Effects of nitro substituents on the properties of a ferroelectric liquid crystalline side chain polysiloxane

Article abstract of DOI:10.1039/a704918h

The syntheses of chiral liquid crystalline side chain polysiloxanes with lateral nitro substituents in the mesogenic core are described. The influence of the substituents and substituent positions on phase behaviour and electro-optical properties are investigated and compared. The lateral nitro groups strongly affect the phase behaviour of the side chain precursors as well as the liquid crystalline polymers. Properties in smectic A and C* phases are discussed with respect to substituent position. One side chain precursor exhibits very large spontaneous polarization of ～700 nC cm^-2.

Full text of DOI:10.1039/a704918h

J O U R N A L O F
Effects of nitro substituents on the properties of a ferroelectric liquid
crystalline side chain polysiloxane
Magnus Svensson,a Bertil Helgee,*a Kent Skarpb and Gunnar Anderssonb
aDepartment of Polymer T echnology, Chalmers University of T echnology, S-412 96 Goteborg,
Sweden
¨
C H E M I S T R Y
bDepartment of Physics, Chalmers University of T echnology, S-412 96 Goteborg, Sweden
¨
The syntheses of chiral liquid crystalline side chain polysiloxanes with lateral nitro substituents in the mesogenic core are
described. The influence of the substituents and substituent positions on phase behaviour and electro-optical properties are
investigated and compared. The lateral nitro groups strongly affect the phase behaviour of the side chain precursors as well as the
liquid crystalline polymers. Properties in smectic A and C* phases are discussed with respect to substituent position. One side
chain precursor exhibits very large spontaneous polarization of ~700 nC cm−2.
The possibility of designing and using chiral liquid crystalline
side chain polymers for technical applications in the future
depends on the understanding of structure–property relation-
ships. A vast number of new molecules, architectures and
concepts of chiral liquid crystalline side chain polymers have
been synthesized, examined and reported1,2 since the first
ferroelectric liquid crystalline polymer was prepared by Shibaev
et al.3 in 1984. Research on the effects of structural changes in
the different parts of the mesogens like the chiral centre,4,5
mesogenic core6–8 and alkyl chain9–11 has been carried out.
The influence of the polymer main-chain on properties has
also been investigated.12–17 Nevertheless, a large part of today’s
information on structure–property relationships in chiral liquid
crystalline polymers relies on extrapolations from research on
low molar mass liquid crystals performed during the 1970s.
Concerning lateral substituents Osman18 has shown that
steric effects and thereby the van der Waals volume of the
substituent are usually more important with respect to phase
behaviour than dipolar interactions. The more detailed influ-
ences of lateral substituents depend on the structure of the
rigid core and the position of the substituent. Generalizations
are difficult, since the effects differ between polar and non-
polar mesogens and probably also change from low molar
mass to polymer liquid crystals. Using electron attracting
substituents Masuda et al.19 showed that substituents in the
centre of the mesogenic core not only increased the intermol-
ecular distance, thus destabilizing the liquid crystal behaviour,
but also hindered the formation of smectic phases. With
multiple substituents where the second and/or third does not
increase the steric effect but adds possibilities for polar inter-
actions the smectic phases can be regained.20 A study by Hird
et al.21 shows the same tendencies.
acid moiety. The nitro group is introduced at the 2- and 3-
position in the phenyl ester and at the 3∞-position in the
biphenyl part. The side-chain precursors are attached to a
poly(dimethyl-co-methylhydrogen)siloxane backbone. It was
found that the introduction of substituents changes the phase
behaviour dramatically depending on position. One of the
side-chain precursors only exhibits a 10 °C temperature interval
of the N*-phase but in the corresponding polymer smectic
phases over a temperature range of 70 °C are again obtained.
Another side-chain precursor shows very large spontaneous
polarization (nearly 700 nC cm−2).25,26 The same system shows
interesting NLO-properties reported earlier,27 where a control-
lable SHG-intensity in the smectic A phase was described for
the first time. For the reference system (without nitro substitu-
ent), both the low molar mass mesogen and the polymeric
liquid crystal have been described in detail previously.24,28
Experimental
Techniques
300 MHz 1H NMR spectra were obtained using a Varian
VXR300 spectrometer. All spectra were run in CDCl or [2H ]
3
6
DMSO solutions. Infrared spectra were recorded on a Perkin-
Elmer 2000 FT-IR spectrophotometer using KBr pellets. The
phase behaviour of the different materials was identified by
combining optical microscopy, differential scanning calor-
imetry and electro-optical measurements. Optical microscopy
was performed using a Mettler FP82HT hot stage, Mettler
FP80HT central processor and an Olympus BH-2 polarizing
microscope. DSC measurements were recorded on a Perkin-
Elmer DSC 7 differential scanning calorimeter.
