Self-organisation through size-exclusion in soft materials

Article abstract of DOI:10.1039/c4tc02991g

A number of materials derived from 4-undecyloxy-4′-cyanobiphenyl but with varying terminal groups were prepared in order to better understand how such a group influences the type, and local structure of mesophases formed. Whereas electron poor terminal groups (fluoroaromatics and halogen atoms) were found to destabilise the smectic A phase through unfavourable electrostatic interactions, bulky silane, siloxane and hydrocarbon groups can be incorporated into the structure of the phase with only minor reductions in clearing point. An increase in the layer spacing of the smectic A_d phase in materials with bulky groups suggests that microphase segregation is not the driving force, but rather exists as a consequence of steric crowding at the smectic layer interface. Electrooptic studies reveal that 'carbosilane' end groups, such as tetramethyldisilapropane, are significantly more electrochemical stable than their siloxane counterparts whilst retaining their desirable thermal properties.

Full text of DOI:10.1039/c4tc02991g

Journal of
Materials Chemistry C
View Article Online
View Journal | View Issue
PAPER
Self-organisation through size-exclusion in soft
materials†
Cite this: J. Mater. Chem. C, 2015, 3,
2380
*
Richard J. Mandle, Edward J. Davis, Constantin-Christian A. Voll, Daniel J. Lewis,
Stephen J. Cowling and John W. Goodby
A number of materials derived from 4-undecyloxy-40-cyanobiphenyl but with varying terminal groups were
prepared in order to better understand how such a group influences the type, and local structure of
mesophases formed. Whereas electron poor terminal groups (fluoroaromatics and halogen atoms) were
found to destabilise the smectic A phase through unfavourable electrostatic interactions, bulky silane,
siloxane and hydrocarbon groups can be incorporated into the structure of the phase with only minor
reductions in clearing point. An increase in the layer spacing of the smectic Ad phase in materials with
bulky groups suggests that microphase segregation is not the driving force, but rather exists as a
consequence of steric crowding at the smectic layer interface. Electrooptic studies reveal that
‘carbosilane’ end groups, such as tetramethyldisilapropane, are significantly more electrochemical stable
than their siloxane counterparts whilst retaining their desirable thermal properties.
Received 30th December 2014
Accepted 23rd January 2015
DOI: 10.1039/c4tc02991g
www.rsc.org/MaterialsC
anisotropy (D3) of these siloxane-terminated materials is at least
an order of magnitude lower than for the parent alkyl materials,
Introduction
and there is no signicant change in order parameter, polar-
isability or dipole moment. The reduction of the dielectric
anisotropy is believed to be due to a large increase, or locking-
in, of the degree of antiparallel associations within the layers of
the smectic phase.8
In recent years there has been a resurgence of interest in elec-
trooptic devices using the smectic A phase,1–15 including dye
doped guest–host systems.4,11,16–18 The relatively low power
consumption, high contrast ratio, fast response time, bistable
switching, and potential outdoor use offered by smectic A
scattering devices makes them highly attractive for a range of
applications including tablet computers, e-readers, electronic
shelf labels (ESLs), and the built environment. Existing mate-
rials, such as 4-octyl-40-cyanobiphenyl (8CB), are far from ideal
due to both high voltages required for switching and limited
device lifetimes.19
Bulky terminal groups such as pentamethyldisiloxane and
pentamethyldisilapropane have been employed extensively in
liquid crystals to give control over directed self-assembly and
self-organization processes,20,21 and to give large reductions in
melting point at a cost of only modest reductions in iso-
tropization point, leading to wider temperature range meso-
phases.22–27 The prevailing hypothesis for this behaviour is
based on silicon containing groups, which are said to ‘micro-
phase segregate’ from aliphatic parts of the molecule due to a
perceived chemical immiscibility.28–30 Such incompatibility
drives the formation of discrete liquid-like layers and promotes
the formation of lower ordered smectic phases (i.e. A and C)
over nematic and higher ordered smectic phases. The dielectric
Turning now specically to the smectic A phase, although
