Synthesis and nonlinear optical properties of side chain liquid crystalline polymers containing azobenzene push-pull chromophores

Article abstract of DOI:10.1002/pola.23776

Azobenzene monomerlc precursors bearing piperazine as donor moiety with different withdrawing groups and derived side chain polymethacrylates have been prepared and characterized. Monomers having terminal cyano or nitro groups, and the corresponding polym

Full text of DOI:10.1002/pola.23776

Synthesis and Nonlinear Optical Properties of Side
Chain Liquid Crystalline Polymers Containing Azobenzene
Push–Pull Chromophores
2
1
RAQUEL ALICANTE,1 RAFAEL CASES,1 PATRICIA FORCEN, LUIS ORIOL,2 BELEN VILLACAMPA
´
´
1Dept. F´ısica de la Materia Condensada, Facultad de Ciencias-Instituto de Ciencia de Materiales de Arago´ n,
Universidad de Zaragoza-CSIC, Pedro Cerbuna 12, Zaragoza 50009, Spain
2Dept. Qu´ımica Orga´nica, Facultad de Ciencias-Instituto de Ciencia de Materiales de Arago´ n, Universidad de
Zaragoza-CSIC, Pedro Cerbuna 12, Zaragoza 50009, Spain
Received 14 September 2009; accepted 5 October 2009
DOI: 10.1002/pola.23776
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: Azobenzene monomeric precursors bearing pipera-
zine as donor moiety with different withdrawing groups and
derived side chain polymethacrylates have been prepared and
characterized. Monomers having terminal cyano or nitro
groups, and the corresponding polymers, exhibited smectic A
phases. Linear and nonlinear optical properties of every mono-
mer and thin films of the cyano polymer (pol-PZ-CN) have been
also studied. UV-vis spectroscopy revealed out-of-plane orienta-
tion in the as prepared films, as confirmed by waveguide refrac-
tive index measurements. Moreover, absorption spectra
indicated the presence of azo aggregates in these films. The ini-
tial molecular arrangement has been modified by applying ther-
mal annealing within the mesophase range and UV-blue
irradiation. Although thermal annealing resulted in a significant
amplification of the out-of-plane optical anisotropy due to ther-
motropic self-organization of side chain azo moieties, irradiation
with 440 nm light induced some disruption of aggregates. The
nonlinear optical response of Corona poled films has been stud-
ied by second harmonic generation measurements, and the
influence of the molecular arrangement on the nonlinear dij
coefficients has been analyzed. The more efficient poling corre-
sponded to preirradiated films. In any case, a noticeable degree
of polar order (70% of the initial d33 value) remained for several
C
V
months after the poling in films kept at RT. 2009 Wiley Period-
icals, Inc. J Polym Sci Part A: Polym Chem 48: 232–242, 2010
KEYWORDS: azo polymers; dij coefficients; EFISH; liquid-crystal-
line polymers (LCP); NLO
INTRODUCTION The study of organic materials with nonlin-
ear optical (NLO) properties involves a great interest for dif-
ferent electrooptical and photorefractive applications. In partic-
ular, polymers provide systems with high flexibility due to its
synthetic versatility and easy processability.1 Nevertheless, one
of the most studied strategies for designing second order
nonlinear polymers not only implies the incorporation of dipo-
lar, highly polarizable donor-p-acceptor molecules (normally
used as NLO-phores) into a macromolecular structure but also
a noncentrosymmetric organization. The experimental tech-
nique generally used for this purpose is based on the electric
field poling. In a typical poling process, thin polymer films are
heated near glass transition temperature (Tg) under a strong
electric field and then cooled to room temperature (RT) with
the field on. After the poling process, NLO chromophores show
a preferential orientation along the field direction, defining an
optical axis perpendicular to the film surface (C1v symmetry).
are stable NLO-phores that can be easily incorporated in
polymeric systems both as guest molecules and as a struc-
tural part of the side or main chain of polymers. On the
other hand, the photoaddressability of these systems can
play an important role in assisting chromophore alignment.3
On illumination with light of appropriate wavelength, azo-
benzene moieties undergo E-Z reversible isomerization and
the molecular motion of NLO-phores is enhanced, so it is
possible to align them along an applied DC electric field at
temperatures well below the Tg of the polymer. This process
that combines light and electrical poling is known as photo-
assisted poling (PAP),4 and it allows inducing NLO properties
on azopolymers while avoiding thermal decomposition.
We have previously reported the photo-induced anisotropy5
and nonlinear optical properties of side chain liquid crystal-
line polymers (SCLCP) having 4-alcoxy-40-cyanoazobenzene
derivatives as chromophores. Concerning nonlinear proper-
ties, a stable polar order and, consequently, NLO response
was achieved in a series of methacrylic homo and copoly-
mers with different azo content.5(b) Higher response is
Azobenzene-containing polymers are a really interesting al-
ternative in nonlinear optics.2 On the one hand, azobenzene
derivatives having appropriate donor and acceptor groups
Correspondence to: B. Villacampa (E-mail: bvillaca@unizar.es)
C
V
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 48, 232–242 (2010) 2009 Wiley Periodicals, Inc.
232
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
Monomer hyperpolarizabilities have been determined in
solution. Thin films of pol-PZ-CN have been prepared and
poled by Corona discharge, to induce second order NLO
properties. For a better understanding of the competition in
alignment between intermolecular and external forces, we
have included some results concerning a previously reported
monomer and the corresponding polymer13 (mon-O-CN and
pol-O-CN, respectively, see Scheme 1), which have been
measured under equivalent experimental conditions. Diverse
treatments have been applied to polymer films of different
thickness, in particular thermal annealing at temperatures
within the mesophase range and/or UV-blue light irradiation.
Special attention has been paid to the role that this previous
processing, which modifies the molecular arrangement and
aggregation state of the films, plays on their final NLO
response, characterized by second order dij coefficients.
EXPERIMENTAL
SCHEME 1 Polymers synthesized for the study.
Synthesis
6-Chlorohexanol, 4-nitroaniline, 4-cyanoaniline, and metha-
cryloyl chloride were purchased from Aldrich and used with-
out previous purification. 4-Aminobenzaldehyde was pre-
pared from N-(4-formylphenyl) acetamide according to a
previously reported method.14
expected to be obtained if more efficient chromophores are
incorporated into the polymer, provided that a efficient polar
orientation of NLO moieties is induced by the electric field.
It has been well established that dipolar interactions
between guests NLO chromophores dispersed in a polymeric
matrix attenuate noncentrosymmetric order.6 Aggregation, as
a consequence of these interactions, is a detrimental factor
for electric poling efficiency and isotropic poled polymer
model7 does not explain the observed NLO macroscopic
response, even at moderate chromophore loading. The role
of dipole moment and chromophore concentration in hinder-
ing polar order has been also discussed in side chain
azopolymers.8 On the other hand, the liquid crystalline char-
acter has been predicted to enhance field-induced polar
order.9 Experimental results supporting this statement were
published by several authors.10 However, it has been also
reported the failure in poling smectic A SCLCP11 because of
the strong dipole–dipole interactions.
Synthesis of Monomers
N-(6-Hydroxyhexyl)-N0-Phenylpiperazine (1)
Phenylpiperazine (31 mmol), 32 mmol of 6-chloro-1-hexanol,
31 mmol of K2CO3, and 1 mmol of KI were mixed in 50 mL
of acetone. The reaction mixture was refluxed overnight,
cooled to room temperature, and finally filtered to remove
inorganic salts. The solvent was removed by evaporation
under vacuum to obtain the crude, which was purified by col-
umn chromatography on silica gel using dichloromethane/
methanol (9:1) as eluent and then washed with hexane.