The nitro group with its large dipole moment and electron
attracting power offers a way to drastically change the electron
distribution in the aromatic core. This has been used to
enhance NLO-effects such as second harmonic generation
(SHG) in liquid crystals and in liquid crystalline polymers.22,23
In the present paper we have studied the effect of introducing
a nitro substituent at various aromatic positions in a 4∞-
alkoxybiphenyl-4-carboxylic acid phenyl ester mesogen. The
nitro group was chosen to increase the spontaneous polariz-
ation in the chiral smectic C phase. We were particularly
interested in how a strong dipole at different locations in the
polarizable aromatic core would change phase behaviour and
electro-optical properties of a ferroelectric liquid crystalline
polymer24 having a very broad smectic C* phase and good
alignment properties. The chiral group is a substituted lactic
Materials
Materials and reagents were of commercial grade quality and
used without further purification unless otherwise noted. Dry
toluene and methylene chloride were obtained by passing the
solvents through a bed of aluminium oxide (ICN Alumina
N-Super I). (+)-2-(4-Hydroxyphenoxy)propionic acid was gen-
erously provided by BASF. Poly(dimethylsiloxane-co-methyl-
hydrosiloxane) with a copolymer ratio 2.7/1 and a degree of
polymerization of ca. 30 (according to manufacturer) and
the hydrosilylation catalyst dicyclopentadienylplatinum()
dichloride were obtained from Wacker Chemie. Size-exclusion
chromatography (SEC) analysis in chloroform of the copoly-
mer performed at 30 °C on a Waters WISP712 instrument
J. Mater. Chem., 1998, 8(2), 353–362
353

O
i
ii
iii
HO
CH3O
CH3O
1
2
O
O
O
iv
v
CH3O
CH3O
HO
OH
OH
OH
3
4
5b
vi
NO2
NO2
O
v
O
O
O
vi
HO
O
OH
OH
OH
5a
6a
6b
NO2
Scheme 1 Reagents: i, MeI, KOH, DMF; ii, AcCl, AlCl , CH Cl ; iii, Br , KOH, 1,4-dioxane; iv, HNO , AcOH; v, HBr, AcOH; vi, undecenyl
bromide, KI, EtOH
3
2
2
2
3
equipped with three commercial Styragel columns and Waters
:
boxylic acid 3 following closely the procedure reported by
Percec et al.29
410 refractive index detector gave M =2.6·103 g mol−1 and
n
:
M =5.4·103 g mol−1 with reference to polystyrene standards.
Synthesis
w
4∞-Methoxy-3∞-nitrobiphenyl-4-carboxylic
acid
4.
4∞-
Methoxybiphenyl-4-carboxylic acid 3 (0.01 mol, 2.28 g) in
acetic acid (50 ml) was refluxed with concentrated nitric acid
(6 ml) for 15 min. The reaction mixture was poured into water
and the precipitate was filtered off. The product recrystallized
from ethanol to yield 2.2 g (80%). d ([2H ] DMSO) 3.98 (s,
The synthesis of polymers 14a–d was carried out according to
the reactions in Schemes 1–4. All structures were verified by
1H NMR spectroscopy and NMR data were in accordance
with the structures in all cases.
H
6
3 H), 7.48 (d, 1 H), 7.85 (d, 2 H), 8.02 (d, 2 H), 8.07 (dd, 1 H),
8.26 (d, 1 H).
4∞-Hydroxybiphenyl-4-carboxylic acid 5a. The acid was syn-
thesized from 4-hydroxybiphenyl via 4-methoxybiphenyl 1, 4-
acetyl-4∞-methoxybiphenyl 2 and 4∞-methoxybiphenyl-4-car-
4∞-Hydroxy-3∞-nitrobiphenyl-4-carboxylic acid 5b. The acid
5b was synthesized as for 5a starting from 4. d ([2H ] DMSO)
H
6
O
Cl
O
O
O
ii
OH
iii
OH
i
HO
OH
+
O
7
8
NO2
O
O
O
iv
v
O
HO
O
HO
O
OH
O
O
O
*
*
O
NO2
NO2
*
NO2
9
10
12b
O
O
v
HO
O
HO
O
OH
O
*
*
12a
ii
O
O
v
HO
O
HO
O
OH
O
*
*
NO2
NO2
11
12c
Scheme 2 Reagents: i, pyridine, CH Cl , N ; ii, HNO , AcOH, <35 °C; iii, (S)-ethyl lactate, PPh , diethyl azodicarboxylate, THF, N ; iv, aq.
KOH, EtOH; v, BuOH, HCl(g)
2
2
2
3
3
2
354 J. Mater. Chem., 1998, 8(2), 353–362

H3C
O
O
O
*
*
O
O
i
6a + 12a
(CH2)9
O
13a
H3C
O
O
O
O
O
i
6a + 12b
(CH2)9
O
NO2
13b
H3C
O
O
O
*
*
O
O
i
6a + 12c
(CH2)9
O
NO2
13c
H3C
O
O
O
O
O
i
6b + 12a
(CH2)9
O
NO2
13d
Scheme 3 Reagents: i, DCC, DMAP, CH Cl
2
2
7.25 (d, 1 H), 7.81 (d, 2 H), 7.95 (dd, 1 H), 8.00 (d, 2 H), 8.22
(d, 1 H), 11.3 (s, 1 H).
CH3
CH3
Si
O
O
+
13a–d
Si
CH3
H
4∞-(Undec-10-enyloxy)biphenyl-4-carboxylic acid 6a. The
hydroxy-acid 5a (0.090 mol, 19.3 g) was dissolved in hot etha-
nol (1500 ml) and water (75 ml) together with potassium
hydroxide 86% (11.5 g) and a few crystals of potassium iodide.
Undec-10-enyl bromide (0.18 mol) was added and the mixture
was refluxed for 24 h. To hydrolyse any ester formed a 10%
solution of potassium hydroxide in 70% ethanol was added
and refluxing was continued for an additional 3 h. The reaction
mixture was allowed to cool and was acidified with concen-
trated hydrochloric acid. The solid formed was recrystallized
from acetic acid and ethanol to yield 22.2 g of product (69%).
n
m
m/n = 2.7
CH3
Si
CH3
O
Si
O
CH3
n
m
H3C
O
O
O
d
([2H ] DMSO) 1.2–1.5 (m, 12 H), 1.72 (quintet, 2 H), 2.00
H
6
(q, 2 H), 4.00 (t, 2 H), 4.96 (m, 2 H), 5.79 (m, 1 H), 7.03 (d, 2
H), 7.67 (d, 2 H), 7.75 (d, 2 H), 7.98 (d, 2 H).
*
O
O
O
3∞-Nitro-4∞-(undec-10-enyloxy)biphenyl-4-carboxylic acid 6b.
The acid 6b was synthesized as for 6a starting from 5b. d
H
([2H ] DMSO) 1.2–1.5 (m, 12 H), 1.73 (quintet, 2 H), 2.00 (q,
6
14a–d
NO2
2 H), 4.21 (t, 2 H), 4.96 (m, 2 H), 5.78 (m, 1 H), 7.47 (d, 1 H),
Scheme 4
7.85 (d, 2 H), 7.98–8.06 (m, 3 H), 8.24 (d, 1 H).