design features that stabilize the SmA phase, over the nematic
phase for example, have been identied, e.g. increasing the
aliphatic chain length, little is known about the implication of
such modications upon the actual structure of the smectic A
phase. This is particularly important when designing materials
to be incorporated into devices. Through the study of a wide
variety of smectogens it is now well established that the majority
of materials known to form the smectic A phase adopt one of
three common phase structures, SmA1, SmA2 or SmAd (see
Fig. 1a–c respectively) depending on the overlap between the
molecules in adjacent layers. Smectic A1, has a monolayer
structuring, whereas the SmA2 phase has a bilayer organization
of the molecules, and the SmAd phase has a partially interdig-
itated bilayer, with a reported 1.4ꢀ the molecular length layer
spacing for smectogenic 4-alkyl-40-cyanobiphenyl materials.31
In some cases the structure of the smectic A phase can be
rationalized in terms of the molecular architecture of the
material in question. For example, cyanobiphenyl-based mate-
rials tend to favour a SmAd interdigitated packing arrangement
as a result of dipole–dipole interactions, antiparallel correla-
tions of neighbouring molecules, as shown in Fig. 1c.31 The
structure then has the aromatic and aliphatic parts separately
overlapping to give a nanostructured system of layers. This
The Department of Chemistry, The University of York, York, North Yorkshire, UK.
E-mail: Richard.Mandle@york.ac.uk
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c4tc02991g
2380 | J. Mater. Chem. C, 2015, 3, 2380–2388
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
Journal of Materials Chemistry C
between the hydroxy and nitrile groups, which favours a
nematic arrangement. Although compound 3 (ref. 33 and 34)
has been synthesised previously, transition temperatures have
not, until now, been reported. This result is in keeping with
previous studies on halogen terminated 4-alkoxy-40-cyanobi-
phenyls, in which the terminal chloro- or bromo-group was
found to reduce clearing points relative to the parent compound
while also suppressing the formation of smectic phases, instead
giving rise to a nematic.35
Calculated electrostatic potential isosurfaces (Fig. 2) reveal
that the polar terminal halogen and hydroxyl units exert no
inuence over the electron distribution in the aromatic core,
and that any electron withdrawing effect of these units persists
only over the nearest 2–3 methylene links in the aliphatic chain.
This short-range electron withdrawing effect is still sufficient to
destabilise the smectic A phase, this likely occurs due to elec-
trostatic repulsion between the now electron-poor methylene
groups at the terminus of the chain. These methylene groups
would be expected to make up the bulk of the smectic A layer
interface, however, given the strong electrostatic repulsion a
layered structure is energetically unfavourable.
Although the change in polarity and electrostatic potential
imparted by these terminal groups, relative to the parent
compound, is signicant, these cannot be taken in isolation
from steric factors. Thus in order to investigate the role of steric
effects in relative isolation from changes in the electrostatic
potentials a number of materials, 5 to 9, with bulky, non-polar
terminal groups were prepared. Their phase behaviours were
examined using polarised optical microscopy (POM) and
differential scanning calorimetry (DSC), and are presented in
Table 2, alongside the thermal behaviour of 4-undecyloxy-4'-
cyanobiphenyl, 11OCB, for comparison.
Fig. 1 Cartoon depicting the idealised structure of; (a) the monolayer
smectic A phase, SmA1, (b) the bilayer smectic A phase, SmA2 and (c)
the interdigitated smectic A phase, SmAd, with the layer spacing (d) and
molecular length (L) shown.
raises the question; does the resultant quadrupolar coupling
dominate over the segregation process or vice versa? In order to
explore this question A number of materials derived from 4-
undecyloxy-40-cyanobiphenyl but with varying terminal groups
were prepared with a view to exploring how the terminal group
(polar to apolar) inuences the type, and local structure of
mesophases formed. Specically this work sought to probe if
microphase segregation is a real phenomenon in the relatively
disordered smectic A phase or if the observed behaviour can be
reconciled through a combination of electrostatic and steric
factors.