Yield: 50%. ELEM. ANAL. (calculated): %C 74.19 (73.42), %H
9.43 (9.99), %N 10.38 (10.68). IR (NaCl, Nujol, cmꢀ1): 3290–
3219 (OH); 1598, 1500 (Ar); 1244 (CAO). 1H-NMR (CDCl3)
d/ppm: 7.28–7.24 (m, 2H), 6.9 (d, J ¼ 8.8 Hz, 2H), 6.8 (t, J ¼
7 Hz, 1H), 3.6 (t, J ¼ 1.6 Hz, 2H), 3.2 (m, 4H), 2.6 (m, 4H),
2.41 (t, J ¼ 8 Hz, 2H), 1.62–1.55 (m, 4H), 1.43–1.37 (m, 4H).
N-(6-Hydroxyhexyl)-N0-[(4-X-Phenyl)Azophenyl] Piperazine,
X ¼ CN (2), NO2 (3), COH (4)
In this work, we present the synthesis, characterization, and
study of the NLO properties of a series of liquid crystalline
monomers and polymers having azobenzene chromophores
with a strong donor group as piperazine (linked by N to azo
unit) and different acceptor groups (Scheme 1). Side-chain
polymethacrylates with similar chromophores directly linked
to the main chain had been studied by Tirelli et al.12 obtain-
ing a weak macroscopic NLO response, fairly lower than the
observed one for the same chromophores dispersed into a
polymeric matrix. As it is expected that the use of flexible
spacers in the side chain would improve the orientability
and consequently the nonlinear response, the piperazine do-
nor group existing in our monomeric precursors is separated
from the methacryloyl moiety by a hexamethylenic spacer.
The configuration of the polymeric system makes a high NLO
response predictable and allows getting a deeper insight into
the influence that liquid crystalline properties have on non-
linear optical response. Thermal and mesomorphic proper-
ties of monomers and derived polymers will be described.
Obtained through diazonium coupling between N-(hydroxy-
hexyl)-N0-phenylpiperazine (1) and the diazonium salt of 4-
cyanoaniline, 4-nitroaniline or 4-aminobenzaldehyde: 4 mmol
of the corresponding aniline was dissolved in a mixture of
water (100 mL) and HCl (45 mL) and cooled in an ice bath
under stirring. A solution of NaNO2 (4 mmol) in water
(10 mL)ꢁ was added drop wise keeping the temperature
under 5 C, and finally N-(hydroxyhexyl)-N0-phenylpiperazine
was added to the mixture. The reaction was running at room
temperature for 24 h. The resulting solution was added to a
Na2CO3 saturated dissolution, and the precipitated was fil-
tered and rinsed with water. The crude product was finally
purified by column chromatography on silica gel using meth-
ylene chloride/methanol (9:1) as eluent.
NLO PROPERTIES OF SIDE CHAIN LIQUID CRYSTALLINE POLYMERS, ALICANTE ET AL.
233

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
(2): Yield: 40%. ELEM. ANAL. (calculated): %C 70.38 (70.56),
%H 7.36 (7.47), %N 17.63 (17.89). IR (NaCl, Nujol, cmꢀ1):
3542 (OH); 2225 (CN); 1600, 1515, (Ar); 1235, 1141 (CAO);
1.88 (s, 3H), 1.61–1.59 (m, 2H), 1.5–1.47 (m, 2H), 1.35–1.33
(m, 4H).
(5): Yield: 52%. ELEM. ANAL. (calculated): %C 70.91 (70.10),
%H 7.30 (7.41), %N 12.60 (12.11). IR (KBr pellet, cmꢀ1):
2816, 2778 (HC¼¼O); 1711 (C¼¼O); 1694 (HC¼¼O); 1632
(C¼¼C); 1596, 1505 (Ar); 1232, 1133 (CAO); 841 (Ar). 1H-
NMR (CDCl3) d/ppm: 10 (s, 1H), 7.94–7.9 (m, 4H), 7.84 (d,
J ¼ 8.8 Hz, 2H), 6. 9 (d, J ¼ 9.2 Hz, 2H), 6.03 (s, 1H), 5.48 (s,
1H), 4.08 (t, J ¼ 6.8 Hz, 2H), 3.37–3.34 (m, 4H), 2.54–2.51
(m, 4H), 2.33 (t, J ¼ 7.2 Hz, 2H), 1.88 (s, 3H), 1.65–1.63 (m,
2H), 1.5–1.47 (m, 2H), 1.35–1.33 (m, 4H).
1
850 (Ar). H-NMR (CDCl3) d/ppm: 7.84–7.83 (m, 4H), 7.7 (d,
J ¼ 8.4 Hz, 2H), 6.9 (d, J ¼ 9.2 Hz, 2H), 3.6 (t, J ¼ 6.8 Hz,
2H), 3.73–3.33 (m, 4H), 2.55–2.52 (m, 4H), 2.34 (t, J ¼ 7.4
Hz, 2H), 1.52–1.49 (m, 4H), 1.33–1.32 (m, 4H) (3): Yield:
54%. ELEM. ANAL. (calculated): %C 64.18 (64.21), %H 7.12
(7.10), %N 16.95 (17.02). IR (NaCl, Nujol, cmꢀ1): 3224 (OH);
1602, 1507, (Ar); 1235, 1141 (CAO); 857 (Ar). 1H-NMR
(CDCl3) d/ppm: 8.3 (d, J ¼ 9 Hz, 2H), 7.95–7 (m, 4H), 6.95
(d, J ¼ 9 Hz, 2H), 3.65 (t, J ¼ 6.4 Hz, 2H), 3.44–3.41 (m, 4H),
2.61–2.58 (m, 4H), 2.4 (t, J ¼ 7.6 Hz, 2H), 1.58–1.55 (m, 4H),
1.4–1.38 (m, 4H). (4): Yield: 40%. ELEM. ANAL. (calculated):
%C 70.18 (70.02), %H 7.52 (7.66), %N 14.55 (14.20). IR
(KBr pellet, cmꢀ1): 3423 (OH); 2779 (HC¼¼O); 1688
(HC¼¼O); 1593, 1506, 1380 (Ar); 1233, 1141 (CAO); 839
N-(6-Methacryloyloxyhexyl)-N0-{4-[40-(2,2-Dicyanovinyl)-
Phenyldiazenyl]phenyl} Piperazine Mon-PZ-DCN
The monomer mon-PZ-DCN was prepared by mixing 1.73
mmol of (5), 2.09 mmol of malononitrile, a few drops of
NaOH 1% in 40 mL of ethanol, and 2 mg of inhibitor (2,6-
di-tert-butyl-p-cresol). The mixture was refluxed overnight.
The solvent was then evaporated in vacuum, and the mono-
mer was purified by column chromatography using methyl-
ene chloride/methanol (9:1) as eluent and finally recrystal-
lized from ethanol.
1
(Ar). H-NMR (CDCl3) d/ppm: 10.07 (s, 1H), 8–7.98 (m, 4H),
7.92–7.9 (d, J ¼ 8 Hz, 2H), 6.98–6.96 (d, J ¼ 8 Hz, 2H),
3.68–3.64 (t, J ¼ 6.8 Hz, 2H), 3.49 (s, 1H), 3.44–3.41 (m,
4H), 2.62–2.59 (m, 4H), 2.43–2.4 (t, J ¼ 7.2 Hz, 2H), 1.59–
1.54 (m, 4H), 1.4–1.39 (m, 4H).
Yield: 73%. ELEM. ANAL. (calculated): %C 70.38 (70.56), %H
6.32 (6.71), %N 16.40 (16.46). IR (KBr pellet, cmꢀ1): 2225
(CN); 1715 (C¼¼O); 1635 (C¼¼C ); 1597, 1513 (Ar); 1383,
1243 (CAO), 837 (Ar). 1H-NMR (CDCl3) d/ppm: 8.3 (d, J ¼
8.4 Hz, 2H), 7.96–7.9 (m, 4H), 6.96 (d, J ¼ 8.8 Hz, 2H), 6.1
(s, 1H), 5.55 (s, 1H), 4.14 (t, J ¼ 6.8 Hz, 2H), 3.44–3.41 (m,
4H), 2.4 (t, J ¼ 7.2 Hz, 2H), 1.95 (s, 3H), 1.70–1.68 (m, 2H),
1.57–1.55 (m, 2H), 1.40–1.38 (m, 4H).