4-Hydroxyphenyl benzoate 7. A mixture of hydroquinone
(0.15 mol, 16.5 g) and pyridine (0.15 mol) in dry methylene
chloride was stirred at room temp. Nitrogen was bubbled
through and benzoyl chloride (0.14 mol, 19.6 g) was added
slowly. After 2 h additional stirring the solvent was evaporated.
The solid was washed with water to remove unreacted hydro-
quinone and then treated with diethyl ether to dissolve the
product; the diester remained as it is less soluble in diethyl
ether. After evaporation of solvent the product was recrys-
tallized from ethanol with increasing amounts of water. Yield
18.0 g (60%). d ([2H ] DMSO–CDCl ) 6.87 (d, 2 H), 7.00 (d,
concentrated nitric acid (11 ml) was added dropwise. The
temperature of the reaction mixture was kept below 35 °C.
After 50 min the reaction mixture was poured into 700 ml of
water. The precipitate was collected and purified on a silica
gel column using light petroleum–ethyl acetate (251) as eluent.
Yield 5.28 g (51%). d (CDCl ) 7.24 (d, 1 H), 7.49 (dd, 1 H),
H
3
7.55 (t, 2 H), 7.67 (m, 1 H), 8.01 (d, 1 H), 8.19 (dd, 2 H), 10.5
(s, 1 H).
3-Nitro-4-[(1R)-1-ethoxycarbonylethoxy]phenyl benzoate 9
[2-(4-benzoyloxy-2-nitrophenoxy)propanoic acid ethyl ester]. 3-
Nitro-4-hydroxyphenyl benzoate 8 (0.016 mol, 4.15 g), S-ethyl
lactate (0.016 mol, 1.89 g) and triphenylphosphine (0.020 mol)
were placed in dry glass equipment with dry THF (100 ml) as
H
6
3
2 H), 7.52 (t, 2 H), 7.65 (t, 1 H), 8.16 (d, 2 H), 9.1 (s, 1 H).
3-Nitro-4-hydroxyphenyl benzoate 8. To hydroquinone
monobenzoate 7 (0.04 mol, 8.6 g) in 100 ml of acetic acid,
J. Mater. Chem., 1998, 8(2), 353–362
355

solvent. N gas was bubbled through the mixture at room
temp. and diethyl azodicarboxylate (0.020 mol) was added
(DMAP) in dry methylene chloride was treated with 4 mmol
(0.83 g) of dicyclohexylcarbodiimide (DCC) at 0 °C. The tem-
perature was allowed to rise to room temp. and stirring was
continued for 3 h. Urea was filtered off and the filtrate was
evaporated to dryness. Column chromatography on silica gel
with light petroleum–ethyl acetate (951) as eluent gave 1.0 g
of pure product (80%). [a] 22−28.7 (CHCl ). d (CDCl ) 0.89
(t, 3 H), 1.2–1.5 (m, 14 H), 1.6 (m, 2H), 1.69 (d, 3 H), 1.80
(quintet, 2 H), 2.02 (q, 2 H), 4.00 (t, 2 H), 4.15 (m, 2 H), 4.82
(q, 1 H), 4.95 (m, 2 H), 5.80 (m, 1 H), 6.98 (d, 2 H), 7.03 (d, 1
H), 7.38 (dd, 1 H), 7.58 (d, 2 H), 7.68 (d, 2 H), 7.78 (d, 1 H),
8.18 (d, 2 H).
2
during 30 min. After 3 h of additional stirring at room temp.
the solvent was evaporated from the reaction mixture. The
resulting solid was dissolved in ethyl acetate and on adding
four times the volume of light petroleum, triphenylphosphine
oxide was precipitated. The solution was evaporated to dryness
and the product recrystallized from light petroleum, 4.76 g
(83%). d (CDCl ) 1.26 (t, 3 H), 1.71 (d, 3 H), 4.23 (m, 2 H),
D
3
H
3
H
3
4.84 (q, 1 H), 7.05 (d, 1 H), 7.40 (dd, 1 H), 7.53 (t, 2 H), 7.67
(t, 1 H), 7.79 (d, 1 H), 8.18 (d, 2 H).
2-(2R)-(4-Hydroxy-2-nitrophenoxy)propanoic acid 10. To
0.008 mol (2.87 g) of 9 in ethanol (120 ml), 0.016 mol of
potassium hydroxide in a few millilitres of water was added.
