Supramolecular approaches to second-order nonlinear optical materials. Self-assembly and microstructural characterization of intrinsically acentric [(aminophenyl)azo]pyridinium superlattices

Article abstract of DOI:10.1021/ja960395f

This paper describes the synthesis and characterization of self-assembled second-order nonlinear optical multilayer materials containing the high-hyperpolarizability [(aminophenyl)azo]pyridinium chromophore. The chromophoric multilayers are assembled on c

Full text of DOI:10.1021/ja960395f

8034
J. Am. Chem. Soc. 1996, 118, 8034-8042
Supramolecular Approaches to Second-Order Nonlinear Optical
Materials. Self-Assembly and Microstructural Characterization
of Intrinsically Acentric [(Aminophenyl)azo]pyridinium
Superlattices
Wenbin Lin,† Weiping Lin,‡ George K. Wong,‡ and Tobin J. Marks*,†
Contribution from the Department of Chemistry, the Department of Physics and Astronomy, and
the Materials Research Center, Northwestern UniVersity, EVanston, Illinois 60208-3113
ReceiVed February 7, 1996X
Abstract: This paper describes the synthesis and characterization of self-assembled second-order nonlinear optical
multilayer materials containing the high-hyperpolarizability [(aminophenyl)azo]pyridinium chromophore. The
chromophoric multilayers are assembled on clean glass or single-crystal silicon substrates via the following iterative
reaction sequence: (1) treatment with 4-ClCH2C6H4SiCl3, 3-BrC3H6SiCl3, or 3-IC3H6SiCl3 to afford a self-assembled
coupling layer; (2) quaternization of 4-[[4-[N,N-bis(hydroxylethyl)amino]phenyl]azo]pyridine in a “topotactic” fashion
to generate an [(aminophenyl)azo]pyridinium chromophore layer; (3) cross-linking with octachlorotrisiloxane to afford
a capping layer (which also regenerates surface hydroxyl groups for subsequent layer deposition). The new
chromophore precursor 4-[[4-[N,N-bis(hydroxylethyl)-amino]phenyl]azo]pyridine was synthesized by diazotization
of 4-aminopyridine followed by coupling with N-phenyldiethanolamine and has been characterized by elemental
analysis, mass spectrometry, and infrared, UV-vis, and NMR spectroscopies. The chromophoric multilayers have
been characterized by X-ray photoelectron and transmission optical spectroscopies, spectroscopic ellipsometry, X-ray
reflectivity, advancing contact angle measurements, and polarized second harmonic generation (SHG). The excellent
structural regularity of the chromophoric multilayers is indicated by the linear dependence of the [(aminophenyl)-
azo]pyridinium chromophore longitudinal HOMO f LUMO charge-transfer excitation absorbance at 572 nm and
the ellipsometry- and X-ray reflectivity-derived multilayer thicknesses on the number of assembled trilayers, while
the uniform polar order of the stacked chromophoric multilayers is evidenced by the quadratic dependence of the
second harmonic generation intensities on the number of trilayers. The [(aminophenyl)azo]pyridinium multilayers
are photochemically stable, have very high structural regularities, and exhibit a second-order nonlinear optical
susceptibility (ø(2)) of ∼3.6 × 10-7 esu (∼150 pm/V) at a fundamental wavelength of 1064 nm. Finally, atomic
force microscopy reveals that the surfaces of the self-assembled multilayers are very smooth (root mean square
roughness ) 12 Å for a 10 trilayer sample), undoubtedly reflecting the high three-dimensional structural regularities
of the individual layers.
Introduction
centrosymmetric architecture.3 Therefore, noncentrosymmetric
organization of high molecular hyperpolarizability constituent
chromophores is an essential requirement for efficient bulk
second-order nonlinear optical materials, and the construction
of such acentric supramolecular assemblies presents a daunting
challenge to conventional synthetic methodologies.4 Of these,
poled polymer2b,c and Langmuir-Blodgett (LB) film transfer
approaches5 have been most commonly employed to prepare
second-order nonlinear optical materials. Although considerable
progress has been made in the fabrication of noncentrosymmetric
assemblies via the above two approaches in recent years,
significant synthetic challenges still remain and have thus far
impeded the realization of optimum materials.2b,c,5 Molecular
self-assembly is an attractive alternative approach to second-
order NLO materials and targets the construction of covalently
linked, intrinsically acentric superlattices containing molecular
chromophoric subunits.6,7
There is great current fundamental scientific and technological
interest in the synthesis and properties of molecule-based
materials having large second-order optical nonlinearities
because they present special conceptual synthetic challenges and
because of their potential applications in second harmonic
generation (SHG), electrooptic, and photorefractive devices.1
Molecule-based second-order nonlinear optical materials offer
many attractions, such as large nonresonant response, ultrafast
response times, low dielectric constants/losses, and intrinsic
architectural tailorability.2 Because second-order nonlinear
optical susceptibility is a third rank tensor, the effect vanishes
if the constituent molecular chromophores are arranged in a
† Department of Chemistry.
‡ Department of Physics and Astronomy.
X Abstract published in AdVance ACS Abstracts, August 1, 1996.
(1) (a) Moerner, W. E.; Silence, S. M. Chem. ReV. 1994, 94, 127-155.
(b) Dalton, L. R.; Harper, A. W.; Ghosn, R.; Steier, W. H.; Ziari, M.;
Fetterman, H.; Shi, Y.; Mustacich, R. V.; Jen, A. K.-Y.; Shea, K. J. Chem.
Mater. 1995, 7, 1060-1081. (c) Dalton, L. R.; Harper, A. W.; Wu, B.;
Ghosn, R.; Laquindanum, J.; Liang, Z.; Hubbel, A.; Xu, C. AdV. Mater.
1995, 7, 519-540. (d) Marder, S. R.; Sohn, J. E.; Stucky, G. D. Materials
for Nonlinear Optics; ACS Symposium Series 455; American Chemical
Society: Washington, DC, 1991.
The field of molecular self-assembly has witnessed tremen-
dous growth in synthetic sophistication and depth of charac-
(3) Prasad, P. N.; Williams, D. J. Introduction to Nonlinear Optical
Effects in Molecules and Polymers; John Wiley & Sons: New York, 1991;
Chapter 4.
(4) Lehn, J.-M. Angew. Chem., Int. Ed. Engl. 1988, 27, 89-112.
(5) (a) Ashwell, G. J.; Jackson, P. D.; Crossland, W. A. Nature 1994,
368, 438-440. (b) Penner, T. L.; Matschmann, H. R.; Armstrong, N. J.;
Ezenyilimba, M. C.; Williams, D. J. Nature 1994, 367, 49-51. (c) Ashwell,
G. J.; Hargreaves, R. C.; Baldwin, C. E.; Bahra, G. S.; Brown, C. R. Nature
1992, 357, 393-395.
(2) (a) Kanis, D. R.; Ratner, M. A.; Marks, T. J. Chem. ReV. 1994, 94,
195-242. (b) Marks, T. J.; Ratner, M. A. Angew. Chem., Int. Ed. Engl.
1995, 34, 155-173 and references therein. (c) Burland, D. M.; Miller, R.
D.; Walsh, C. A. Chem. ReV. 1994, 94, 31-75.
