The dramatic influence of the location of bend and of lateral fluoro substitution on the mesomorphic properties of angular chiral esters based on a 1,3-disubstituted benzene ring

Article abstract of DOI:10.1039/b923267b

The synthesis and mesomorphic properties of a range of 3-ring and 4-ring chiral esters based on a 1,3-disubstituted benzene unit are detailed. These materials all deviate from the usual linear molecular architecture of liquid crystals, and hence are all angular in nature. Some of these materials have the bend right at the end of the molecule where the chain deviates from the normal linear arrangement, and hence a 'hockey stick' molecular architecture is perhaps an accurate description. Other materials have a genuine bent-core construction where the bend is towards the centre of the molecule, and hence are best termed as 'boomerang' shape. In all cases, interesting comparisons of mesophase morphology and transition temperatures were found, both between the various angular materials and with their linear analogues. In particular, the influence on transition temperatures of lateral fluoro substitution in the novel angular materials was found to be wholly different to that found in the known linear analogues. The work forms part of a larger on-going research programme to investigate the mesomorphic and chirality-dependent properties of angular liquid crystals. The research revealed that no mesomorphism in parent compounds is possible when the bend is close to the centre of the molecule, however, lateral fluoro substitution of such compounds facilitates the generation of liquid crystal phases. Where the molecular bend is as a consequence of the terminal unit at the end of the core, then surprisingly high clearing points resulted, and such materials were found to show the potential for a high tilt angle, and a strong tendency towards helical mesophases. Lateral fluoro substitution of these latter examples resulted consistently in significantly higher clearing points, which is in marked contrast to the behaviour reported previously in known liquid crystals of linear molecular architectures. The Royal Society of Chemistry.

Full text of DOI:10.1039/b923267b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
The dramatic influence of the location of bend and of lateral fluoro
substitution on the mesomorphic properties of angular chiral esters based on
a 1,3-disubstituted benzene ring†
*
Kenneth M. Fergusson and Michael Hird
Received 6th November 2009, Accepted 27th January 2010
First published as an Advance Article on the web 22nd February 2010
DOI: 10.1039/b923267b
The synthesis and mesomorphic properties of a range of 3-ring and 4-ring chiral esters based on a 1,3-
disubstituted benzene unit are detailed. These materials all deviate from the usual linear molecular
architecture of liquid crystals, and hence are all angular in nature. Some of these materials have the
bend right at the end of the molecule where the chain deviates from the normal linear arrangement, and
hence a ‘hockey stick’ molecular architecture is perhaps an accurate description. Other materials have
a genuine bent-core construction where the bend is towards the centre of the molecule, and hence are
best termed as ‘boomerang’ shape. In all cases, interesting comparisons of mesophase morphology and
transition temperatures were found, both between the various angular materials and with their linear
analogues. In particular, the influence on transition temperatures of lateral fluoro substitution in the
novel angular materials was found to be wholly different to that found in the known linear analogues.
The work forms part of a larger on-going research programme to investigate the mesomorphic and
chirality-dependent properties of angular liquid crystals. The research revealed that no mesomorphism
in parent compounds is possible when the bend is close to the centre of the molecule, however, lateral
fluoro substitution of such compounds facilitates the generation of liquid crystal phases. Where the
molecular bend is as a consequence of the terminal unit at the end of the core, then surprisingly high
clearing points resulted, and such materials were found to show the potential for a high tilt angle, and
a strong tendency towards helical mesophases. Lateral fluoro substitution of these latter examples
resulted consistently in significantly higher clearing points, which is in marked contrast to the behaviour
reported previously in known liquid crystals of linear molecular architectures.
acute at ꢀ120ꢁ, and most commonly based on a 1,3-disubstituted
Introduction
benzene ring with a symmetrical or near-symmetrical molecular
architecture, where flexible linking groups are symmetrically
organized away from a centrally located core (most notably
a resorcinol unit) has uncovered a whole new classification of
mesomorphism termed ‘banana’ phases. Such materials have
been designated ‘banana’ liquid crystals, and despite being
composed of achiral molecules, several of the so-called banana
liquid crystal phases show a remarkable property of reduced
symmetry mesophases, although an equal number of oppositely
handed domains are always present.13–16 Naturally, this discovery
catalyzed an enormous amount of fundamental research in this
exciting area, which is still very topical.