The mixture was stirred overnight at room temp. Ethanol was
evaporated and the product was extracted from methylene
chloride with water. Water was evaporated and the solid was
dissolved in ethyl acetate. The insoluble part was filtered off
and the filtrate was evaporated to dryness and used without
further purification. d ([2H ] DMSO) 1.47 (d, 3 H), 4.89 (q,
4-[(1R)-1-Butoxycarbonylethoxy]-2-nitrophenyl 4∞-(undec-
10-enyloxy)biphenyl-4-carboxylate 13c. From 6a and 12c as for
13b. [a] 22+21.7 (CHCl ). d (CDCl ) 0.93 (t, 3 H), 1.24–1.43
(m, 12 H), 1.48 (m, 2 H), 1.64 (m, 2 H), 1.69 (d, 3 H), 1.82
(quintet, 2 H), 2.05 (q, 2 H), 4.02 (t, 2 H), 4.21 (t, 2 H), 4.82
(q, 1 H), 4.97 (m, 2 H), 5.82 (m, 1 H), 7.01 (d, 2 H), 7.23 (dd,
1 H), 7.30 (d, 1 H), 7.6 (d+d, 3 H), 7.70 (d, 2 H), 8.22 (d, 2 H).
D
3
H
3
H
6
1 H), 7.01 (dd, 1 H), 7.07 (d, 1 H), 7.19 (d, 1 H), 9.9 (s, 1 H).
4-[(1R)-Butoxycarbonylethoxy]phenyl-3∞-nitro-4∞-(undec-10-
enyloxy)biphenyl-4-carboxylate 13d. From 6b and 12a as for
13b. [a] 22+16.0 (CHCl ). d (CDCl ) 0.92 (t, 3 H), 1.2–1.55
2-(2R)-(4-Hydroxy-3-nitrophenoxy)propanoic acid 11. To a
D
3
H
3
solution
of
2-(2R)-(4-hydroxyphenoxy)propanoic
acid
(m, 16 H), 1.64 (d, 3 H), 1.88 (quintet, 2 H), 2.05 (q, 2 H), 4.17
(m, 4 H), 4.75 (q, 1 H), 4.97 (m, 2 H), 5.82 (m, 1 H), 6.94 (d, 2
H), 7.14 (d, 2 H), 7.18 (d, 1 H), 7.69 (d, 2 H), 7.80 (dd, 1 H),
8.13 (d, 1 H), 8.26 (d, 2 H).
(0.027 mol, 5.0 g) in acetic acid (100 ml) concentrated nitric
acid (1.5 ml) was added dropwise. The temperature of the
reaction mixture was kept below 35 °C. After 30 min the
reaction mixture was poured into water. The product was
extracted from the aqueous phase using diethyl ether. No
further purification was done. Yield 5.1 g (80%). d ([2H ]
Polymers 14a–d. Dry equipment was used. 0.26 g of poly(di-
methyl-co-methylhydrogen)siloxane (2.751) (corresponding to
1.0 mmol Si-H) and 1.1 mmol of 13a-d were dissolved in dry
toluene (2 ml). Catalyst solution [0.8 ml; dicyclopentadienyl-
platinum() chloride in dry toluene; 0.1 mg ml−1] was added
and the reaction flask was sealed with a septum and heated to
110 °C. After 2 d an additional 0.8 ml of catalyst solution was
added and heating was continued. The reaction was monitored
by IR spectroscopy and further addition of catalyst continued
until the SiMH signal at ca. 2155 cm−1 was constant relative
to the CNO signal at ca. 1740 cm−1. The reaction mixture
was added dropwise to methanol (400–800 ml) during vigorous
stirring. The product was collected by centrifugation.
Reprecipitation from chloroform in methanol was continued
until no free side-chain could be detected by thin layer chroma-
tography. Treatment of the product by dissolving it in ethyl
acetate and passing the solution through a 0.2 mm Teflon filter
often reduced the amount of remaining SiMH drastically. Yield
0.5–0.8 g. 1H NMR spectra of the polymers show broader
signals than those of the side chain precursors, but are other-
wise very similar except that the signals from methyl groups
attached to silicon at d 0–0.16 and a methylene group attached
to silicon at d 0.5 are present, while the olefinic protons at d
4.95–4.97 and d 5.80–5.82 are absent. A SiMH signal appeared
at d 4.7, equivalent to 10–15% unreacted hydrogens.
H
6
DMSO) 1.53 (d, 3 H), 4.89 (q, 1 H), 7.11 (d, 1 H), 7.24 (dd, 1
H), 7.39 (d, 1 H), 10.6 (s, 1 H).
Butyl 2-(2R)-(4-hydroxyphenoxy)propanoate 12a. A solution
of 2-(2R)-(4-hydroxyphenoxy)propanoic acid (0.01 mol, 1.9 g)
in butanol (150 ml) was treated with hydrogen chloride gas
until no further heat was evolved. The reaction mixture was
evaporated to dryness. The product was separated by column
chromatography on silica gel with light petroleum–ethyl acet-
ate (251) as eluent. Yield 1.6 g (64%). [a] 22+33.9 (CHCl ).
D
3
d
(CDCl ) 0.88 (t, 3 H), 1.30 (m, 2 H), 1.6 (d+m, 5 H), 4.14
H
3
(m, 2 H), 4.65 (q, 1 H), 6.72 (m, 4 H).
Esters 12b and c were synthesized as for 12a starting from
10 and 11 respectively.
Butyl 2-(2R)-(4-hydroxy-2-nitrophenoxy)propanoate 12b.
[a] 22−84 (CHCl ). d (CDCl ) 0.90 (t, 3 H), 1.31 (m, 2 H),
1.60 (m, 2 H), 1.66 (d, 3 H), 4.16 (m, 2 H), 4.74 (q, 1 H), 6.94
D
3
H
3
(d, 1 H), 6.97 (dd, 1 H), 7.32 (d, 1 H).
Butyl 2-(2R)-(4-hydroxy-3-nitrophenoxy)propanoate 12c.
[a] 22+77.1 (CHCl ). d (CDCl ) 0.92 (t, 3 H), 1.35 (m, 2 H),
D
3
H
3
1.64 (d+m, 5 H), 4.18 (m, 2 H), 4.73 (q, 1 H), 7.10 (d, 1 H),
7.27 (dd, 1 H), 7.48 (d, 1 H), 10.3 (s, 1 H).
14a [a] 22+13.8 (CHCl ).
The last reaction step to the mesogenic side-chain precursors
13a–d was similar for all and is described for 13b.
D
3
14b [a] 22−10.0 (CHCl ).
D
3
14c [a] 22+15.3 (CHCl ).
4-[(1R)-Butoxycarbonylethoxy]phenyl
enyloxy)biphenyl-4-carboxylate 13a. From 6a and 12a as for
4∞-(undec-10-
D
3
14d [a] 22+10.0 (CHCl ).
D
3
13b. [a] 22+18.0 (CHCl ). d (CDCl ) 0.90 (t, 3 H), 1.2–1.5
D
3
H
3
(m, 14 H), 1.6 (d+m, 5 H), 1.81 (quintet, 2 H), 2.04 (q, 2 H),
4.00 (t, 2 H), 4.17 (m, 2 H), 4.74 (q, 1 H), 4.97 (m, 2 H), 5.81
(m, 1 H), 6.92 (d, 2 H), 7.00 (d, 2 H), 7.13 (d, 2 H), 7.58 (d, 2
H), 7.68 (d, 2 H), 8.21 (d, 2 H).