S0002-7863(96)00395-2 CCC: $12.00 © 1996 American Chemical Society

Acentric [(Aminophenyl)azo]pyridinium Superlattices
J. Am. Chem. Soc., Vol. 118, No. 34, 1996 8035
terization over the past decade.8 Numerous self-assembled
systems have been investigated, including disulfides and thiols
on gold,9 carboxylic acids on metal oxides,10 trichloro- and
trialkoxysilanes on oxides,11 and phosphonates on metal phos-
phonate surfaces.7,12 Spontaneous and sequential adsorption of
appropriately derivatized adsorbates onto substrates in a self-
limiting fashion can yield thin films with uniform polar orders
in individual layers and therefore represents an attractive
approach to the construction of intrinsically acentric chro-
mophoric superlattices. Unfortunately, while the routes to self-
assembled monolayers are very well developed, few efficient
routes to structurally regular self-assembled multilayers have
been devised.13 Among them, self-assembled siloxane and metal
phosphonate systems are promising candidates for constructing
acentric superlattices. Several years ago, we reported synthetic
routes to the first SHG-active multilayer structures using an
attractive covalent siloxane self-assembly approach; highly
regular and highly nonlinear (ø(2) ∼ 200 pm/V at ω0 ) 1064
nm) self-assembled stilbazolium (I) multilayers were constructed
in a regular, layer-by-layer fashion.6a Katz et al. later showed
that acentric chromophoric multilayers could also be obtained
using a zirconium phosphate/phosphonate self-assembly ap-
proach.7
Recently, we reported a new “topotactic” pathway for the
preparation of self-assembled stilbazolium (I) and [(aminophe-
nyl)ethynyl]pyridinium (II) multilayers with greatly improved
synthetic efficiency, structural regularity, and second-order
nonlinear optical response.14 A full account of the synthesis,
microstructures, and NLO response characteristics of these
materials will be presented elsewhere.14d Our studies also
indicated that both the chemical and photophysical properties
of these self-assembled materials strongly depend on the
chromophore molecular architectures. For example, the stil-
bazolium monolayers not only exhibit packing geometries very
different from those of the [(aminophenyl)ethynyl]pyridinium
monolayers15 but also exhibit ø(2) responses up to 10× larger
than those of the latter.6c,15 Therefore, it would be of great
interest to investigate the properties of self-assembled super-
lattices in which the chromophore conjugative pathway, hyper-
polarizability, shape, and potential chemical/photochemical
reactivity were further modified. To this end, we have designed
a new [(aminophenyl)azo]pyridinium chromophore (III) for
incorporation into siloxane self-assembled multilayer structures.
This azo chromophore can potentially exhibit additional ad-
vantages such as enhanced thermal, oxidative, and photochemi-
cal stability. For example, previous studies have suggested that
diphenylazo NLO chromophores are more thermally stable than
the stilbene analogs.16 In addition, the [(aminophenyl)azo]-
pyridinium molecules should be resistant to degradative [2 +
2] cycloaddition pathways that are potential complications in
some stilbazolium systems and that would result in diminished
NLO activity.17 We present in this paper the synthesis of the
new [(aminophenyl)azo]pyridine chromophore precursor and its
efficient self-assembly into [(aminophenyl)azo]pyridinium su-
perlattices. We report detailed characterization of these self-
assembled [(aminophenyl)azo]pyridinium mono- and mutlilayers
using X-ray photoelectron and transmission optical spec-
troscopies, spectroscopic ellipsometry, X-ray reflectivity, ad-
vancing contact angle measurements, and polarized second
harmonic generation (SHG) measurements. It will be seen that
these readily synthesized superlattices exhibit very high second-
order nonlinear response and high structural regularity as well
as high chemical/photochemical stability.18
(6) (a) Li, D.; Ratner, M. A.; Marks, T. J.; Zhang, C.; Yang, J.; Wong,
G. K. J. Am. Chem. Soc. 1990, 112, 7389-7390. (b) Yitzchaik, S.; Roscoe,
S. B.; Kakkar, A. K.; Allan, D. S.; Marks, T. J.; Xu, Z.; Zhang, T.; Lin,
W.; Wong, G. K. J. Phys. Chem. 1993, 97, 6958-6960. (c) Roscoe, S. B.;
Yitzchaik, S.; Kakkar, A. K.; Marks, T. J.; Lin, W.; Wong, G. K. Langmuir
1994, 10, 1337-1339. (d) Yitzchaik, S.; Marks, T. J. Acc. Chem. Res.
1996, 29, 197-202.
Experimental Section
(7) (a) Katz, H. E.; Scheller, G.; Putvinscki, T. M.; Schilling, M. L.;
Wilson, W. L.; Chidsey, C. E. D. Science 1991, 254, 1485-1487. (b) Katz,
H. E.; Wilson, W. L.; Scheller, G. J. Am. Chem. Soc. 1994, 116, 6636-
6640.
(8) (a) Ulman, A. An Introduction to Ultrathin Organic Films; Academic
Press: New York, 1991; Part 3. (b) Swalen, J. D.; Allara, D. L.; Andrade,
J. D.; Chandross, E. A.; Caroff, S.; Israelachvili, J.; McCarthy, T. J.; Murray,
R.; Pease, R. F.; Rabolt, J. F.; Wynne, K. J.; Yu, H. Langmuir 1987, 3,
932-950. (c) Dubois, L. H.; Nuzzo, R. G. Annu. ReV. Phys. Chem. 1992,
43, 437-463. (d) Bell, C. M.; Yang, H. C.; Mallouk, T. E. In Materials
Chemistry: An Emerging Discipline; Interrante, L. V., Casper, L. A., Ellis,
A. B., Eds.; Adv. Chem. Ser. 245; American Chemical Society: Washington,
D.C., 1995; pp 212-230.
(9) (a) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481-
4483. (b) Nuzzo, R. G.; Fusco, F. A.; Allara, D. L. J. Am. Chem. Soc.
1987, 109, 2358-2368. (c) Bain, C. D.; Whitesides, G. M. Angew. Chem.,
Int. Ed. Engl. 1989, 28, 506-512. (d) Bain, C. D.; Troughton, E. B.; Tao,
Y.-T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989,
111, 321-335.
(10) (a) Allara, D. L.; Atre, S. V.; Elliger, C. A.; Snyder, R. G. J. Am.
Chem. Soc. 1991, 113, 1852-1854. (b) Allara, D. L.; Nuzzo, R. G.
Langmuir 1985, 1, 45-52.
(11) (a) Wasserman, S. R.; Tao, Y.-T.; Whitesides, G. M. Langmuir 1989,
5, 1074-1087. (b) Maoz, R.; Sagiv, J. Langmuir 1987, 3, 1034-1044. (c)
Pomerantz, M.; Segmuller, A.; Netzer, L.; Sagiv, J. Thin Solid Films 1985,
132, 153-162.
(12) (a) Zeppenfeld, A. C.; Fiddler, S. L.; Ham, W. K.; Klopfenstein, B.
J.; Page, C. J. J. Am. Chem. Soc. 1994, 116, 9158-9165. (b) Cao, G.;
Hong, H.-G.; Mallouk, T. E. Acc. Chem. Res. 1992, 25, 420-425. (c)
Thompson, M. E. Chem. Mater. 1994, 6, 1168-1175.
(13) Thoden van Velzen, E. U. Ph.D. Thesis, University of Twente, 1994,
Chapter 1.
The synthesis, purification, and characterization of the reagents (3-
iodopropyl)trichlorosilane (3-IC3H6SiCl3) and 4-[[4-[N,N-bis(hydrox-
ylethyl)amino]phenyl]azo]pyridine can be found in the supporting
information.
(14) (a) Lin, W.; Yitzchaik, S.; Lin, W.; Malik, A.; Durbin, M. K.;
Richter, A. G.; Wong, G. K.; Dutta, P.; Marks, T. J. Angew. Chem., Int.
Ed. Engl. 1995, 34, 1497-1499. (b) Yitzchaik, S.; Lin, W.; Marks, T. J.;
Lin, W.; Wong, G. K. Polym. Mater. Sci. Eng. 1995, 72, 217-218. (c)
Lin, W.; Marks, T. J.; Yitzchaik, S.; Lin, W.; Wong, G. K. Mat. Res. Soc.
