Synthesis, Spectroscopic Characterization, and Electro-Optical Properties of Noncentrosymmetric Azobenzene/Zirconium Phosphonate Multilayer Films

Article abstract of DOI:10.1021/ac960756+

Ultrathin noncentrosymmetric nonlinear optical films based on zirconium phosphonate (ZP) self-assembled multilayers that incorporate the asymmetric azobenzene chromophore[5-[4-[[4-[(6-hydroxyhexyl)sulfonyl]phenyl]azo]phenyl]pentoxy] phosphonic acid (HAPA)

Full text of DOI:10.1021/ac960756+

Anal. Chem. 1997, 69, 240-248
Synthesis, Spectroscopic Characterization, and
Electro-Optical Properties of Noncentrosymmetric
Azobenzene/Zirconium Phosphonate Multilayer
Films
Dennis G. Hanken, Roberta R. Naujok, Jesse M. Gray, and Robert M. Corn*
Department of Chemistry, University of WisconsinsMadison, 1101 University Avenue, Madison, Wisconsin 53706
noncentrosymmetric material and is described by the electro-optic
coefficient, r.7 Current electro-optical devices are typically based
on the inorganic crystal lithium niobate, LiNbO3, which has an
electro-optic coefficient, r33, of 31 pm/ V.8 Most of the research
on molecular alternatives to LiNbO3 have focused on the incor-
poration of NLO organic chromophores into polymer networks.
These NLO polymers are poled in a strong electric field at elevated
temperatures in order to create a noncentrosymmetric film. The
electro-optic coefficients generated by such films vary from 1 to
55 pm/ V in the presence of the poling fields but can decay
significantly with time.9-14
An alternative to poled polymer films are ultrathin noncen-
trosymmetric organic films that have been created one monolayer
at a time with a self-assembly process. Marks et al. have shown
that noncentrosymmetric NLO films of bifunctional stilbazole
chromophores can be formed by covalent coupling reactions
utilizing an organosilane-based chemistry.15-17 A second method
for creating noncentrosymmetric self-assembled multilayers uses
a zirconium phosphonate (ZP) ligand coupling chemistry first
reported by Mallouk and co-workers for centrosymmetric films18-20
and subsequently modified by Katz et al. to create noncentrosym-
metric multilayers.21-24 Ultrathin noncentrosymmetric films formed
from ZP multilayers with NLO chromophores are more chemically
and thermally stable than most NLO polymer films,22 but spec-
Ultrathin noncentrosymmetric nonlinear optical films
based on zirconium phosphonate (ZP ) self-assembled
multilayers that incorporate the asymmetric azobenzene
chromophore [5-[4-[[4-[(6-hydroxyhexyl)sulfonyl]phenyl]-
azo]phenyl]pentoxy]phosphonic acid (HAP A) are synthe-
sized and constructed. The ZP film structure and multi-
layer deposition chemistry are characterized by a com-
bination of polarization/ modulation Fourier transform
infrared reflection absorption spectroscopy, surface plas-
mon resonance (SP R) measurements, and optical second
harmonic generation (SHG). SP R measurements on the
HAP A multilayer films yield an average monolayer thick-
ness of 2 7 ( 0 .5 Å. The resonant SHG at 3 6 5 nm from
ultrathin HAP A ZP films on silica surfaces increases
quadratically with the number of self-assembled HAP A
monolayers, and an analysis of the polarization depen-
dence of the surface SHG yields an orientation parameter
D ) 0 .7 9 ( 0 .0 3 corresponding to an average tilt angle
of 2 7 ( 2 ° for the azobenzene chromophores with respect
to the surface normal. An electro-optic coefficient r 3 3 for
the HAP A monolayers of 1 1 pm/ V at 6 3 2 .8 nm is
obtained from SP R modulation experiments of ZP films
on gold substrates that have been incorporated into air-
gap capacitors. SP R modulation experiments are then
performed on a HAP A monolayer in an in situ electro-
chemical environment in order to determine that a modu-
lation of (2 5 mV corresponds to a change in the electric
field strength of 1 × 1 0 4 V/ cm within the ultrathin organic
film at the electrode surface.
(7) Shen, Y. R. The Principles of Nonlinear Optics; Wiley: New York, 1984.
(8) Yariv, A.; Yeh, P. Optical Waves in Crystals; Wiley: New York, 1984.
(9) Ahlheim, M.; Barzoukas, M.; Bedworth, P. V.; Blanchard-Desce, M.; Fort,
A.; Hu, Z.-Y.; Marder, S. R.; Perry, J. W.; Runser, C.; Staehelin, M.; Zysset,
B. Science 1 9 9 6 , 271, 335.
(10) Chen, T.-A.; Jen, A. K.-Y.; Cai, Y. J. Am. Chem. Soc. 1 9 9 5 , 117, 7295.
(11) Robello, D. R.; Dao, P. T.; Schildkraut, J. S.; Scozzafava, M.; Urankar, E. J.;
Willand, C. S. Chem. Mater. 1 9 9 5 , 7, 284.
(12) Verbiest, T.; Burland, D. M.; Jurich, M. C.; Lee, V. Y.; Miller, R. D.; Volksen,
W. Science 1 9 9 5 , 268, 1604.
(13) Sekkat, Z.; Kang, C.-S.; Aust, E. F.; Wegner, G.; Knoll, W. Chem. Mater.
1 9 9 5 , 7, 142.
(14) Singer, K. D.; Kuzyk, M. G.; Sohn, J. E. J. Opt. Soc. Am. B 1 9 8 7 , 4, 968.
(15) Li, D.; Ratner, M. A.; Marks, T. J.; Zhang, C.; Yang, J.; Wong, G. K. J. Am.
Chem. Soc. 1 9 9 0 , 112, 7389.
(16) 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. 1 9 9 3 , 97, 6958.
(17) Kakkar, A. K.; Yitzchaik, S.; Roscoe, S. B.; Kubota, F.; Allan, D. S.; Marks,
T. J.; Lin, W.; Wong, G. K. Langmuir 1 9 9 3 , 9, 388.
A wide variety of novel organic materials have been designed
and synthesized for use in nonlinear optics.1-3 One important
application of these nonlinear optical (NLO) materials is the
creation of noncentrosymmetric thin organic films for electro-
optical devices.4-6 The electro-optical or Pockel’s effect results
from a change in the index of refraction of a material upon
application of an external electric field; this response requires a
(1) Prasad, P.; Williams, D. J. Introduction to Nonlinear Optical Effects in
Molecules and Polymers; Wiley: New York, 1991.
(2) Marder, S. R., Stucky, G. D., Sohn, J. E., Eds. Materials for Nonlinear Optics:
Chemical Perspectives; ACS Symposium Series 455; American Chemical
Society: Washington, DC, 1991.
(3) Kanis, D. R.; Ratner, M. A.; Marks, T. J. Chem. Rev. 1 9 9 4 , 94, 195.
(4) Zyss, J. Molecular Nonlinear Optics; Academic Press: San Diego, CA, 1994.
(5) Burland, D. M.; Miller, R. D.; Walsh, C. A. Chem. Rev. 1 9 9 4 , 94, 31.
(6) 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. 1 9 9 5 ,
7, 1060.
(18) Lee, H.; Hong, H. G.; Mallouk, T. E.; Kepley, L. J. J. Am. Chem. Soc. 1 9 8 8 ,
110, 618.
(19) Lee, H.; Mallouk, T. E.; Kepley, L. J.; Hong, H. G.; Akhter, S. J. Phys. Chem.
1 9 8 8 , 92, 2597.