From the point of view of device applications, our aim was to
create new families of materials that would supress melting
point, widen N and SmA mesomorphic ranges, and when mixed
with standard device materials would retain favourable physical
properties. When formulating mixtures it is imperative to mix
only materials of the same phase type else the individual
components will separate out due to immiscibility.
Compounds 5 and 6, which bear tert-butyl and trimethylsilyl
terminal groups respectively, have near identical melting and
clearing points, as might be expected considering the relatively
similar sizes and polarities of the two end groups. Relative to
11OCB, both 5 and 6 have marginally lower clearing points
ꢁ
(4–6 C), however there is a sharp reduction in melting point.
Results and discussion
In the rst study we show the results accruing from the
attachment of polar units to the terminal positions of the
methyleneoxy chains in the u-hydroxy- and the u-halogeno-
methylenoxy-4'-cyanobiphenyls (see structure in table). The
transition temperatures and associated enthalpies of the
compounds 1 to 4 prepared for this investigation were
determined.
When positioned at the terminus of the aliphatic chain in
4-alkoxy-40-cyanobiphenyls both chloro- and bromo-substitu-
ents suppress the smectic A phase, while retaining the nematic
phase. The same effect is observed for the polar hydroxyl
terminated compound 1, in which the clearing point is notably
higher than 11OCB, however the iodo-terminated compound 4
is non-mesogenic. Compound 1 has been reported previously,32
with the stabilisation of the nematic phase relative to the parent
Fig. 2 Electrostatic potential isosurfaces (kJ molꢂ1) calculated at the
B3LYP/6-311++G** level of theory on geometry optimised at the
same level of theory for compounds 1 (top), 2 (upper middle), 3 (lower
material being due to intermolecular hydrogen bonding middle) and 4 (bottom).
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. C, 2015, 3, 2380–2388 | 2381

View Article Online
Journal of Materials Chemistry C
Paper
Table 1 Transition temperatures (ꢁC) and associated enthalpies of transition [kJ molꢂ1] for compounds 1–4, with those of 11OCB for
comparison. Total molecular volumes (Volrel) were computed using MOPAC2012 with the PM3 method and are given relative to 11OCB
No.
X ¼
Volrel
Cr
SmA
N
Iso
11OCB
H
1.000
1.030
1.069
1.086
1.105
C
C
C
C
C
71.5 [24.01]
91.9 [50.87]
78.7 [46.02]
77.4 [56.68]
87.5 [75.92]
C
—
—
—
—
87.5 [3.45]
—
—
C
C
C
C
C
1
2
3
4
–OH
–Cl
–Br
–I
—
—
—
—
C
(C
(C
—
96.4 [1.81]
69.1) [0.95]
67.2) [0.62]
—
Table 2 Transition temperatures (ꢁC) and associated enthalpies of transition [kJ molꢂ1] for compounds 5–9, with those of 11OCB for
comparison. Total molecular volumes (Volrel) were computed using MOPAC2012 with the PM3 method and are given relative to 11OCB
No.