N-(6-Methacryloyloxyhexyl)-N0-[4-(40-X-Phenyldiazenyl)-
phenyl] Piperazine, X ¼ CN (mon-PZ-CN), NO2
(Mon-PZ-NO2), COH (5)
The monomers mon-PZ-CN, mon-PZ-NO2 and the precursor
(5) were obtained by condensation of the corresponding
compound (2), (3), or (4) and methacryloyl chloride in
freshly distilled THF: 1.5 mmol of hydroxy derivative (3),
(4), or (5) and 2.2 mmol of methacryloyl chloride were dis-
solved in 250 mL of distilled THF, to which triethylamine
(2.19 mmol) and 2 mg of inhibitor (2, 6-di-tert-butyl-p-cre-
sol) were added. The reaction mixture was refluxed under
argon for 24 h and then cooled. Water (200 mL) was added,
and the solution was extracted with ethyl acetate, the or-
ganic layer was dried with anhydrous MgSO4. The solvent
was removed by evaporation in vacuum. The resulting prod-
uct was purified from column chromatography on silica gel
using methylene chloride/methanol (9:1) as eluent. The
monomers were finally recrystallized from ethanol and dried
under vacuum.
Synthesis of Polymers
Polymers were synthesized by free radical polymerization.
The corresponding monomer and AIBN (2.5% w/w in the
case of pol-PZ-CN, 3% w/w in the case of pol-PZ-NO2 and
5% w/w in the case of pol-PZ-DCN, see text), as a thermal
initiator, were dissolved in 10 mL of freshly distilled DMF.
Oxygen was removed by three freeze-pump-thaw cycles by
applying vacuum and backfilling with argon. The mixture
was immersed into a silicon oil bath, preheated to 70 ꢁC.
The reaction was maintained from 48 to 72 h, and the reac-
tion was then cooled to room temperature and poured into
ethanol. The solid was isolated by filtration. The polymers
were thoroughly washed by using a Soxlhet extractor.
Mon-PZ-CN. Yield: 60%. ELEM. ANAL. (calculated): %C 70.68
(70.56), %H 7.30 (7.24), %N 15.02 (15.24). IR (KBr pellet,
cmꢀ1): 1719 (C¼¼O); 1635 (C¼¼C); 1561, 1506 (Ar); 1235,
1140 (CAO); 848 (Ar). 1H-NMR (CDCl3) d/ppm: 7.84–7.83
(m, 4H), 7.69 (d, J ¼ 7.6 Hz, 2H), 6.89 (d, J ¼ 8.4 Hz, 2H),
6.03 (s, 1H), 5.48 (s, 1H), 4.08 (t, J ¼ 6.8 Hz, 2H), 3.37–3.34
(m, 4H), 2.54–2.51 (m, 4H), 2.33 (t, J ¼ 7.2 Hz, 2H), 1.88 (s,
3H), 1.63–1.61 (m, 2H), 1.5–1.47 (m, 2H), 1.35–1.33 (m, 4H).
Pol-PZ-CN
Yield: 40%. ELEM. ANAL. (calculated): %C 70.71 (70. 56), %H
7.32 (7.24), %N 15.13 (15.24). IR (KBr pellet, cmꢀ1): 2222
(CN); 1723 (C¼¼O); 1597, 1507, (Ar); 1233, 1155 (CAO);
846 (Ar). 1H-NMR (CDCl3) d/ppm: 7.86–7.85 (m, 4H), 7.73–
7.72 (m, 2H), 6.91–6.90 (m, 2H), 3.97–3.95 (m, 2H), 3.37–
3.36 (m, 4H), 2.57–2.55 (m, 4H), 2.38–2.35 (m, 2H), 1.64–
0.86 (m, 13H). Mn ¼ 10,400, Mw/Mn ¼ 1.2.
Mon-PZ-NO2. Yield: 80%. ELEM. ANAL. (calculated): %C 64.95
(65.12), %H 6.75 (6.94), %N 14.18 (14.60). IR (KBr pellet,
cmꢀ1): 1718 (C¼¼O); 1635 (C¼¼C); 1598, 1515 (Ar); 1235,
1141 (CAO); 857 (Ar). 1H-NMR (CDCl3) d/ppm: 8.27–8.25
(d, J ¼ 9.2 Hz, 4H), 7.88–7.83 (m, 4H), 6. 9 (d, J ¼ 9.2 Hz,
2H), 6.03 (s, 1H), 5.48 (s, 1H), 4.08 (t, J ¼ 6.8 Hz, 2H), 3.38–
3.35 (m, 4H), 2.54–2.51 (m, 4H), 2.33 (t, J ¼ 7.2 Hz, 2H),
Pol-PZ-NO2
Yield: 42%. ELEM. ANAL. (calculated): %C 65.21 (65.12), %H
6.89 (6.94), %N 14.12 (14.60). IR (KBr pellet, cmꢀ1): 1721
(C¼¼O); 1598, 1514, (Ar); 1235, 1140 (CAO); 857 (Ar). 1H-
NMR (CDCl3) d/ppm: 8.34–8.31 (m, 2H), 7.93–7.89 (m, 4H),
234
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
6.97–6.90 (m, 2H), 3.99–3.95 (m, 2H), 3.42–3.39 (m, 4H),
2.59–2.58 (m, 4H), 2.40–2.38 (m, 2H), 1.67–0.88 (m, 13H).
Mn ¼ 2300; Mw/Mn ¼ 1.7.
Optical absorption measurements were performed in a Cary
500 UV-Vis-NIR spectrophotometer. Linear (vertical/horizon-
tal) polarization of the measuring beam was achieved using
a Glan-Thompson prism. The films were held in a rotatory
mount (around a vertical axis, perpendicular to the incidence
plane), and the absorption spectra were registered at normal
and oblique incidence (0ꢁ and 45ꢁ, respectively).
Pol-PZ-DCN
Yield: 35%. ELEM. ANAL. (calculated): %C 70.23 (70.56), %H
6.41 (6.71), %N 16.26 (16.46). IR (KBr pellet, cmꢀ1): 2189
(CN); 1725 (C¼¼O); 1672 (C¼¼C); 1597, 1507, (Ar); 1230,
1139 (CAO), 823 (Ar).
Nonlinear Optical Measurements
Nonlinear optical characterization was carried out in a NLO
Spectrometer from Sopra. Molecular hyperpolarizabilities (lb
values) were measured in solution by using the electric field
induced second harmonic (EFISH) generation technique at
1.06 and 1.91 lm. The latter was obtained from a H2 Raman
shifter pumped by a Q-switched Nd:YAG laser emitting at
1.06 lm. The laser repetition rate was 10 Hz and the pulse
width 8 ns. The fundamental light was split in two beams.
The less intense one was directed to a N-(4-nitrophenyl)-(L)-
prolinol (NPP) powder sample whose SH signal was used as
a reference to correct laser fluctuations. The other beam was
focused onto a wedge shaped EFISH cell, containing the liq-
uid sample (NLO-phore solution). The voltage applied across
the cell electrodes (separated 2 mm) was 4.5 kV. The SH
light was measured with two different photomultipliers,
Hamamatsu R2949 for visible (532 nm) and R406 for near
infrared (954 nm), respectively, and appropriate interference
filters were used to remove the residual excitation light
beyond the film and the NPP reference.
Characterization Techniques
Elemental analyses were performed using a Perkin–Elmer
240C microanalyzer. IR spectra were obtained on a Nicolet
Avatar 360-FTIR. 1H-NMR spectra were measured at room
temperature with a Bruker AV-400 spectrometer (spectrum
of pol-PZ-DCN was not registered because of its poor solu-
bility). Molecular weights and polydispersities were mea-
sured by GPC using a Waters 600E HPLC (Waters Styragel
columns HR2 and HR4) with a UV Waters 991 photodiode
array detector. THF was used as the eluent, and standard
samples of polystyrene were used. Thermogravimetric analy-
sis (TGA) was performed using a TA Q5000IR instrument at
a heating rate of 10 ꢁC minꢀ1 under a nitrogen atmosphere.