For over 100 years now, and more particularly in the last 40
years, research into liquid crystal materials has been continu-
ously uncovering novel, interesting anisotropic electric and
optical properties.1–3 These special properties have been tailored
and fine-tuned to meet the exact requirements for device appli-
cations, in particular to enable the high quality liquid crystal
displays seen in the vast array of electronic consumer goods.4–8
The vast majority of research into liquid crystals concerned
linear, rod-like molecular architectures because these materials
provide the optimum basis for the generation of liquid crystal
phases. Nevertheless, many angular liquid crystals have been
synthesized, such as those based on 2,5-disubstituted five-
membered heterocyclic rings, including oxadiazoles,9 thio-
phene,10,11 selenophene,10 and thiadiazole.12 These heterocyclic
systems have a rather shallow angle of bend (ꢀ160ꢁ), and so their
near-linear shape ensures the generation of calamitic liquid
crystal phases such as smectic C, smectic A and nematic. More
recent research into materials where the molecular bend is more
In chiral liquid crystals, the handedness of the molecular
chirality confers reduced symmetry properties on the liquid
crystal phase. These reduced symmetry properties are seen in
many liquid crystal phases, including the chiral nematic (chole-
steric),17–19 the chiral smectic C phase (including the sub-phases
of ferroelectric, antiferroelectric, and others),20–23 and twist-grain
boundary phases.24,25 The reduced symmetry properties of chiral
liquid crystals are fascinating academically, and also of great
technological importance, from the small quantity of chiral
material employed in the twisted nematic (TN) display to ensure
a uniform twist direction, to the larger quantities used in the
super twisted nematic (STN) display to generate twist angles of
Department of Chemistry, University of Hull, Hull, UK HU6 7RX. E-mail:
m.hird@hull.ac.uk
† Electronic supplementary information (ESI) available: Full
experimental procedures. See DOI: 10.1039/b923267b
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 3069–3078 | 3069

View Article Online
up to 270ꢁ, to the chiral material used to confer the ferroelectric
properties to a smectic C phase in the ferroelectric liquid crystal
(FLC) display.26
Synthesis of materials
Generally the synthesis of the materials (Schemes 1–7) was
relatively straightforward, but nevertheless involves some inter-
esting chemistry in the synthesis of the required intermediates,
and is necessarily extensive given the wide range of final prod-
ucts. Extensive use was made of sequential benzyl protections,
The aims of this research in angular molecular architectures
are many and wide-ranging, and have origins in basic structure–
property relationships and in the development of ferroelectric
materials of high dielectric biaxiality.27 Hence this area of
research is completely different to that of the so-called banana
liquid crystals, which generate mesophases that are of a different
category to those reported here. Ferroelectric liquid crystal
displays are recognized as the future for LCOS microdisplay
formats because of their very fast switching and high resolu-
tion.23,28,29 Antiferroelectric liquid crystals have shown promise
for display applications,22,23 in particular the device format that
requires a tilt angle of 45ꢁ.30 Hence, there is a strong need for
research in the tailoring of chiral materials of specific properties
(e.g., low or high spontaneous polarization, low or high tilt angle,
and specific phase sequences) to optimize both ferroelectric and
antiferroelectric device applications. Whilst angular materials
could not be expected to provide as optimum a basis for the
generation of liquid crystal phases as those with linear, rod-like
molecular architectures, the angular structure combined with
molecular chirality would certainly be expected to generate some
unusual, perhaps beneficial properties which could be used
advantageously in liquid crystal mixtures for applications,
and maybe even pave the way for the development of new
applications.
DCC–DMAP (DCC
¼
N,N0-dicyclohexylcarbodiimide and
DMAP ¼ 4-(N,N-dimethylamino)pyridine) esterifications and
deprotections, which is the usual method for the clean and effi-
cient synthesis of multi-ester materials.
As can be seen, this esterification strategy was employed in
Scheme 1 for the synthesis of liquid crystalline compounds 1H
and 1F. A benzyl protection of the angular-shaped 3-hydrox-
ybenzoic acid (15) generated compound 16H that enabled the
chiral terminal unit to be introduced through a DCC–DMAP
esterification to give compound 17H. Subsequent removal of the
benzyl protecting group of compound 17H provided phenol 18H
that allowed a further esterification with 4-benzyloxybenzoic
acid (19) to extend the core to compound 20H. Deprotection of
the angular 2-ring ester (20H) gave phenol 21H, which was
esterified with the simple biphenylcarboxylic acid 22 to generate
the desired final angular liquid crystalline ester 1H. In the case of
the fluorinated analogue (1F) an identical synthetic strategy was
employed, except for the necessary prior synthesis of the tri-
fluorinated benzyl-protected carboxylic acid 16F. Acid 16F was
synthesized through the benzyl-protection of the trifluorophenol
Scheme 1 Reagents and conditions: 1A: BnCl, K2CO3, butanone; 1B: (i) n-BuLi, THF; (ii) (MeO)3B; (iii) 10% HCl; 1C: BnCl, NaOH, EtOH, H2O; 1D:
(S)-2-octanol, DCC, DMAP, DCM; 1E: H2, 10% Pd/C, EtOAc; 1F: DCC, DMAP, DCM.