Sympos. Proc., in press. (d) Lin, W.; Malinsky, J. E.; Marks, T. J.; Lin,
W.; Wong, G. K. Manuscript in preparation.
(15) Roscoe, S. B.; Kakkar, A. K.; Marks, T. J.; Malik, A.; Durbin, M.
K.; Lin, W.; Wong, G. K.; Dutta, P. Langmuir, in press.
(16) (a) Moylan, C. R.; Swanson, S. A.; Walsh, C. A.; Thackara, J. I.;
Twieg, R. J.; Miller, R. D.; Lee, V. Y. Nonlinear Optical Properties of
Organic Materials VI. In SPIE Proc. 1993, 2025, 192-201. (b) Twieg,
R. J.; Burland, D. M.; Hedrick, J. L.; Lee, V. Y.; Miller, R. D.; Moylan, C.
R.; Volksen, W.; Walsh, C. A. Mat. Res. Soc. Sympos. Proc., in press.
(17) (a) Zhao, X.-M.; Perlstein, J.; Whitten, D. G. J. Am. Chem. Soc.
1994, 116, 10463-10467. (b) Takagi, K.; Shibata, S.; Oguri, S.; Sawaki,
Y.; Suzuoki, Y.; Segi, T. Chem. Lett. 1993, 2103-2106. (c) Usami, H.;
Takagi, K.; Sawaki, Y. J. Chem. Soc., Faraday Trans. 1992, 88, 77-81.
(d) Takagi, K.; Suddaby, B. R.; Vadas, S. L.; Backer, C. A.; Whitten, D.
G. J. Am. Chem. Soc. 1986, 108, 7865-7867. (e) Quina, F. H.; Whitten,
D. G. J. Am. Chem. Soc. 1977, 99, 877-883.
(18) After completion of this work, a communication on Langmuir-
Blodgett films based on lanthanide salts of the (E)-1-methyl-4-[[4-
(dihexadecylamino)phenyl]azo]pyridinium chromophore appeared. See:
Gao, L. H.; Wang, K. Z.; Huang, C. H.; Zhao, X. S.; Xia, X. H.; Li, T. K.;
Xu, J. M. Chem. Mater. 1995, 7, 1047-1049.

8036 J. Am. Chem. Soc., Vol. 118, No. 34, 1996
Lin et al.
Scheme 1. Synthesis and Purification of the Chromophore Precursor 4-[[4-[N,N-Bis(hydroxylethyl)amino]phenyl]azo]pyridine, 1
Construction of Chromophoric Superlattices. Sodium lime glass,
quartz, or silicon wafer substrates were cleaned by immersion in
“piranha” solution at 80 °C for 1 h. After being cooled to room
temperature, they were rinsed repeatedly with deionized water and then
subjected to an RCA-type cleaning protocol (room temperature, 40 min).
They were then washed with deionized water and dried under vacuum
immediately before coupling agent deposition.
(i) Self-Assembly of 4-ClCH2C6H4SiCl3. Under inert atmosphere,
freshly cleaned glass or silicon single crystal substrates were immersed
in a 1:100 (v/v) solution of 4-ClCH2C6H4SiCl3 in heptane for 20 min,
washed with pentane, and sonicated in acetone for 1 min. The
3-BrC3H6SiCl3 and 3-IC3H6SiCl3 coupling layers were chemisorbed
similarly.
(ii) Self-Assembly of the 4-[[4-[N,N-Bis(hydroxylethyl)amino]-
phenyl]azo]pyridinium Chromophore. The silylated substrates (after
step i treatment) were spin-coated at 4000 rpm with a 5 mM solution
of 4-[[4-[N,N-bis(hydroxylethyl)amino]phenyl]azo]pyridine in methanol
inside a class 100 clean hood (Envirco) and then heated at 110 °C in
a vacuum oven (0.4 Torr) for 13 min. The samples were cooled to
room temperature and then washed with methanol and acetone to
remove any residual 1.
(iii) Self-Assembly of Octachlorotrisiloxane. The glass and silicon
single crystal substrates after step i and step ii treatment were immersed
in a 1:150 (v/v) solution of octachlorotrisiloxane in heptane for 30 min,
washed with pentane, and sonicated in acetone for 1 min.
Physical Measurements. Full details of the physical measurements
are given in the supporting information. Polarized second harmonic
generation measurements were made in the transmission mode using
the 1064 nm output of a Q-switched Nd:YAG laser operated at 10 Hz
with a pulse width of 3 ns. The details of this setup can be found
elsewhere.6b Contact angles were measured on a standard goniometric
bench fitted with a Teflon micrometer syringe. X-ray photoelectron
spectra were recorded on a VG Scientific Ltd., Escalab MK II
instrument. Details of the spectral acquisition and data analysis
techniques are described elsewhere.6c X-ray reflectivity measurements
were performed at Beam Line X23B of the National Synchrotron Light
Source. Data acquisition and analysis procedures are described
elsewhere.19 Ellipsometric data were obtained on a Sopra MOSS ES
4 G spectroscopic ellipsometer. The data were modeled in the
multilayer regression mode with the number of layers set to 2; a
refractive index of 1.7 was assumed for the self-assembled multilayers,
while the literature values of the refractive indices for silicon and silicon
oxide were used for these materials.20 Atomic force microscopic images
were recorded using a Nanoscope II microscope with A and D scanners
(Digital Instruments, Inc.). All of the AFM samples were prepared on
Si(100) single crystal substrates having a root mean square (rms)
roughness of 2.0 Å.
Results
(1) Synthesis of 4-[[4-[N,N-Bis(hydroxylethyl)amino]phe-
nyl]azo]pyridine, 1. The new chromophore precursor 4-[[4-
[N,N-bis(hydroxylethyl)amino]phenyl]azo]pyridine was prepared
by diazotization of 4-aminopyridine to form a 4-pyridyldiazo-
nium intermediate followed by coupling with N-phenyldietha-
nolamine (Scheme 1). Because the 4-pyridyldiazonium inter-
mediate has very limited stability in solution,21 it was generated
in situ and immediately coupled with N-phenyldiethanolamine
to form 1 in modest yields. The product 4-[[4-[N,N-bis-
(hydroxylethyl)amino]phenyl]azo]pyridine, however, is difficult
to separate from the N-phenyldiethanolamine starting material.
To facilitate this separation, the mixture was treated with tert-
butyldimethylchlorosilane, which derivatized the hydroxyl
groups to siloxy functionalities.22 The resulting 4-[[4-[N,N-bis-
[(tert-butyldimethylsiloxy)ethyl]amino]phenyl]azo]pyridine, 2,
can be easily separated from N-phenyl[bis[(tert-butyldimethyl-
siloxy)ethyl]]amine by chromatography on silica gel with a
mixture of hexane and ethyl acetate as the eluant. Compound
2 can be readily transformed back to chromophore precursor 1
by treatment with tetra-n-butylammonium fluoride in tetrahy-
drofuran. Both 1 and 2 have been characterized by elemental
analysis, mass spectrometry, and infrared, UV-vis, and NMR
spectroscopies (see the supporting information for details).
Spectroscopic data for 1 and 2 are consistent with those reported
for the related known compound, 4-[[4-(N,N-dimethylamino)-
phenyl]azo]pyridine.23 The synthesis and purification of 1 are
summarized in Scheme 1.
(2) Preparation and Characterization of Self-Assembled
[(Aminophenyl)azo]pyridinium Monolayers. As illustrated
in Scheme 2, the construction of [(aminophenyl)azo]pyridinium
monolayers is based on chemisorptive siloxane self-assembly
strategies first developed by Sagiv.11 Cleaned, hydroxylated
substrates such as glass or single-crystal silicon wafers were
(19) Shih, M. C.; Peng, J. B.; Huang, K. G.; Dutta, P. Langmuir 1993,
9, 776-782.