Mesogenic behavior was evaluated by polarized-light optical
microscopy using an Olympus BH-2 polarizing microscope
fitted with a Linkam THMS600 hot stage. Thermal transitions
were determined by differential scanning calorimetry (DSC)
using a TA DSC Q-1000 instrument under a nitrogen atmos-
phere with samples sealed in aluminum pans. Glass transi-
tion temperatures (Tg) were determined at the midpoint of
the baseline jump, and the melting or mesophase-isotropic
transition temperatures were read at the maximum of the
corresponding peaks. Powdered samples of about 3 mg were
used.
Polymer films were Corona poled to break the natural cen-
trosymmetry of these systems. The films were held on a
thermoregulated fixed stage for ‘‘in situ’’ SHG measurements
at 1.91 lm during the poling process. The angle of incidence
of the fundamental beam on the film was about 40ꢁ. A posi-
tive electric discharge was applied across the film using a
standard Corona poling setup. A DC voltage of þ5 kV was
applied to a needle, perpendicular to the film. The distance
between the needle and the film surface was about 1 cm.
The poling process was performed as follows. The polymer
films were heated at a rate of 5 ꢁC/min up to the optimal
poling temperature with the electric field on and maintained
in these conditions for 40 min. After several trials to avoid
thermal decomposition and to control reproducibility of the
process, 120 ꢁC for Pol-PZ-CN and 95 ꢁC for Pol-O-CN (g
56 ꢁC SmA 163 ꢁC I) were chosen as poling temperatures.
Then, the films were cooled down to RT in the presence of
the poling field, and finally, it was turned off. A quite sharp
decay followed by a slower decrease of the SH signal is
observed after switching the field off. This behavior has been
associated with the neutralization of charges trapped at the
film surface during Corona poling and with different relaxa-
tion processes.15
Film Preparation and Optical Measurements
Thin films for optical measurements were produced by cast-
ing a solution of the polymer in dichloromethane onto clean
ITO (Indium Tin Oxide) coated glass substrates. Before film
casting and as the last step of the cleaning process, sub-
strates were ozone exposed in a UV photo reactor. Among all
the polymers synthesized for this study, only pol-PZ-CN
samples had suitable quality for optical measurements. In
the case of pol-PZ-NO2, brittle films with a poor optical
quality were obtained probably due to its low molecular
weight. On the other hand, the low solubility of pol-PZ-DCN
precluded film processing. Pol-PZ-CN and pol-O-CN films
were dried at 40 ꢁC overnight to eliminate the residual sol-
vent and then stored in the dark. Two groups of films with
different thickness, about 0.6 lm and 1.2 lm thick, respec-
tively, were prepared. Refractive indices measurements were
performed in a Metricon instrument in which a prism cou-
pler was used to obtain guided wave spectra at different
wavelengths: 633, 780 nm, and 1.31 lm. The use of light of
TE and TM polarization allowed us to determine nE (ordi-
nary index, in-plane), nM (extraordinary index, normal to the
film surface) and film thickness. An independent thickness
measurement was performed using a Veeco Dektak3ST con-
tact profilometer.
Once the films were poled as just described, Maker fringes
were measured to obtain second order nonlinear coefficients
(dij). The poled films were held on a stage which rotates
around a vertical axis, perpendicular to light’s plane of inci-
dence so as to detect the SH intensity generated by the film
as a function of the incidence angle for the two mentioned
NLO PROPERTIES OF SIDE CHAIN LIQUID CRYSTALLINE POLYMERS, ALICANTE ET AL.
235

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
FIGURE 1 Synthetic pathway
of monomers and polymers.
RESULTS AND DISCUSSION
polarizations of the fundamental light. In general, poled films
show
C
symmetry, and the nonlinear response is
1v
Synthesis and Characterization
described by two independent dij coefficients, assuming
Kleinmann symmetry conditions. So, two measurements
using S and P polarized fundamental light are enough to
obtain the experimental d31 and d33 values. In this work, the
d31 coefficient has been deduced from the fitting of the har-
monic signal obtained under excitation with S polarized light,
and the d33 value has been calculated from measurements
with P polarized light and using the d31 previously obtained.
The analysis of the experimental curves (averaged from sev-
eral runs at different points in the poled area) allowed us to
obtain the nonlinear coefficients using a X-cut quartz crystal
(d11 ¼ 0.35 pm/V at 1.91 lm) as a reference. The estimated
error in dij values is 615%. Because of the noticeable differ-
ence between indices nM and nE measured in the films, the
experimental Maker fringes were analyzed considering the
work of Hayden and Herman,16 whose model takes into
account film birefringence to calculate dij values. The refrac-
tive indices at 954 nm and 1.91 lm have been extrapolated
from the guided mode measurements at 633, 780 nm, and
1.31 lm.
The monomeric dyes were prepared by azo coupling of N-
(hydroxyhexyl)-N0-phenylpiperazine with the diazonium salts
of 4-cyanoaniline, 4-nitroaniline and 4-aminobenzaldehyde
and further esterification of the resulting azobenzene with
methacryloyl chloride (Fig. 1). In the case of the dicyanovinyl
derivative, the azomonomer (5), having a formyl group, was
first synthesized and finally condensed with malonitrile
(Knoevenagel reaction) to yield the monomer mon-PZ-DCN.
Attempts to introduce the dicyanovinyl group in previous
step lead to low yields.
It has been previously reported that N-methacryloyl-N0-phe-
nylpiperazine and N0-phenylazophenyl derivatives are not
prone to polymerize by a free radical polymerization.17 The
steric hindrance of the N,N-dialkylmethacrylamide group was
claimed as the reason for this failure of the polymerization.
In our approach, the reactive group was separated from the
piperazine ring by a flexible hexamethylenic spacer, which it
has been widely used in the synthesis of reactive monomers
for preparing side chain liquid crystalline polymers.
236
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
TABLE 1 Thermal and Mesomorphic Properties of Monomers
Compound
Phase Transitions (ꢁC)a
mon-PZ-CN
K 120 (23.7) SmA 164 (0.4)
N 179 (0.2) I polymerization
mon-PZ-NO2
K 128 (22.7) SmA 179 (3.6)
I polymerization
mon-PZ-DCN
K 136 (16.7) polymerization
a
Phase transitions were read at the onset temperature of peaks corre-
sponding to first heating run at 10 ꢁC/min. DH are indicated in brackets
(kJ/mol).
Nevertheless, a relatively high amount of thermal initiator is
required to polymerize these monomers with yields around
40%. However, similar monomers but without piperazine
groups (ether linkage instead of the piperazine) were suc-
cessfully polymerized in good yields and with high molecular
weights under similar conditions.5(a,b) The expected chemical
structure of the synthesized polymers was confirmed by ele-
mental analysis, infrared spectroscopy, and NMR. Polymer
having a dicyanovinyl group pol-PZ-DCN was insoluble in
common organic solvents once purified. Nevertheless, IR
spectrum of this compound exhibits a band at 1672 cmꢀ1
corresponding to the terminal double bond, which remains
when the polymer was heated up to temperatures around
225 ꢁC. This result ruled out crosslinking of the terminal
group as the cause of the insolubility of this material.
FIGURE 2 DSC trace corresponding to the first heating scan
(10 ꢁC/min) of pol-PZ-CN.
phic phases. Thus, pol-PZ-CN and pol-PZ-NO2 exhibit a
mesomorphic phase having textures, which are indicative of
a SmA phase. Both polymers are semicrystalline as prepared,
but in the second heating scan only a glass transition is
observed (when they are previously heated up to tempera-
tures below thermal decomposition). Nevertheless, both
polymers show evidences of decomposition above ꢂ200 ꢁC,
as it can be detected by DSC and optical microscopy. In fact,
an exotherm is observed around this temperature (see Fig. 2
as an example), and a darkening of the polymeric sample is
observed under the microscope. However, pol-PZ-DCN did
not exhibit softening or evidences of a fluid phase, and it
decomposes above ꢂ160 ꢁC according to DSC traces (a
broad exotherm is observed) and the optical microscopy
observations.