Sample preparation and electric measurements
For the study of physical properties such as electro-optic effects
in smectic liquid crystals, the achievement of well aligned
samples is essential. Our standard shear cell,30 built for
measurements on low molar mass ferroelectric liquid crystals,
has proved very useful for obtaining aligned polymer samples
and was used for all polymeric liquid crystals and some low
4-[(1R)-Butoxycarbonylethoxy]-3-nitrophenyl 4∞-(undec-10-
enyloxy)biphenyl-4-carboxylate 13b. From 6a and 12b. Dry
glassware was used. A mixture of 2 mmol (0.73 g) of 12b,
2 mmol (0.57 g) of 6a and 15 mg of dimethylaminopyridine
356
J. Mater. Chem., 1998, 8(2), 353–362

Fig. 1 Experimental set-up and cell geometry
molar mass liquid crystal compounds in this study. The sample
is applied to the lower glass plate and melted into the isotropic
phase to cover the whole glass area. Then the upper glass plate
is put on and the shear cell is assembled. Orientational shear
was applied in the smectic A phase close to the isotropic phase.
A special electrode pattern ensures that the active area is
always 16.8 mm2, independent of shear position. Spacers con-
Fig. 2 Phase behaviour of side-chain precursors and polymers
third chiral group (compound 12b) was synthesized via a
Mitsunobu reaction of (S)-ethyl lactate with nitrohydroquinone
monobenzoate. The reaction was performed under standard
Mitsunobu conditions and proceeds with inversion of con-
figuration.31 In the final step of the side chain precursor
synthesis, the biphenylcarboxylic acids were esterified with the
chiral propanoic ester derivatives using dicyclohexylcarbodiim-
ide (Scheme 3). The side chain liquid crystalline polymers were
finally obtained by a hydrosilylation reaction (Scheme 4). The
extent of reaction was monitored by FT-IR. Attempts to
quantify the optical purity of the mesogens by 1H NMR using
a chiral shift reagent, (+)-praseodymium tris(3-heptafluorobu-
tyrylcamphorate), have been made. Addition of the chiral shift
reagent caused no observable separation of peaks, although
this has been reported by others.32 Materials up to three years
old show no reduction in the specific optical rotations due to
racemization.
sist of thermally evaporated 2 mm SiO , and the orientation of
2
ˆ
the smectic layers is such that the layer normal (k) is perpen-
˚
dicular to the shear direction, cf. Fig. 1. A 1000 A protective
SiO layer covers the ITO electrodes. The shear alignment cell
is mounted in a Mettler FP52 hot stage for temperature
2
control and put in a Zeiss Photomicroscope equipped with a
fast electro-optic recording system. The sample temperature is
independently measured by a Pt 100 resistor element placed
in the sample holder close to the sample. For some of the low
molar mass samples good alignment could also be achieved in
standard commercial surface-coated cells with 4×4 mm active
area and 4 mm thickness, obtained from EHC.
The measurement of the spontaneous polarization P was
done with either the bridge method or the triangular wave
s
method. The measured polarization is taken from a series of
hysteresis curves, where the reading of the amplitude of the
loop directly gives the spontaneous polarization by eqn. (1),
Liquid crystal properties
The system used as reference24 (13a, 14a) is easy to handle
during physical measurements and shows typical ferroelectric
behaviour. The introduction of a strong electron attracting
group at different aromatic positions in the mesogenic core
was thought to give indications of the importance of electric
dipoles in the mesogen. The influence on spontaneous polariz-
ation in the chiral smectic C phase was of special interest.
1 kDU
P =
C
(1)
s
1
2A
G
where A is the active sample area, k is a correction factor for
the SiO protective layer (k=1.1 for the shear cell glasses), G
is the gain of the instrumental amplifier, DU is the amplitude
2
of the loop and C is the reference capacitor in series with the
ferroelectric LC sample. In order to measure the smectic tilt
1
Phase behaviour. The nitro substituent affected the phase
behaviour markedly for the low molar mass materials as can
be seen in Table 1 and Fig. 2. The lateral substituent lowers
the clearing temperature in all cases, a well-known effect.
Furthermore the nitro group alters the type of liquid crystalline
phase that appears, depending on substituent position. Only
in mesogen 13b is the smectic C* phase retained and a
comparison of ferroelectric behaviour can be made with the
reference system, 13a. In mesogen 13c the smectic phases are
lost and only in a narrow temperature interval does a chiral
nematic phase remain. The nitro group ortho to the central
ester linkage probably induces greater order and thereby causes
crystallization. This can be explained by an increase in attract-
ive forces by dipole–dipole interactions between mesogens or
by a decrease in intramolecular mobility on introduction of
the nitro group in this position. The observed drastic lowering
of the clearing temperature would at first make the former
explanation less probable but dipole–dipole interactions are
known to be strongly temperature dependent33 and rotation
angle H in the C* phase, we applied a low frequency square
wave field and determined the two extinction orientations by
rotating the sample between crossed polarizers. To investigate
the electroclinic effect, measurements of induced tilt angle and
optical response time were carried out as a function of applied
field and temperature.