X ¼
Volrel
Cr
SmA
Iso
11OCB
5
6
H
1.000
1.190
1.229
C
C
C
71.5 [24.01]
54.9 [25.55]
55.1 [17.50]
C
C
C
87.5 [3.45]
83.3 [6.04]
81.7 [4.22]
C
C
C
–C(CH3)3
–Si(CH3)3
7
8
1.445
1.404
C
C
37.7 [9.40]
C
C
73.7 [3.13]
70.1 [5.82]
C
C
31.7 [21.17]
9
1.361
C
22.4 [5.41]
C
73.1 [2.48]
C
Incorporation of signicantly larger pentamethyldisiloxane, of 1.47 molecular lengths. However, for materials with bulky
pentamethyldisilapropane and butyldimethylsilane groups (7, 8 end groups the layer spacing increases with decreasing
and 9 respectively) gives signicantly reduced melting points temperature, giving signicantly larger values of D/L The mean
relative to the parent compound 11OCB. The transition and maximum layer spacings (Lmean and Lmax respectively),
temperatures of compound 7 are in good agreement with calculated molecular length and the D/L ratio for both mean
literature values.22
and maximum layer spacings (Dmean/L and Dmax/L respectively)
It should also be noted that the compounds in Table 2 are given in Table 3. A plot of the D/L ratio (Fig. 3) as a function
possess terminal groups that are signicantly larger than those of reduced temperature demonstrates that the layer spacing
in the hydroxy/halogen terminated compounds 1–4, and hence increases almost linearly for 5–9, with that of 9 being notably
have larger calculated molecular volumes. So it is perhaps lower than that of 5–8, yet the ratio remains almost constant for
surprising at rst sight to nd that the smectic A phase is not 10OCB and 11OCB. 2D diffraction patterns for magnetically
entirely suppressed for compounds 5–9 whose molecular aligned samples of 11OCB, 7 and 9 at reduced temperatures of
ꢁ
structures occupy a greater volume in comparison to the 20 C are given in Fig. 4.
materials with polar end groups in Table 1. In order to assess
The strength of the dipole–dipole interactions between
how these sterically demanding groups inuence the layered adjacent cyanobiphenyl units are unchanged by the incorpora-
structure of the smectic A phase the mean layer spacing in the tion of various bulky groups, hence, the biphenyl–biphenyl
smectic A phases of compounds 5–9, as well as 11OCB and distance within each layer is expected to be unchanged. The
10OCB for comparison, were measured as a function of reduced increase in layer spacing therefore necessitates that the distance
temperature by small angle X-ray scattering (SAXS) for a sample between biphenyl units in adjacent layers must grow larger. One
aligned with a 1 T magnetic eld.
plausible mechanism is that the bulky end groups can be
The mean smectic A layer spacing was found to be essentially accommodated easily upon immediately entering the SmA
temperature invariant for 10OCB and 11OCB, being an average phase due to the highly diffuse layer interfaces. Upon further
2382 | J. Mater. Chem. C, 2015, 3, 2380–2388
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
Journal of Materials Chemistry C
Table 3 Mean and maximum smectic A layer spacings (dmean and dmax), molecular length from B3LYP/6-311++G**optimised geometry (L) and
D/L for mean and maximum layer spacing values for compounds 5–9, 10OCB and 11OCB
˚
˚
˚
L (A)
No.
X ¼
Dmean (A)
Dmax (A)
D/Lmean
D/Lmax
10OCB
11OCB
H
H
35.308
37.087
43.679
44.646
47.379
49.468
47.071
35.390
37.300
44.160
45.210
48.580
50.850
48.020
24.002
25.275
27.767
28.287
29.973
31.136
31.881
1.471
1.467
1.573
1.578
1.581
1.589
1.476
1.474
1.475
1.590
1.598
1.621
1.633
1.504
5
6
7
8
9
–C(CH3)3
–Si(CH3)3
–Si(CH3)2–O–Si(CH3)3
–Si(CH3)2–CH2–Si(CH3)3
–Si(CH3)2–(CH2)3CH3
Fig. 5, as a function of the dihedral angle and using these values
to determine the relative probability for each conformer.
Given that there is near complete rotational freedom about
the carbon–silicon and carbon–oxygen bonds in the tempera-
ture range at which the smectic A layer spacing was measured
(320–350 K), as shown in Fig. 6, the increase in layer spacing is
not likely to be a consequence of conformer distribution of the
end group. Indeed, the aliphatic chains as a whole are unlikely
to experience any signicant change in conformer distribution
over the temperature range in question, given the near total
conformational freedom that they enjoy. The possibility that
layer expansion is a consequence of a decrease in the core-to-
core overlap can therefore be discounted, as this would also
occur in 10OCB and 11OCB and thus give a temperature
dependant variation in layer spacing. Likewise, the distribution
of angles between the molecular long axis and the director (i.e.
the smectic A layer normal) will narrow with decreasing
temperature irrespective of the presence of a bulky group, so
this cannot account for the observed layer expansion. The
increase in layer spacing as a function of temperature can be
accounted for by considering the interface that exists between
the diffuse smectic layers and how a sterically demanding bulky
group is incorporated into this structure.