3070 | J. Mater. Chem., 2010, 20, 3069–3078
This journal is ª The Royal Society of Chemistry 2010

View Article Online
13, followed by a low-temperature lithiation of compound 14 and
Scheme 4 outlines the synthesis of the genuine bent-core
materials 4H and 4F, which are isomers of compounds 3H and
3F respectively, and have the rigid section of the molecular
architecture as the majority of the molecular length. Here the
non-fluorinated precursor (35H) is commercially available, but
the trifluorinated analogue (35F) was generated from compound
33, where the acidic proton ortho to a fluoro substituent was
exploited with lithium diisopropylamide (LDA), an essential
alternative to butyllithium that tolerates the presence of the
bromo substituent. A Suzuki coupling of the resultant bromoaryl
esters (35H and 35F) with the available biphenylboronic acid 36
ultimately furnished the required angular terphenyl intermediate
acids (38H and 38F). These carboxylic acids (38H and 38F) were
then esterified with the chiral phenol 23 in the usual manner to
generate the desired final products 4H and 4F.
subsequent carboxylation.
A
similar strategy was employed for the synthesis of
compounds 2H and 2F with genuine bent-core molecular archi-
tectures (Scheme 2). In the case of esterification of the angular
phenol 25H with acid 22 a substantial quantity of the known
3-ring compound (8H)10 was isolated in addition to the desired
product (compound 2H). This unexpected by-product resulted
from the trans-esterification competing with the desired esterifi-
cation of compound 25H. The analogous trifluorinated product
(2F) did not suffer from the same problem, showing that the
fluorinated phenol (25F) is more reactive than the non-fluori-
nated analogue (25H). We have found tꢁhrough subsequent
research that the use of low temperature (0 C) circumvents the
problem, as indeed does the reverse mode of construction of the
multi-ester compound (see Scheme 6).
Scheme 5 begins the synthetic routes to the analogous 3-ring
angular esters. Phenols 18H and 18F were available from
previous syntheses, and were esterified with the biphenyl acid 22
to yield the desired final products (5H and 5F) with the chiral
centre in a conformationally flexible location at the end of the
core. Scheme 6 shows the synthesis of 3-ring esters (compounds
6H and 6F) with three ester groups to provide greater confor-
mational flexibility than in compounds 5H and 5F, which possess
just two ester groups. Esterification of acid 39 and phenol 40
generated the benzyl-protected esters-carboxylic acid (41), which
on deprotection yielded the required ester-carboxylic acid 42.
Esterification of ester-carboxylic acid 42 with angular chiral
phenols 18H and 18F provided the desired angular final products
6H and 6F.
Scheme 3 outlines the synthesis of genuine bent-core mate-
rials (3H and 3F) where there is a direct aryl–aryl bond at the
bent-core unit, and hence with one less ester group, the
conformational flexibility is reduced in comparison with
compounds 2H and 2F. Central to the synthesis is the genera-
tion of angular boronic acids 28H and 28F. Both boronic acids
were generated by low-temperature lithiation followed by
treatment with trimethyl borate, however, in the non-fluori-
nated case (28H) a bromo-lithium exchange was employed,
whereas in the case of the trifluorinated boronic acid (28F) the
acidic proton ortho to a fluoro substituent was exploited. Suzuki
couplings of these boronic acids with the chiral aryl bromide 30
provided the desired angular biaryl units (31H and 31F), which
were subsequently deprotected to provide the required phenols
32H and 32F. These phenols were then esterified in the usual
manner with carboxylic acid 22 to provide the desired final
products (3H and 3F).
Scheme 7 outlines the synthesis of the novel 4-ring linear esters
7H and 7F, which were required in order to make effective
comparisons with the analogous 4-ring angular and bent-core
materials. These linear materials were synthesized through the
Scheme 2 Reagents and conditions: 2A: DCC, DMAP, DCM; 2B: H2, 10% Pd/C, EtOAc.
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 3069–3078 | 3071

View Article Online
Scheme 3 Reagents and conditions: 3A: BnCl, K2CO3, butanone; 3B: (i) n-BuLi, THF; (ii) (MeO)3B; (iii) 10% HCl; 3C: (S)-2-octanol, DCC, DMAP,
DCM; 3D: Pd(PPh3)4, Na2CO3, DME, H2O; 3E: H2, 10% Pd/C, EtOAc; 3F: DCC, DMAP, DCM.
Scheme 4 Reagents and conditions: 4A: (i) LDA, THF; (ii) CO2, THF; (iii) 10% HCl; 4B: EtOH, H2SO4; 4C: Pd(PPh3)4, Na2CO3, DME, H2O; 4D: (i)
KOH, EtOH, H2O; (ii) 36% HCl; 4E: DCC, DMAP, DCM.
3072 | J. Mater. Chem., 2010, 20, 3069–3078
This journal is ª The Royal Society of Chemistry 2010

View Article Online
are 23ꢁ and 130 nC cmꢂ2 respectively.32 Now the broader picture
is being reported with respect to the influence of the position of
the bend within the molecular structure and of the influence of
lateral fluoro substituents at the location of the bend on meso-
morphic behaviour. As far as possible, the mesomorphic
behaviour is also compared with analogous linear compounds.