(20) Collins, R. W.; Allara, D. L.; Kim, Y.-T.; Lu, Y.; Shi, J. In
Characterization of Organic Thin Films; Ulman, A., Ed.; Butter-
worth-Heinemann: Boston, 1995; pp 35-55.
(21) Kalatzis, E. J. Chem. Soc., B 1967, 273-277.
(22) Greene, T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic
Synthesis; 2nd ed.; John Wiley & Sons: New York, 1991; pp 77-81.
(23) (a) Kubota, H.; Idei, M.; Motomizu, S. Analyst 1990, 115, 1109-
1115. (b) Pentimalli, L. Tetrahedron 1960, 9, 194-201.

Acentric [(Aminophenyl)azo]pyridinium Superlattices
J. Am. Chem. Soc., Vol. 118, No. 34, 1996 8037
Scheme 2. Construction of Self-Assembled 4-[[4-[Bis(hydroxylethyl)amino]phenyl]azo]pyridinium multilayersa
a Conditions: step i, 1/100 (v/v) [4-(chloromethyl)phenyl]trichlorosilane in dry heptane for 20 min; step ii, spin-coated from 5 mM CH3OH
solution at 4000 rpm and baked at 110 °C/0.4 Torr for 13 min; step iii, 1/150 (v/v) octachlorotrisiloxane in dry heptane for 30 min.
first treated with 4-ClCH2C6H4SiCl3 to afford a monolayer of
coupling agent. The [(aminophenyl)azo]pyridinium chro-
mophore layer was deposited upon this coupling layer via the
quaternization of [(aminophenyl)azo]pyridine. Finally, the
chromophoric layer was planarized/crosslinked with octachlo-
rotrisiloxane; this step also regenerates hydroxyl groups on
which subsequent layers can be iteratively constructed.
(a) Self-Assembled Coupling Layers. Under rigorously
anhydrous/anaerobic conditions, the freshly cleaned substrates
were treated with a 1:100 (v/v) solution of 4-ClCH2C6H4SiCl3
in dry heptane for 20 min, followed by washing with copious
amounts of pentane and acetone to remove physisorbed coupling
agent. This treatment (step i, Scheme 2) is known to afford a
densely packed, self-assembled monolayer of benzylic chloride
coupling agent on the surface.15 The self-assembled coupling
layer was characterized by advancing aqueous contact angle (θa)
measurements and ellipsometry. The contact angle changes
from 15° for SiO2 to 70° for the silylated surface. This
hydrophobic change in θa is consistent with the presence of
benzylic chloride functionalities on the surface.24 Ellipsometric
measurements reveal a thickness of ∼10 Å for the self-
assembled coupling agent.25 This thickness agrees well with
that of a self-assembled 4-ClCH2C6H4SiCl3 monolayer reported
by Koloski et al.26 and is consistent with that expected for a
chemisorbed monolayer of 4-ClCH2C6H4SiCl3 from molecular
modeling.
In order to further characterize the self-assembled monolayers
using X-ray photoelectron spectroscopy (XPS), monolayers were
also prepared using 3-BrC3H6SiCl3 and 3-IC3H6SiCl3 as cou-
pling agents.27 XPS spectra of the 3-BrC3H6SiCl3-derived
coupling layer films exhibit a Br 3d signal at 70.5 eV (Figure
Figure 1. (A) Br 3d XPS spectra of self-assembled 3-BrC3H6SiCl3
1); this binding energy is consistent with an alkyl bromide
coupling and [(aminophenyl)azo]pyridinium chromophore monolayers.
Bottom spectrum: after coupling agent deposition. Top spectrum: after
(24) Li, D.; Neumann, A. W. In Characterization of Organic Thin Films;
coupling agent and chromophore deposition. (B) I 3d XPS spectra of
self-assembled 3-IC3H6SiCl3 coupling and [(aminophenyl)azo]pyri-
dinium chromophore monolayers. Bottom spectrum: after coupling
agent deposition. Top spectrum: after coupling agent and chromophore
deposition.
Ulman, A., Ed.; Butterworth-Heinemann: Boston, 1995; pp 165-192.
(25) All the ellipsometry-derived thicknesses reported in this contribution
are, within experimental error, consistent with those derived from inde-
pendent X-ray reflectivity measurements. Lin, W.; Malik, A.; Malinsky,
J. E.; Durbin, M. K.; Dutta, P.; Marks, T. J. Manuscript in preparation.
(26) Koloski, T. S.; Dulcey, C. S.; Haralson, Q. J.; Calvert, J. M.
Langmuir 1994, 10, 3122-3133.
(27) Our previous XPS studies6c on ion exchange processes of self-
assembled stilbazolium monolayers showed that the complicated Cl 2p line
shape resulting from overlap of the Cl 2p1/2 and Cl 2p3/2 peaks can
complicate the deconvolution of Cl 2p signals and therefore obscure shifts
in Cl 2p binding energies.
moiety.28 Interestingly, no Cl 2p signal is present in the
spectrum, suggesting that the labile Si-Cl bonds have been
completely hydrolyzed to form densely packed siloxane linkages
on the surface. Similarly, the XPS spectra of the corresponding

8038 J. Am. Chem. Soc., Vol. 118, No. 34, 1996
Lin et al.
propyl iodide (3-IC3H6SiCl3) coupling layer films exhibit an I
3d5/2 signal at 620.3 eV, which is also consistent with an alkyl
iodide functionality. Thus, these XPS results further support
the formation of self-assembled coupling layers.
(b) Self-Assembled [(Aminophenyl)azo]pyridinium Chro-
mophore Layer. The deposition of the [(aminophenyl)azo]-
pyridinium chromophore layers has been achieved in a “topo-
tactic” fashion (see the Discussion section).14 The substrates
terminated with a coupling layer were spin-coated with a thin
layer (∼500 Å thick) of 4-[[4-[N,N-bis(hydroxylethyl)amino]-
phenyl]azo]pyridine, followed by brief heating in vacuo to effect
quaternization and removal of excess chromophore precursor 1
(110 °C/0.4 Torr). These samples were then washed with
methanol and acetone to ensure the complete removal of excess
1 on the surfaces. This procedure yields a self-assembled
monolayer of the [(aminophenyl)azo]pyridinium chromophore
layer via the quaternization of [(aminophenyl)azo]pyridine with
the benzylic chloride functionality. This [(aminophenyl)azo]-
pyridinium self-assembly process has been monitored by
transmission optical spectroscopy, advancing aqueous contact
angle measurements, ellipsometry, polarized second harmonic
generation, and XPS.
After step ii, the films exhibit three optical absorption maxima
at 278, 450, and 573 nm. The new absorption maximum at
573 nm can be assigned to a longitudinal HOMO f LUMO
charge transfer excitation in the quaternized [(aminophenyl)-
azo]pyridinium chromophore.29 This result suggests that 1
undergoes quaternization with the benzylic chloride to form a
self-assembled monolayer of the 4-[[4-[N,N-bis(hydroxylethyl)-
amino]phenyl]azo]pyridinium chromophore on the surface. This
conclusion is further supported by the advancing aqueous contact
angle of the surface, which shows the expected hydrophilic
change with θa falling to 40° upon [(aminophenyl)azo]pyri-
dinium chromophore deposition. Consistent with this observa-
tion, ellipsometric measurements also show an increase in film
thickness from ∼10 to ∼18 Å after quaternization step ii.