(20) Hong, H. G.; Sackett, D. D.; Mallouk, T. E. Chem. Mater. 1 9 9 1 , 3, 521.
(21) Putvinski, T. M.; Schilling, M. L.; Katz, H. E.; Chidsey, C. E. D.; Mujsce, A.
M.; Emerson, A. B. Langmuir 1 9 9 0 , 6, 1567.
(22) Katz, H. E.; Scheller, G.; Putvinski, T. M.; Schilling, M. L.; Wilson, W. L.;
Chidsey, C. E. D. Science 1 9 9 1 , 254, 1485.
240 Analytical Chemistry, Vol. 69, No. 2, January 15, 1997
S0003-2700(96)00756-1 CCC: $14.00 © 1997 American Chemical Society

interfacial electrostatic fields.27 The electric field strength and
field profile within ultrathin organic films at charged interfaces is
currently not well-characterized, even though the variation of the
potential through the ultrathin dielectric layer controls the
electrochemical behavior of species incorporated into the
interface.28-30 One NLO method that has been employed is the
characterization of charged interfaces with electric field-induced
second harmonic generation (EFISH).27,31-33 In principle, the
interfacial electric fields can also be monitored via the molecular
linear electro-optical response. To date, only a few attempts to
measure the electric field strength within adsorbed monolayers
at charged electrode surfaces based on electrochromic shift effects
have been reported.34-37 In this paper we report the first
application of the linear electro-optical effect to study in situ the
electrostatic fields within ultrathin organic monolayers at elec-
trochemical interfaces. The electro-optical responses of the
noncentrosymmetric ZP azobenzene monolayers incorporated into
air-gap capacitors and electrochemical interfaces are compared
in order to determine the electric field strength within the ZP
films at the electrode surface.
Figure 1. Schematic diagram for the construction of noncentrosym-
metric ZP multilayer films. The ZP films are created on vapor-
deposited gold substrates that have been first primed with a self-
assembled monolayer of phosphorylated MUD. Exposure of the
zirconated surface to a solution of the nonlinear optical compound
HAPA following the procedure of Katz and co-workers led to
noncentrosymmetric multilayer films that exhibited large nonlinear
optical responses.
EXPERIMENTAL CONSIDERATIONS
Materials. 2,4,6-Collidine (99%), phosphorus oxychloride
(99%), zirconyl chloride octahydrate (98%), anhydrous octane
(99+%), anhydrous tetrahydrofuran (THF) (99.9%), dihexadecyl
phosphate (DHP) (99%), sodium sulfate, potassium carbonate, and
lithium aluminum hydride (1 M in THF) were obtained from the
Aldrich Chemical Co. Other reactants and solvents included (10-
carbomethoxydecyl)dimethylchlorosilane (United Chemical Tech-
nologies), tetrabutylammonium bromide (99+%) (Fluka), meth-
ylene chloride, UV grade acetonitrile (Burdick and Jackson), ethyl
acetate (Fisher), and absolute ethanol (Pharmco). All chemicals
were used as received. Mercaptoundecanol (MUD) and 1,10-
decanediylbis(phosphonate) (DBP) were synthesized as described
previously.38 All aqueous solutions were prepared from 18 MΩ
Millipore-filtered water.
troscopic measurements have shown that the ZP films are not as
highly ordered and oriented as other self-assembled systems such
as alkanethiol monolayers.25 We have recently reported the
incorporation of azobenzene chromophores into ZP multilayers
for the creation of ultrathin films of variable index of refraction.26
In this paper, we describe the synthesis, structural character-
ization, and nonlinear optical response of noncentrosymmetric ZP
azobenzene multilayer films that incorporate the asymmetric
nonlinear optical chromophore [5-[4-[[4-[(6-hydroxyhexyl)sul-
fonyl]phenyl]azo]phenyl]pentoxy]phosphonic acid (HAPA) (Fig-
ure 1). The structural characterization of these ultrathin films is
accomplished using a combination of optical techniques: polariza-
tion/ modulation Fourier transform infrared reflection absorption
spectroscopy (PM-FTIRRAS), surface plasmon resonance (SPR)
measurements of film thickness, and optical second harmonic
generation (SHG). An electro-optic coefficient for the ultrathin
films is measured with a series of modulated SPR experiments
from ZP monolayers that have been incorporated into air-gap
capacitors.
A second application area for NLO-active organic molecules
has been in the creation of optical probes for the study of
interfacial processes. SHG measurements of NLO-active organic
compounds that have been incorporated into solid/ liquid, liquid/
liquid, and liquid/ air interfaces have yielded valuable information
on adsorption, diffusion, aggregation, and reaction at these
interfaces.7,27 In an electrochemical environment, the use of
nonlinear optical methods can also provide information about the
HAP A Synthesis. The synthesis of HAPA pictured in Figure
2 is described in detail.
(1 ) 4 -[(6 -Hydroxyhexyl)thio]nitrobenzene. In a 250 mL
round-bottom flask containing 150 mL of absolute ethanol was
placed 2.17 g (14 mmol) of 4-nitrothiophenol (Fluka). The mixture
was heated with stirring to dissolve all of the solid. To this
solution was slowly added 5.2 mL (14 mmol) of 21 wt % sodium
ethoxide in ethanol (Aldrich). The resulting deep red solution
was stirred in a nitrogen atmosphere for 10 min. Next, 1.70 mL
(13 mmol) of 6-bromo-1-hexanol (Aldrich) was added to the
solution. The mixture was refluxed for 12 h until the reaction
was complete as determined by TLC in ethyl acetate. The solution
was concentrated to half the volume by rotary evaporation and
(27) Corn, R. M.; Higgins, D. A. Chem. Rev. 1 9 9 4 , 94, 107.
(28) Smith, C. P.; White, H. S. Anal. Chem. 1 9 9 2 , 64, 2398.
(29) Gao, X.; White, H. S. J. Electroanal. Chem. 1 9 9 5 , 389, 13.
(30) Basame, S. B.; White, H. S. J. Phys. Chem. 1 9 9 5 , 99, 16430.
(31) Lantz, J. M.; Baba, R.; Corn, R. M. J. Phys. Chem. 1 9 9 3 , 97, 7392.
(32) Lantz, J. M.; Corn, R. M. J. Phys. Chem. 1 9 9 4 , 98, 9387.
(33) Lantz, J. M.; Corn, R. M. J. Phys. Chem. 1 9 9 4 , 98, 4899.
(34) Schmidt, P. H.; Plieth, W. J. J. Electroanal. Chem. 1 9 8 6 , 201, 163.
(35) Pope, J. M.; Tan, Z.; Kimbrell, S.; Buttry, D. A. J. Am. Chem. Soc. 1 9 9 2 ,
114, 10085.
(23) Katz, H. E.; Wilson, W. L.; Scheller, G. J. Am. Chem. Soc. 1 9 9 4 , 116, 6636.
(24) Katz, H. E. Chem. Mater. 1 9 9 4 , 6, 2227.
(25) Ulman, A. An Introduction to Ultrathin Organic Films; Academic: New York,
1991.
(36) Franzen, S.; Goldstein, R. F.; Boxer, S. G. J. Phys. Chem. 1 9 9 0 , 94, 5135.
(37) Lockhart, D. J.; Hammes, S. L.; Franzen, S.; Boxer, S. G. J. Phys. Chem.
1 9 9 1 , 95, 2217.
(26) Hanken, D. G.; Corn, R. M. Anal. Chem. 1 9 9 5 , 67, 3767.
(38) Frey, B. L.; Hanken, D. G.; Corn, R. M. Langmuir 1 9 9 3 , 9, 1815.