Results and Discussion
Synthesis
The synthetic route to these side chain liquid crystalline
polymers consists of four parts as already shown in Schemes
1–4. The synthesis of 4∞-hydroxybiphenyl-4-carboxylic acid9,29
and its etherification also including the 3∞-nitro derivatives
were straightforward (see Scheme 1). Scheme 2 shows the route
to the chiral groups of which two were made from (+)-2-(4-
hydroxyphenoxy)propionic acid kindly supplied by BASF. The
J. Mater. Chem., 1998, 8(2), 353–362
357

about the molecular long axis could be restricted by steric or
electronic interactions involving the nitro group or by an
increased moment of inertia. The location of the substituent
towards the centre of the mesogenic core hinders the formation
of layered phases. With a different chiral group Walba22
described similar behaviour although the smectic phases were
not completely lost. With the nitro group in the biphenyl as
in 13d a broad smectic A phase and a narrow nematic region
are observed (Fig. 2). Compared to 13a the core is broadened
by the nitro substituent which according to Goodby dis-
favours formation of a smectic C phase.34 It also reduces the
anisotropy of molecular polarizability and increases the
polarizability perpendicular to the long molecular axis. This
is known to diminish liquid crystalline order,6 as is also seen
in system 13d.
The changes in phase behaviour when the side-chain precur-
sors are attached to the polysiloxane backbone follows the
general expectations. All clearing temperatures are raised by
25–45 °C and crystallization is prohibited (Fig. 2). The stability
of the smectic phases is increased at the expense of the nematic
phases which disappears. In the polysiloxane systems, micro-
phase separation of main chain and side chains exerts an extra
driving force for this formation of layered structures. In the
case of 13b the smectic C* phase cannot be detected when
going to polymer 14b which is somewhat unexpected. Naciri
Fig. 3 Spontaneous polarization of side-chain precursors versus tem-
perature with reference to S –S * phase transition: (+) 13a and (#) 13b
A
C
et al.26 reported a smectic C* phase ranging from T to 37 °C
for a polymer similar to 14b with one methylene unit less in
which have the polar lateral substituents at the end of the stiff
core. The formation of distinct smectic layers is more favour-
able than polar interactions between mesogens giving a small
overlap of the stiff cores. The intermesogenic interactions are
reduced and a nitro substituent at the end of the core in the
polymeric liquid crystals would favour a smectic A phase over
a smectic C*. In the case of 14c the formation of distinct
smectic layers arranges the mesogens to give large overlap of
the stiff cores although a disturbing substituent is positioned
in the central part of the core. Without the restricting polymer
sublayers this substituent causes a longitudinal shift of the
mesogens to give only a nematic phase, 13c. Liquid crystalline
phase behaviour seems to depend on a very delicate balance
between attractive and repulsive intermesogenic forces. The
distance between the mesogens, which is a key parameter, is
affected by geometry, electron distribution and polarizability
of the mesogen. All these factors change when substituents are
introduced or varied and this of course complicates the under-
standing of lateral substituent effects in liquid crystals.
g
the spacer and a different co-polymerization ratio in the
siloxane backbone. It is possible that polymer 14b possesses a
smectic C* phase but that it has a very short helical pitch
which we have not succeeded in unwinding. Therefore the
sample behaves as a smectic A phase. Comparing polymer 14a
without a lateral substituent in the mesogenic core with 14b–d
containing a nitro group, it can be seen that the clearing
temperatures are lowered by the substituent effect. In 14b and
14d the isotropic transition is moved by 15 and 20 °C, respect-
ively. In 14c the substituent in the central part of the core
gives a more drastic effect and the lowering of the transition
temperature is 60 °C. Another striking difference is that 14b
and 14d only exhibit very broad smectic A phases while 14c
has a broad smectic C* phase as well as a smectic A phase. In
order to rationalize this let us consider the layer structures of
the polymers in relation to the low molar mass mesogens. In
most cases of smectic phases of low molar mass mesogens, the
layer structure is not especially well defined. This is seen in X-
ray studies which then only give the first order reflection. The
mesogens are thus able to make use of the most favourable
intermesogenic interactions regardless of how much the mesog-
ens overlap each other. In a polymer liquid crystal with smectic
phases there is a much more distinct layer structure. This is
evident from X-ray diffraction studies where many polymeric
smectic A phases show more than one order of reflection.35 In
a polysiloxane system the backbone is confined to sublayers
between the liquid crystalline layers because of the microphase
separation. This reduces the freedom of mesogen arrangement.
The intermesogenic interactions that were favourable in the
low molar mass smectic phase may not give the lowest energy
arrangement anymore. This can be the case in 14b and 14d
Electro-optical properties, SmC* phase. The desired evalu-
ation of the influence on ferroelectric behaviour exerted by the
nitro group in the mesogenic core becomes rather limited since
the smectic C* phase is present only in 13b and 14c of the
nitro-containing systems. Nevertheless interesting observations
can be made. In Fig. 3 the spontaneous polarizations (P ) of
13a and 13b are shown and one particularly interesting effect
of the nitro substituent can be seen. To our knowledge a P
s
s
value of ca. 700 nC cm−2 is the highest reported for a system
with one chiral centre. One possible explanation could be the
increased double bond character of the phenyl ether bond
caused by the nitro group in an ortho position. This restricts
rotational mobility around the ether linkage.
Table 1 Phases and transition temperatures
phase sequence/°C
material
13a
13b
13c
13d
14a
14b
14c
14d
S
30
K
S *
74
S
106
I
I
X
C
A
−5
S *
52
S
70
I
C
A
35
N*
45
−5
−5
S
S *
59
105
N*
S
115
S
110
70
130
I
70
I
I
I
A
C
A
0
S *
S
61
A
0
I
C
A
−10
S
A
358
J. Mater. Chem., 1998, 8(2), 353–362

Fig. 6 Tilt angle of polymers versus temperature with reference to
S –S * phase transition: (6) 14a and (&) 14c
Fig. 4 Spontaneous polarization of polymers versus temperature with
reference to S –S * phase transition: (6) 14a and (&) 14c
A
C
A
C
smectic A transition polymer 14c speeds up and response times
as low as 10 ms are recorded. The greater temperature depen-
dence of dipole–dipole interactions is a possible explanation
for these latter observations.