Figure 2. SHG intensity as a function of fundamental light (λ0 ) 1064
nm) incident angle from a glass slide surface having a self-assembled
[(aminophenyl)azo]pyridinium monolayer on both sides. The coupling
agent is 4-ClCH2C6H4SiCl3.
provides further information on the quaternization process: the
two peaks at 620.3 and 618.2 eV have an intensity ratio of ∼1:
9. This indicates that ∼90% of the propyl iodide functionalities
comprising the coupling layer have undergone reaction with 1
to form the anchored 4-[[4-[N,N-bis(hydroxylethyl)amino]-
phenyl]azo]pyridinium chromophore on the surface. The pres-
ence of the 4-[[4-[N,N-bis(hydroxylethyl)amino]phenyl]azo]-
pyridinium chromophore on the surface can be further inferred
from the N 1s XPS spectrum, which shows a broad unsym-
metrical peak centered around 400.5 eV. The N 1s signal can
be deconvoluted into three peaks with binding energies of 399.4,
400.5, and 402.0 eV.30 These peaks are assigned to the amino,
azo, and pyridinium nitrogen atoms, respectively.28
The most convincing evidence for the deposition of aligned
4-[[4-[N,N-bis(hydroxylethyl)amino]phenyl]azo]pyridinium chro-
mophore molecules comes from polarized second harmonic
generation measurements, which show a large increase in the
second harmonic response in the quaternized, bilayer film. Such
a large second harmonic response can only be explained by the
presence of the high-â 4-[[4-[N,N-bis(hydroxylethyl)amino]-
phenyl]azo]pyridinium chromophore chemisorbed onto the
surface.29 Figure 2 shows angle-dependent SHG response data
from a sample coated with the 4-[[4-[N,N-bis(hydroxylethyl)-
amino]phenyl]azo]pyridinium chromophore on both sides of the
glass substrate, which were prefunctionalized with the
4-ClCH2C6H4SiCl3 coupling agent. Qualitatively similar SHG
angle dependence was observed for bilayer samples with the
3-BrC3H6SiCl3 and 3-IC3H6SiCl3 coupling agents. The char-
acteristic SHG interference pattern in Figure 2 is due to the
phase difference between two SHG waves generated at either
side of the sample. The presence of near-zero intensity minima
in Figure 2 indicates that the quality and uniformity of the
monolayers on both sides of the glass substrate are nearly
identical. This result provides indirect evidence for the good
structural regularity of the chromophoric monolayers prepared
with the present self-assembly technique. Furthermore, the
experimental angle-dependent SHG data can be analyzed
according to literature methods.31 The solid curve in Figure 2
XPS spectra of the chromophoric monolayers with n-propyl
bromide and n-propyl iodide coupling layers provide further
evidence for the chemisorptive deposition of the 4-[[4-[N,N-
bis(hydroxylethyl)amino]phenyl]azo]pyridinium chromophore
on the surface. Thus, the [(aminophenyl)azo]pyridinium chro-
mophore monolayers with the 3-BrC3H6SiCl3 and 3-IC3H6SiCl3
coupling agents exhibit optical spectra and SHG responses (Vide
infra) similar to those of the chromophoric monolayers with
the 4-ClCH2C6H4SiCl3 coupling layer. In bilayer samples
prepared with 3-BrC3H6SiCl3 as the coupling agent, the XPS
spectrum shows a Br 3d binding energy of 68.0 eV upon the
quaternization of 1 (Figure 1A). This 2.5 eV shift to lower
binding energy is consistent with the formation of Br- anions.28
However, overlap of the Br 3d3/2 and Br 3d5/2 peaks precludes
the deconvolution of the Br 3d signals, and therefore, it is not
straightforward to quantitatively determine the extent of the
quaternization process. This difficulty was overcome using
3-IC3H6SiCl3 as the coupling agent. Here, the I 3d5/2 XPS
spectrum of the 3-IC3H6SiCl3 coupling layer exhibits two distinct
peaks upon quaternization (Figure 1B). While the minor peak
has a binding energy of 620.3 eV, the major peak has shifted
to a lower binding energy of 618.2 eV. These two peaks can
be assigned to unquaternized propyl iodide and quaternized
iodide anion, respectively. The integration of these two peaks
shows the fit to ø(z2z)z/ø(2) ) 3. Assuming a unidimensional
zyy
(30) We assume that the N1s binding energies of the two azo N atoms
of the [(aminophenyl)azo]pyridinium chromophore are sufficiently similar
so that only three N1s peaks are expected. We employed the binding
energies and linewidths (FWHM) of the N1s peaks of the amino and
pyridinium nitrogen atoms of self-assembled stilbazolium monolayers in
deconvolution of the present XPS data.14c
(31) (a) Ashwell, G. J.; Hargreaves, R. C.; Baldwin, C. E.; Bhara, G. S.;
Brown, C. R. Nature 1992, 357, 393-395. (b) Lupo, D.; Prass, W.;
Scheunemann, U.; Laschewsky, A.; Ringsdorf, H. J. Opt. Soc. Am. 1988,
5B, 300-308.
(28) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D.
Handbook of X-ray Photoelectron Spectroscopy; Perkin Elmer: Eden Praire,
MN, 1992.
(29) ZINDO/SOS calculations show that the HOMO f LUMO charge
transfer excitation red-shifts by ∼100 nm as 1 undergoes quaternization to
calc
vec
form the [(aminophenyl)azo]pyridinium chromophore.
â
) 984 ×
10-32 cm5 esu-1 (λo ) 1064 nm) by the ZINDO/SOS formalism.

Acentric [(Aminophenyl)azo]pyridinium Superlattices
J. Am. Chem. Soc., Vol. 118, No. 34, 1996 8039
from standing wave X-ray fluorescence measurements33 as well
as with that predicted by molecular modeling for a regular self-
assembled bilayer. This surface coverage also compares favor-
ably with that determined for LB films containing similar
stilbazolium chromophores.31a The calculated chromophore
surface density of the present system is significantly larger than
that derived for the self-assembled stilbazolium bilayers prepared
using earlier solution phase deposition techniques (∼1.1 × 1014
molecules/cm2).6b This result suggests that the “topotactic”
pathway provides greater structural regularity/packing densities
in the self-assembled [(aminophenyl)azo]pyridinium monolayers
so that a greater number of chromophore molecules can be
accommodated per unit area.
(d) Self-Assembled Capping Layers. The final (capping)
step of the deposition sequence (step iii in Scheme 2) involves
treatment of the chromophoric bilayers with octachlorotrisilox-
ane. This capping step provides lateral structural stabilization/
planarization via interchromophore cross-linking and also
regenerates hydroxyl groups for the subsequent iterative super-
lattice assembly process. The deposition of capping layers
yields the expected hydrophilic change in θa; the advancing
aqueous contact angle of the surface decreases from 40° to 17°
after capping layer chemisorption. This result is consistent with
the regeneration of surface hydroxyl groups as a result of the
capping layer deposition. Ellipsometric measurements show that
the thickness of the film after capping layer deposition has
increased to ∼25 Å.
Figure 3. In situ monitoring of the quaternization/chromophore
chemisorption step ii in Scheme 2 by transmission second harmonic
generation at λ0 ) 1064 nm. The coupling agent is 4-ClCH2C6H4-
SiCl3. Temperature ramp: 60-150 °C, 1 °C/min; 150-200 °C, 1.5
°C/min.
chromophore (i.e., with a single dominant â tensor component)
and minimal dispersion, the average orientation angle of the
chromophore dipoles with respect to the surface normal (tilt
angle) can then be calculated from the expression ø(z2z)z/ø(2) ) 2
zyy
cot2 ψj. The tilt angles for the present self-assembled chro-
mophoric superlattices fall in the range of 38-42°, comparing
favorably with those calculated for the self-assembled and
Langmuir-Blodgett films of a related stilbazolium chro-
mophore.6,31
(3) Synthesis and Characterization of Self-Assembled
[(Aminophenyl)azo]pyridinium Multilayer Films. The con-
struction of self-assembled multilayers involves repetitive
deposition of the aforementioned coupling, chromophore, and
capping layers (Scheme 2). The present “topotactic” approach
has very high efficiency: the total reaction time for constructing
a complete trilayer is about 1 h. The degree to which the present
self-assembly approach produces a regular superlattice without
loss of acentricity (randomization of chromophore orientation)
and diminution of SHG efficiency was investigated as a function
of the number of added layers (steps i-iii) by ellipsometry,
X-ray reflectivity, transmission optical spectroscopy, and SHG.
Figure 4 shows the dependence of ellipsometry- and X-ray
reflectivity-derived multilayer thicknesses on the number of
trilayers. The superlattice thicknesses exhibit a smooth, mono-
tonic increase as a function of the total number of assembled
layers. The ellipsometric measurements give a thickness of ∼25
Å for each trilayer, while the X-ray reflectivity measurements
give a trilayer thickness of 24.2 ( 1 Å. The inset of Figure 4
shows the X-ray reflectivity data for a 10-trilayer self-assembled
[(aminophenyl)azo]pyridinium multilayer film. The closely
spaced minima in the data are due to the interference between
the reflections from the film-substrate and film-air interfaces,
from which the thickness of the 10-trilayer film is determined
to be 240 ( 1 Å. This thickness is further supported by the
appearance of the Bragg peak at Kz ) 0.24 Å-1, which is due
to scattering from individual layers; the interlayer spacing
calculated from this peak position is 24 ( 1 Å. The detailed
analyses of these X-ray reflectivity data will be reported
elsewhere.19,25 Considering the differences in radiation wave-
length as well as in analysis assumptions and procedures, the
ellipsometric and X-ray reflectivity film thickness results can
be considered to be in good agreement. The linear dependence
of multilayer thicknesses (determined from both spectroscopic
ellipsometry and X-ray reflectivity) on the number of assembled
In order to better understand the chromophore self-assembly
process, the quaternization step (step ii, Scheme 2) was also
studied using an in situ SHG technique. In this experiment,
glass substrates that had been coated with self-assembled
4-ClCH2C6H4SiCl3 coupling layers were spin-coated with thin
layers of 1 and placed in the path of the Nd:YAG laser. The
second-harmonic generation responses of these samples were
then measured in situ while the temperature was ramped at
various rates. Figure 3 shows that the onset temperature for
chemisorptive chromophore formation is ∼110 °C. The quat-
ernization of 4-[[4-[N,N-bis(hydroxylethyl)amino]phenyl]azo]-
pyridine by the benzylic chloride functionality is quite facile
under these conditions. Figure 3 also shows that the quater-
nization process is complete by ∼140 °C and that the resulting
monolayer is stable up to 190 °C in air under these conditions.
(c) Chromophore Surface Density Calculations. By cali-
brating the SHG data against quartz, a bulk second-order
nonlinear susceptibility ø(z2z)z of 5 × 10-7 esu is obtained for the
self-assembled [(aminophenyl)azo]pyridinium bilayers. Chro-
mophore surface densities (Ns) in these bilayers can then be
estimated from the surface NLO susceptibilities and the
calcd
zzz
calculated molecular hyperpolarizability â
(at 1064 nm)
using eq 1.6b,c,32 The surface susceptibility (ø(s2))
is the
(1)
(ø(s2))
) Ns cos3 ψh âcalcd
zzz
(2)
zzz
product of bulk nonlinear optical susceptibility ø and the
calcd
zzz
bilayer thickness. Using the ZINDO-derived â
value of
983.70 × 10-32 esu at 1064 nm, an average tilt angle of 40°,
and an estimated bilayer thickness of 18 Å, the chromophore
surface density is calculated to be ∼2.0 × 1014 molecules/cm2.
This value corresponds to an average area (“footprint”) of 49
Å2/molecule, which is in agreement with experimental results
(33) We have recently measured the surface Br coverage of the self-
assembled stilbazolium monolayers using standing wave X-ray fluorescence
techniques at NSLS. The surface Br coverage is on the order of 40-50
Å2/atom. Lin, W.; Lee, T.-L.; Lyman, P. F.; Lee, J.; Bedzyk, M. J.; Marks,
T. J., submitted for publication.
(32) Heinz, T. F.; Tom, H. W. K.; Shen, Y. R. Phys. ReV. A 1983, 28,
1883-1885.

8040 J. Am. Chem. Soc., Vol. 118, No. 34, 1996
Lin et al.
introduced in step iii. This level of second-order nonlinearity
is comparable to or greater than those of the most optically
nonlinear siloxane and zirconium phosphonate/phosphate self-
assembled chromophoric multilayers reported to date.6,7,14 The
SHG efficiency of the self-assembled [(aminophenyl)azo]-
pyridinium multilayers also rivals those of the most efficient
poled polymers and LB films reported to date.1,2,5
We also carried out initial photostability studies on the self-
assembled [(aminophenyl)azo]pyridinium multilayers. It is
found that the second harmonic response of the [(aminophenyl)-
azo]pyridinium multilayers exhibits negligible decay after
exposure to fluorescent light for over 1 month. It is reasonable
that the photostability is mainly a result of the resistance of the
azo linkage to degradative [2 + 2] cycloaddition or oxidative
processes that would yield SHG-inactive products.17
(4) Atomic Force Microscopy Studies. Atomic force
microscopy has proven to be a powerful technique for determin-
ing molecular scale ordering of Langmuir-Blodgett films34 and
self-assembled monolayers.35 The microstructural regularity of
the monolayers can be inferred from the surface morphologies
of these samples. To our knowledge, however, there have been
only two atomic force microscopic studies of self-assembled
metal phosphonate multilayers,36 and no AFM study on self-
assembed siloxane multilayers has been reported. This is
presumably due to the lack of synthetic approaches to these
structures. The present multilayer systems provide an interesting
opportunity for atomic force microscopic investigations of
siloxane-based self-assembled multilayers.
Figure 4. Ellipsometry- and X-ray reflectivity-derived thicknesses for
the self-assembled chromophoric multilayers of Scheme 2 as a function
of the number of trilayers. The inset shows the X-ray reflectivity data
for a 10-trilayer self-assembled chromophoric multilayer. Each trilayer
is composed of the sequence steps i-iii. The coupling agent is
4-ClCH2C6H4SiCl3.
Figure 6A shows an AFM image of a 4-ClCH2C6H4SiCl3 self-
assembled monolayer. The coupling layer surface is essentially
featureless with an rms roughness of 3.5 Å for a scan area of 1
× 1 µm2. In comparison, the AFM image of the cleaned Si-
(100) substrate (not shown) exhibits an rms roughness of 2.0
Å. In contrast to the growth of self-assembled long chain
n-alkyltrichlorosilane films,35a the 4-ClCH2C6H4SiCl3 mono-
layers prepared by the present procedure give no evidence by
AFM of island growth, and the film morphologies do not appear
to depend on the reaction times. This discrepancy is probably
because the growth rate for 4-ClCH2C6H4SiCl3 monolayers is
very rapid. For example, a complete layer of 4-ClCH2C6H4-
SiCl3 can be deposited in less than 20 min under the present
conditions as judged from the SHG response; in contrast, we
find that a complete layer of 4-ClCH2C6H4C2H4SiCl3 can only
be achieved after 24 h reaction.
Figure 5. Optical absorbance at 572 nm and square root of the 532
nm second harmonic generation intensity (λ0 ) 1064 nm) for the self-
assembled chromophoric multilayers of Scheme 2 as a function of the
number of trilayers. Each trilayer is composed of the sequence steps
i-iii. The coupling agent is 4-ClCH2C6H4SiCl3.
trilayers is consistent with the uniform buildup of the chro-
mophoric superlattices.
An AFM image for a self-assembled [(aminophenyl)azo]-
pyridinium bilayer prepared with a 4-ClCH2C6H4SiCl3 coupling
layer is shown in Figure 6B. The film is also almost featureless,
and the rms roughness has increased only slightly to 4.0 Å.
After the deposition of the octachlorotrisiloxane capping layer,
the films show essentially the same surface morphology with
an rms roughness of ∼5 Å. The most impressive image,
however, is that shown in Figure 6C that illustrates the surface
morphology of a 10-trilayer (30 layer) self-assembled [(ami-
Further evidence for the buildup of regular superlattice layers
comes from the linear dependence of the 572 nm optical
absorbance on the number of assembled trilayers (Figure 5).
The absorption maximum at 572 nm is assigned to the
[(aminophenyl)azo]pyridinium chromophore longitudinal HOMO
f LUMO charge-transfer excitation.29 This linear dependence
suggests that equal amounts of chromophore are deposited in
constructing each trilayer and that a uniform chromophore
orientation is maintained in the self-assembled multilayers.
A final test of superlattice regularity comes from the SHG
response as a function of the number of trilayers. Because the
film thicknesses are small compared to the wavelength of the
incident light, the intensity of the SHG light from a regular
multilayer structure should scale quadratically with the number
of trilayers. The observed linear dependence of the square root
of the SHG intensity on the number of trilayers indicates both
structural regularity in layer thicknesses as well as uniform
alignment of the chromophore molecules (Figure 5). From the
slope, ø(2) ∼ 3.6 × 10-7 esu (∼150 pm/V) can be deduced (we
take the chromophoric trilayer thickness to be 25 Å); this is
slightly lower than the aforementioned monolayer result (Vide
supra) due to the added thickness imparted by the siloxane layer
(34) (a) Meyer, E.; Howald, L.; Overney, R. M.; Heinzelmann, H.;
Frommer, J.; Guntherodt, H.-J.; Wagner, T.; Schier, H.; Roth, S. Nature
1991, 349, 398-400. (b) Weisenhorn, A. L.; Egger, M.; Ohnesorge, F.;
Gould, S. A. C.; Heyn, S.-P.; Hansma, H. G.; Sinsheimer, R. L.; Gaub, H.
E.; Hansma, P. K. Langmuir 1991, 7, 8-12.
(35) (a) Bierbaum, K.; Grunze, M.; Baski, A. A.; Chi, L. F.; Schrepp,
W.; Fuchs, H. Langmuir 1995, 11, 2143-2150. (b) Caldwell, W. B.;
Campbell, D. J.; Chen, K.; Herr, B. R.; Mirkin, C. A.; Malik, A.; Durbin,
M. K.; Dutta, P.; Huang, K. G. J. Am. Chem. Soc. 1995, 117, 6071-6081.
(c) Fujii, M.; Sugisawa, S.; Fukada, K.; Kato, T.; Shirakawa, T.; Seimiya,
T. Langmuir 1994, 10, 984-987. (d) Alves, C. A.; Smith, E. L.; Porter,
M. D. J. Am. Chem. Soc. 1992, 114, 1222-1227.
(36) (a) Yang, H. C.; Aoki, K.; Hong, H.-G.; Sackett, D. D.; Arendt, M.
F.; Yau, S.-L.; Bell, C. M.; Mallouk, T. E. J. Am. Chem. Soc. 1993, 115,
11855-11862. (b) Byrd, H.; Snover, I. L.; Thompson, M. E. Langmuir
1995, 11, 4449-4453.

Acentric [(Aminophenyl)azo]pyridinium Superlattices
J. Am. Chem. Soc., Vol. 118, No. 34, 1996 8041
demonstrate that the present multilayers have very smooth
surfaces as a result of the high structural regularity within each
individual layer.
Discussion
Efficient “Topotactic” Deposition of the [(Aminophenyl)-
azo]pyridinium Chromophore. Our previous studies on the
solution phase deposition of stilbazolium chromophores indi-
cated that the rate-limiting step in the multilayer deposition
process is the chromophore precursor quaternization step (step
ii, Scheme 2). For example, a complete layer of the stilbazolium
chromophore can only be deposited after 100 h of reaction at
typical reagent concentrations and 60 °C. Moreover, long
reaction times can result in significant side reactions of the
chromophore precursor. Adsorption of these contaminants onto
the coupling agent-terminated surfaces results in films with
diminished structural regularity and second-order optical non-
linearity. The efficient “topotactic” approach reported here
dramatically increases the rate of the chromophore deposition
and results in self-assembled chromophoric films with greatly
enhanced structural regularity and second-order optical nonlin-
earity.
In contrast to the previous slow solution phase deposition of
the stilbazolium chromophore, the present assembly of the
[(aminophenyl)azo]pyridinium chromophore illustrates the great
advantage of the “topotactic” multilayer self-assembly approach
(which is also applicable to the stilbazolium system).14 In this
new methodology, the substrates terminated with coupling layers
are spin-coated with a thin layer of chromophore precursor 1
and then briefly heated at 110 °C under vacuum. UV-vis, XPS,
SHG, advancing contact angle measurements, and thickness data
all indicate that a complete layer of [(aminophenyl)azo]-
pyridinium chromophore can be deposited in each cycle of this
procedure. We hypothesize that the ease of the quaternization
reaction of 1 is a result of several factors including higher
effective concentration of 1 on the surface, higher deposition
temperatures, and adequate mobility of 1 on the surface.
Because chromophore preccursor 1 is spin-coated on the
surface at high speed (4000 rpm), crystallization of 1 as a result
of solvent evaporation is greatly depressed and uniform coverage
of 1 on the surface becomes possible. This drastically increases
the effective surface concentration of 1 as compared to the
solution deposition, which is inevitably limited by the modest
solubility of 1.37 Equally importantly, higher deposition tem-
peratures become possible because the quaternization step is
solvent-free in the present approach. Higher deposition tem-
peratures accelerate the quaternization process. The mobility
of 1 on the surface under deposition conditions may also play
a significant role in the [(aminophenyl)azo]pyridinium chro-
mophore assembly process. Thermal gravimetric analysis shows
that 1 is volatile under the multilayer deposition conditions.
Therefore, the facile assembly of the [(aminophenyl)azo]-
pyridinium chromophore may be in part due to the gas-surface
reaction between 1 and self-assembled coupling agents. How-
ever, an alternative interfacial solid state reaction in which the
presence of the coupling agent lowers the melting point and
enhances the mobility of 1 seems more likely to be the
predominant self-assembling pathway. Thus, in situ SHG
experiments carried out at atmospheric pressure indicate that
the onset temperature for quaternization is ∼110 °C, which is
considerably lower than the melting point of 1, 150.3 °C.
Because any volatization process should play at most a minor
Figure 6. AFM images at 1 × 1 µm2 scan area of 1-based self-
assembled chromophoric mono- and multilayers. (A) Self-assembled
monolayer composed of 4-ClCH2C6H4SiCl3 coupling agent. (B) Self-
assembled bilayer composed of 4-ClCH2C6H4SiCl3 coupling agent and
4-[[4-[bis(hydroxylethyl)amino]phenyl]azo]pyridinium chromophore.