Analytical Chemistry, Vol. 69, No. 2, January 15, 1997 241

green, and all solid material dissolved. This mixture was stirred
at 0 °C for an additional 10 min. In a second 25 mL flask, 0.24 g
(2.5 mmol) of phenol (Mallinckrodt) was dissolved in 10 mL of
1% aqueous NaOH and cooled to 0 °C in an ice bath. The
diazonium solution was slowly added to the phenolate solution,
resulting in a bright red solution. The solution was stirred at 0
°C for 1 h, during which a brown precipitate formed, and then
was stirred for 12 h at room temperature. The mixture was added
to a separatory funnel containing 10% aqueous HCl and ethyl
acetate. The organic layer was washed with 3 × 50 mL portions
of 10% aqueous HCl and with 2 × 50 mL portions of saturated
aqueous NaCl. The ethyl acetate was dried with Na2SO4, and the
solvent was removed by rotary evaporation to yield 0.76 g (91%)
of an orange-brown solid: 1H NMR (acetone-d6), δ 1.2-1.8 (m, 8
H), 3.1 (t, 2 H), 3.5 (t, 2 H), 7.0 (d, 2 H), 7.6 (dd, 4 H), 8.1 (d, 2
H).
(4 ) Diethyl [5 -[4 -[[4 -[(6 -Hydroxyhexyl)thio]phenyl]azo]-
phenyl]pentoxy]phosphonate. In a 100 mL round-bottom flask
containing 70 mL of absolute ethanol was placed 0.39 g (1.2 mmol)
of 4-[[4-[(6-hydroxyhexyl)thio]phenyl]azo]phenol. The mixture
was heated with stirring to dissolve all of the solid. To this
solution was slowly added 0.43 mL (1.2 mmol) of 21 wt % sodium
ethoxide in ethanol. The deep red solution was stirred in a
nitrogen atmosphere for 10 min. Next, 0.32 g (1.1 mmol) of
diethyl (5-bromopentyl)phosphonate (Sigma) was added to the
solution. The mixture was refluxed for 48 h until the reaction
was complete as determined by TLC in ethyl acetate. The solution
was concentrated to half the volume by rotary evaporation and
added to a separatory funnel containing 10% aqueous K2CO3 and
ethyl acetate. The organic layer was washed with 7 × 50 mL
portions of 10% aqueous K2CO3 and with 2 × 50 mL portions of
saturated aqueous NaCl. The ethyl acetate was dried with Na2-
SO4, and the solvent was removed by rotary evaporation to yield
0.56 g (87%) of a red paste.
(5 ) Diethyl [5 -[4 -[[4 -[(6 -Hydroxyhexyl)sulfonyl]phenyl]-
azo]phenyl]pentoxy]phosphonate. In a 100 mL round-bottom
flask containing 40 mL of methanol was placed 0.56 g (1.1 mmol)
of diethyl [5-[4-[[4-[(6-hydroxyhexyl)thio]phenyl]azo]phenyl]-
pentoxy]phosphonate. The solution was cooled to 0 °C in an ice
bath, and 2 g (3.3 mmol) of oxone (potassium peroxymonosulfate)
(Aldrich) dissolved in 40 mL of water was slowly added.40 The
slurry was stirred at 0 °C for 10 min and then was stirred at room
temperature. The reaction was complete in 3 h as determined
by TLC (1:1 ethyl acetate/ acetone). The mixture was then added
to a separatory funnel containing 10% aqueous HCl and ethyl
acetate. The organic layer was washed with 3 × 50 mL portions
of water and with 2 × 50 mL portions of saturated aqueous NaCl.
The ethyl acetate was dried with Na2SO4, and the solvent was
removed by rotary evaporation to yield 0.57 g (96%) of a red paste.
(6) [5-[4-[[4-[(6-Hydroxyhexyl)sulfonyl]phenyl]azo]phenyl]-
pentoxy]phosphonic Acid. In a 100 mL round-bottom flask
containing 30 mL of dry methylene chloride was placed 0.57 g
(1.0 mmol) of diethyl [5-[4-[[4-[(6-hydroxyhexyl)sulfonyl]phenyl]-
azo]phenyl]pentoxy]phosphonate. To this stirred solution was
added 0.86 mL (6.5 mmol) of bromotrimethylsilane (Aldrich).41
The solution was stirred at room temperature under a nitrogen
atmosphere for 24 h and was then refluxed for 2 h. The solvent
Figure 2. Reaction scheme for the synthesis of the nonlinear optical
compound[5-[4-[[4-[(6-hydroxyhexyl)sulfonyl]phenyl]azo]phenyl]pentoxy]-
phosphonic acid (HAPA). Reaction steps 1-6 are outlined in detail
in the text.
added to a separatory funnel containing 10% aqueous K2CO3 and
ethyl acetate. The organic layer was washed with 5 × 50 mL
portions of 10% aqueous K2CO3 and with 2 × 50 mL portions of
saturated aqueous NaCl. The ethyl acetate was subsequently
dried with Na2SO4, and the solvent was removed by rotary
evaporation to yield 3.02 g (91%) of light yellow solid: 1H NMR
(acetone-d6), δ 1.2-1.8 (m, 8 H), 3.1 (t, 2 H), 3.5 (t, 2 H), 7.2 (d,
2 H), 8.0 (d, 2 H).
(2 ) 4 -[(6 -Hydroxyhexyl)thio]aniline. In a 100 mL round-
bottom flask containing 30 mL of absolute ethanol was placed 1.5
g (6 mmol) of 4-[(6-hydroxyhexyl)thio]nitrobenzene. To the
solution was added 6.6 g (30 mmol) of SnCl2‚2H2O (Mallinck-
rodt).39 The mixture was refluxed for 12 h until the reaction was
complete as determined by TLC in ethyl acetate. The ethanol
was removed by rotary evaporation. The resulting yellowish oil
was added to a separatory funnel containing 10% aqueous K2CO3
and ethyl acetate. The aqueous layer was maintained above pH
8 and extracted with 5 × 50 mL portions of ethyl acetate. The
ethyl acetate was dried with Na2SO4, and the solvent was removed
by rotary evaporation to yield 1.14 g (86%) of white solid: 1H NMR
(CDCl3), δ 1.2-1.8 (m, 8 H), 3.1 (t, 2 H), 3.4 (t, 2 H), δ 4.2 (s, 2
H), 6.7 (d, 2 H), 7.4 (d, 2 H).
(3 ) 4 -[[4 -[(6 -Hydroxyhexyl)thio]phenyl]azo]phenol. In a
25 mL flask, 0.56 g (2.5 mmol) of 4-[(6-hydroxyhexyl)thio]aniline
was dissolved in 11 mL of hot 10% aqueous HCl with stirring. The
solution was rapidly cooled to 0 °C in an ice/ salt bath which then
became a fine milky suspension. To this mixture, 0.17 g (2.5
mmol) of sodium nitrite (Fisher), dissolved in 2 mL of water, was
slowly added in order to keep the solution temperature below 10
°C. Upon addition of the sodium nitrite, the solution turned light
(40) Trost, B. M.; Curran, D. P. Tetrahedron Lett. 1 9 8 1 , 22, 1287.
(41) McKenna, C. E.; Higa, M. T.; Cheung, N. H.; McKenna, M.-C. Tetrahedron
Lett. 1 9 7 7 , 18, 155.
(39) Bellamy, F. D.; Ou, K. Tetrahedron Lett. 1 9 8 4 , 25, 839.