The thermal and mesomorphic properties of monomers are
collected in Table 1. As it can be seen, all the monomers ex-
hibit mesomorphism except in the case of mon-PZ-DCN,
which polymerizes just above the melting transition and pre-
vents mesomorphism to be adequately characterized. Mon-
PZ-CN and mon-PZ-NO2 also polymerize around the isotrop-
ization transition temperature and exhibit a SmA phase.
Mon-PZ-CN exhibits a short range of nematic phase as well.
Thermal properties of polymers are gathered in Table 2.
Polymers exhibit a lower thermal stability than analogous
polymers having an oxygen atom instead of the piperazine
group, and all polymers present weight loss at temperatures
above 200 ꢁC. This fact limits the observation of mesomor-
Optical Characterization
UV-vis spectra of piperazine containing monomers have been
measured in CH2Cl2 solutions, showing an intense band in
the blue region, assigned to p–p* transitions of trans azo iso-
mers. The wavelengths of maximum absorption, kmax, are
430, 460, and 490 nm for CN, NO2, and DCN acceptors,
respectively. kmax for mon-O-CN in the same solvent is
365 nm. As expected, the substitution with stronger donor
and acceptor terminal groups results in red shifted absorp-
tion bands.
TABLE 2 Thermal Stability and Transition Temperatures
of Polymers
TGA
Previously to the study of NLO response, the optical proper-
ties of pol-PZ-CN have been characterized and UV-vis spec-
tra of thin fresh films have been measured at different inci-
dence angles. As it can be seen in Figure 3, the spectrum at
normal incidence shows a broad band with a marked peak
at about 380 nm. If this spectrum is compared with that of a
polymer solution, shown in the figure as well, a clear band
broadening and blue shift of the absorption maximum are
observed in the film. Hypsochromic shifts in azopolymer
spectra are usually attributed to the aggregation of azoben-
zene chromophores. Similar absorption spectra, with a peak
in the higher energy side of the band, have been reported by
Decomposition
Temp. (ꢁC)a
Phase Transitions
(ꢁC)b
Polymer
pol-PZ-CN
245
235
240
g 77 K 114 SmA
decomp t > 200
pol-PZ-NO2
pol-PZ-DCN
a
g 60 K 118 SmA
decomp t > 175
g 83 K 150 decomp
t > 170
Onset of the decomposition process as detected on the thermogravi-
metric curve.
b
Data corresponding to the first heating scan.
NLO PROPERTIES OF SIDE CHAIN LIQUID CRYSTALLINE POLYMERS, ALICANTE ET AL.
237

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
in several side chain azopolymers that side mesogens tended
to form homeotropic domains when the sample was ther-
mally annealed without being previously photooriented.21 To
obtain a deeper insight about molecular arrangement in the
thermally annealed films, UV-vis measurements were carried
out after different time intervals at 120 ꢁC. The absorbance
spectra of a fresh film and the corresponding to the annealed
one for 1 min at 120 ꢁC are shown in Figure 4(a). The ab-
sorbance of the latter is clearly less intense, as expected if
azo chromophores are oriented out-of-plane, but differences
in the band shape are also observed. In particular, the band
associated with azo aggregates has a higher relative weight
with respect to the isolated chromophore’s one. This spectral
feature is stressed for rough thermal conditions. Aggregation
favored by thermally enhanced homeotropic alignment has
been reported.18,21(c,d),22
FIGURE 3 UV-vis spectra of pol-PZ-CN corresponding to CH2Cl2
solution and as prepared film. Measurements taken at normal
and 45ꢁ incidence are shown.
The influence of UV-blue preirradiation on the optical prop-
erties of pol-PZ-CN has been explored. The effect of light in
azo aggregate’s breakage has been widely studied.18,23 The
E-Z photoisomerization cycles of azo chromophores destroy
several authors and associated with antiparallel H azo aggre-
gates.18,19 On the other hand, higher absorption is observed
when UV-vis spectra are taken at 45ꢁ when compared with
those measured at normal incidence, once they are corrected
by the longer path length at oblique incidence. So, the ab-
sorbance increase at 45ꢁ indicates an anisotropic molecular
arrangement in the fresh films, where azo chromophores are
rather oriented out-of-plane (homeotropically). This fact was
confirmed by refractive index measurements, which revealed
a difference Dn ¼ nM ꢀ nE of about 0.1 at 633 nm, as col-
lected in Table 3. Indices corresponding to pol-O-CN films,
in which a slighter homeotropic tendency was observed, are
also included in Table 3. The initial homeotropic orientation
experienced a noticeable increase on heating to temperatures
within the mesophase range. Although the increase rate
depends on the temperature, similar final values (up to 0.35)
were achieved on heating pol-PZ-CN films. This fact can be
associated with the thermotropic cooperative motion of azo
groups, extensively described in the literature.20 Most of the
related published works concern the thermal amplification
of the photoinduced in-plane anisotropy that occurs when
photo-oriented azo LC polymer films are annealed at temper-
atures in the mesophase. However, it has been also described
TABLE 3 Refractive Indices Values (Averaged from Several
Films) of the Polymers, Obtained by Waveguide Method at
633 nm After Different Treatments
Sample
Treatment
nE
nM
pol-PZ-CNa,b
pol-PZ-CNa
pol-PZ-CNa
pol-PZ-CNb
pol-O-CN
pol-O-CN
a
Untreated
1.63
1.56
1.65
1.67
1.63
1.55
1.72
1.89
1.76
1.75
1.64
1.82
Annealed at 120 ꢁC
Preirrad. 440 nm depol
Preirrad. 440 nm depol
Fresh
FIGURE 4 (a) UV-vis spectra of a film of pol-PZ-CN before and
after annealing at 120 ꢁC for 1 min, measured at normal inci-
dence. (b) UV-vis spectra of a pol-PZ-CN film before and after
irradiation with light, measured at normal incidence.
Annealed at 90 ꢁC
Film thickness: 1.2 lm.
Film thickness: 0.6 lm.
b
238
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
TABLE 4 Wavelength of Maximum Absorption, kmax and Values of lb in CH2Cl2, Determined from
EFISH Measurements at 1.91 lm, and 1.06 lm When Possible. lb(0) Values have been Deduced
Using a Two Level Dispersion Model
kmax
lb
lb(0)
(10ꢀ48 esu)
lb
lb(0)
(10ꢀ48 esu)
Compound
(nm)
(10ꢀ48 esu)a,b
(10ꢀ48 esu)a,c
mon-PZ-CN
mon-PZ-NO2
mon-PZ-DCN
mon-O-CN
a
430
460
490
365
950
280
340
440
750
90
260
320
510
70
d
d
–
–
d
d
–
–
180
80
Experimental uncertainty in lb values is estimated to be 615%.
Measurements performed at 1.06 lm.
b
c
Measurements performed at 1.91 lm.
d
The red shifted absorption spectra precluded the use of the 1.06 nm excitation.
the aggregate phase and the azo groups become isolated.
Bearing in mind that the presence of nonpolar aggregates is
detrimental for polar ordering, we planned to modify the
aggregation state to study its influence on the poling process.