In Fig. 4 the spontaneous polarization of the polymers are
presented and 14c shows values up to twice as high as for 14a
at the same reduced temperatures. Measured tilt angles are a
few degrees higher for the nitro containing systems as presented
in Fig. 5 and 6. When comparing spontaneous polarization
and tilt angle for a system, they usually exhibit the same
temperature behaviour, but as is seen in the figures this does
not apply for 13a and 14a. The analysis of 13c and 14c was
limited by the easily hydrolysed central ester linkage in the
mesogen.
Electro-optical properties, SmA phase. Some details of the
electro-optical behaviour in the smectic A phase have been
examined for the side-chain precursors 13a, b, d. In Fig. 9 the
temperature dependence of the induced tilt shows the usual
steep increase close to the S –S * transition for 13a and b,
A
C
while 13d has a smaller temperature dependence because of
the different phase sequence. The variations in induced tilt
with applied field are shown in Fig. 10 at 15 °C below the
Optical response times (t ) are also of interest since they
relate to collective side-chain dynamics, and from Fig. 7 and 8
r
it is clear that for 13b and 14c with a nitro substituent in the
mesogen, t have a stronger temperature dependence than 13a
Iso–S transition temperature. Linear dependencies were
A
r
obtained for all three systems.
and 14a, respectively. 13b with a very high P has response
times in the 100 ms region while 13a is ten times faster. It is
s
The induced tilt as a function of temperature for the polymers
(Fig. 11) does not give the usual behaviour for 14d. The curve
can be divided into three parts with different slopes. As the
evident that the stronger mesogenic interactions in 13b, which
are responsible for the high spontaneous polarization, are
slowing the reorientation of the side-chains. The polymers are
slower and show t in the millisecond range but near the
r
Fig. 7 Response times in the smectic C* phase of side-chain precursors
versus temperature with reference to S –S * phase transition, at a field
A
C
Fig. 5 Tilt angle of side-chain precursors versus temperature with
reference to S –S * phase transition: (+) 13a and (#) 13b
of 8 V mm−1. For comparison the measured values of 13b were
recalculated to this field (t 3 1/E). (+) 13a and (#) 13b
A
C
r
J. Mater. Chem., 1998, 8(2), 353–362
359

Fig. 10 Induced tilt as a function of applied field for side-chain
precursors at 15 °C below the I–S phase transition (T−T =−15 °C):
Fig. 8 Response times in the smectic C* phase of polymers versus
temperature with reference to S –S * phase transition, at a field of
A
tr
A
C
(+) 13a, (#) 13b and (1) 13d
8 V mm−1. For comparison the measured values of 14c were recalcu-
lated to this field (t 3 1/E). (6) 14a and (&) 14c.
r
Other interesting properties. The side-chain precursor 13b
with its very large spontaneous polarization was of course of
interest for NLO measurements. The results from these investi-
gations have been reported earlier.27 From the SHG-intensity
phenomenon remains in a plot of the response time as a
function of inverse temperature (Fig. 12) it cannot be regarded
as a viscosity effect, if the viscosity is assumed to follow an
Arrhenius type of behaviour. Are there changes in the meso-
genic interactions through the temperature span of the smectic
A phase? In Fig. 13 polymers 14b and d show large electroclinic
coefficients at low temperatures. This could indicate a smectic
C* phase at lower temperatures, but this has not been con-
firmed. Further investigations of the polymers show response
times independent of applied field (Fig. 14). At the same
reduced temperatures the response times were within the same
order of magnitude for polymers 14a, b and d. When comparing
14b and 14d at low temperatures (Fig. 13 and 14), polymer
14b shows larger electroclinic coefficient and faster response.
As the phase sequences are identical these observations can be
explained by the difference in position of the nitro group which
may cause a larger transverse dipole moment in 14b compared
to 14d.
in the smectic C* phase, a d value of 0.055 pm V−1 was
eff
estimated. Furthermore, for the first time a field-controllable
SHG-intensity in the smectic A* phase dependent on the square
of the applied electric field was reported.
Conclusions
We have studied the effect of introducing nitro substituents
in the mesogenic core, and have found it to be considerable.
Drastic changes in phase behaviour with substituent position
are observed. In one position the nitro group reduces the
transition temperatures for the low molar mass compound
and more than doubles the maximum spontaneous polariz-
ation to a value of ca. 700 nC cm−2. Moving the nitro
substituent one position in the aromatic ring gives a mesogen
Fig. 9 Induced tilt as a function of temperature with reference to I–S
phase transition, for side-chain precursors. Applied field 7.5, 16.5 and
12.5 V mm−1 for (+) 13a, (#) 13b and (1) 13d, respectively.
Fig. 11 Induced tilt as a function of temperature with reference to
I–S phase transition, for polymers. Applied field 16.2, 53.5 and
23.8 V mm−1 for (6) 14a, ($) 14b and (2) 14d, respectively.
A
A
360
J. Mater. Chem., 1998, 8(2), 353–362

104
103
102
101
100
10-1
10-2
10-3
2.4
2.6
2.8
3.0
T–1/10–3 K–1
3.2
3.4
3.6
Fig. 12 Response times as a function of inverse temperature. Applied
field 53.5 and 23.8 V mm−1 for ($) 14b and (2) 14d respectively. The
lines are linear regressions of the different parts of the 14d plot.
Fig. 14 Response times in the smectic A phase of polymers: (6) 14a,
($) 14b and (2) 14d. T refers to the I–S transition. The lines are
only a guide to the eye.
tr
A
Financial support from The Swedish Natural Science Research
Council, The Swedish Research Council for Engineering
Sciences and The Swedish Defence Research is gratefully
acknowledged.
with no smectic phases. The position of the lateral nitro
substituent has a more pronounced effect in the low molar
mass compounds than in the polymer liquid crystals. In the
polymers the phase separation of the polymer backbone is a
driving force for smectic layer formation and this reduces the
effect of the substituents.
References
The presence of a nitro substituent results in larger tilt
angles and stronger temperature dependence of the response
time in the smectic C* phase. The comparison of the investi-
gated properties in the smectic A phase is complicated by the
differences in phase sequence of the systems caused by the
nitro substituents. The nitro group introduces both steric and
electronic effects. Apart from changing the geometry of the
mesogen it also changes the dipole moment and the polariz-
ability of the mesogen. Depending on substituent position the
total effect on phase behaviour and electro-optical properties
varies over a considerable range.
1
2
V. P. Shibayev and S. V. Byelyayev, Polym. Sci. USSR, 1990,
32(12), 2384.
P. Le Barny and J. C. Dubois, in Side Chain L iquid Crystal
Polymers, ed. C. B. McArdle, Blackie, Glasgow, and Chapman and
Hall, New York, 1989, p. 130.
V. P. Shibaev, M. V. Kozlovsky, L. A. Beresnev, L. M. Blinov and
N. A. Plate´, Polym. Bull., 1984, 12, 299.
N. Shiratori, A. Yoshizawa, I. Nishiyama, M. Fukumasa,
A. Yokoyama, T. Hirai and M. Yamane, Mol. Cryst. L iq. Cryst.,
1991, 199, 129.
3
4
5
6
J. W. Goodby and I. Nishiyama, J. Mater. Chem., 1993, 3(2), 149.
G. W. Gray, in Adv. L iquid Crystals, ed. G. H. Brown, Academic
Press, New York, 1976, p. 1.
7
8
E. M. Averyanov, L iq. Crystals, 1987, 2(4), 491.
G. W. Gray, J. S. Hill and D. Lacey, Makromol. Chem., 1990,
191, 2227.
9
G. W. Gray, J. B. Hartley and B. Jones, J. Chem. Soc., 1955, 1412.
10 T. Inukai, S. Saitoh, H. Inoue, K. Miyazawa, K. Terashima and
K. Furukawa, Mol. Cryst. L iq. Cryst., 1986, 141, 251.
11 M. Svensson, B. Helgee, K. Skarp, G. Andersson and D. Hermann,
Ferroelectrics, 1996, 181, 319.
12 N. F. Cooray, M.-a. Kakimoto, Y. Imai and Y.-i. Suzuki, Polym. J.,
1993, 25, 863.
13 H. Stevens, G. Rehage and H. Finkelmann, Macromolecules, 1984,
17, 851.
14 V. Percec and A. Keller, Macromolecules, 1990, 23, 4347.
15 M. Dumon, H. T. Nguyen, M. Mauzac, C. Destrade, M. F. Achard
and H. Gasparoux, Macromolecules, 1990, 23(1), 355.
16 H. Poths, E. Wischerhoff, R. Zentel, A. Schonfeld, G. Henn and
F. Kremer, L iq. Cryst., 1995, 18(5), 811.
¨
17 R. Zentel and H. Poths, L iq. Cryst., 1994, 16(5), 749.
18 M. A. Osman, Mol. Cryst. L iq. Cryst., 1985, 128, 45.
19 Y. Masuda, Y. Sakurai, H. Sugiura, S. Miyake, S. Takenaka and
S. Kusabayashi, L iq. Cryst., 1991, 10(5), 623.
20 C. J. Booth, J. W. Goodby, J. P. Hardy, O. C. Lettington and
K. J. Toyne, J. Mater. Chem., 1993, 3(9), 935.
21 M. Hird, K. J. Toyne, P. Hindmarsh, J. C. Jones and V. Minter,
Mol. Cryst. L iq. Cryst., 1995, 260, 227.
22 D. M. Walba, M. B. Ros, N. A. Clark, R. Shao, K. M. Johnson,
M. G. Robinson, J. Y. Liu and D. Doroski, Mol. Cryst. L iq. Cryst.,
1991, 198, 51.
23 M. Ozaki, M. Sakuta, K. Yoshino, B. Helgee, M. Svensson and
K. Skarp, Appl. Phys. B, 1994, 59, 601.
Fig. 13 Induced tilt as a function of applied field for polymers:
(6) 14a, ($) 14b and (2) 14d. T refers to the I–S transition. The
lines are only a guide to the eye.
tr
A
J. Mater. Chem., 1998, 8(2), 353–362
361

24 B. Helgee, T. Hjertberg, K. Skarp, G. Andersson and F. Gouda,
L iq. Cryst., 1995, 18(6), 871.
30 K. Skarp and G. Andersson, Ferroelectrics L ett., 1986, 6, 67.
31 O. Mitsunobu, Synthesis, 1981, 1.
25 M. Svensson, B. Helgee and K. Skarp, Conference paper (presen-
tation), in International conference on liquid crystal polymers, 1994,
Beijing, China.
26 J. Naciri, B. R. Ratna, S. Baral-Tosh, P. Keller and R. Shashidhar,
Macromolecules, 1995, 28, 5274.
27 K. Kobayashi, T. Watanabe, S. Uto, M. Ozaki, K. Yoshino, M.
Svensson, B. Helgee and K. Skarp, Jpn. J. Appl. Phys., 1996, 35, L104.
28 R. Shashidhar, J. Naciri, G. P. Crawford and B. R. Ratna,
Ferroelectrics, 1993, 148, 297.
32 C. J. Booth, G. W. Gray, K. J. Toyne and J. Hardy, Mol. Cryst.
L iq. Cryst., 1992, 210 31.
33 M. A. Osman, Z. Naturforsch. T eil A, 1983, 38a, 693.
34 J. W. Goodby, in Ferroelectricity and related phenomena, vol. 7,
Ferroelectric liquid crystals: Principles, properties and applications,
ed. G. W. Taylor, Gordon and Breach, 1991, p. 99.
35 P. Davidson and A. M. Levelut, L iquid Crystals, 1992, 11(4), 469.
29 V. Percec, Q. Zheng and M. Lee, J. Mater. Chem., 1991, 1(4), 611.
Paper 7/04918H; Received 11th July, 1997
362
J. Mater. Chem., 1998, 8(2), 353–362