242 Analytical Chemistry, Vol. 69, No. 2, January 15, 1997

was removed by rotary evaporation, and the excess bromotrimeth-
ylsilane was removed under vacuum. Cold 1% aqueous HCl was
added to the red paste, and a fine brown precipitate formed. The
solid was collected with a glass filter frit and was washed
thoroughly with water and ethyl ether. The product was dried
under vacuum to yield 0.41 g (80%) of a brown-orange solid: 1H
NMR (DMSO-d6), δ 1.2-1.75 (m, 14 H), 1.87 (m, 2 H), 3.45 (pt,
4 H), 4.2 (t, 2 H), 7.25 (d, 2 H), 8.0 (d, 2 H), 8.17 (dd, 4 H); IR
(KBr) 1602, 1583, 1253, 1137, 1003 cm-1; UV/ visible max (EtOH)
362 nm.
P olarization/ Modulation Fourier Transform Infrared Re-
flection Absorption Spectroscopy (P M-FTIRRAS) Measure-
ments on Gold Substrates. All PM-FTIRRAS spectra were
obtained on a Mattson RS-1 spectrometer without a background
or reference spectrum by utilizing real-time interferogram sam-
pling electronics that have been described previously.42,43 The
differential reflectance (% ∆R/ R) spectra were converted to
absorbance units for comparison with other spectra. Sample
spectra in the CH stretching region (3100-2700 cm-1) were
obtained with an InSb detector, and spectra in the mid-infrared
region (1800-850 cm-1) were obtained with a narrow-band
HgCdTe detector. All infrared spectra were acquired at 2 cm-1
resolution from 1024 interferometer scans.
SP R and Electro-Optical Measurements on Gold Sub-
strates. The experimental apparatus for conventional scanning
surface plasmon resonance measurements was described in detail
in a previous paper.44 This apparatus measures the reflectivity of
p-polarized light from a HeNe laser (632.8 nm, 1 mW, Newport
Corp.) from the BK7/ Au/ ZP film sample as a function of the
incident angle, θ. For air-gap electro-optical SPR measurements,
the sample setup is pictured in the inset of Figure 7. The ZP-
coated thin gold samples (470 Å) were placed in optical contact
with a BK7 right-angle prism (Melles Griot) using an index-
matching fluid (Type NVH, Cargille Labs). This gold sample was
spaced from a second gold electrode (1200 Å of gold deposited
on a fused-silica window) by a thin sheet of Mylar dielectric film
(DuPont, 14C gauge). A small square was cut out of the Mylar
to produce a thin air gap. The air gap was aligned with the
incident HeNe laser beam to allow the surface plasmon wave to
be evanescent in air upon excitation. The thickness of this gap,
assuming the sample to be a parallel-plate capacitor, was estimated
from a measurement of the capacitance (BK Precision 820
capacitance meter, Dynascan Corp.). The value for the air gap
was calculated to be ∼15 µm for the sample in Figure 7. An ac
electric field with a peak-to-peak amplitude of 10-50 V at a
frequency of 1-10 kHz (Model 501 ac power source, Elgar Corp.)
was applied to the sample cell to produce an oscillating electric
field on the order of 104 V/ cm. This modulation produced a
corresponding modulation in the reflectivity due to the Pockel’s
effect which was detected by use of a photodiode (Hamamatsu)
and a lock-in amplifier (SR510, Stanford Research Systems).
For in situ electro-optical measurements, the sample setup is
pictured in the inset of Figure 8. The ZP-coated thin gold samples
(470 Å) formed on SF10 glass slides (Schott) were placed in
optical contact with an SF10 equilateral prism (Ealing Electro-
Optics, Inc.) using an index-matching fluid (1-iodonaphthalene,
Aldrich). The sample cell consisted of a Teflon electrochemical
cell that was pressed against the ZP-coated thin gold sample by
use of a viton o-ring for a water-tight seal. The electrochemical
cell was made to incorporate two electrodes and a nitrogen line
for solution purging. A three-electrode assembly was formed with
the ZP-coated gold sample as the working electrode, a saturated
calomel reference electrode, and a platinum counter electrode.
The potential was controlled by a potentiostat (EG&G Model 173
potentiostat/ galvanostat) and a voltage controller (EG&G Model
175 Universal Programmer). All in situ electrochemical experi-
ments were carried out using 0.2 M tetrabutylammonium bromide
as the electrolyte. Cyclic voltammograms were recorded in the
potential region -0.4 to 0.7 V vs SCE. In this region, no faradaic
processes occurred for the ZP samples. The potential was fixed
to 0.0 V vs SCE, and a small sinusoidal modulation ((25 mV, 1
kHz, Model 5100B function generator, Krohn & Hite) was applied
to the electrochemical cell. This modulation produced a corre-
sponding modulation in the reflectivity due to the Pockel’s effect
which was detected by use of a photodiode and a lock-in amplifier.
Theoretical Fresnel fits to the electro-optical data were calcu-
lated by subtracting two surface plasmon theory curves which
differed only by small changes in the real component of the
dielectric constant of the ZP film. Air-gap experiments were fit
using a five-phase model (BK7, Au, phosphorylated MUD, HAPA,
air).26,44,45 In the in situ case, a background signal due to the
electro-optical modulation of the surface gold layer (0.5 Å) was
compensated for in the theory calculations by using a seven-phase
model [SF10, Au, Au (0.5 Å), phosphorylated MUD, HAPA, DHP,
water]. This background signal was not observed within the
signal-to-noise ratio in the air-gap electro-optical experiments.
UV/ Visible Absorption and Ellipsometry Measurements
on Fused-Silica Substrates. UV/ visible absorption spectra were
obtained on a Hewlett-Packard Model 8452A diode array spec-
trophotometer. Micromolar solutions of HAPA in ethanol exhib-
ited an absorption maximum at 362 nm. The absorption is
associated with the main charge transfer transition for this
molecule. On fused-silica substrates, the position of this transition
red-shifted slightly to 365 nm upon incorporation into the ZP
multilayer films, indicating that the HAPA chromophores do not
significantly interact within the films. The amount of absorbance
at 365 nm increased in a linear fashion with the number of HAPA
layers for all ZP multilayer samples. Ellipsometry measurements
for the HAPA films on fused-silica substrates were obtained with
a Rudolph Research Model AutoEL II ellipsometer employing 633
nm light. Theoretical Fresnel fits to the data were calculated as
described previously.26
Second Harmonic Generation Measurements on Fused-
Silica Substrates. Second harmonic generation measurements
for ZP films deposited onto a primer layer on fused-silica
substrates were made by employing a Nd:YAG pumped dye laser
(10 ns pulse width, 10 Hz repetition rate) and an experimental
configuration described previously.46 The fundamental input
wavelength of 730 nm was focused onto the silica surface at an
incident angle of 60°, and the second harmonic light at 365 nm
was measured in reflection from the top silica surface. Filters
and a 0.175 M monochromator were used to remove any reflected
fundamental light. The SHG signal was detected with a photo-
(42) Green, M. J.; Barner, B. J.; Corn, R. M. Rev. Sci. Instrum. 1 9 9 1 , 62, 1426.
(43) Barner, B. J.; Green, M. J.; Saez, E. I.; Corn, R. M. Anal. Chem. 1 9 9 1 , 63,
55.
(45) Hansen, W. N. J. Opt. Soc. Am. 1 9 6 8 , 58, 380.
(46) Higgins, D. A.; Byerly, S. K.; Abrams, M. B.; Corn, R. M. J. Phys. Chem.
1 9 9 1 , 95, 6984.
(44) Jordan, C. E.; Frey, B. L.; Kornguth, S.; Corn, R. M. Langmuir 1 9 9 4 , 10,
3642.
Analytical Chemistry, Vol. 69, No. 2, January 15, 1997 243

multiplier tube, and the output was averaged with a boxcar
averager. Orientation measurements for the HAPA films, which
were determined from polarization dependence experiments, are
obtained by measuring the intensity of the s-polarized and
p-polarized SHG signals as a function of the polarization of the
incident fundamental light.47-49
RESULTS AND DISCUSSION
A. Noncentrosymmetric HAP A Multilayer Films at Gold
Surfaces. Ultrathin (<50 nm) noncentrosymmetric organic films
were formed on gold substrates by the sequential deposition of
self-assembled ZP monolayers of the molecule HAPA (see Figure
1). The HAPA molecule contains an azobenzene NLO chro-
mophore with sulfonyl electron-accepting and phenyl ether
electron-donating groups (with an absorption maximum at 365
nm) and is terminated with a phosphonic acid on one end and an
alcohol functional group on the other. This organic chromophore
was specifically chosen over others that contain amine electron-
donating substituents because its absorption maximum is blue-
shifted considerably from the visible region. This reduced the
possibility of resonant contributions to the electro-optical measure-
ments. A noncentrosymmetric HAPA multilayer is formed by the
method outlined by Katz et al.,21-24 in which HAPA adsorbs as a
self-assembled monolayer on a Zr4+ ion-coated substrate via the
phosphonic acid group, and then the terminal alcohol groups on
the other end of the molecule are converted to phosphates which
can bind a new monolayer of Zr4+ ions for the adsorption of a
subsequent HAPA monolayer. Films of up to 10 HAPA mono-
layers were created by this sequential self-assembly process.
These ultrathin noncentrosymmetric HAPA films are used in
section C for the study of electric fields within organic films at
electrode surfaces via the electro-optical effect.
Figure 3. Mid-infrared region of the PM-FTIRRAS spectrum of a
seven-layer ZP/HAPA film on a vapor-deposited thin gold film. The
gold surface was primed with a self-assembled monolayer of phos-
phorylated MUD before the formation of the ZP films. The frequencies
and band assignments for the spectrum are listed in Table 1. The
inset graph plots the intensity of the 1140 cm-1 symmetric sulfonyl
stretching band as a function of the number of HAPA monolayers
during the formation of a 10-layer ZP film. The straight line is a linear
fit to the data. This linear dependence is observed for all of the bands
listed in Table 1.
Table 1. Vibrational Band Frequencies and
Assignments for the HAPA Multilayers65-68
frequency
(cm-1
)
band
assignment
HAP A Film
2930 ( 1
2855
1599
1585
νa(CH2)
νs(CH2)
methylene stretch
methylene stretch
ring stretch
ring stretch
ring stretch
ν8a(CC)a
Multilayer films of HAPA were formed on vapor-deposited gold
substrates using a primer monolayer and deposition conditions
that were optimized previously.26,38 The primer layer for the gold
substrates consisted of a self-assembled monolayer of MUD, in
which the terminal hydroxy group is converted to a phosphate
species via a surface reaction.38 This primer layer is then exposed
to an aqueous solution of Zr4+ ions. As reported previously, PM-
FTIRRAS measurements show that this primer monolayer has a
well-organized zirconium phosphate structure that can serve as
the basis for the creation of ZP multilayer films.38 A HAPA
monolayer is self-assembled onto this primer layer by exposure
to a 0.5 mM ethanolic HAPA solution over a period of 16 h. This
monolayer adsorption is followed by phosphorylation of the
terminal alcohol groups with phosphorus oxychloride and then
ν
ν
8b(CC)a
19a(CC)a
1500
1469
δ(CH2), ν19b(CC)a
CH2 scissoring deformation
ring stretch
1400
1298
1254
1140
1088
δ(R-CH2)
R-CH2 scissoring deformation
sulfonyl stretch
phenyl ether stretch
sulfonyl stretch
phosphonate/ phosphate
stretch
νa(SO2)
νa(COC)
νs(SO2)
νa(PO33-), νa(PO4
)
2-
a
Frequency modes and assignments follow the numbering conven-
tion of Wilson for p-substituted benzenes and azobenzenes.50,69,70
The PM-FTIRRAS spectra of the ultrathin films yield informa-
tion about the molecular structure and orientation within the self-
assembled multilayer. In previous papers, we have demonstrated
that PM-FTIRRAS is particularly well-suited to studying the
formation of ZP multilayer films on gold surfaces.26,38 The mid-
infrared region of the PM-FTIRRAS spectrum for a seven-layer
HAPA film self-assembled onto a primer layer on a gold surface
is shown in Figure 3. The frequencies and assignments of the
various infrared bands are listed in Table 1. Absorption bands at
1599, 1585, and 1500 cm-1 are assigned to the azobenzene
functionality by comparison with bulk infrared spectra.50 The
absorption band at 1088 cm-1 has been observed previously in
other ZP multilayer films and is attributed to the phosphate and
phosphonate groups in the HAPA multilayer.26,38 Two additional
subsequent exposure to an aqueous Zr4+ solution to form a Zr4+
-
terminated monolayer. Repetition of these steps leads to the
formation of multilayer HAPA films. For the electro-optical studies
at electrochemical surfaces, a terminal monolayer of DHP was
self-assembled onto the ZP monolayer in order to produce a very
hydrophobic surface that would reduce the penetration of elec-
trolyte into the film. Each of the deposition steps in the multilayer
film formation process was verified with a combination of PM-
FTIRRAS and SPR measurements.
(47) Mizrahi, V.; Sipe, J. E. J. Opt. Soc. Am. B 1 9 8 8 , 5, 660.
(48) Heinz, T. F. In Nonlinear Surface Electromagnetic Phenomena; Ponath, H.
E., Stegeman, G. I., Eds.; North-Holland: Amsterdam, 1991.
(49) Corn, R. M.; Higgins, D. A. In Characterization of Organic Thin Films; Ulman,
A., Ed.; Butterworth-Heineman: Woburn, MA, 1995.
(50) Klima, M. I.; Kotov, A. V.; Gribov, L. A. J. Struct. Chem. 1 9 7 2 , 13, 987.
244 Analytical Chemistry, Vol. 69, No. 2, January 15, 1997

prominent bands appear at 1298 and 1140 cm-1 and are assigned
as the sulfonyl antisymmetric and symmetric stretching modes,
respectively. The presence of these two bands in the PM-
FTIRRAS spectrum suggests that the NLO chromophores in the
HAPA multilayer film are not oriented completely normal to the
gold surface. If they were, then these vibrational modes would
be parallel to the gold surface and would not be observed due to
the surface selection rules at metal surfaces.51 The appearance
of these bands indicates that there is a significant amount of tilting
of the azobenzene chromophores in the HAPA multilayer film.
This effect has been observed previously in other azobenzene/
ZP films26 and has been attributed to the inability of these films
to form highly packed ZP multilayers due to the mismatch of the
azo dye substituent size and the lattice spacing of the inorganic
regions in the ZP film.52 The amount of tilt does not change as a
function of the number of HAPA monolayers since the relative
band intensities in the infrared spectrum of the HAPA film did
not change during the multilayer deposition process. In Figure
3, the inset graph plots the intensity of the 1140 cm-1 band as a
function of the number of HAPA monolayers during the formation
of a multilayer film. The linear increase in the intensity of this
band indicates that each additional self-assembly step creates an
additional HAPA monolayer with an equivalent molecular structure
and packing density. This linear increase was observed for all of
the bands listed in Table 1.
Figure 4. Shifts in the SPR angle (∆θ) measured during the
formation of a 10-monolayer ZP/HAPA film onto a primer layer on a
thin (470 Å) gold film that is in optical contact with a BK7 prism. The
reflectivity of p-polarized light at 632.8 nm from the BK7/Au/HAPA
sample is monitored as a function of incident angle, θ (see inset).
The shifts ∆θ were determined from SPR reflectivity curves with the
SPR angle of the primer layer, denoted as the “0 layer” surface and
defined as ∆θ ) 0. The three lines in the figure are the results of
complex Fresnel calculations using an index of refraction of 1.65 (as
determined from ellipsometric experiments) for the ZP multilayer and
a HAPA monolayer thickness of 26, 27, or 28 Å. From the
experimental data, a HAPA monolayer thickness of 27 ( 0.5 Å is
determined.
In conjunction with the PM-FTIRRAS measurements, SPR
measurements were used to provide quantitative information on
the thickness of the HAPA multilayer films. We have demon-
strated previously the utility of the SPR method for assessing film
thicknesses of ZP multilayers.26 Following the approach used in
our previous paper, the shifts that occurred in the SPR angle (∆θ)
were measured during the sequential deposition of a HAPA
multilayer film on a thin (470 Å) gold substrate. Figure 4 plots
the shifts in the SPR angle that were obtained after the addition
of each self-assembled HAPA monolayer in a 10-layer HAPA film.
Five phase complex Fresnel calculations are used to interpret
these shifts in SPR angle;26,44,45 these calculations used a value of
1.65 for the index of refraction of the ultrathin HAPA multilayer,
which was determined from ellipsometric measurements of HAPA
films on fused-silica substrates (see below). This index of
refraction is consistent with that observed for similar azobenzene
ZP films.26 With these Fresnel calculations, the shifts in the SPR
angle can be used to ascertain an average film thickness for the
HAPA monolayers; the solid and dashed lines in Figure 4 are
theoretical curves generated from the Fresnel calculations for
monolayer thicknesses of 26, 27, and 28 Å. The fit to the
experimental data indicates that the HAPA monolayers in the ZP
multilayer films exhibit a thickness of 27 ( 0.5 Å. A space filling
model of the HAPA multilayer ZP film in which the azobenzene
functionality is oriented perpendicular to the surface and the two
B. Noncentrosymmetric HAP A Multilayer Films at Silica
Surfaces. In conjunction with the HAPA multilayers formed on
gold substrates, noncentrosymmetric HAPA multilayers were
formed on fused-silica substrates in order to characterize the linear
and nonlinear optical properties of the thin films. The primer layer
on the transparent substrates consisted of a self-assembled
monolayer of (10-carbomethoxydecyl)dimethylchlorosilane that
was chemically converted to a phosphate species on the surface
as detailed previously.26 This phosphate-terminated surface was
then exposed to an aqueous solution of Zr4+. Once the initial Zr4+
coated surface was created, the adsorption of the HAPA mono-
layers was performed in an fashion identical to those on the gold
substrates. The characterization of the ZP multilayer films on the
silica substrates employed a combination of UV/ visible absorption,
ellipsometry, and SHG measurements.
-
The UV/ visible absorption spectrum for a four-layer HAPA film
is plotted in Figure 5. A large charge transfer absorption band is
observed at 365 nm; this absorption maximum is red-shifted by
only 3 nm from its value in dilute ethanolic solutions. The small
shift indicates that there is very little interaction between the NLO
chromophores in the ZP multilayer films. As detailed in previous
papers, the sensitivity of the surface plasmon resonance technique
depends on both the thickness and index of refraction of the
absorbed film.26,44 Therefore, different combinations of thickness
and index of refraction can be used to fit measured shifts in the
SPR angle. To assist with the interpretation of the SPR measure-
ments at gold surfaces, ellipsometric measurements at 633 nm
were performed on 5, 8, and 10-multilayer HAPA films on silica
substrates in order to determine a film index of refraction of 1.65
( 0.04. A thickness per HAPA monolayer of 27 ( 6 Å was also
determined from the ellipsometric data. The variability of these
thickness values on silica substrates was significantly greater than
methylene chains are in all-trans configurations predicts a Zr4+
/
HAPA monolayer thickness of approximately 31-32 Å.53,54 The
experimentally determined value of 27 Å is in agreement with
the chromophore tilting observed in the PM-FTIRRAS spectra of
the multilayer films.
(51) Pearce, H. A.; Sheppard, N. Surf. Sci. 1 9 7 6 , 59, 205.
(52) Alberti, G.; Costantino, U.; Marmottini, F.; Vivani, R.; Zappelli, P. Angew.
Chem., Int. Ed. Engl. 1 9 9 3 , 32, 1357.
(53) Dines, M. B.; DiGiacomo, P. M. Inorg. Chem. 1 9 8 1 , 20, 92.
(54) Zeppenfeld, A. C.; Fiddler, S. L.; Ham, W. K.; Klopfenstein, B. J.; Page, C.
J. J. Am. Chem. Soc. 1 9 9 4 , 116, 9158.
Analytical Chemistry, Vol. 69, No. 2, January 15, 1997 245

Figure 5. Surface UV/visible absorption spectrum for a four-layer
ZP/HAPA film formed onto a primer layer on a fused-silica surface.
The absorption maximum at 365 nm is associated with the charge
transfer transition for this molecule. The absorption maximum is
slightly shifted from the solution spectrum, indicating only a small
amount of interaction between HAPA chromophores takes place in
the ZP film.
Figure 6. Observed second harmonic generation intensity obtained
for a five-layer ZP/HAPA film. The HAPA multilayer for these
measurements is self-assembled onto a phosphorylated silane primer
layer on a fused-silica substrate. A fundamental input wavelength
of 730 nm is employed to be in resonance with the HAPA charge
transfer electronic transition at the second harmonic wavelength of
365 nm. The SHG signal at 365 nm is obtained in a reflection
geometry from the top surface. (See inset). The solid line is a
quadratic fit to the observed data.
that observed for the ZP multilayer films created on gold
surfaces.26 This difference reflects the variability in the primer
layer formation on the fused-silica substrates.
angle θ of 27 ( 2° from the surface normal if all of the HAPA
molecules in the film have the same tilt angle (i.e., if cos3 θ ≈
cos3 θ and cos θ ≈ cos θ ). The orientation parameter and
tilt angle are consistent with the average thickness as determined
from SPR experiments and the observed PM-FTIRRAS spectra of
HAPA multilayers on gold surfaces.
The noncentrosymmetric nature of the HAPA multilayer films
on silica substrates was assessed with resonant optical SHG
measurements. SHG measurements have been made previously
on ZP multilayer films.22,23 The SHG measurements in this work
are performed at a fundamental wavelength of 730 nm. At this
wavelength, the measurements are in resonance with the HAPA
charge transfer transition at the second harmonic wavelength of
365 nm. We have previously used the resonant-enhanced SHG
from similar NLO azobenzene chromophores to study adsorption
to liquid/ liquid interfaces.55,56 Figure 6 plots the measured second
harmonic signal as a function of the number of HAPA monolayers
in a five-layer ZP film on a fused silica substrate. The solid line
in the figure is a quadratic fit to the data that arises from the
dependence of the SHG signal on the square of the surface
concentration of oriented SHG-active chromophores at the inter-
face. These SHG measurements clearly indicate that the self-
assembly of the HAPA multilayers produces a uniformly noncen-
trosymmetric ultrathin organic film.