As shown in Table 1, compound 1H has a linear shape for the
core structure, but the structure is bent at the end of the core
through a conformationally flexible ester link. This flexible ester
link facilitates the generation of conformers that confer a near-
linear structure, and in combination with four benzene rings, it is
perhaps no surprise that compound 1H exhibits mesomorphism
to a high temperature. Nevertheless, in comparison with the
novel linear analogue (compound 7H), synthesised to put the
results into perspective (Table 3), the clearing point of compound
1H is reduced by 81 ꢁC. More significantly, the rich meso-
morphism in the form of the SmC*anti and SmC*ferri, sub-
phases seen in the linear four-ring analogue (compound 7H) has
been completely eliminated in the angular molecular architecture
(1H). The generation of these so-called sub-phases of these ester
molecular architectures is well-documented (for example, in
compounds 8H, 8F and 9H, Table 3),10 hence, given the low
melting point and high clearing temperature of the angular
analogue (compound 1H), it might have been expected that the
sub-phases would remain. Indeed some reports suggest that
a bent molecular architecture actually favours the generation of
the antiferroelectric (SmC*anti) phase.33,34
Scheme 5 Reagents and conditions: 5A: DCC, DMAP, DCM.
usual sequence of benzyl protection, esterification and depro-
tection.
Melting points, mesophase morphology and transition
temperatures
In comparison with linear molecular architectures, angular
analogues are obviously not so conducive to mesomorphism, and
hence four-ring structures (Table 1) were targeted to maximize
the chances of obtaining liquid crystalline phases. However,
some appropriate three-ring analogues (Table 2) were targeted to
test the limits of mesomorphism from these angular molecular
architectures. In order to put the mesomorphic properties of the
angular materials into perspective, make some meaningful
comparisons, and draw conclusions, Table 3 details two novel
linear four-ring analogues and a few known materials of
appropriate molecular structure.
As can be seen from the transition temperatures for compound
9H,10 the slight bend (150ꢁ) caused by the thiophene ring does
still support the SmC*anti phase, and in comparison with the
linear analogue (compound 8H),10 the phase stability is reduced,
but to a much lesser extent than the reduction of the SmC*ferro
phase stability. The melting point of compound 1H (38 ꢁC) is
remarkably low for a compound with four benzene rings and
three ester groups, and despite this low melting point the
SmC*anti phase is not exhibited. The formation of the SmC*anti
phase is driven by the interaction of the terminal chains, and in
the case of compound 1H the chiral terminal chain is located
laterally at a 120ꢁ angle from the linear core which is too severe
for favourable interactions for the SmC*anti phase.
Compound 1H has been reported previously in a recent
communication, and some very special properties were dis-
cussed.31 Most significantly in terms of a novel phase structure
(termed SmA*helical) arising as a consequence of the bend and
chirality causing a marked tendency to twist. Additionally,
compound 1H was found to show a most unusual behaviour in
the combination of a very high tilt angle (39ꢁ) and a very low
value of spontaneous polarization (50 nC cmꢂ2); the corre-
sponding values for the known linear material (compound 8H)
Scheme 6 Reagents and conditions: 6A: DCC, DMAP, DCM; 6B: H2, 10% Pd/C, EtOAc.
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 3069–3078 | 3073

View Article Online
Scheme 7 Reagents and conditions: 7A: DCC, DMAP, DCM; 7B: H2, 10% Pd/C, EtOAc.
Table 2 Transition temperatures (ꢁC) of 3-ring angular esters
Another very significant point in comparing the angular
material (compound 1H) with the linear analogue (compound
7H, Table 3) is that the molecular bend reduces the SmA* phase
stability by 81 ꢁC, yet the SmC* phase stability is reduced by just
66 ꢁC. Hence it appears that a bent molecular architecture
favours the formation of tilted phases, such as the SmC*, over
Compound structure and transition temperatures/ꢁC
Table 1 Transition temperatures (ꢁC) of 4-ring angular esters
Compound structure and transition temperatures/ꢁC
and above orthogonal phases, such as the SmA*, which is
perhaps expected given the obvious link between bend and tilt.
Although compound 1H exhibits a smectic A phase with
a helical structure (SmA*helical), the trifluorinated analogue
(compound 1F) shows a conventional smectic A phase (SmA*),
as evidenced by optical polarizing microscopy. However,
surprisingly the three lateral fluoro substituents of compound 1F
enhance the clearing point (smectic A phase stability) in
comparison with the non-fluorinated parent system (compound
1H). The influence of just one lateral fluoro substituent in linear
mesogens causes a marked reduction in transition temperatures,
and only in very rare examples where the lateral fluoro substit-
uent is shielded by the core structure is the clearing point
increased, and even so by only a couple of degrees. Hence, here
the fact that three lateral fluoro substituents have increased the
clearing point (SmA* phase stability) by 9 ꢁC is truly remarkable.