So, pol-PZ-CN films were irradiated with light from a Hg
lamp using a band-pass filter centered at 440 nm (45 mW/cm2,
BW: 100 nm). UV-vis absorption spectrum [Fig. 4(b)] at nor-
mal incidence shows the increase and red shifting of the
absorption band, and aggregate’s peak at 380 nm is not
observed. The change in the shape evidences that a noticea-
ble percentage of aggregates absorbing at 380 nm are broken
on irradiation.
by applying the thermal poling process described in the ‘‘Ex-
perimental’’ section (up to 120 ꢁC). Quite high SH intensity
was detected during the ‘‘in situ’’ Corona poling process and
a remarkable signal remained after switching the field off. To
evaluate the temporal stability of the induced polar order,
we measured the harmonic signal at different time intervals
after finishing the poling process. In Figure 5, we show the
evolution of the squared effective nonlinear coefficient, deff
,
normalized to the value measured about 2 h after switching
the field off, for a variety of measured films. It can be seen
in the figure that after a few days, a quite stable value of
harmonic signal is measured, which stays essentially con-
stant several weeks after the poling. This indicates that a no-
ticeable degree of polar order remains in the films kept at
RT. Concerning the thermal stability of the electric field
induced polar ordering, we show in Figure 6 a typical evolu-
tion with temperature of SH signal. These measurements
were carried out at least 2 weeks after finishing the poling
process. Thermal stability was checked by heating samples
near the mesophase range from RT at a rate of 5 ꢁC/min. A
decrease of the SH signal well above Tg was observed. The
NLO Properties
The lb values of monomeric precursors were obtained from
EFISH measurements in CH2Cl2 at 1.91 lm. Molecular hyper-
polarizabilities of mon-PZ-CN and mon-O-CN (both showing
blue shifted absorption bands when compared with mon-
PZ-NO2 and -DCN monomers) have also been measured at
1.06 lm. Slight absorption effects were observed in mon-PZ-
CN EFISH fringes, so absorption coefficient a2x was mea-
sured in solution and taking into account in data process-
ing.24 Experimental lb and lb(0) values deduced by using a
simple two level dispersion model25 are shown in Table 4.
Very close lb(0) values have been obtained from measure-
ments performed at different wavelengths as predicted by
two level model for rod shape dipolar chromophores. As
expected, piperazine group gives rise to an enhancement of
the molecular NLO response, which also increases with the
strength of the acceptor moiety (DCN > NO2 > CN). The
well-known standard Disperse Red 1 was measured in the
same conditions obtaining lb ¼ 680 ꢃ 10ꢀ48 and lb(0) ¼
470 ꢃ 10ꢀ48 esu. These results agree with those previously
reported for a series of push–pull azobenzene dyes bearing a
dialkylamino donor group and the mentioned electron with-
drawing moieties among others.26
The second order NLO properties of pol-PZ-CN films have
been evaluated by second harmonic (SH) generation meas-
urements. This polymer does not show an isotropic state
under 200 ꢁC, decomposition temperature (Table 2), so it
was not possible to study thermally isotropized pol-PZ-CN
films. Freshly made films, of about 1.2 lm thick, were poled
FIGURE 5 Temporal stability of SHG signal for several poled
pol-PZ-CN films. The squared effective NLO coefficients are
normalized to the value measured about 2 h after switching
the field off.
NLO PROPERTIES OF SIDE CHAIN LIQUID CRYSTALLINE POLYMERS, ALICANTE ET AL.
239

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
any spacer. The authors attributed the hardly effective poling
to the hindered mobility of the nonlinear chromophores due
to the rigidity of the side-chain structures.
As we have mentioned previously, the initial homeotropic ori-
entation in pol-PZ-CN fresh films can be thermally enhanced
by annealing in the mesophase. Some authors have stated
that axial orientation of the NLO moieties can improve elec-
tric field-induced polar ordering,10,28 when compared with
initially isotropic samples. However, we have clearly observed
that in our system the homeotropic ordering is accompanied
by an increase of aggregation. Therefore, the effect of the
enhanced out-of-plane ordering of thermally annealed poly-
mer samples on the NLO response was studied. The samples
were annealed at 120 ꢁC for 30 min (out-of-plane birefrin-
gence Dn (0.30)). Afterward, a poling process in the same
conditions as the freshly made films was carried out, obtain-
ing SHG signals clearly lower than those observed in the for-
mer. These results are consistent with the detrimental effect
of the aggregation, which partially inhibits the noncentrosym-
metric alignment under our experimental conditions.
FIGURE 6 Second harmonic signal versus temperature for
poled pol-PZ-CN film.
films were maintained at 110 ꢁC for a short time, just until
the disappearance of the harmonic signal. Under these condi-
tions, new poling processes gave rise to NLO responses simi-
lar to those obtained for the first time.
The influence of the different strength of chromophore’s
dipole moment in NLO properties was more extensively
studied, by comparing the results obtained for pol-PZ-CN
with pol-O-CN films oriented in the same conditions. The
lower dij values obtained (Table 5) for pol-O-CN are a result
of the lower lb value of the corresponding chromophore.
The effect of the enhanced out-of-plane ordering of thermally
annealed films in NLO properties was also studied on pol-O-
CN resulting quite different to the behavior in pol-PZ-CN
films. Although a noticeable axial order is also achieved by
annealing the films at 95 ꢁC (Dn ꢄ 0.20), the obtained SHG
signals are similar for both of the initial states (untreated
and annealed). These results could be associated with the
lower mon-O-CN dipole moment values and consequently
weaker aggregation tendency in pol-O-CN when compared
with pol-PZ-CN. In both cases, homeotropic arrangement
leading to an antiparallel alignment does not improve subse-
quent electric field induced polar alignment.
The stable nonlinear coefficients of pol-PZ-CN, gathered in
Table 5, were obtained from the fitting of the Maker fringes
measured beyond 2 weeks after finishing the poling process.
As correspond to C symmetry, the obtained Maker fringes
1v
were identical when the sample was rotated around an axis
perpendicular to the film. In the case of films of about
1.2 lm thick, the nonlinear coefficients were d31 ¼ 1 pm/V
and d33 ¼ 8 pm/V. These values are consistently higher than
those obtained for a series of azo polymers with a weaker
NLO chromophore (as mon-O-CN but with a spacer of two
carbon atoms instead of six) previously studied in our
group.5(b) A value of d33 ¼ 26 pm/V, larger than the
obtained for Pol-PZ-CN, was reported for a side chain poly-
methacrylate functionalized with a similar azo chromophore,
4-dialkylamino-40cyanoazobenzene.27 Nevertheless, the mea-
surements were performed using 1.06 lm fundamental
wavelength, so the reported d33 value is likely resonantly
enhanced due to the proximity of the harmonic wavelength
(532 nm) to the absorption of the side chain chromophore.
On the other hand, clearly lower d33 values were reported12
for polymethacrylates functionalized (ca., 30%) with similar
or stronger NLO moieties, linked to the main chain, without
A comment should be done at this point about macroscopic
response comparison between recently poled pol-PZ-CN (d33
¼ 11 pm/V) and pol-O-CN (d33 ¼ 6.5 pm/V). The ratio
between their respective coefficients is below the predicted
TABLE 5 Nonlinear Coefficients Measured for Polymer Films Poled After Different Treatments. The Coefficients have been
Determined Beyond 2 Weeks After Finishing the Poling Process. The Refractive Index Values at 953 (n2x) and 1907 nm (nx)
Used in the Fitting Procedure were Extrapolated from Mode Coupling Measurements Performed at 633 and 1308 nm
Refractive Indexes
dij (pm/V)
d33
8.0
Polymer Films
Treatment
nEx
nMx
n2Ex
n2Mx
d31
pol-PZ-CN (1.2 lm)
pol-PZ-CN (1.2 lm)
pol-PZ-CN (0.6 lm)
pol-PZ-CN (0.6 lm)
pol-O-CN (1.2 lm)
Untreated
1.55
1.53
1.55
1.53
1.56
1.70
1.73
1.72
1.78
1.70
1.56
1.55
1.56
1.54
1.57
1.73
1.76
1.74
1.80
1.73
1
Preirrad. 440 nm depol
Untreated
1
12
10
14
1.3
1.1
0.5
Preirrad. 440 nm depol
Untreated
3.5
240
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
one using the expression d33 ¼ N bz*zz hcos3 hi.7 N is the
number density, bz*zz is the microscopic hyperpolarizability
incorporating local field factors, and hcos3 hi is the axial
order parameter, being h the angle between chromophore
axis and poling direction. Different models have been used
to analytically evaluate the order parameter, from the sim-
plest isotropic poled polymer7 one to effective field
approaches,6 which include the effect of a chromophore on
another. In the last, dipole moment and distance between
chromophores are two key parameters, leading to a reduc-
tion of the maximum achievable polar order. In our case, the
intensity of dipole–dipole interactions in pol-O-CN and pol-
PZ-CN is probably different and side chain chromophores
tend to adopt antiparallel conformations, with closeness
depending on their dipole moment. These internal forces
work in opposition to the electric field external one, hinder-
ing at different extents polar orientation, and contributing to
the observed difference between the two polymers. The
assumption of stronger interactions in pol-PZ-CN is sup-
ported by the relevant presence of H aggregates, even in
fresh films, (Fig. 3) which becomes more important after
thermal treatments. It has been reported that these nonpolar
aggregates, caused by the strong dipole-dipole interaction,
cannot be oriented by the applied electric field.6
0.3 for the irradiated ones have been measured at 633 nm, in
the poled zone. However, we have found that this birefrin-
gence in pol-PZ-CN is caused by both the polar order and the
homeotropic ordering tendency shown by nonpolar aggre-
gates. In fact, clearly bigger Dn values have been measured
out of the poled zone, which has been subjected to the same
thermal conditions. On the other hand, SHG signal time decay
does not go along with a significant Dn decrease and, in addi-
tion, just after thermal removal of the SHG signal (Fig. 6) a
consistent decrease of Dn has not been detected.