In addition to the magnitude of the SHG from the HAPA
multilayer films, the polarization dependence of the signal from
the multilayers was used to examine the average orientation of
the NLO chromophores in the films. This polarization depen-
dence was determined by measuring the intensity of the s-
polarized and p-polarized SHG signals as a function of the
polarization of the incident fundamental light.47-49 An analysis of
this polarization dependence (not shown) resulted in the deter-
mination of an orientation parameter, D ) cos3 θ / cos θ , where
θ is the angle between the primary axis of the NLO chromophore
(in this case the long axis of the HAPA molecule) and the surface
normal.57 A value of D ) 0.79 ( 0.03 was obtained for the HAPA
multilayer films on silica substrates; this value corresponds to an
C. Electro-Optic Coefficient of a HAP A Monolayer. De-
termination of the linear electro-optical properties of a HAPA
monolayer film was accomplished by the use of modulated surface
plasmon resonance measurements. This method has been used
previously for determination of the electro-optic coefficients of
ultrathin Langmuir-Blodgett NLO organic films by Cross and co-
workers,58,59 and micrometer-thick NLO poled polymer films by
Sekkat et al.13 In these experiments, the ultrathin NLO organic
film is deposited onto a thin (470 Å) gold film that serves as one
electrode of an air-gap capacitor (see the inset in Figure 7). An
external ac voltage (on the order of 10-50 V with a frequency of
1-10 kHz) is applied to the capacitor in order to create an
oscillating electrostatic field on the order of 104 V/ cm that
modulates the index of refraction of the ultrathin NLO organic
film via the electro-optical effect. This change in the index of
refraction results in a shift of the SPR angle that can be observed
by detecting the modulated change in reflectivity as a function of
incident angle, θ.
Figure 7 plots the modulated change in reflectivity (∆%R) as
a function of incident angle near the surface plasmon angle (θsp)
for a single HAPA monolayer film. This signal was obtained from
application across the air gap of a (15 V sinusoidal voltage at a
frequency of 10 kHz. The thickness of the air gap in the capacitor
in this experiment was determined from capacitance measure-
(57) Heinz, T. F.; Tom, H. W.; Shen, Y. R. Phys. Rev. A 1 9 8 3 , 28, 1883.
(58) Cross, G. H.; Girling, I. R.; Peterson, I. R.; Cade, N. A.; Earls, J. D. Electron.
Lett. 1 9 8 6 , 22, 1111.
(59) Cross, G. H.; Girling, I. R.; Peterson, I. R.; Cade, N. A.; Earls, J. D. J. Opt.
Soc. Am. B 1 9 8 7 , 4, 962.
(55) Naujok, R. R.; Higgins, D. A.; Hanken, D. G.; Corn, R. M. J. Chem. Soc.,
Faraday Trans. 1 9 9 5 , 91, 1411.
(56) Naujok, R. R.; Paul, H. J.; Corn, R. M. J. Phys. Chem. 1 9 9 6 , 100, 10497.
246 Analytical Chemistry, Vol. 69, No. 2, January 15, 1997

from øzzz(2) and the tilt angle of the chromophore units by assuming
a number density for the HAPA film on the Au surface.55 The
estimated value for âzzz of 9.3 × 10-29 esu at 632.8 nm, while larger
than sum-over-state (SOS) and finite-field (FF) theoretical calcula-
tions made at 1907 nm for similar NLO chromophores, is
reasonable given the dispersion relationship of â vs wavelength.14,61
The magnitude of the electro-optic coefficient in these ZP films is
less than LiNbO3 but is within the range of the values observed
in NLO poled polymer films.5,9-14 Moreover, the noncentrosym-
metric character of the HAPA monolayer did not decrease with
time, as is often the case for the polymer systems.
D. Electric Field Measurements at Electrochemical
Surfaces. As an example of the ultility of the noncentrosymmetric
HAPA multilayer films, a HAPA monolayer was formed at a gold
electrode surface in order to measure the electrostatic fields in
an electrochemical environment. Electric fields at electrochemical
interfaces are thought to vary from 103 to 108 V/ cm, depending
on the exact nature of the electrode surface.62,63 Previous optical
studies have used electroreflectance,34 fluorescence,35-37 and SHG
measurements27,31-33 to examine the electric fields at electro-
chemical interfaces. When the electrode is coated with an
ultrathin organic film, the electric fields are expected to be
significantly lower than in the case of a simple metal/ electrolyte
interface.28-30 Nevertheless, these electric field strengths can be
monitored via the electro-optical effect for the noncentrosymmetric
HAPA multilayers.
A rough schematic diagram of the experiment is pictured in
the inset of Figure 8. Instead of an air-gap capacitor, a Teflon
cell is pressed against the HAPA-modified gold slide with an o-ring
seal. A three-electrode potentiostat with the 470 Å thin gold film
as the working electrode is used to modulate the potential and
interfacial electric fields. Cyclic voltammetry of the HAPA-
modified gold slides in 0.2 M tetrabutylammonium bromide
exhibited no faradaic processes in the range from -0.4 to 0.7 V
vs SCE. For the SPR modulation experiments, the electrode
potential was fixed to 0.0 V vs SCE, and a small sinusoidal wave
form was applied ((25 mV at 1 kHz) to the electrochemical
system.
The in situ modulated surface plasmon spectra that are
observed from this modulation are plotted in Figure 8. Two
modulation experiments are shown in the figure: one for a ZP
multilayer that contains HAPA and one that does not. The open
circles are the data obtained for a ZP multilayer that does not
contain HAPA, but instead consists of a primer monolayer of
phosphorylated MUD, followed by two layers of DBP, and one
capping layer of DHP. The modulation in the SPR signal in these
data results from electro-optical contributions of the gold electrode
surface. This effect was not observed in the air-gap experiment
to within the signal-to-noise ratio of the experiment, but changes
in the optical constants of metal electrode surfaces have been
observed previously in electroreflectance measurements.62,64 The
solid line that fits these data is a seven-phase Fresnel calculation
Figure 7. Differential reflectivity (∆%R) obtained by modulated
surface plasmon resonance measurements in air for a ZP film
consisting of a monolayer ZP/HAPA film formed onto a primer layer
on a thin gold film. This signal is attributed to a modulation in the
dielectric constants of the HAPA monolayer through the nonlinear
electro-optical (Pockel’s) effect. The sample for this experiment
shown in the inset of the figure consists of a thin Mylar dielectric film
which is pressed between the ZP-modified gold sample and a second
gold electrode. A small square is cut out of the Mylar so the surface
plasmon wave is evanescent in air upon excitation. The thickness
of the resulting air gap is calculated to be 15 µm. Electrical contact
is made to the two gold surfaces, and a sinusoidal wave form is
applied across the gap ((15 V, 10 kHz). The differential reflectivity
caused by this modulation is recorded as a function of the incident
angle θ. The solid line in the figure is a complex Fresnel fit to the
data obtained by varying the real component of the dielectric constant
of the HAPA layer.
ments to be 15 µm. A static or zero frequency dielectric constant
of 6 was assumed for the ZP/ HAPA monolayer films. This value
was chosen to match the experimentally determined dielectric
constant for similar NLO azo dye multilayer ZP films reported by
Katz et al.24,60 The conditions of this experiment correspond to
an electric field strength of 3.3 × 103 V/ cm within the ZP/ HAPA
monolayer film. The change in reflectivity at 10 kHz was obtained
by lock-in amplifier detection of the photodiode current (see the
Experimental Section for more details) and has the functional form
of the derivative of the standard surface plasmon reflectivity curve.
The magnitude of the differential reflectivity (∆%R) increased
linearly with the applied ac voltage for the HAPA sample as
expected and was essentially constant over the frequency range
of 1-10 kHz. In addition, no signal was observed in the absence
of the HAPA monolayer. The solid line in Figure 7 is a five-phase
Fresnel calculation in which the real part of the index of refraction
for the HAPA monolayer is varied by ∆n ) 8.0 × 10-6. A similar
change in the SPR reflectivity can be obtained by a change in the
thickness of the film with applied voltage (∆d), but we assume
that all changes are due to index of refraction effects via the
electro-optic coefficient. A change in ∆n of 8.0 × 10-6 corresponds
to an electro-optic coefficient r33 of 11 pm/ V at 632.8 nm.13 This
value of r33 is an off-resonance measurement since no absorption
of light in the monolayer film occurs at this wavelength. From
the value of r33, the surface nonlinear susceptibility øzzz(2)(-ω;ω,0)
for the ZP film is calculated to be 41 pm/ V (9.7 × 10-8 esu).13,14
The HAPA molecular hyperpolarizability âzzz can then be estimated
(61) Ulman, A.; Willand, C. S.; Ko¨ hler, W.; Robello, D. R.; Williams, D. J.; Handley,
L. J. Am. Chem. Soc. 1 9 9 0 , 112, 7083.
(62) Kolb, D. M. In Surface Polaritons; Agranovich, V. M., Mills, D. L., Eds.;
North-Holland: Amsterdam, 1982.
(63) Adamson, A. W. Physical Chemistry of Surfaces, 5th ed.; Wiley: New York,
1990.
(64) Kolb, D. M. In Spectroelectrochemistry: Theory and Practice; Gale, R. J., Ed.;
Plenum Press: New York, 1988.
(65) Bellamy, L. J. The Infra-red Spectra of Complex Molecules; Wiley: New York,
1975.
(60) Katz, H. E.; Schilling, M. L. Chem. Mater. 1 9 9 3 , 5, 1162.
Analytical Chemistry, Vol. 69, No. 2, January 15, 1997 247

monolayer and the modulation of the index of refraction of the
gold surface into account. This theoretical fit yields a value for
the change in index of refraction for the HAPA monolayer of ∆n
) 3.3 × 10-5
. This value is a factor of 4 greater than that
measured in the air-gap experiment and therefore corresponds
to a change in electric field strength during potential modulation
of roughly 1 × 104 V/ cm within the ultrathin organic film. This
value is close to that expected for a potential modulation of (25
mV across an ultrathin organic film with a thickness of 70 Å
(which is the thickness expected for the MUD/ HAPA/ DHP
multilayer). The observation of an electric field strength within
the multilayer of 1 × 104 V/ cm indicates that there is only minimal
electrolyte penetration into this ZP multilayer in the electrochemi-
cal environment. The magnitude of the electric field strength is
similar to that determined by Pope et al. for fluorescent hemi-
cyanine chromophores imbedded in self-assembled monolayers
on roughened silver electrodes, although their measurements
actually probed the diffuse region of the interfacial electric double
layer.35
Figure 8. Differential reflectivity (∆%R) obtained by modulated
surface plasmon resonance measurements made in situ. The sample
for this experiment shown in the inset of the figure consists of a Teflon
cell which is pressed against the ZP-modified gold sample. A three-
electrode assembly is made with the ZP-modified gold sample as
the working electrode. A potentiostat controlled the applied potential
relative to SCE. The electrolyte consisted of 0.2 M tetrabutylammo-
nium bromide in water. A small sinusoidal wave form ((25 mV, 1
kHz) was then applied to the cell at a potential where no electro-
chemistry took place. The open circles are obtained for a ZP film
consisting of two monolayers of DBP, and one monolayer of DHP
formed onto a primer layer on a thin gold film. This signal results
from a modulation in the gold dielectric constants. The open squares
are obtained for a ZP film consisting of one monolayer of HAPA and
one monolayer of DHP formed onto a primer layer on a thin gold
film. The solid lines for both sets of data are complex Fresnel fits
obtained by varying the real component of the dielectric constant of
the HAPA layer and the gold.
SUMMARY AND CONCLUSIONS
In summary, the experiments described in this paper demon-
strate that ultrathin noncentrosymmetric ZP films that incorporate
the nonlinear optical chromophore HAPA can be formed on gold
and fused silica substrates. A combination of PM-FTIRRAS, SPR,
and SHG measurements reveal that the multilayer HAPA films
self-assemble in a reproducible manner with an index of refraction
of 1.65 ( 0.04 at 632.8 nm, an interlayer spacing of 27 ( 0.5
Å/ layer, and an average azobenzene chromophore tilt angle of
27 ( 2° from the surface normal. An electro-optic coefficient for
the ultrathin HAPA multilayer films (r33) of 11 pm/ V at 632.8 nm
was determined from electric field-modulated surface plasmon
resonance measurements in air-gap capacitors. These preliminary
results indicate that the self-assembly technique may serve as an
alternative route to the formation of noncentrosymmetric thin films
for electro-optical applications, as has been suggested by other
researchers.21-24,60
Noncentrosymmetric HAPA films were also shown to be
valuable probes of the electrostatic fields in organic monolayers
at electrochemical interfaces. A comparison of the in situ electro-
optical response of ZP films at electrode surfaces with the results
obtained from air-gap capacitors resulted in an in situ measure-
ment of 1 × 104 V/ cm for the electric field strength within the
organic thin film. This field strength suggests that for this
particular ZP film the electrolyte ions that form the electrochemical
double layer at the interface do not penetrate into the organic
film and reside primarily on the surface of the multilayer. Future
experiments will focus on the systematic incorporation of HAPA
chromophores into self-assembled multilayers in order to deter-
mine the electric field profile within various ultrathin organic films
at electrochemical interfaces.
which assumes that the real part of the index of refraction of the
first 0.5 Å of the gold electrode surface changes upon modulation
of the electrode potential by an amount ∆n ) 1.6 × 10-3. The
thickness of this layer is taken as 0.5 Å in order to approximate
the Thomas-Fermi screening depth of the metal.64 Increasing
this thickness reduced the size of ∆n but had no effect on the
shape of the calculated fit. Note that, for these data, the minimum
in ∆%R occurs at the surface plasmon angle θsp. This is in contrast
to the air-gap capacitor data for the HAPA monolayer (Figure 7).
The second SPR modulation experiment plotted in Figure 8
(the open squares) was obtained for a single-monolayer HAPA
film formed on a phosphorylated MUD primer monolayer and
capped with a DHP monolayer. The differential reflectivity curve
for this sample is significantly different in both shape and
magnitude as compared to the curve observed for the DBP
sample, with the minimum and maximum in ∆%R occurring above
and below θsp as in the case for the air-gap capacitor experiments.
The solid line that fits the data is a seven-phase Fresnel fit taking
both the modulation of the index of refraction of the HAPA
ACKNOWLEDGMENT
The authors gratefully acknowledge the support of the National
Science Foundation in these studies. The authors also thank Dr.
J. W. Taylor and his research group for the use of the ellipsometer
in these studies.
(66) Socrates, G. Infrared Characteristic Group Frequencies; Wiley: Chichester,
England, 1980.
(67) Lin-Vien, D.; et al. The Handbook of Infrared and Raman Characteristic
Frequencies of Organic Molecules; Academic: Boston, 1991.
(68) Thomas, L. C. Interpretation of the Infrared Spectra of Organophosphorus
Compounds; Heyden: London, 1974.
Received for review July 26, 1996. Accepted August 1,
1996.X
(69) Sverdlov, L. M.; Kovner, M. A.; Krainov, E. P. Vibrational Spectra of
Polyatomic Molecules; Wiley: New York, 1974.
AC960756+
(70) Bouquet, M.; Chassaing, G.; Corset, J.; Favrot, J.; Limouzi, J. Spectrochim.
Acta 1 9 8 1 , 37A, 727.
X Abstract published in Advance ACS Abstracts, September 15, 1996.
248 Analytical Chemistry, Vol. 69, No. 2, January 15, 1997