ꢁ
However, the SmC* phase stability is reduced (by 6 C), which
again is somewhat unusual since the influence of lateral fluoro
substituents in linear systems usually is to enhance the tendency
for molecular tilting. The comparison between the two four-ring
linear analogues (7H and 7F, Table 3) shows the typical influence
3074 | J. Mater. Chem., 2010, 20, 3069–3078
This journal is ª The Royal Society of Chemistry 2010

View Article Online
Table 3 Transition temperatures (ꢁC) of esters for comparison with their angular analogues
Compound structure and transition temperatures/ꢁC
of lateral fluoro substituents, in this case two of them next to each
other, with reductions in SmA* and SmC*ferro phase stability
by 18 ꢁC and 27 ꢁC respectively. When the lateral fluoro
substituents are in a more central location of the core
(compound 7Fa) the reduction of phase stability is even more
marked, with reductions in SmA* and SmC*ferro phase stability
four rings. Given that the parent system (compound 2H) is
non-mesogenic, it is not possible to comment accurately on the
influence of the lateral fluoro substituents of compound 2F on
liquid crystal phase stability. Nevertheless, since compound 2H
supercools to 40 ꢁC, it is clear that the lateral fluoro substituents
of compound 2F have increased the clearing point, which
supports the results discussed above for the angular compounds
(1H and 1F).
ꢁ
ꢁ
by 24 C and 44 C respectively, again a typical effect in linear
structures. Hence, the influence of lateral fluoro substituents of
angular materials on mesomorphic properties is significantly
different to that seen in the linear analogues (see below for
further discussion). At this point it should be mentioned that the
structure of compound 43 in a previous publication27 was
wrongly reported as being identical to compound 2F (Table 1),
however, the true structure is that of compound 7Fa shown in
Table 3.
Compound 3H is also a genuine bent-core material with the
bend towards the centre of the molecule, however, in this case an
ester group has been removed in order to reduce the conforma-
tional flexibility, which also reduces the molecular length. Given
that the more flexible system (compound 2H) is non-mesogenic,
it is not surprising that compound 3H is also non-mesogenic,
especially since the melting point is several degrees higher.
However, it is rather surprising, given this reduced conforma-
tional flexibility, that the fluorinated analogue 3F still exhibits
the SmC* phase (28.9 ꢁC) and the SmA* phase (31.9 ꢁC),
although both phases are monotropic. Given that the parent
system (compound 3H) crystallizes on cooling at 65 ꢁC, then the
influence of the three lateral fluoro substituents on meso-
morphism cannot be discerned. The reduced conformational
flexibility of compound 3F when compared with compound 2F
means less chance of linear conformers and hence the reduced
SmA* phase stability (clearing point) of 18 ꢁC. However, the
SmC* phase stability is only reduced by 6 ꢁC, which provides
further indication that a more bent-core structure favours the
tilted smectic C phase over and above the smectic A phase.
Compound 4H is isomeric with compound 3H, but has both
conformationally flexible ester groups at one side of the bend and
the other side of the bend is a very rigid terphenyl structural
moiety. Hence there would be no chance of mesomorphism in
compound 4H, particularly given the very high melting point of
109.8 ꢁC and a crystallization point on cooling of 95 ꢁC.
However, the influence of three lateral fluoro substituents in the
On moving the bend towards the middle of the molecule to
create a genuine bent-core material (compound 2H), all liquid
crystal phases have been eliminated, despite the material super-
ꢁ
cooling down to 40 C. Hence liquid crystal phase stability has
been reduced by at least 116 ꢁC in comparison with the angular
ꢁ
material 1H, and by a massive 200 C when compared with the
linear analogue 7H. Despite conformationally flexible ester
groups on either side of the molecular bend, there is much less
scope for linear conformers in compound 2H than for compound
1H, and hence the liquid crystalline nature is destroyed. The
introduction of three lateral fluoro substituents into the genuine
bent-core structure (compound 2F) causes a significant reduction
in melting point (compared with compound 2H), which contrasts
with the influence of the lateral fluoro substituents in the angular
compound 1F, and reflects the unpredictable effect of lateral
fluoro substituents on melting point. The very low melting point
of compound 2F (38 ꢁC) allows the exhibition of an enantiotropic
SmA* phase to 49.8 ꢁC and a monotropic SmC* phase to
34.5 ꢁC, and this mesomorphism represents the first reported for
a genuine bent-core (1,3-disubstituted phenylene) material of just
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 3069–3078 | 3075

View Article Online
other systems discussed above (compounds 1F, 2F and 3F) gave
a good reason to be hopeful of mesomorphism from compound
4F, however, a high melting point (84.6 ꢁC) prevents enantio-
tropic mesomorphism, and a crystallization temperature of 65 ꢁC
on cooling precludes monotropic mesomorphism.
an increase in clearing point through three lateral fluoro
substituents is unique. One argument to consider in explaining
this unique situation is based on well-known examples of linear
compounds where a lateral fluoro substituent that occupies free
space at the end of a core tends to depress clearing point to
a much lesser extent than when a lateral fluoro substituent
occupies a central core location. Clearly the molecular bend in
compounds 1–6 necessarily creates substantial free space at the
location of the bend, certainly more so than in any position of the
comparable linear materials (compounds 7 and 8). Filling this
free space with lateral fluoro substituents to generate compounds
1F–6F would be expected to enhance the intermolecular attrac-
tions and facilitate lamellar packing, and hence enhance smectic
phase stability. However, there is a certain degree of steric
protrusion from the lateral fluoro substituents, but this is shel-
tered by the angular molecular architecture. Perhaps of more
significance for angular molecular shapes, the space-filling effect
of the lateral fluoro substituents in all the angular materials
serves to smooth out the molecular bend and hence effectively
straighten the molecules to some extent. Such straightening of
the molecules would account for higher clearing points, and the
obvious link between bend and tilt accounts for the lower
SmC* phase stability of the straighter fluorinated materials
(compounds F) compared with the parent systems (compounds
H). Significantly, the reduced bend of trifluorinated compound
1F, in comparison with the parent compound 1H, reduces the
tendency towards twisting and a conventional chiral smectic A
(SmA*) phase is generated rather than the SmA*helical phase.
The 3-ring analogues (5H and 5F) serve to illustrate the limits
of mesomorphism of angular molecular architectures (Table 2).
For the linear systems (Table 3), the clearing point of the 3-ring
ꢁ
system (compound 8H) is 108 C lower than that of the 4-ring
analogue (compound 7H), so mesomorphism from the 3-ring
angular system (compound 5H) would be expected, provided
that the melting point is sufficiently low. Given the very low
melting point of the 4-ring angular system (compound 1H), then
the melting point of compound 5H should perhaps also be low.
However, rather surprisingly, the melting point of the compound
ꢁ
5H is 7 C higher than that of the 4-ring analogue (compound
1H), but nevertheless still exhibits the SmC* and SmA* phases
enantiotropically. Perhaps even more surprisingly, the molecular
bend is more supportive of mesomorphism in the 3-ring system
(5H), where the clearing point is reduced by 74 ꢁC, compared
with the reduction of 81 ꢁC in the 4-ring ꢁanalogue (1H).
Although this difference is somewhat small (7 C), it should be
the other way around, since it would be expected that a larger
4-ring system would be more tolerant of the disruption to
mesomorphism caused by a molecular bend at the end of the core
than a shorter 3-ring analogue.
The influence of the three lateral fluoro substituents on the
mesomorphism of the 3-ring system (5F) is similar to the
surprising result seen for the 4-ring analogue (1F), in that the
lateral fluoro substituents actually enhance the SmA* phase
stability (clearing point), and an identical increase of 9 ꢁC is seen.
However, the reduction in the SmC* phase stability (15 ꢁC)
caused by the three lateral fluoro substituents is much greater for
the 3-ring system than for the 4-ring analogue.
Tilt angle and spontaneous polarization
The tilt angle and spontaneous polarization as a function of
temperature have already been reported for compound 1H, but
are illustrated here in Fig. 1 and 2 respectively for comparison
with the behaviour of the trifluoro-substituted analogue
(compound 1F) illustrated in Fig. 3 and 4 respectively.
Another 3-ring structure (compound 6H), with an additional
ester linkage, was prepared to examine the effect of a greater
flexibility within the core when compared with compound 5H. As
can be seen from Table 2, the additional ester group hardly
makes any difference to the melting point and mesomorphic
properties, despite extending the linear section of the molecular
core, however, some subtle points are worthy of discussion. The
slight reduction in the SmA* phase stability is probably the net
result of the generation of additional bent conformers out-
weighing the extension of the linear core, and the slight increase
in the SmC* phase stability is most likely caused by the enhanced
lateral polarity from the additional ester group. The fluorinated
analogue (6F) is rather more disadvantaged in terms of clearing
point than the less flexible analogue (5F), perhaps confirming
that the additional ester confers more possible bent conformers
than linear conformers.
The tilt angle behaviour of compound 1H (Fig. 1) is very much
first order in nature with a very steep rise in tilt angle on cooling,
to reach a very high value of 39ꢁ. In contrast the tilt angle of
compound 1F (Fig. 3) shows a more second order behaviour of
a gradual increase as the temperature is reduced, to attain
a maximum value of 26ꢁ. The unusual tilt angle behaviour of
compound 1H can be attributed to the tilted, helical nature of the
smectic A phase (SmA*helical) from which the material is cooled
into the SmC* phase, which in turn is a result of the bent
So why do the angular compounds (1F, 5F and 6F) and the
genuine bent-core compounds (2F and 3F) with lateral fluoro
substituents show higher clearing points (SmA* phase stability)
than the respective parent systems (1H, 5H, 6H, 2H and 3H)? As
mentioned above, except in very rare cases, the influence of one
lateral fluoro substituent on clearing point of linear structures is
to reduce the clearing point quite markedly, although lateral
fluoro substituents are known to enhance the phase stability of
banana liquid crystals.14,35,36 However, the situation here of
Fig. 1 The tilt angle values of compound 1H as a function of reduced
temperature in the SmC*ferro phase.
3076 | J. Mater. Chem., 2010, 20, 3069–3078
This journal is ª The Royal Society of Chemistry 2010

View Article Online
the layers rather than the perpendicular direction that is
measured.