(C) A 10-trilayer [(aminophenyl)azo]pyridinium chromophore multi-
layer. Each trilayer is composed of the sequence steps i-iii. The
coupling agent is 4-ClCH2C6H4SiCl3.
nophenyl)azo]pyridinium multilayer sample. Although evenly
distributed features are evident on the surface, the surface rms
roughness is only 12 Å. Considering that such a 10-trilayer
sample has an average vertical thickness of 250 Å, the 12 Å
average thickness variation (e5%) is rather small. Unfortu-
nately, this result is not informative in comparing the morphol-
ogies of the present multilayers to those of metal phosphonate
multilayers on gold substrates. In the latter, surface roughness
could not be estimated because of the polycrystalline nature of
the gold substrates.36a Nonetheless, the AFM results clearly
(37) Although higher solubility is obtainable in polar solvents, significant
side reactions of the chromophore occur, which in turn result in multilayers
with diminished structural regularity and second-order response. See:
Yitzchaik, S.; Kakkar, A. K.; Roscoe, S. B.; Orihashi, Y.; Marks, T. J.;
Lin, W.; Wong, G. K. Mol. Cryst. Liq. Cryst. 1994, 240, 9-16.

8042 J. Am. Chem. Soc., Vol. 118, No. 34, 1996
Lin et al.
role under these conditions, the ease of the quaternization
reaction of 1 on the silylated surface at temperatures below the
melting of 1 is more reasonably explained in terms of an
interfacial solid state reaction. As a result of the “topotactic”
nature of the aminophenylazopyridinium chromophore deposi-
tion, the present self-assembly approach has very high ef-
ficiency: the total reaction time for constructing a complete
trilayer is about one hour.
Structural Regularity of Self-Assembled [(Aminophenyl)-
azo]pyridinium Superlattices. As mentioned in the Introduc-
tion, efficient routes to structurally regular self-assembled
multilayers are sparse. The expeditious assembly of the present
mutlilayer system enables us to examine the characteristics of
reaction chemistry, film uniformity, and microstructural acen-
tricity of the self-assembled superlattices using a variety of
spectroscopic and structural techniques.
decline. We attribute this fall to the loss of microstructural
acentricity as a result of randomization of chromophore align-
ment, perhaps induced by chromophore dipole-dipole repul-
sions. In this respect, the present system offers advantages over
SHG-active zirconium phosphate/phosphonate assemblies de-
veloped by Katz and co-workers. In this only other known self-
assembled SHG-active multilayer material, the aforementioned
“self-healing” step is not implemented and other measures are
needed to produce defect-free, regular structures.39
Conclusions
We have successfully fabricated regular superlattices contain-
ing [(aminophenyl)azo]pyridinium second-order nonlinear opti-
cal chromophore using siloxane self-assembly techniques. The
[(aminophenyl)azo]pyridinium multilayers can be assembled on
hydroxylated substrates in an efficient, “topotactic” fashion via
the iterative reaction sequence of a coupling layer, a chro-
mophore layer, and a capping layer. The [(aminophenyl)azo]-
pyridinium mono- and multilayers have been characterized by
X-ray photoelectron and transmission optical spectroscopies,
spectroscopic ellipsometry, X-ray reflectivity, advancing contact
angle measurements, and polarized second harmonic generation
(SHG) measurements. These spectroscopic and structural data
indicate that the [(aminophenyl)azo]pyridinium multilayers have
very good structural regularity and very high second-order
nonlinear optical susceptibility (ø(2) ∼ 3.6 × 10-7 esu or 150
pm/V at ω0 ) 1064 nm). Atomic force microscopy also reveals
that the [(aminophenyl)azo]pyridinium multilayers have very
smooth surface morphologies as a result of the high structural
regularity. These multilayers are photochemically stable and
thermally robust. Because these self-assembled chromophoric
superlattices are intrinsically acentric, i.e., no electric-field poling
is required to establish bulk second-order nonlinearity (as in
poled polymer SHG materials), integration of these self-
assembled SHG materials into other device structures should
be possible. We are currently pursuing the fabrication of active
devices such as low-loss planar waveguides for frequency
doubling using self-assembled chromophoric multilayers.40 We
are also extending the present synthetic methodologies to the
preparation of other heterostructural assemblies.
The combination of UV-vis and XPS spectroscopies indi-
cates that the degree of surface quaternization of chromophore
precursor 1 (to form anchored [(aminophenyl)azo]pyridinium
chromophore) is ∼90% of the available surface alkyl halide
groups. This extent of quaternization is entirely reasonable since
the [(aminophenyl)azo]pyridinium chromophore molecule is
considerably larger than the 3-IC3H6SiCl3 coupling agent, and
complete quaternization of the packed surface propyl iodide
groups does not appear physically (sterically) reasonable.
Interestingly, because the subsequent capping step (step iii,
Scheme 2) cross-links/planarizes the surface, any “defect”
caused by incomplete conversion in the quaternization step will
not catastrophically propagate. We will return to this point later.
The uniform growth of the self-assembled multilayers is
indicated by the linear dependence of the [(aminophenyl)azo]-
pyridinium chromophore longitudinal HOMO f LUMO charge-
transfer absorbance at 572 nm on the number of assembled
trilayers as well as by the linear dependence of the ellipsometry-
and X-ray reflectivity-derived multilayer thicknesses on the
number of assembled trilayers. The most impressive feature,
however, is the presence of “Bragg” peaks in the multilayer
film XRR data. These features are due to scattering by the
individual superlattice layers. The presence of the “Bragg”
peaks necessarily requires that the individual layers have similar
spacings as well as electron density profiles, both of which are
a result of structural uniformity of the multilayers.38 The
mutlilayer structural uniformity is further evidenced in the AFM
images that show that the present multilayer films have very
smooth surface morphologies.
The most important issue concerns the microstructural
acentricity of the multilayer films, because noncentrosymmetric
organization is a prerequisite for bulk second-order nonlinear
optical effects. The quadratic dependence of the SHG response
of the multilayers on the number of assembled trilayers clearly
indicates that microstructural acentricity is maintained in the
multilayer films. It is reasonable to propose that the “self-
annealing” nature of the siloxane capping step (step iii, Scheme
2) in large part assists in stabilizing the microstructural
acentricity. In fact, previous studies of self-assembled stilba-
zolium multilayers have indicated that this structural stabiliza-
tion/planarization step is crucial in maintaining the acentricity
of the chromophore microstructure in the resulting multilayers.14c
Although XRR and optical spectroscopy indicates the uniform
build-up of the stilbazolium chromophoric superlattice when
capping step iii is omitted, the intensity of SHG response no
longer depends quadratically on the number of layers. As the
number of layers exceeds three, the SHG intensity begins to
Acknowledgment. This research was supported by the NSF
through the Northwestern Materials Research Center
(DMR9120521) and by the Office of Naval Research. We thank
Dr. A. Israel for ZINDO/SOS calculations of linear and
nonlinear optical properties of the chromophores, A. Malik, J.
Malinsky, and Prof. P. Dutta for assistance with the X-ray
reflectivity measurements, and S. Roscoe and Dr. S. Yitzchaik
for helpful discussions. W.B.L. thanks the National Science
Foundation for a postdoctoral fellowship.
Supporting Information Available: Full experimental
details regarding synthesis and physical measurements (6 pages).
This material is contained in many libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
JA960395F
(39) Schilling, M. L.; Katz, H. E.; Stein, S. M.; Shane, S. F.; Wilson,
W. L.; Buratto, S.; Ungashe, S. B.; Taylor, G. N.; Putvinski, T. M.; Chidsey,
C. E. D. Langmuir 1993, 9, 2156-2160.
(40) (a) Lundquist, P. M.; Lin, W.; Yitzchaik, S.; Wong, G. K.; Marks,
T. J. Appl. Phys. Lett. in press. (b) Yitzchaik, S.; Lundquist, P. M.; Lin,
W.; Wong, G. K.; Marks, T. J. SPIE Proc. 1995, 2285, 282-289.
(38) If interlayer spacings and/or electron density profiles vary signifi-
cantly from trilayer to trilayer, destructive interference will occur. Con-
sequently, no Bragg diffraction peaks will be present in the X-ray reflectivity
curves.