Microscopy studies of the nematic NTB phase of 1,11-di-(1′′-cyanobiphenyl-4-yl)undecane

Article abstract of DOI:10.1039/c3tc32137a

A detailed microscopy study of the mesophases of 1,11-di-(1′′- cyanobiphenyl-4-yl)undecane (CB11CB) was undertaken, with an emphasis on attempting to relate the recent helical twist-bend model of the N_TB phase to the observed optical textures. Our ability to draw a freestanding film indicates that the phase is unlikely to be nematic and possesses long range periodicity. No electrooptic response could be detected in the N_TB phase, with or without addition of a chiral dopant although dielectric breakdown and space charge were observed at high voltage leading to field dependent pulsing of the sample. No optical rotation associated with a macroscopic helical structure could be discerned in the N_TB phase. However, this observation could be taken to be in keeping with the extremely short pitch of a proposed twist-bend model.

Full text of DOI:10.1039/c3tc32137a

Journal of
Materials Chemistry C
View Article Online
View Journal | View Issue
PAPER
Microscopy studies of the nematic NTB phase of
1,11-di-(100-cyanobiphenyl-4-yl)undecane†
Cite this: J. Mater. Chem. C, 2014, 2,
556
Richard J. Mandle, Edward J. Davis, Craig T. Archbold, Stephen J. Cowling
*
and John W. Goodby
A detailed microscopy study of the mesophases of 1,11-di-(100-cyanobiphenyl-4-yl)undecane (CB11CB) was
undertaken, with an emphasis on attempting to relate the recent helical “twist-bend” model of the NTB phase
totheobservedopticaltextures. Ourabilitytodrawa freestanding filmindicatesthatthephaseis unlikelytobe
nematic and possesses long range periodicity. No electrooptic response could be detected in the NTB phase,
with or without addition of a chiral dopant although dielectric breakdown and space charge were observed at
high voltage leading to field dependent pulsing of the sample. No optical rotation associated with a
macroscopic helical structure could be discerned in the NTB phase. However, this observation could be
taken to be in keeping with the extremely short pitch of a proposed “twist-bend” model.
Received 29th October 2013
Accepted 23rd November 2013
DOI: 10.1039/c3tc32137a
www.rsc.org/MaterialsC
the NTB phase formed by the dimesogen 1,7-di-(100-cyanobiphenyl-
4-yl)heptane (CB7CB). The results obtained from these studies, in
Introduction
particular from freeze-fracture TEM (FFTEM), X-ray investigations
and H NMR studies, indicate that the twist-bend nematic NTB
Over recent years dimeric liquid crystals have been found to
exhibit a variety of fascinating mesomorphic properties. For
example, they have been shown to possess novel exoelectric
properties,1–4 to stabilise Blue phases over wide temperature
ranges,5–8 and to exhibit unusual nematic (oen termed NX)
phases.9,10 Of particular interest are dimesogens that have
methylene spacer units sandwiched between aromatic moieties
which carry polar terminal groups, as depicted by structure (I) for
the 1,u-di-(100-cyanobiphenyl-4-yl) alkanes (CBnCBs), see Fig. 1.
When the methylene chain is of an even parity the mesophase
behaviour of these materials is classical. However, dimesogens
that have an odd number of methylene units exhibit a transition
from a nematic phase to a lower-temperature mesophase that
has recently been called a twist-bend nematic phase (NTB).11–23
Many studies regarding the physical properties and structural
features of the twist-bend modication of the nematic phase have
been performed in recent years, with the majority focussing on
2
mesophase has a structure in which the constituent the molecules
intercalate to give a spiralling organization which is spatially
homogeneous, i.e. the molecular arrangement in the NTB phase is
nematic-like rather than lamellar.21–23 Despite the spatially
homogeneous nature of the NTB phase, its associated defect
textures are strikingly similar to those observed for smectic mes-
ophases as they show both focal-conic and parabolic defects.14
Clearly, such defects can be formed only when a periodic structure
is present; a criterion that is met in the NTB phase not by the
presence of a lamellar molecular arrangement but instead by the
presence of a helical molecular arrangement, which in the case of
CB7CB has an extremely tight pitch (ꢀ8 nm).21–23
CB7CB, however, exhibits a monotropic N–NTB phase tran-
sition, which hinders detailed examinations of its defect
textures by thermal polarized optical microscopy between glass
slides or via free-standing lms. Conversely, the structurally
related
material
1,11-di-(100-cyanobiphenyl-4-yl)undecane
(CB11CB) exhibits an enantiotropic NTB phase.14 Therefore, it
has the potential to provide access to the study of free-standing
lms, which could prove to be diagnostic in the determination
of the mesophase characterization/structure of the NTB phase.
Thus, in the following we examine the bulk properties of the
NTB phase of CB11CB and relate the experimental results
obtained to the proposed theoretical models for its structure.
Fig. 1 The molecular structures of CB7CB and CB11CB.
Materials and methods
1,11-Di-(100-cyanobiphenyl-4-yl)undecane and its bromo-
substituted precursors were prepared via literature methods.11
Department of Chemistry, University of York, Heslington, York, North Yorkshire, UK.
E-mail: john.goodby@york.ac.uk
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c3tc32137a
556 | J. Mater. Chem. C, 2014, 2, 556–566
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
to be compound 1 (Rf ¼ 0.85, DCM, yield ¼ 100 mg; 10%),
compound 2 (Rf ¼ 0.75, DCM, yield ¼ 77 mg; 8.4%) and CB11CB
(Rf ¼ 0.70, DCM, yield ¼ 472 mg; 57%). Compound 2 and
CB11CB were individually recrystallised from toluene to give the
compounds as ne white crystals.
CB11BBr (2)
1H NMR (400 MHz, CDCl3): 1.16–1.34 (14H, m, –CH2–(CH2)7–
CH2–), 1.51–1.61 (4H, m, Ar–CH2–CH2–(CH2)7–CH2–CH2–Ar),
2.55 (2H, t, J ¼ 7.9, Ar–CH2–(CH2)10–Ar), ꢃ2.58 (2H, t, J ¼ 7.9 Ar–
CH2–(CH2)10–Ar), 7.17 (2H, ddd, J ¼ 1.5, J ¼ 2.1, J ¼ 8.2, Ar), 7.21
(2H, ddd, J ¼ 1.8, J ¼ 2.1, J ¼ 8.2, Ar), 7.36 (2H, ddd, J ¼ 1.5, J ¼
2.1, J ¼ 8.5, Ar), 7.39 (2H, ddd, J ¼ 1.8, J ¼ 2.4, J ¼ 8.2, Ar), 7.43
(2H, ddd, J ¼ 1.8, J ¼ 2.1, J ¼ 8.2, Ar), 7.47 (2H, ddd, J ¼ 1.8, J ¼
2.4, J ¼ 8.5, Ar), 7.59 (2H, ddd, J ¼ 1.2, J ¼ 2.4, J ¼ 8.5, Ar), 7.64
(2H, ddd, J ¼ 1.2, J ¼ 2.4, J ¼ 8.5, Ar) ppm.
Scheme 1 The final step in the synthesis of CB11CB.
The target material CB11CB was determined to be >99% pure by
reverse phase HPLC analysis. The nal step of this synthetic
pathway involves the Rosenmund–von Braun cyanation of the
dibromo-compound 1, yielding the desired product (CB11CB)
in addition to the monocyanated derivative (2) (Scheme 1).
All commercial materials employed in the synthetic proce-
dures were used as received, without further purication, with
the exception of solvents, which were dried by passage through
a column of activated alumina. NMR spectra were recorded on a
JEOL ECX spectrometer operating at 400 MHz (1H) or 100.5 MHz
(13C), with the residual protic solvent used as the internal stan-
dard. Mass spectra were recorded on a Bruker micrOTOF MS-
Agilent series 1200LC spectrometer. FTIR spectroscopy was
performed using a Shimadzu IR Prestige-21 with Specac Golden
Gate diamond ATR IR insert. High-performance liquid chro-
13C NMR (100.5 MHz, CDCl3): 29.30, 29.34, 29.47, 29.53,
29.60, 31.40, 31.45, 35.58, 35.61, 110.47, 119.04, 121.14, 126.73,
127.03, 127.45, 128.52, 128.93, 129.16, 131.77, 132.54, 136.41,
137.25, 140.02, 142.54, 143.78, 145.59 ppm. IR: 516, 555, 632,
702, 725, 765, 810, 964, 1002, 1080, 1112, 1180, 1296, 1357,
1481, 1604, 2222, 2846, 2916, 3032 cmꢃ1. MS-ESI+ (m/z):
586.2080 [C36H38BrNNa]+ assay (HPLC): 99.45%.
CB11CB
1H NMR (400 MHz, CDCl3): 1.14–1.33 (14H, m, –CH2–(CH2)7–
CH2–), 1.51–1.61 (4H, quintet, J ¼ 7.6, Ar–CH2–CH2–(CH2)7–
CH2–CH2–Ar), 2.58 (4H, t, J ¼ 7.6 Ar–CH2–(CH2)9–CH2–Ar), 7.22
(4H, ddd, J ¼ 1.8, J ¼ 2.1, J ¼ 8.2, Ar), 7.44 (4H, ddd, J ¼ 1.8, J ¼
2.1, J ¼ 8.2, Ar), 7.59 (4H, ddd, J ¼ 1.2, J ¼ 2.1, J ¼ 8.5, Ar), 7.64
(4H, ddd, J ¼ 1.2, J ¼ 2.1, J ¼ 8.5, Ar) ppm.
matography was performed on
a Shimadzu Prominence
13C NMR (100.5 MHz, CDCl3): 29.27, 29.44, 29.51, 29.58,
31.36, 35.58, 110.44, 119.00, 127.40, 127.00, 129.12, 132.50,
143.73, 145.30 ppm. IR: 516, 555, 640, 725, 771, 810, 848, 964,
1002, 1080, 1118, 1180, 1219, 1288, 1365, 1465, 1489, 1664,
1681, 1743, 1913, 2222, 2846, 2916, 3032 cmꢃ1. MS-ESI+ (m/z):
533.2911 [C37H38N2Na]+ Assay (HPLC): 99.84%.
modular HPLC system comprising a LC-20A liquid chromato-
graph, a DGU-20A5 and DGU-20B degasser, a SIL-20A auto-
sampler, a CBM-20A communication bus, a CTO-20A column
oven, and a SPO-20A dual wavelength UV-vis detector, oper-
ating, in this instance, at 230/255 nm. The column used was an
Alltech C18 bonded reverse-phase silica column with a 5 mm
pore size, an internal diameter of 10 mm and a length of
250 mm, with acetonitrile used as the mobile phase.
Physical studies
Polarized optical microscopy was performed on a Zeiss Axi-
oskop 40Pol microscope using a Mettler FP82HT hotstage
controlled by a Mettler FP90 central processor. Photomicro-
graphs were captured with aid of an InnityX-21 MP digital
Synthesis of 1-(100-bromobiphenyl-4-yl)-11-(100-cyanobiphenyl-
4-yl)undecane, 2 (CB11BBr), and 1,11-di-(100-cyanobiphenyl-4-
yl)undecane (CB11CB)
1,11-Di-(100-bromobiphenyl-4-yl)undecane, compound 1 (1 g, camera. Differential scanning calorimetry was performed on a
1.618 mmol) and copper cyanide (435 mg, 4.854 mmol) were Mettler DSC822e tted with an autosampler operating with
dissolved in anhydrous N-methyl-2-pyrollidone (45 ml) and Mettler Stare soware and calibrated using an indium standard
heated, with stirring and under an atmosphere of dry nitrogen, (onset ¼ 156.55 ꢄ 0.2 C, DH ¼ 28.45 ꢄ 0.40 J gꢃ1) under an
ꢁ
to 190 ꢁC for 3 h. The solution was cooled to room temperature, atmosphere of dry nitrogen.
and iron chloride (1.62 g, 10 mmol) in aqueous HCl (50 ml)
In order to assess the behaviour of the materials under
added. Water (200 ml) was added, and the resultant suspension applied electric elds, cells were placed within a Metter FP82HT
was washed with DCM (5 ꢂ 25 ml). The combined organic hotstage, which was controlled by a Mettler FP90 temperature
extracts were washed with water (3 ꢂ 100 ml), dried over controller. The hotstage was mounted on a Zeiss Universal
magnesium sulphate and the solvent removed in vacuo to give polarizing microscope, tted with an 8/0.2 objective and ꢂ10
an off white solid. TLC analysis showed three major compo- wideeld eyepieces. The waveform was generated by a Hewlett
nents, these were separated via gradient elution ash chroma- Packard 33120 A arbitrary waveform generator and amplied by
tography (2 : 1 hexane–DCM / 1 : 2 hexane–DCM) and found a linear ꢂ20 amplier that was custom built by QinetiQ. The
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 556–566 | 557

View Article Online
Journal of Materials Chemistry C
Paper
Table 1 Transition temperatures (ꢁC) for materials 1, 2 and CB11CB
transition temperatures obtained by thermal, polarized-light
microscopy. The comparative cooling cycles in Fig. 2(a) show
that the Iso. Liq. to N and N to NTB transitions, although broad,
are rst order in nature for compound 2, whereas for CB11CB
the NTB to N transition appears very broad, with second order
character and an unusual shape. Moreover, the enthalpy and
the entropy for the N to NTB transition of CB11CB are much
smaller than those for compound 2, whereas the reverse is the
case for the clearing points. Thus, the most interesting features
of this comparative study are; (i) the similarities between the
transition enthalpies and entropies for the clearing point and
the N to NTB transitions for compound 2, and the large differ-
ences for the same transitions for CB11CB, and (ii) the apparent
second order transition for the N to NTB phase change for
CB11CB.
No.
X
Y
Transition temperatures/ꢁC
1
Br
Br
Cr 122.5 Iso. Liq.
2
CN
CN
Br
CN
Cr 100.2 (NTB 85.7 N 96.5) Iso. Liq.
Cr 99.9 NTB 108.6 N 125.5 Iso. Liq.
CB11CB
Table 2 The enthalpies (kJ molꢃ1) and entropies associated with each
transition for materials 1, 2 and CB11CB
DH
DS/R
Cr–N
In order to gauge the nature for the N to NTB transition for
CB11CB, the scan rate was varied from 0.5 to 10 C minꢃ1, as
Compd
Cr–N
NTB–N
N–I
N
TB–N
N–I
ꢁ
shown in Fig. 2(b). The N to NTB phase transition appears to be
rst order at low scan rates, and seemingly second order for
scan rates above 5 ꢁC minꢃ1. This behaviour is progressive and
unusual because the peak shape would be expected to sharpen
for a rst order phase transition as the scan rate is increased.
Furthermore, if the peak shape for the lowest scan rate is
examined it appears that there might be two events occurring at
the phase transition, i.e. at a scan rate of 0.5 ꢁC minꢃ1 there is a
possible shoulder on the peak corresponding to the N–NTB
phase transition, see Fig. 2(b). Additionally, there appears to be
a slight increase in the transition temperatures as the scan rate
1
2
39.86
44.84
28.87
—
0.85
0.0351
—
1.17
1.54
12.12
14.44
9.31
—
0.284
0.011
—
0.387
0.465
CB11CB
electrical response from the cell was rescaled by a nano-current
amplier (20 kU or 100 kU impedance) and fed into a Hewlett
Packard 54600B oscilloscope, which was connected to a PC. The
antiparallel buffed polyimide cells used (Linkam) had spacings
of approx. 5 ꢄ 0.1 mm, and were constructed from ITO-coated
glass. In all cases, the cells were lled by capillary action at
atmospheric pressure with the sample in the isotropic liquid.
is increased. For the high scan rate (10 C minꢃ1) the N to NTB
ꢁ
transition of CB11CB appears almost second order, and is
reminiscent of a glass transition. Interestingly, the transition
shows reversibility, and mirroring of peak shape on heating
Results and discussion
The nematic to isotropic liquid and the nematic to the twist-
bend nematic transition temperatures were determined by
thermal polarized light microscopy, whereas the melting points
were determined by differential scanning calorimetry. The
entropies and enthalpies of all of the transitions were deter-
mined by differential scanning calorimetry. The values for these
processes are given in Tables 1 and 2 for compounds 1, 2 and
CB11CB.
The rst point to note from the data in the tables is that only
the di-nitrile material CB11CB exhibits enantiotropic liquid
crystal phases, compound 2 exhibits a monotropic nematic and
NTB phase transition on cooling. Compound 1 is non-meso-
genic, melting directly into the isotropic liquid. This can be
ascribed to the fact that for CB11CB the cyanobiphenyl units
will tend to overlap, whereas for the bromobiphenyl compound
the units do not, and as a consequence the clearing point is
highest for CB11CB and potentially virtual for compound 1, i.e.
below the melting and recrystallisation temperatures.
Differential scanning calorimetry
The differential thermograms for the cooling cycles for the
mesomorphic compounds 2 and CB11CB are shown together in
Fig. 2(a), and with various cooling rates for the N to NTB tran-
Fig. 2 (a) DSC thermograms of compound 2 (top) and CB11CB
(bottom) at a cooling rate of 10 ꢁC minꢃ1 and; (b) DSC thermograms for
sition of CB11CB in Fig. 2(b). The DSC results conrm the the N–NTB transition of CB11CB at various cooling rates.
558 | J. Mater. Chem. C, 2014, 2, 556–566
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
with respect to the thermograms shown on cooling. This the NTB phase. Overall, the sequence of textures shown in Fig. 3
comparative behaviour for the various scan rates suggests, in indicates that the higher temperature nematic phase appears
addition to thermodynamic behaviour, kinetic processes are normal, whereas the lower temperature NTB phase does not
occurring at the phase transition. Such processes could be exhibit any textures that are naturally associated with the
associated with the speed for the formation of complex bulk nematic phase. Thus, in addition to the short-range order ever
organisational structures. Hence, the shoulder on the peak present in the nematic phase, there is an additional long-range
corresponding to the N–NTB phase transition observed at low ordering within the NTB phase, and furthermore, this ordering
scan rates could be a consequence of a kinetic aspect of the continues to grow as the temperature is reduced, yielding
N–NTB phase transition.
further changes in the optical texture of the sample.
In addition to exhibiting paramorphotic, quasi-schlieren
nematic textures, the NTB phase also exhibits focal-conic
defects. Fig. 4(a) shows the focal-conic defect texture of the NTB
phase where both elliptical and hyperbolic lines of optical
discontinuity are seen as crosses. The ellipses and hyperbolae
shown in the gure have a dened mathematical confocal
relationship, where the confocal relationship results from
periodic ordering associated with the structure of the meso-
phase through the formation of “Dupin Cyclides”. This rela-
tionship indicates that the phase, unlike the conventional
nematic phase, has long range periodic ordering. In conven-
tional liquid crystal phases this periodic order may be due to
layering, a helical macrostructure, or periodicity associated with
lattices of defects.
Another texture that is associated with periodic ordering is
the parabolic defect texture. This pattern is commonly exhibited
by smectic phases, but has not been reportedly observed for
chiral nematic phases or phases that possess lattices of defects,
for example Blue phases. Fig. 4(b) shows the parabolic defect
texture found at the edge of a coverslip and slide for the NTB
phase. In this defect two parabolas (see white arrows in gure)
are positioned at right-angles to one another in a confocal
arrangement, and thus two lines of optical discontinuity are
observed, one a parabola, and the other a straight line which is
the edge of the other parabola.24 While in a smectic phase the
parabolas are constructed through periodic ordering of layers, it
is also plausible that they could arise from a lattice of defects or
a helical macrostructure, such as that proposed for the NTB
phase. This defect texture complements those described previ-
ously above for focal-conic defects, and conrms that the NTB
phase possesses long range periodic ordering.
Polarized optical microscopy
In the following, the defect textures obtained for a specimen of
CB11CB contained within a 9 mm cell treated to give homeo-
tropic alignment are discussed. On cooling from the isotropic
liquid to the nematic phase of CB11CB a schlieren texture is
formed rst, with no evidence of a homeotropic texture being
formed. The schlieren texture of the nematic phase exhibits two
and four brush defects clearly shown at a temperature of 107 ꢁC
in Fig. 3(a). Cooling of the phase into the NTB phase progresses
via a change in the correlation length from short to long range,
as shown in Fig. 3(b) (106.8 ꢁC), for the phase transition at 106.5
ꢁC (Fig. 3(c)). The correlation length is clearly shown by the
edges between the light and dark textures, and in the steps
associated with fractal-like patterns in the schlieren brushes.
Further cooling into the temperature range of the NTB phase
(106.2 ꢁC) at a rate of 0.1 ꢁC minꢃ1 shows that the schlieren
texture has broken up into light and dark domains that are
separated by grain boundaries, i.e., the NTB phase does not
exhibit a schlieren texture because the director eld is not
continuous as in an elastic uid but rather is discontinuous as
in a so crystal. A further reduction in temperature results in a
change in birefringence and further breaking up of the textures
into smaller domains; see textures at 105.0 and 100.7 ꢁC in
Fig. 3(e) and (f) respectively. It should also be noted that no
change in texture was observed upon annealing the sample in
Mechanical shearing of the NTB phase of CB11CB shows no
formation of schlieren defects, or Brownian motion typical of a
nematic phase. Instead the sheared texture that forms is similar
to that obtained when a smectic A phase is sheared, as shown in
Fig. 5. The centre of the specimen (top le of Fig. 5) exhibits a
Fig. 3 Polarized optical micrographs showing the textures observed
for (a) the nematic phase at 107 ꢁC; (b) the nematic phase at 106.8 ꢁC;
(c) the nematic to NTB transition at 106.5 ꢁC; (d) the NTB phase at
106.2 ꢁC; (e) the NTB phase at 105 ꢁC and; (f) the NTB phase at 100.7 ꢁC.
The textures were obtained at a cooling rate of 0.1 ꢁC minꢃ1 with the
specimen contained in a 9 mm cell treated to give homeotropic Fig. 4 Photomicrographs (ꢂ100) showing (a) focal-conic texture and
alignment.
(b) parabolic defects of the NTB phase of CB11CB.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 556–566 | 559

View Article Online
Journal of Materials Chemistry C
Paper
Initial drawing of the lm in the NTB phase gave rise to a
texture in which it was possible to see defects that resemble focal-
conic domains, as shown in Fig. 7(a). On heating the lm these
defects began to reorganise, or the molecules began to reorient
within the lm, to give rise to a texture exhibiting long-range
correlations in the plane of the lm; see the lower portion of
Fig. 7(b), and the lm thinned. Once the temperature approached
the NTB to N phase transition (i.e., the value determined by DSC) a
large degree of turbulence was observed, the material became
much less birefringent and formed a patchwork schlieren-like
texture, as shown in Fig. 7(c). Further heating of less than a
degree produced a schlieren texture that was remarkably similar
to that of a smectic C phase, see Fig. 7(d) and (e). The lm was
subsequently found to be only stable for a few degrees above the
NTB to N phase transition, whereupon it ruptured, i.e., when it
entered into a normal nematic phase. However, cooling the lm
before it ruptured lead again to turbulence and large uctuations
in the lm. Finally, instead of readopting the focal-conic texture
of the virginal lm, cooling gave rise to paramorphosis of the
schlieren texture with long-range correlations occurring in the
plane of the preparation as shown in Fig. 7(f). These correlations
were similar to those observed for the homeotropically, aligned
specimen in 9 mm cell, shown in Fig. 3.
Thus, studies of free-standing lms of the NTB phase are
unequivocal. A lm cannot be formed in a normal nematic
phase, nor to our knowledge by a chiral nematic phase, or in any
phase in which periodicity is not supported by a layered struc-
ture. For example, when the NTB phase is heated into the
nematic phase the lm ruptures. Thus, our ability to create free-
standing lms in the NTB phase points to the phase not being
nematic-like. In the NTB there is much motion similar that
experienced for TGB and ferrielectric phases. This motion
indicates that there could be two processes ongoing; one asso-
ciated with the surface of the lm and the other with the bulk.
At lower temperatures in the NTB phase the turbulence lessens
and long-range correlations build up that are in the plane of the
lm, and which do not cause rupture; these are properties
typically associated with a so crystal-like phase.
Fig. 5 Optical texture of the NTB phase of CB11CB after mechanical
shearing (x100).
homeotropic texture, whereas at the edges of the coverslip an
oily-streak texture possessing concentric tubes is obtained
similar to that found for lamellar lyophases and chiral
nematics. However, the viscosity of the NTB phase was also
found to be relatively high; with no restorative ow on
displacement of the coverslip being further indicative of the
long-range order within the phase. These combined observa-
tions are atypical of conventional nematic or chiral nematic
phases, but more in keeping with smectic phases.
Free standing lm
One of the requirements for the formation of a free standing
lm is that the phase needs to be smectic in nature where the
layered structure helps to create the surface tension that aids in
the suspension of the lm across an aperture. One of the big
challenges with the CBnCB series of compounds is that the NTB
phase is usually monotropic, which means that crystallization is
prone to occur during the process of drawing a lm. However,
due to the enantiotropic nature of the NTB phase of CB11CB,
lm drawing produced a number of examples of freely sus-
pended lms of various thicknesses. The bulk appearance of a
lm is shown in Fig. 6. Once a lm was formed it became
obvious that the transition between the NTB and N phases was
complex because the lm becomes extremely turbulent
(see Fig. 6, right, and Video in the ESI†).
Fig. 7 Photomicrographs (ꢂ100) of the free standing film of CB11CB
pulled in the NTB phase. (a) Virginal texture of the NTB phase of the
Fig. 6 A free standing film of CB11CB pulled over a 1.5 mm circular drawn film; (b) reorganisation of the film as it is heated towards the
aperture in an aluminium plate, and a photo of the turbulence TB–N phase transition; (c) reorganization of the film to give a schlieren
N
occurring in the film at the transition between the NTB phase and the N texture on heating; (d and e) the NTB phase just below the transition to
phase. The exposure time of the photograph is approximately the nematic phase and; (f) the paramorphosis of the NTB texture formed
500 milliseconds.
on cooling from near to the phase transition to the nematic phase.
560 | J. Mater. Chem. C, 2014, 2, 556–566
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
Table 3 Transition temperature (ꢁC) of mixtures of CB11CB with chiral
dopant BE8OF2N
Fig. 9 Photomicrographs (ꢂ100) showing; (a) the chiral nematic
phase viewed through crossed polars; (b) analyzer rotated 40ꢁ to the
left; (c) analyzer rotated 40ꢁ to the right.
Wt%
BE8OF2N
Cr
NTB
N
Iso
0
6
16
ꢅ
ꢅ
ꢅ
99.9
100.0
99.9
ꢅ
108.6
100.2
82.0
ꢅ
ꢅ
ꢅ
125.5
116.5
97.6)
ꢅ
ꢅ
ꢅ
ꢅ
(ꢅ
Studies with the addition of a chiral dopant
Compound CB11CB was doped with the chiral material
BE8OF2N at concentrations of 6 and 16 wt%. Interestingly, the
Fig. 10 Photomicrograph (ꢂ100) of the homeotropic texture with oily
streaks, (a) and the planar texture upon reheating of NTB phase from a
sheared texture, (b).
N
TB phase remained in both mixtures, even though the isotropic
liquid to nematic and nematic to NTB phase transitions were
depressed at high dopant concentration such that the phases
became monotropic (see Table 3).
Upon cooling, typical textures were observed for the chiral
nematic phase of both of the mixtures with the phase devel-
oping from the isotropic liquid as striated droplets that evolved
into a Grandjean planar texture (Fig. 8(a)). On further cooling,
the NTB phase developed from the chiral nematic phase to give
an optical texture that displayed long-range correlations, see
Fig. 8(b). The texture developed parabolic defects at lower
temperatures (Fig. 8(c)). Upon reheating the nematic phase
reappeared via a tight ngerprint texture (see Fig. 8(d)). Thus,
the chiral dopant is not being excluded from the NTB phase as
no segregation and re-adsorption was observed (Fig. 9).
In order to sense the chirality of the mesophases the analyzer
was rotated to observe colour dispersion associated with a
matching helical twist. Fig. 11(a–c) shows the Grandjean planar
texture of the chiral nematic phase viewed through crossed-
polars (a), upon 40ꢁ rotation of the analyzer to the le (b), and
40ꢁ rotation of the analyzer to the right (c). These textures clearly
demonstrate that the chiral nematic phase has a le-handed
helix associated with the stereochemical structure of the
dopant. No difference in the texture could be detected with
respect to the direction of the optical rotation for the NTB phase
of either of the mixtures. This indicates that a macroscopic helix
is not present in the doped mixtures. However, as a nanoscale
helix would impart only a very slight optical rotation per
micron, the detection of such a structure is impractical using
polarised optical microscopy. Suppression of macroscopic hel-
icity in lamellar phases has been previously shown to occur at
the change over from smectics to so crystals, e.g. at a heli-
electric smectic C* phase to a crystal J* as reported by Brand
and Cladis.25
Shearing of the NTB phase of CB11CB doped with BE8OF2N
gave a homeotropic texture with bright oily streaks, as shown in
Fig. 10(a). As with unsheared samples, no macroscopic helix
could be observed via rotation of the analyser. However, when
the sheared texture was reheated to the NTB to N* phase tran-
sition, reorientation of the sample was observed just before the
phase transition to give a broken streaky planar texture just
prior to the formation of the Grandjean texture of the chiral
nematic phase, see Fig. 10(b). At this point the colour dispersion
associated with a helical macrostructure returns. Thus, the NTB
phase does not appear to exhibit a Grandjean planar texture
even when doped with a chiral material. Thus, we conclude that
the locked-in structure of the NTB phase is strong enough to
suppress the induction of a helical macrostructure.
Fig. 8 Photomicrographs (ꢂ100) showing; (a) chiral nematic planar
texture; (b) chiral nematic to NTB phase transition; (c) NTB phase
showing parabolic defects; (d) NTB to nematic phase transition
observed on heating.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 556–566 | 561

View Article Online
Journal of Materials Chemistry C
Paper
Studies in aligned cells and electrical
field studies
Compound CB11CB was ow-lled into both homogeneous and
homeotropic cells by capillary action while the sample was in
the isotropic liquid. The homogeneous cells used were 5 mm
antiparallel buffed polyimide coated cells purchased from
Linkam and the homeotropic cells were 9 mm cells purchased
from Displaytech. The homogeneous cells gave a well-aligned
nematic phase, but upon cooling into the NTB phase a striped
texture evolved where the stripes appeared along the rubbing
direction of the cell, see Fig. 11(a–d). The stripes in this texture
resemble chevron defects that are normally associated with
chiral smectic C phases, and the texture becomes increasingly
grainy upon further cooling with the formation of defects that
resemble parabolic deformations. Clearly there is a preferred
orientation of the liquid crystal within the cell as demonstrated
by dark and bright states achieved by rotation of the sample
between crossed polars. Conversely, the sample did not align
well in the homeotropic cell, and consequently a schlieren
texture was observed for the nematic phase, suggesting that
there is random orientation of the director within the cell. The
NTB phase appears in the cell as a patchwork paramorphotic
texture possessing long-range correlations, and upon cooling
the texture becomes increasingly grainy and broken, see
Fig. 4(a–c).
Fig. 12 Photomicrograph (ꢂ100) of William's domains observed in the
nematic phase of CB11CB confined in a 5 mm homogeneous aligned
cell with a square waveform 20 V mmꢃ1 at 1 kHz, (a); and with a square
waveform 20 V mmꢃ1 at 540 Hz, (b).
temperatures a higher voltage and frequency was required to
observe the domains than at lower temperatures. The William's
domains appeared as a series of diagonal stripes, which were
oriented at an angle of approximately 60ꢁ or 120ꢁ relative to
each other (depending on which side of the domain is consid-
ered), but both lie at an angle away from the rubbing direction
of the cell (see Fig. 12(a)). The domains were well dened at
higher voltage and frequency and increasingly became more
dynamic as the frequency was reduced (Fig. 12(b)). On cooling
into the NTB phase it was not possible to observe any form of
switching process. Indeed, on application of a large voltage
(20 V mmꢃ1) the liquid crystal reoriented irreversibly from a
homogeneous to homeotropic arrangement within the cell, in
keeping with previous studies.14 This suggests reorganization in
the phase structure of the material in a similar way to layer
reorientation in smectic phases.
The cell was revisited aer a short period of time and re-
examined in the NTB phase using the experimental conditions
reported by Panov et al.14 In this article a small linear optical
response was reported in the NTB phase at voltages close to 20 V
mmꢃ1 in a 5 mm cell under the application of a sawtooth wave-
form. The initial response observed was consistent with the
virginal study whereby reorientation to give a homeotropic
arrangement occurred. Once aligned homeotropically, the
Under the application of an applied electrical eld, the
nematic phase of CB11CB in the homogeneous aligned cell
displayed typical characteristics of a low molecular weight
material. Depending on the voltage applied to the cell it was
possible to observe a Fredericks transition at low frequency and
at higher frequencies it was also possible to observe William's
domains.26 The appearance of the William's domains was
dependent on the voltage and frequency of the applied eld, as
well as the temperature of the nematic phase. At higher
Fig. 11 Photomicrographs (ꢂ100) of the NTB phase at 90 ꢁC in a Fig. 13 Oscilloscope traces showing the applied waveforms and the
homogeneous cell showing the striped defects occurring along the electrical responses of the NTB phase at 100 ꢁC. The applied wave-
rubbing direction of the cell after cooling from the nematic phase into forms examined were sinusoidal (a), square (b), triangular (c) and
the NTB phase at 1 ꢁC minꢃ1 (a and b) and at 0.2 ꢁC.
sawtooth (d).
562 | J. Mater. Chem. C, 2014, 2, 556–566
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
Fig. 15 The energy minimised architecture of CB11CB at absolute
zero in the gas phase computed at the B3LYP/3-21G level of theory.
Fig. 14 Oscilloscope trace showing the space charge observed in a 5
mm cell containing CB11CB in the NTB phase with the application of a
units from the biphenyl cores are likely to be in a trans confor-
mation and this will have the effect of stabilising the bent
molecular shape. However, linear and hairpin structures should
not be overlooked, although probably the most important
feature of the architecture, in relation to the formation of the
NTB phase, is the bent shape in the all-trans ground state.
Indeed, in unpublished work we have prepared over 80 materials
that exhibit the NTB phase, see Fig. 16, and the results show that
the incorporation of polar groups, the magnitude and direction
of the dipole moment, and alkoxy versus methylene spacer units
are not important in the formation of the NTB phase.27 Over-
whelmingly our results show that it is the molecular shape,
rather than electronic properties, which is important for the
occurance of the NTB phase. A similar observation is demon-
strated by compound 2, CB11BBr, which still exhibits an NTB
phase even though one of the nitrile units has been replaced by a
bromide substituent. Furthermore, the dimers CB11CB and
compound 2 could be considered to be supermolecular mate-
rials, for which the molecular topology is the dominating feature
in mesophase formation, i.e., self-assembly processes may be
involved in the self-organization of the NTB phase.28,29
Using the bent molecular architecture as a template it is
possible to examine how the molecules might pack in two-
dimensional space. For cyanobiphenyl-terminated systems it is
known that antiparallel overlapping associations of the cyano-
biphenyl units form the basis of mesophase stability. If this
interaction is used to investigate mesophase organization only
two packing arrangements are possible in 2D; one where the
molecules form a quasi-layered structure via intercalation and
an anticlinic arrangement of the biphenyl cores, as shown in
Fig. 17(a). The alternative is a synclinic arrangement of the
molecules, in which case a circular organisation of the mole-
cules results, as shown in Fig. 17(b). Both of these arrangements
are unacceptable for condensed mesophase formation because
they possess substantial void space. If we now look at the
packing of the molecules in 3D the problems of the ensuing
void volumes can be overcome. In the anticlinic structure the
sawtooth waveform at 20 V mmꢃ1
.
sample was examined using a sinusoidal, square, triangular and
sawtooth waveform (representative examples of each are shown
in Fig. 13 along with the electrical response from the liquid
crystal). In all cases there was no switching observed at low
voltage across a range of frequencies from mHz up to MHz. An
increase in the voltage did not yield any switching behaviour
with a square or sinusoidal waveform but a small effect was
observed with the triangular and sawtooth waveform at
20 V mmꢃ1 and 200 mHz although this was also accompanied by
the observation of a large space charge associated with an ionic
response in the cell, see Fig. 14. With each increase in applied
voltage it was possible to observe a small change located around
air bubbles in the cell, the frequency of this effect correspond-
ing to the electrical detection of space charge. Hence, we
interpret this response to be ionic in its origin. Aer continued
exposure of the cell to high voltages, it was apparent that the
material had begun to degrade, suggesting that the response is
due to ionic breakdown.
Similarly there is no electrooptic response in the NTB phase
doped with the chiral material. Furthermore, there is no
observation of any dielectric or exoelectric response that
would be expected from a nematic phase, or alternatively by a
chiral lamellar-type of phase such as a ferroelectric, antiferro-
electric or ferrielectric phase, indicating the NTB phase has too
high a viscosity or is too highly ordered for the applied electric
eld to overcome. The only apparent electrooptic responses that
have been achieved have been where there is the potential for
injection of space charge or rearrangement of the mesophase
structure.
Molecular modelling
In order to evaluate the potential possibilities of how the inclusion of gauche conformers in the methylene chain would
molecules might pack together the structure of CB11CB was shorten the overall molecular length, and at the same time will
modelled in the gas phase at absolute zero computed at the ll the void volume. This could be achieved by spiralling the
B3LYP/3-21G level of theory. The energy minimised molecular packing of the molecules as proposed by Chen et. al.21 It should
structure of CB11CB is shown in Fig. 15. The bend angle be also be noted that Memmer observed the NTB phase in
between the two arms of the material was determined to be simulations consistent with the model proposed by Chen et.
111ꢁ, which is similar to the value obtained via molecular al.30 This arrangement is then similar to the “clock model”
modelling for CB7CB,12 the dipole moment dipole moment was proposed for the structures of the ferrielectric phases where
found to be 5.521 D, and its anisotropy of polarizability, Da, there is a three layer molecular twist.31 For the circular
3
˚
equal to 31.8 A . In the condensed state, the rst few methylene arrangement of the molecules, the packing of bent molecules
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 556–566 | 563

View Article Online
Journal of Materials Chemistry C
Paper
Fig. 16 Families of materials that exhibit the NTB phase.27
Structural studies on NTB phase conclude that the phase is
not layered, and that the periodicity seen in FFTEM images is
˚
due to a helical superstructure a pitch of ꢀ80 A made up of ꢀ3
molecules per twist.12,21 A simple, but obvious observation is
that the NTB phase is an intercalated phase where the dimers
overlap with one another, to form a twist where the overlapping
does not produce a layered bulk structure, and hence the phase
is nematic-like. As the biphenyl arms of the mesogens are tilted
at an angle to the heliaxis the phase has been termed a heli-
conical nematic phase. However, for a model with a twist
composed of approximately three molecules along the heliaxis
one might expect when rotation of the heliaxes of subsequent
Fig. 17 Packing of molecules in a 2D plane where the packing alter-
nates from one molecule to the next (a), and where the molecules
pack in the same orientation (b).
cannot be sustained, and so the curvature of the packing is
expelled into the third dimension, thereby forming a twisted
structure, as proposed by Memmer et. al.30 Such a cylindrical
arrangement is similar to that of a double twist cylinder found
in Blue phases. Interestingly, these two extremes of packing
molecules together to ll the void volume are closely related. For
example, in order to ll the void volume at the centre of the
cylinder the molecules must become oriented so that their long
axes are almost parallel to the cylinder axis forming a twisted
disclination, whereas at the external surface of the cylinder the
long axes of the molecules will be perpendicular to the axis, as
in a double twist cylinder, see Fig. 18. Another way to express
this concept is that the bend is expelled as twist into the third
dimension, which will be related to the magnitude and rela-
tionship of the elastic coefficients k33 and k22.
Fig. 18 The twist associated with the 3D packing of the molecules.
564 | J. Mater. Chem. C, 2014, 2, 556–566
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
layers of heliconical twists that take the structure into three large increases in viscosity at the phase transition. In one
dimensions, i.e., the heliconical structure would be a sub-unit example a nematic glass has been reported that was formed in a
of a supramolecular organization which would have its heliaxis TN device for which the optical guiding of light had been
perpendicular to those of the heliconical axes.
Heliconical nematic phases have already been described by structural integrity as the nematic phase itself.36
Meyer,32 and Patel and Meyer33 demonstrated exoelectricity
However, glassication is not the only possible explaination
retained, i.e., the nematic glass was found to have the same
when an electric eld applied to a chiral nematic phase could for the high viscosity of the NTB phase. The intercalation of the
induce the molecules to tilt away from the heliaxis, thereby molecules will have the effect of forming supramolecular
forming a new heliaxis tilted at an angle to the original one. ribbons in a similar way to how polymers and biological mate-
However, in our case no electrooptic response was obtained for rials form spiraling strands or bers, see Fig. 19. This has also
the doped NTB phase. An electrical response in the doped been considered by Chen et. al.21 for single helices, but complex
nematic phase may be because the pitch is in the micron intertwined spiraling structures, as in collagen, have not been
regime, but for the NTB phase the lack of a response maybe due considered. These types of structural constructs, however, are
to the nanoscale structuring. Interestingly, for chiral biological not the same as in a heliconical organization that has been
structures such as the hardened outer wing covers of scarab developed to reect the continuous twisting structure of a chiral
beetles, Bouligand34 showed that cuts through the short-pitch nematic phase. Nevertheless, the intercalating of the bent
helical superstructure could produce bend-splay director elds. molecules along the heli-axis producing a brous phase would
There has been a recent report suggesting this is also the case also substantially raise viscosity, and reduce responses to
with the NTB phase,35 for which the bend-splay mode would applied external elds. It is also known that the self-assembly of
produce layering. However, the structure of the “chiral nematic molecules can form the basis for phases with complex super-
phases” of scarab beetles is frozen-in, and the structure is in fact structures that occur between conventional phases in a ther-
quasi-crystalline. This could also be the case for the NTB phase modynamic ordering sequence of phase transitions, for
of CB11CB.
example the cubic D phase and the S4 columnar phase can be
If we now consider the t of the experimental results to a found to exist between the smectic A and smectic C phases of
so-crystal/glassy model with
following observations can be made:
a
locked-in structure, the nitro- and cyano-substituted biphenyl carboxylic acids. Thus, it
would be possible for the NTB phase to be sandwiched between a
ꢅ The X-ray diffraction would show a pattern for a nematic typical nematic and smectic phases in a phase sequence. Indeed
not a lamellar phase.
this has been found to be the case in our more extensive study of
ꢅ The phase would look like a so smectic crystal phase in materials that exhibit NTB phases.27
the microscope.
Through the extensive physical property examination we
ꢅ The phase would exhibit long range correlations.
ꢅ The textures would show grain boundaries.
ꢅ No switching would be necessarily expected in applied
electrical elds.
have given to the NTB phase of CB11CB we can conclude the
following:
1. The physical properties of the phase are not similar to
those a nematic phase, and appear more in keeping with those
ꢅ Large voltages would be expected break up the structure or of a so crystal phase.
cause dielectric breakdown.
2. If the phase is glassy, its structure may be similar to a
ꢅ No induced helicity would be seen with the addition of nematic phase in terms of anisotropy and molecular
chiral dopants.
organization.
ꢅ No switching in the chiral doped systems would be found,
3. The NTB phase has been reported to be a conglomerate,
possessing two mirror image helical forms.21 By implication,
unless structural break-up occurs.
ꢅ The phase would be very viscous, and show only viscous ow. this observation also suggests the phase has long range order,
ꢅ No Brownian motion would be observed. and could be crystalline in nature, however, this cannot be in
ꢅ Kinetic processes would also be observed at the N to NTB terms of positional ordering of the constituent molecules.
transition.
4. In a locked-in form, the structural repeat is not molecular,
ꢅ There would be a time dependency with respect to reor- but related to helical pitch, and it doesn't really matter how the
ganization at the phase transition.
ꢅ The phase transition could look glassy for fast scan rates.
ꢅ It would be possible to draw a free-standing lm; this
would not be possible for a normal nematic phase.
All of these observations suggest that the NTB phase
possesses a structure that is at least to some degree xed or
locked-in, for example as in a so-crystal or glassy nematic-like
phase, and not that of a conventional nematic phase which is
responsive to external stimuli. If we examine the possibilities of
nematic glass formation by related materials such as the alkyl
cyanobiphenyls, and associated mixtures, we nd that many
Fig. 19 Arrangement of helical fibrous structure to give a glassy or soft
exhibit nematic glasses at low temperatures with accompanying crystal-like phase.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 556–566 | 565

View Article Online
Journal of Materials Chemistry C
Paper
´
molecules are arranged. In this sense the phase has a superstruc- 13 C. S. P. Tripathi, P. Losada-Perez, C. Glorieux, A. Kohlmeier,
ture and is similar to a Blue phase which has a lattice based on the
organization of defects rather than ordering of the molecules.
M. G. Tamba, G. H. Mehl and J. Leys, Phys. Rev., 2011, 84E,
041707.
5. In comparison, all forms of nematic phases respond to 14 V. P. Panov, R. Balachandran, M. Nagaraj, J. K. Vij,
applied electric elds, whereas the NTB phase of CB11CB only
exhibits a transient response due to space charge.
M. G. Tamba, A. Kohlmeier and G. H. Mehl, Appl. Phys.
Lett., 2011, 99, 261903.
6. With over 80 materials exhibiting the NTB phase, it is clear 15 L. Bequin, J. W. Emsley, M. Lelli, A. Lesage, G. R. Luckhurst,
that the phenomenon is common to materials with bent gross
structures, and therefore it is likely that NTB phases also will be
B. A. Timimi and H. Zimmermann, J. Phys. Chem. B, 2012,
116, 7940–7951.
found in conventional bent core systems. For example, similar 16 P. A. Henderson and C. T. Imrie, Liq. Cryst., 2011, 38, 1407–1414.