Summary
A wide range of novel angular chiral esters has been prepared.
Where mesomorphism is generated in the angular materials it
always takes the form of a chiral smectic C phase and a chiral
smectic A phase, but chiral smectic C sub-phases such as anti-
ferroelectric and ferrielectric phases, seen in the analogous linear
materials, are not exhibited. The incorporation of lateral fluoro
substituents fills space at the location of the bend, which tends to
reduce the degree of bend, and hence clearing points are
enhanced in comparison to the parent systems, which contrasts
markedly with the effect seen in linear molecules.
Fig. 2 The spontaneous polarization (Ps) values of compound 1H as
a function of reduced temperature in the SmC*ferro phase.
A bend at the end of the molecular core generates the highest
clearing points because the angle of bend is not as marked as
when the bend is at a mid-core position. Nevertheless, the bend at
the end of the core does cause a reduction of clearing point by
81 ꢁC (comparing compound 1H with compound 7H), and as
a consequence of the bend and chirality causing a marked
tendency to twist, a helical chiral smectic A (SmA*helical) phase
is generated. The incorporation of three lateral fluoro substitu-
ents (compound 1F) fills free space at the bend, effectively
providing a straighter molecular structure, and this enhances the
clearing temperature, but precludes the formation of the
SmA*helical phase and reduces the phase stability of the SmC*
phase.
Fig. 3 The tilt angle values of compound 1F as a function of reduced
temperature in the SmC*ferro phase.
The genuine bent-core materials, where the bend is in the
centre of the molecules, have a more marked bend, and hence the
parent systems (compounds 2H and 3H) do not show liquid
crystal phases. However, compounds 2F and 3F benefit from the
three lateral fluoro substituents as outlined above, and also from
lower melting points to exhibit the SmC* phase and the SmA*
phase, which is the first report of liquid crystal phases from
genuine bent-core materials with just four rings.
Experimental
Structural information of materials was obtained by 1H and 13
C
NMR spectroscopy (JEOL Eclipse 400 spectrometer), by mass
spectrometry (Finnigan-MAT 1020 spectrometer), and by
elemental analysis (Fisons EA1108 CHN). Specific optical
rotation, [a]D, for the final compounds (1 to 11) was determined
Fig. 4 The spontaneous polarization (Ps) values of compound 1F as
a function of reduced temperature in the SmC*ferro phase.
ꢁ
at 26 C in chloroform using a Bendix-NPL Automatic Polar-
imeter Type 143A. The purity of all final compounds (1 to 11)
was checked by HPLC analysis (Merck-Hitachi with Merck RP
18 column, Cat. No. 16 051) and were all found to be >99.9%
pure. The progress of some reactions and the purity of some
intermediates were analyzed by Gas Liquid Chromatography
(GLC) using a Varian CP3380 gas chromatograph with a 10 m,
0.25 mm internal diameter, 0.12 mm fused silica capillary
column. All melting points, transition temperatures and meso-
phase morphologies were determined using an Olympus BH-2
polarizing microscope in conjunction with a Mettler FP52
heating stage and FP5 temperature controller, and these values
were confirmed using differential scanning calorimetry (Perkin-
Elmer DSC-7 and Mettler-Toledo DSC822e). The spontaneous
polarization values and tilt angles were evaluated as a function of
molecular structure. Conversely, the rather normal tilt angle
behaviour of the fluorinated analogue (compound 1F) is a result
of the less angular molecular structure, brought about by the
lateral fluoro substituents when compared with the parent system
(compound 1H).
The spontaneous polarization behaviour on cooling is much
the same for both compounds 1H and 1F (Fig. 2 and 4 respec-
tively), with a gradual second order manifestation. However, the
maximum value attained is just 50 nC cmꢂ2, which is much
smaller than the typical value for materials such as compounds
8H and 8F (around 130 nC cmꢂ2). This low value of spontaneous
polarization can be attributed to the angular nature of molecules
causing the direction of polarization to be more in the plane of
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 3069–3078 | 3077

View Article Online
12 T. Geelhaar, Ferroelectrics, 1988, 85, 329–349.
13 T. Niori, T. Sekine, J. Watanabe, T. Furukawa and H. Takezoe,
J. Mater. Chem., 1996, 6, 1231–1233.
14 G. Pelzl, S. Diele and W. Weissflog, Adv. Mater., 1999, 11, 707–724.
15 A. R. Reddy and C. Tschierke, J. Mater. Chem., 2006, 16, 907–961.
16 M. Hird, Liq. Cryst. Today, 2005, 14, 9–21.
17 H. L. de Vries, Acta Crystallogr., 1951, 4, 219.
18 J. L. Fergason, Mol. Cryst., 1966, 1, 293–307.
19 J. L. Fergason, N. N. Goldberg and R. J. Nadalin, Mol. Cryst., 1966,
1, 309–323.
temperature in electro-optical cells (5 mm thickness with anti-
parallel rubbed polyimide alignment layers) using equipment
purchased from Instec Inc. (USA), employing a triangular wave
of 10 V mmꢂ1 at a frequency of 30 Hz.