The lower Dn values measured in the poled zone, and the
optical spectra, suggest that the electric field inhibits (at
least partially) the antiparallel axial ordering tendency of azo
chromophores30 leading to a certain degree of polar order.
Finally, it is worth to say that, although poling of pol-PZ-CN
presents some of the handicaps reported for smectic A poly-
mer, quite high and stable nonlinear response was achieved.
These results would be most likely overcome by using
copolymers with reduced piperazine moiety content.
CONCLUSIONS
The molecular nonlinearities of a series of chromophores
bearing piperazine as donor group and several acceptor
groups have been measured. lb values have shown the
expected tendency considering the strength of the withdraw-
ing groups (DCN > NO2 > CN).
Considering the later, we planned to reduce aggregation in
pol-PZ-CN, to achieve a higher polar order. Firstly, the dis-
ruption of aggregates was carried out by the irradiation of
1.2 lm thick films with UV-blue light (Hg lamp). Samples
were subsequently poled following the same described pro-
cess for fresh and annealed films. A faster increase and an
enhanced NLO response were observed ‘‘in situ’’ during the
poling process, supporting our assumption about the hin-
dered polar ordering of aggregates. As it can be seen in Table
5, the dij values measured confirm the difference. As the
response after irradiation was higher, we tried to optimize
our results using thinner films. Freshly made and irradiated
samples of about 0.6 lm thick were poled. High enough NLO
efficiency was achieved by reducing thickness and by irradi-
ating. The obtained dij coefficients for recently poled preirra-
diated thin films were d33 ¼ 17 and d31 ¼ 1.3 pm/V, and in
the case of as prepared ones, the results were d33 ¼ 13 pm/
V and d31 ¼ 1.7 pm/V. In Table 5, we report the stable dij val-
ues and the refractive indices extrapolated from measure-
ments in the irradiated and poled zone. These results show
that larger d33 values and d33/d31 ratios are obtained on irra-
diation. This ratio is predicted to be 3 by the thermodynamic
model for isotropic materials poled under low electric field
conditions.7 Ratio values higher than 3, as ours, have been
mainly associated with higher polar order due to intense
poling electric fields29 or to chromophore interactions deal-
ing to favor high axial ordered states, as several authors
reported in side chain liquid crystalline polymers.10,28 As for
the field strength, low content DR1 doped PMMA were poled
in the same conditions, giving rise to d33/d31 ꢅ 3. On the
other hand, the existence of a noticeable axial order has been
demonstrated by the aforementioned out-of-plane birefrin-
gence values. Dn values as high as 0.2 for untreated films and
Smectic liquid crystal polymer having a terminal cyano
group in the side chain chromophore (pol-PZ-CN) formed
thin films with good optical quality. However, formation of
aggregates proceeds from film production by casting, as it
can be detected in the absorption spectra. These azobenzene
aggregates are enhanced by heating up to temperatures
within the mesophase range of the polymer. On the other
hand, irradiation with UV-vis light has been associated with
their breakage. Corona poling process induces stable polar
order of chromophores and consequently thin films show a
settled NLO response. The axial order achieved is due to
both polar and nonpolar arrangement of the NLO moieties,
as a result of the competition between external electric field
effect and interchromophore interactions. The obtained dij
values are lower than the potentially achievable ones, taking
into account the lb values of the chromophores. Aggregation
favored by dipole–dipole interactions represents a major fac-
tor in determining the macroscopic response of the system.
This statement is supported by the noticeable increase of the
NLO response observed when, to disrupt aggregates, thin
pol-PZ-CN films are irradiated before the electric field poling
process. Moreover, it has been reinforced by the lower SHG
obtained when the out-of-plane arrangement is forced by the
previous annealing at the mesophase.
The stability of the light improved nonlinear response has
been checked several months after the poling, remaining at
least the 50% of the initial harmonic signal.
Financial support from MICINN-FEDER (MAT2008-06522-
C02-01 and -02) and Gobierno de Arago´n-Fondo Social
NLO PROPERTIES OF SIDE CHAIN LIQUID CRYSTALLINE POLYMERS, ALICANTE ET AL.
241

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
11 Rau, I.; Kajzar, F. Thin Solid Films 2008, 516, 8880–8886.
Europeo (E39) is gratefully acknowledged. R. Alicante thanks
the Spanish Government for a grant (FPI program, No.
BES2006-12104).
12 Tirelli, N.; Altomare, A.; Solaro, R.; Ciardelli, F.; Follonier, S.; Bos-
shard Ch.; Gu¨nter, P. Polymer 2000, 41, 415–421.
13 Altomare, A.; Andruzzi, L.; Ciardelli, R.; Solaro, F.; Tirelli, N. Macro-
mol Chem Phys 1999, 200, 601–608.
REFERENCES AND NOTES
14 Woo, H. Y.; Shim, H. K.; Lee, K. S. Macromol Chem Phys 1998, 199,
1 (a) Burland, D. M.; Miller, R. D.; Walsh, C. A. Chem Rev 1994, 94,
31–75; (b) Nalwa, H. S.; Miyata, S. In Nonlinear Optics of Organic Mol-
ecules and Polymers; CRC Press: Boca Rato´n, 1997; Chapter 7,
pp 128–309; (c) Samyn C.; Verbiest, T.; Persoons, A.; Macrom Rapid
Commun 2000, 21, 1–15; (d) Cho, M. J.; Choi, D. H.; Sullivan, P. A.;
Akelaitis, A. J. P.; Dalton, L. R. Prog Polym Sci 2008, 33, 1013–1058.
1427–1433.
15 (a) Hampsch, H.; Torkelson, J.; Bethke, S.; Grubb, J. J Appl Phys
1990, 67, 1037–1040; (b) Wang, H.; Jarnegin, R. C.; Samulski, E. T.
Macromolecules 1994, 27, 4705–4713; (c) Weyrauch, T.; Haase, W. In
Relaxation Phenomena. Liquid Crystals, Magnetic Systems, Polymers,
High-Tc Superconductors, Metallic Glasses; Haase, W.; Wro´bel, S., Eds.;
Springer: Berlin, 2003; Chapter 12, pp 667–691.
2 (a) Delaire, J. A.; Nakatani, K. Chem Rev 2000, 100, 1817–1845;
(b) Yesodha, S K; Sadashiva-Pillai, C K; Tsutsumi, N Prog Polym Sci
2004, 29, 45–74.
16 Herman, W. N.; Hayden, L. M. J Opt Soc Am B 1995, 12, 416–427.
17 Tirelli, N.; Suter, U. W.; Altomare, A.; Solaro, R.; Ciardelli, F.; Follon-
3 (a) Kumar, G. S.; Neckers, D. C. Chem Rev 1989, 89, 1915–1925;
(b) Xie, S.; Natansohn, A.; Rochon, P. Chem Mater 1993, 5, 403–411;
(c) Natansohn, A.; Rochon, P. Adv Mater 1999, 11, 1387–1391;
(d) Sekkat, Z. In Photoreactive Organic Thin Films; Sekkat, Z.; Knoll,
W., Eds.; Academic Press: San Diego, 2002; Chapter 8, pp 271–287;
(e) Natansohn, A.; Rochon, P. Chem Rev, 2002, 102, 4139–4176.
ier, S.; Bosshard, Ch.; Gu¨nter, P. Macromolecules 1998, 31, 2152–2159.