´
´
debates have been ongoing for oxadiazole systems that have 17 D. O. Lopez, N. Sebastian, M. R. de la Fuente, J. C. Martınez-
´
´
been reported to exhibit biaxial nematic phases. It would not be
surprising if some of the smectic phases of these materials are
in fact NTB phases.37
Garcıa, J. Salud, M. A. Perez-Jubindo, S. Diez-Berart,
D. A. Dunmur and G. R. Luckhurst, J. Chem. Phys., 2012,
137, 034502.
Finally we note that the phase has a novel structure in rela- 18 L. Beguin, J. W. Emsley, M. Lelli, A. Lesage, G. R. Luckhurst,
tion to liquid crystals, and may be this represents a paradigm
shi in our understanding of complex liquid crystal forms, and
therefore it deserves a new name, maybe a TB phase.
B. A. Timimi and H. Zimmermann, J. Phys. Chem. B, 2012,
116, 7940–7951.
19 I. Dozov, Europhys. Lett., 2001, 56, 247.
20 P. J. Barnes, A. G. Douglass, S. K. Heeks and G. R. Luckhurst,
Liq. Cryst., 1993, 13, 603–613.
21 D. Chen, J. H. Porda, J. B. Hooper, A. Klittnick, Y. Shen,
M. R. Tuchband, E. Korblova, D. Bedrov, D. M. Walba,
M. A. Glaser, J. E. Maclennan and N. A. Clark, Proc. Natl.
Acad. Sci. U. S. A., 2013, 110, 15931–15936.
Acknowledgements
The authors thank the Engineering and Physical Sciences
Research Council (EPSRC) for nancial support, Grant refer-
ence EP/J007714/1.
22 C. Greco, G. R. Luckhurst and A. Ferrarini, Phys. Chem.
Chem. Phys., 2013, 15, 14961–14965.
23 C. Meyer, G. R. Luckhurst and I. Dosov, Phys. Rev. Lett., 2013,
111, 067801.
24 C. S. Rosenblatt, R. Pindak, N. A. Clark and R. B. Meyer,
J. Phys. II, 1997, 38, 1105–1115.
25 H. R. Brand, P. E. Claddis and P. L. Flinn, Phys. Rev. A: At.,
Mol., Opt. Phys., 1985, 31, 361–367.
References
1 B. Musgrave, P. Lehmann and H. J. Coles, Liq. Cryst., 1999,
26, 1235–1249.
2 A. E. Blatch, M. J. Coles, B. Musgrave and H. J. Coles, Liq.
Cryst., 2003, 401, 161–169.
3 B. J. Broughton, M. J. Clarke, A. E. Blatch and H. J. Coles,
J. Appl. Phys., 2005, 98, 134109.
26 R. Williams, J. Chem. Phys., 1963, 39, 384–388.
4 A. Ferrarini, C. Greco and G. R. Luckhurst, J. Mater. Chem., 27 E. J. Davis, R. J. Mandle, C. T. Archbold, C. C. Voll, S. Lobato,
2007, 17, 1039–1042.
M. Davey, S. J. Cowling, J. W. Goodby, unpublished results.
5 T. Izumi, S. Kang, T. Niori, Y. Takanishi, H. Takezoe and 28 J. W. Goodby, I. M. Saez, S. J. Cowling, J. S. Gasowska,
J. Watanabe, Jpn. J. Appl. Phys., Part 1, 2006, 45, 1506–1514.
6 C. V. Yelamaggad, N. L. Bonde, A. S. Achalkumar,
D. S. S. Rao, S. K. Prasad and A. K. Prajapati, Chem. Mater.,
2007, 19, 2463–2472.
R. A. MacDonald, S. Sia, P. Watson, K. J. Toyne, M. Hird,
R. A. Lewis, S.-E. Lee and V. Vaschenko, Liq. Cryst., 2009,
36, 567–605.
¨
29 J. W. Goodby, I. M. Saez, S. J. Cowling, V. Gortz, M. Draper,
7 U. S. Hiremath, G. M. Sonar, D. S. S. Rao and
C. V. Yelamaggad, J. Mater. Chem., 2011, 21, 4064–4067.
A. W. Hall, S. Sia, G. Cosquer, S.-E. Lee and E. P. Raynes,
Angew. Chem., Int. Ed., 2008, 47, 2754–2787.
8 G. Shanker and C. V. Yelamaggad, New J. Chem., 2012, 36, 30 R. Memmer, Liq. Cryst., 2002, 29, 483–496.
918–926.
31 L. S. Matkin, H. F. Gleeson, P. Mach, C. C. Huang, R. Pindak,
9 R. W. Date, G. R. Luckhurst, M. Shuman and J. M. Seddon,
J. Phys. II, 1995, 5, 587–605.
G. Sraijer, G. Pollmann, J. W. Goodby, M. Hird and A. J. Seed,
Appl. Phys. Lett., 2000, 76, 1863–1865.
10 V. P. Panov, R. Balachandran, V. K. Vij, M. G. Tamba, 32 R. B. Meyer, Appl. Phys. Lett., 1968, 12, 281–282.
A. Kohlmeier and G. H. Mehl, Appl. Phys. Lett., 2013, 101, 33 J. S. Patel and R. B. Meyer, Phys. Rev. Lett., 1987, 58, 1538–1540.
234106.
34 Y. Bouligand, M.-O. Soyer and S. Puiseux-Dao, Chromosoma,
1968, 24, 251–287.
11 V. P. Panov, M. Nagaraj, J. K. Vij, Yu. P. Panarin,
´
A. Kohlmeier, M. G. Tamba, R. A. Lewis and G. H. Mehl, 35 V. Borshch, Y.-K. Kim, J. Xiang, M. Gao, A. Jakli, V. P. Panov,
Phys. Rev. Lett., 2010, 105, 167701.
12 M. Cestari, S. Diez-Berart, D. A. Dunmur, A. Ferrarini,
M. R. de la Fuente, D. J. B. Jackson, D. O. Lopez,
J. K. Vij, C. T. Imrie, M.-G. Tamba, G. H. Mehl and
O. D. Lavrentovich, 2013, preprint at arXiv:1309.1452
[cond-mat.so.
G. R. Luckhurst, M. A. Perez-Jubindo, R. M. Richardson, 36 J. O. Kessler and E. P. Raynes, Phys. Lett. A, 1974, 50, 335–336.
¨
J. Salud, B. A. Timimi and H. Zimmermann, Phys. Rev. E: 37 V. Gortz, C. D. Southern, N. Roberts, H. F. Gleeson and
Stat., Nonlinear, So Matter Phys., 2011, 84, 031704.
J. W. Goodby, So Matter, 2009, 5, 463–471.
566 | J. Mater. Chem. C, 2014, 2, 556–566
This journal is © The Royal Society of Chemistry 2014