Compounds 13, 15, 19, 26, 29, 33, 39 and 40, and other simple
starting materials and solvents are commercially available.
Compounds 22,10 23H,37 and 3638 were prepared as described
previously.
20 A. D. L. Chandani, Y. Ouchi, H. Takezoe and A. Fukuda, Jpn.
J. Appl. Phys., 1989, 28, L1261.
21 R. B. Meyer, L. Liebert, L. Strzelecki and P. Keller, J. Phys. Lett.,
1975, 36, L69–L71.
Full preparation details and compound characterization can
be found in the ESI†.
22 A. Fukuda, Y. Takanishi, T. Isozaki, K. Ishikawa and H. Takezoe,
J. Mater. Chem., 1994, 4, 997–1016.
Acknowledgements
23 S. T. Lagerwall, Ferroelectrics, 2004, 301, 15–45.
24 J. W. Goodby, I. Nishiyama, A. J. Slaney, C. J. Booth and
K. J. Toyne, Liq. Cryst., 1993, 4, 37–66.
25 J. W. Goodby, M. A. Waugh, S. M. Stein, E. Chin, R. Pindak and
J. S. Patel, J. Am. Chem. Soc., 1989, 111, 8119–8125.
26 J. W. Goodby, J. Mater. Chem., 1991, 1, 307–318.
27 M. Hird, J. W. Goodby, N. Gough and K. J. Toyne, J. Mater. Chem.,
2001, 11, 2732–2742.
We thank the Department of Chemistry, University of Hull for
providing research facilities, and the Engineering and Physical
Sciences Research Council (EPRSC) for financial support.
References
1 P. J. Collings and M. Hird, Introduction to Liquid Crystals Chemistry
and Physics, Taylor and Francis, London, 1997.
2 Handbook of Liquid Crystal Research, ed. P. J. Collings and J. S. Patel,
Oxford University Press, Oxford, 1997.
3 D. Demus, J. W. Goodby, G. W. Gray, H.-W. Spiess and V. Vill,
Handbook of Liquid Crystals, Wiley-VCH, Chichester, 1998.
4 M. Hird, Chem. Soc. Rev., 2008, 36, 2070–2095.
5 S. M. Kelly, Flat Panel Displays Advanced Organic Materials, Royal
Society of Chemistry, Cambridge, 2000.
28 T. D. Wilkinson, W. A. Crossland and A. B. Davey, Ferroelectrics,
2002, 278, 799–804.
29 R. Zhang and H. Hua, Appl. Opt., 2008, 47, 2888–2896.
30 K. D’have, P. Rudquist, M. Matuszczyk, S. T. Lagerwall, H. Pauwels
and R. Dabrowski, Proc. SPIE, 2000, 3955, 33–44.
31 K. M. Fergusson and M. Hird, Adv. Mater., 2007, 19, 211–214.
32 W. K. Robinson, H. F. Gleeson, M. Hird, A. J. Seed and P. Styring,
Ferroelectrics, 1996, 178, 249–266.
33 I. Nishiyama, Adv. Mater., 1994, 6, 966–970.
6 D. Pauluth and K. Tarumi, J. Mater. Chem., 2004, 14, 1219–1227.
7 M. Bremer and L. Lietzau, New J. Chem., 2005, 29, 72–74.
8 M. Bremer, M. Klassen-Memmer and K. Tarumi, Adv. Mater., 2004,
16, 1882–1885.
34 Y. Ouchi, Y. Yoshioka, H. Ishii, K. Seki, M. Kitamura, R. Noyori,
Y. Takanishi and I. Nishiyama, J. Mater. Chem., 1995, 5, 2297–
2304.
35 C. Tschierske and G. Dantlgraber, Pramana, 2003, 61, 455–481.
36 R. Amaranatha Reddy, V. A. Raghunathan and B. K. Sadashiva,
Chem. Mater., 2005, 17, 274–283.
37 A. I. Stipetic, J. W. Goodby, M. Hird, Y. M. Raoul and
H. F. Gleeson, Liq. Cryst., 2006, 33, 819–828.
9 L. A. Madsen, T. J. Dingemans, M. Nakata and E. T. Samulski, Phys.
Rev. Lett., 2004, 92, 1455051–1455054.
10 A. J. Seed, M. Hird, P. Styring, H. F. Gleeson and J. T. Mills, Mol.
Cryst. Liq. Cryst., 1997, 299, 19–25.
11 M. Hird, K. J. Toyne, J. W. Goodby, G. W. Gray, V. Minter,
R. P. Tuffin and D. G. McDonnell, J. Mater. Chem., 2004, 14,
1731–1743.
38 G. W. Gray, M. Hird and K. J. Toyne, Mol. Cryst. Liq. Cryst., 1991,
195, 221–237.
3078 | J. Mater. Chem., 2010, 20, 3069–3078
This journal is ª The Royal Society of Chemistry 2010