18 (a) Menzel, H.; Weichart, B.; Schmidt, A.; Paul, S.; Knoll, W.;
Stumpe, J.; Fischer, T. Langmuir 1994, 10, 1926–1933; (b) Stumpe, J.;
Fischer, Th.; Menzel, H. Macromolecules 1996, 29, 2831–2842;
(c) Geue, Th.; Ziegler, A.; Stumpe, J. Macromolecules 1997, 30, 5729.
4 (a) Sekkat, Z.; Dumont, M. Appl Phys B, 1992, 54, 486–489; (b) Blan-
chard, P. M.; Mitchell, G. R. Appl Phys Lett 1993, 63, 2038–2040; (c) Sek-
kat, Z.; Kang, C. S.; Aust, E. F.; Wegner, G.; Knoll, W. Chem Mater 1995, 7,
142–147; (d) Sekkat, Z.; Preˆtre, P.; Knoesen, A.; Volksen, W.; Lee, V. Y.;
Miller, R. D.; Wood, J.; Knoll, W. J Opt Soc Am B 1998, 15, 401–413.
19 (a) Lagugne´-Labarthet, F.; Freibert, S.; Pellerin, C.; Pe´zolet, M.;
Natansohn, A.; Rochon, P. Macromolecules 2000, 33, 6815–6823; (b)
Freibert, S.; Lagugne´-Labarthet, F.; Rochon, P.; Natansohn, A. Can J
Chem 2004, 82, 1–10.
20 (a) Fischer, Th.; La¨sker, L.; Czapla, S.; Ru¨ bner, J.; Stumpe, J. Mol
Cryst Liq Cryst 1997, 298, 213–220; (b) Meier, J. G.; Ruhmann, R.;
Stumpe, J. Macromolecules 2000, 33, 843–850; (c) Kidowaki, M.;
Fujiwara, T.; Morino, S.; Ichimura, K.; Stumpe, J. Appl Phys Lett
2000, 76, 1377–1379.
5 (a) Sa´nchez, C.; Cases, R.; Alcala´, R.; Lo´pez, A.; Quintanilla, M.; Oriol,
L.; Millaruelo, M. J. Appl Phys 2001, 89, 5299–5306; (b) Rodrı´guez, F.
J.; Sa´nchez, C.; Villacampa, B.; Alcala´, R.; Cases, R.; Millaruelo, M.; Oriol
L. Polymer 2004, 45, 2341–2348.
6 (a) Dalton, L. R.; Harper A. W.; Robinson B. H. Proc Natl Acad Sci USA
1997, 94, 4842–4847; (b) Dalton, L. R.; Steier, W. H.; Robinson, B. H.;
Zang, C.; Ren, A.; Garner, S.; Chen, A.; Londergan, T.; Irwin, L.; Carlson, B.;
Fifield, L.; Phelan, G.; Kincaid, C.; Amend, J.; Jen, A. J. Mater Chem 1999,
9, 1905–1920; (c) Wu¨thner, F.; Wortmann, R.; Meerholz, K. Chem Phys
Chem 2002, 3, 17–31; (d) Pereverzev, Y. V.; Prezhdo, O. V.; Dalton, L. R.
Chem Phys Chem 2004, 5, 1821–1830; (e) Yager, K. G.; Barrett, C. J. In
Polymeric Nanostructures and Their Applications; Nalwa, H. S., Ed.;
American Scientific Publishers: New York, 2006; Chapter 8, pp 1–38.
21 (a) Han M.; Morino S.; Ichimura K. Macromolecules 2000, 33,
6360–6371; (b) Cojocariu, C.; Rochon, P. Pure Appl Chem 2004, 76,
1479–1497; (c) Cojocariu, C.; Rochon, P. Macromolecules 2005, 38,
9526–9538; (d) Bobrovsky, A.; Shibaev, V.; Hamplova, V.; Kaspar, M.;
Glogarova, M. Monatsh Chem 2009, 140, 789–799.
22 Geue, T. M.; Pietsch, U.; Haferkorn, J. J.; Stumpe, J.; Date, R. W.;
Fawcett, A. H. Mol Cryst Liq Cryst 1999, 329, 113–119.
23 Rodrı´guez, F. J.; Sa´nchez, C.; Villacampa, B.; Alcala´, R.; Cases, R.; Col-
lados, M. V.; Hvilsted, S.; Strange, M. Polymer 2004, 45, 6003–6012.
7 Williams, D. J. In Nonlinear Optical Properties of Organic Molecules
and Crystals; Chemla, D. A.; Zyss, J., Eds.; Academic Press: Orlando, FL,
1987; Vol. 1, Chapter II-7, pp 405–434.
24 Oudar J. L. J Chem Phys 1977, 67, 446–457.
25 Oudar, J. L.; Chemla, D. S. J Chem Phys 1977, 66, 2664–2668.
8 (a) Beltrani, T.; Bo¨sch, M.; Centore, R.; Concilio, S.; Gu¨nter, P.; Sirigu,
A. J Polym Sci Pol Chem 2001, 39, 1162–1168; (b) Carella, A.; Casal-
boni, M.; Centore, R.; Fusco, S.; Noce, C.; Quatela, A.; Peluso, A.; Sirigu,
A. Opt Mater 2007, 30, 473–477; (c) Lanzi, M.; Paganin, L.; Cesari, G.
Macromol Chem Phys 2008, 209, 375–384.
26 Tirelli, N.; Altomare, A.; Solaro, R.; Ciardelli, F.; Meier, U.; Bosshard
Ch.; Gu¨nter, P. J Prakt Chem 1998, 340, 122–128.
27 S’Heeren, G.; Persoons, A.; Rondou, P.; Wiersman, J.; Van Beylen,
M.; Samyn, C. Makromol Chem 1993, 194, 1733–1744.
28 Gonin, D.; Guichart, B.; Noe¨l, C. Macromol Symp 1995, 96,
9 (a) Meredith, G. R.; Vandusen, J. G.; Williams, D. J. In Nonlinear Optical
Properties of Organic and Polymeric Materials; Williams, D. J., Ed.; Wash-
ington: ACS Symposium Series, 1983; Chapter 5, pp 109–133; (b) Singer,
K. D.; Kuzyk, N. G.; Sohn, J. E. J Opt Soc Am B 1987, 4, 968–976; (c) Van
der Vorst, C. P. J. M.; Picken, S. J. J Opt Soc Am B 1990, 7, 320–325.
185–206.
29 (a) Eich, M.; Sen, A.; Looser, H.; Bjorklund, G. C.; Swalen, J. D.;
Twieg, R.; Yoon, D. Y. J. Appl Phys 1989, 66, 2559–2567; (b) Rodri-
guez, V.; Lagugne´-Labarthet, F.; Sourisseau C. Appl Spectrosc 2003, 59,
322–328.
10 (a) Amano, M.; Kaino, T.; Yamamoto, F.; Takeuchi, Y. Mol Cryst Liq
Cryst 1990, 182A, 81–90; (b) Dubois, C.; Le Barry, P.; Mauzac, M.;
No¨el, C. Acta Polym 1997, 48, 47–87; (c) Kajzar, F.; Noe¨l, C. Adv Mater
Opt Electron 1998, 8, 247–262.
30 (a) Wortmann, R.; Ro¨sch, U.; Redi-Abshiro, M.; Wu¨rthner, F. Angew
Chem Int Ed 2003, 42, 2080–2083; (b) Vembris, A.; Rutkis, M.; Lai-
zane, E. Mol Crys Liq Crys 2008, 485, 125–132.
242
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA