.
ΓS(P)
)
cosa1,int ln
- cosa2,int ln
(9)
[
]
I1(E ) 0)
I2(E ) 0)
Acknowledgment. This work was supported by the VW
foundation, the Fonds der Chemischen Industrie and the BMBF
where d is the device thickness (105 µm).
(Liebig grant for F.W.), the DFG (Habilitation grant for F.W.,
In degenerate four-wave mixing (DFWM) the grating recorded by
2
research grant Wu 317/1-2 and Graduiertenkolleg “Molecular
Organization and Dynamics at Interfaces” at the University of
Ulm), the Bavarian government through the program “Neue
Materialien”, and BASF AG. We thank Prof. P. B a¨ uerle, Prof.
H.-U. Siehl, Dr. G. H o¨ hne, and A. Sautter (all University of
Ulm) as well as Prof. C. Br a¨ uchle, R. Bittner, and Y. De Nardin
s-polarized writing beams was read out by a weak (IR,ext ) 3 mW/cm )
p-polarized beam (λ ) 790 nm) counter-propagating to write beam
L
“
1”. The transmitted and diffracted beams were measured by two
independent detectors. The internal diffraction efficiency ηP,int was
calculated according to:
(
all University of Munich) for fruitful discussions and their
ηP,int ) IR,diffr/(IR,diffr + IR,transm
)
(10)
support.
This definition eliminates the field-dependent absorption and reflection
losses and allows a more straightforward evaluation of the data.
Supporting Information Available: Further experimental
details on the synthesis of ATOP dyes, cyclic voltammetry,
dimerization studies of ATOP as well as ATOP-TNFM complex
formation studies both studied in dilute solution; the glass
transition temperatures and absorption coefficients of all PR
composites; and the single-crystal X-ray structure determination
of ATOP-10 tables of atomic positional parameters, bond
distances, bond angles, anisotropic thermal parameters, hydrogen
atom positions, and data collection parameters (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
58
According to the coupled-wave theory,
ηP,int can be approximated by:
2
ηP,int ≈ sin (CFWM∆nP)
(11)
where CFWM ) π d/[λ
L
cos(R1,int)] is a geometrical constant and ∆n
P
the amplitude of the PR index modulation.
All field-dependent steady-state data were obtained by gradually
increasing the electric field in steps of 250 V. For each step, the data
were taken after this field had been applied for sufficiently long time
(
typically 2-5 min) to allow the devices to reach quasi-steady-state
JA002321G
conditions.
(59) Mecher, E.; Bittner, R.; Br a¨ uchle, C.; Meerholz, K. Synth. Met. 1999,
(58) Kogelnik, H. Bell Syst. Technol. J. 1969, 2909-2947.
102, 993-996.