Fig. 3 Plot of molecular length (L) over smectic A mean layer spacing
(d) as a function of reduced temperature (TSmA–Iso ꢂ T) for compounds
5–9, 10OCB and 11OCB. The molecular lengths were determined
from geometry optimised at the B3LYP/6-311++G**level of theory.
cooling the layer interfaces are expected to grow more dened,
however doing so leads to steric crowding and the layer spacing
must increase in order to accommodate the bulky groups.36
The space lling aspects of the terminal end groups were
investigated further by calculating the steric energies of the end
groups using the PM3 semi-empirical method. It was found that
the barrier to rotation about the carbon–silicon and oxygen–
silicon bonds is signicantly lower than that of carbon–carbon
bonds, so much so that there is effectively free rotation at
ambient temperature. This was conrmed by computing the
steric energy (PM3) of the silane/siloxane fragments, shown in
It has been suggested previously that bulky groups disrupt
the smectic layer interfaces through steric crowding.20,34,36 As
compounds 5–9 are topologically quite different to the nOCB
Fig. 4 2D X-ray diffraction patterns taken at a reduced temperature of 20 ꢁC for magnetically aligned samples of 11OCB (a), compound 7 (b) and
compound 9 (c).
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. C, 2015, 3, 2380–2388 | 2383

View Article Online
Journal of Materials Chemistry C
Paper
Fig. 5 Histogram plots of dihedral angle probability as a function of temperature for the siloxane/silane fragments as determined from steric
energies calculated using the PM3 semi-empirical method.
In order to assess switching behaviour and electrochemical
stability compounds 7 and 8 were doped with 0.09 wt% tetrabu-
tylammonium bromide (TBAB) and the resulting mixtures lled
into 50 mm cells with ITO electrodes. The samples were heated
into the isotropic liquid, cooled to a reduced temperature of 4 ꢁC
and subjected to application of an electric eld of 93 V amplitude
frequencies of 1 Hz and 1 kHz used to obtain transparent and
scattering states respectively. As shown in Fig. 6, the pentam-
Fig. 6 Left, photomicrographs (ꢀ100) showing the transparent (a) and
ethyldisiloxane terminated material 8 was found to be severely
scattering (b) states of the switching process of compound 8 at 66 ꢁC
electrochemically unstable, with the sample in the electrode area
with a driving voltage of 93 V and frequencies of 1 Hz and 1 kHz
being transformed into a yellowish isotropic liquid. Compound
7 however, with a tetramethyldisilapropane end group, was found
to be stable to these conditions. Although the threshold voltage
for switching is too high for device use the enhanced stability
offered by this end group is a potentially important step in the
respectively. Right, photograph of cells of compounds 8 (c) and 7 (d)
illustrating that while compound 8 is stable 7 suffers from extensive
electrochemical decomposition to the point that the sample within the
electrode area is rendered isotropic.
design of materials for smectic
A devices. Furthermore,
materials, due to the presence of bulky terminal groups, for the
layer spacing to remain constant as a function of temperature
(or indeed, order parameter) would require the energetically
unfavourable packing of these bulky groups. To avoid this, the
interlayer spacing increases as a function of decreasing
temperature, allowing the interlayer interfaces to remain
diffuse so as to accommodate the bulky group. In compound 9
the bulkiest part of the chain (the dimethylsilyl unit of the
butyldimethylsilyl end group) is not at the terminus, hence,
there is less disruption to the layer interface and the layer
spacing is smaller, although still temperature dependent. In
this structural arrangement a degree of “microphase segrega-
tion” appears to exist, however, this is a consequence of the
need to pack bulky groups within the layer interface, rather than
the driving force behind the lamellar structure. The argument
that microphase segregation in liquid crystals is due to ‘chem-
ical immiscibility’ between hydrocarbon and siloxane/silane
units is further debunked by considering that compounds 5 and
6, which have a tert-butyl group and a trimethylsilyl group
respectively, show nearly identical layer expansion. This
demonstrates that the segregation of the bulky groups is near
identical, despite their different chemical makeup.
compound 8 has only a slight reduction in clearing point relative
to 7 and possesses a lower melting point. It is well known that the
materials used in electrooptic devices based on the smectic A
phase experience limited device lifetimes, and methods have been
proposed to extend this.19,37 From this study however it appears
that carbosilane units, such as the tetramethyldisilapropane
group used in compound 7 offer far greater lifetimes than their
siloxane counterparts at the cost of only slight reductions in
clearing point.
In our third study materials containing terminal units that
are both polar and bulky were prepared in order to assess how a
combination of steric and electronic effects would impact on
the smectic layer interface and thus the thermal stability of the
smectic A phase. To this end, materials with either phenoxy,
uorinated phenoxy, peruorinated phenoxy, and [2,2]-para-
cyclophanyloxy end groups were synthesised, their phase
behaviour was examined using polarised optical microscopy
(POM) and differential scanning calorimetry (DSC) and the
results obtained for their phase behaviours are presented
together in Table 4, alongside the transition temperatures of
11OCB for comparison.
2384 | J. Mater. Chem. C, 2015, 3, 2380–2388
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
Journal of Materials Chemistry C
Table 4 Transition temperatures (ꢁC) and associated enthalpies of transition [kJ molꢂ1] for compounds 10–15, with those of 11OCB given for
comparison. Total molecular volumes (Volrel) were computed using MOPAC2012 with the PM3 method and are given relative to 11OCB
No.
X ¼
Volrel
Cr
SmA
N
Iso
11OCB
10
H
1.000
1.224
C
C
71.5 [24.01]
84.6 [46.14]
C
87.5 [3.45]
—
—
C
C
—
—
(C
62.1) [0.90]
11
1.254
C
62.3 [52.39]
—
—
(C
55.1) [0.98]
C
12
13
1.246
1.250
C
C
76.2 [54.15]
87.1 [51.62]
—
—
—
—
—
—
C
C
(C
77.8) [0.48]
14
15
1.321
1.561
C
C
59.0 [50.50]
57.1 [58.72]
(C
46.1 [0.48]
C
50.1) [0.48]
32.9) [3.32]
C
C
—
—
(C
Compared with 11OCB, all of the materials were nemato- on the position of the uorine atom within the ring, alluding to
genic except for compound 12, which was non-mesogenic. None the importance of repulsive electrostatic interactions in desta-
of the materials were smectogenic except for the per- bilising the smectic state as described previously. Conversely,
uoroophenoxy material 14, which has been reported previ- compound 14, with a pentauorophenoxy-terminal group,
ously, with the transition temperatures in this work being in exhibits monotropic nematic and smectic A phases, despite
good agreement with literature values.38 All of the materials being notably larger than the phenoxy and monouorophenoxy
were found to exhibit lower clearing points than the parent groups. It is possible for this material that the electron decient
compound 11OCB. The largest end group employed in this peruoro-ring associates strongly with phenyl rings, as
work, [2,2]-paracyclophane (compound 15) gives dramatic observed in other systems, and hence the smectic phase
reductions in clearing point, probably as to the large steric bulk becomes stabilised. Thus to rationalise the presence of a
of this group cannot be incorporated into the layer interface and smectic A phase in 14 and its absence in 10 to 13, electrostatic
so only a monotropic nematic phase was observed. The melting potential isosurfaces were computed for end group fragments of
and clearing points of the mono-uorinated analogues (11 to compounds 10 to 15 at the B3LYP/6-311++G** level of theory
13) of the phenoxy-terminated compound 10 depend strongly and are presented in Fig. 7.
As expected, the phenoxy (fragment a) and [2,2]-para-
cylophanyloxy (fragment f) groups are broadly similar in terms
of calculated ESP isosurfaces. For fragments b to d, corre-
sponding to compounds 11 to 13 respectively, the electrostatic
potentials are heavily dependent on the position of the uorine
atom within the ring. The aromatic carbon atoms and terminal
methylene unit for the pentauorophenoxy fragment (e) are
signicantly more electron poor than the phenoxy fragment (a),
which is why fragments of type (e) strongly associate with
aromatic rings. Furthermore, the smectic A phase appears to be
able to tolerate the electrostatic repulsion that is likely to result
from interactions between electron poor pentauorophenoxy
groups, whereas the electrostatic repulsion between the rela-
tively electron rich phenoxy end group seems to be sufficient to
destabilise the phase.
Fig. 7 Electrostatic potential isosurfaces (kJ molꢂ1) calculated at the
B3LYP/6-311++G** level of theory on geometry optimised at the same
By mixing materials with electron rich and electron poor end
level of theory for end group fragments of compounds 10–15, phenoxy
groups, or indeed chemically synthesising materials bearing
(a), 2-fluorphenoxy (b), 3-fluorophenoxy (c), 4-fluorophenoxy (d), tetra-
fluorophenoxy (e) and [2,2]-paracyclophanyloxy (f).
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. C, 2015, 3, 2380–2388 | 2385

View Article Online
Journal of Materials Chemistry C
Paper
formation of smectic A phases in these materials should be
questioned. Given that the aliphatic chains are not likely to
undergo a signicant change in length or conformer distribution
over the temperature range in question, it would appear likely that
the layers and layer interfaces grow more dened as the temper-
ature is reduced, and this leads to increased spacing between the
aromatic regions of adjacent layers.
2. The classication of smectic A phases into three categories,
SmA1, SmA2 or SmAd is still acceptable for the monolayer and
bilayer types because the two phases exist for individual
compounds in reentrant morphology and can be clearly identied
via X-ray diffraction. However, the possible variation in the layer
spacing for the SmAd phase suggests that it is perhaps not a
separate phase requiring classication. The fact that the original
identication of the SmAd phase was made on the alkyl- and
alkoxy-cyanobiphenyl where the layer spacing was typically 1.4ꢀ
the molecular length, produced an opportunity for the introduc-
tion of a sub-classication which was perhaps not necessary.
3. Calculations of conformational freedom, coupled with
X-ray diffraction in the smectic A phase imply, but do not
conrm, that the terminal bulky groups are in dynamic motion,
even at ambient temperature. The smectic A phase therefore
should be considered as a condensed phase where the mole-
cules are uctuating rapidly on very short time scales. A result of
this behaviour is that it is highly unlikely that smectic A phases
made up of rod-like molecules will be biaxial or polar, unless
they are subjected to an applied external eld. By extension, the
same will also apply to nematic phases.
4. Electrostatic potentials cannot be considered in isolation
from steric effects. At the molecular level larger molecular
volumes appear on the whole to support smectic formation,
whereas situations where the molecules are more rod-like, i.e.
have a higher molecular aspect ratio, support nematic phase
behaviour. However, smectic materials incorporating bulky
terminal groups will experience disruption at the layer inter-
faces through steric crowding, leading to lower clearing points
and melting points relative to materials lacking sterically
demanding end groups.
Fig. 8 Gibbs phase diagram for binary mixtures of compound 10 and
14. The dotted line shows denotes the smectic A to nematic transition
temperature in 14, aiding visualisation of the stabilisation effect.
one of each group, the unfavourable electrostatic repulsions will
be reduced and would be expected to lead to stabilisation of the
smectic A phase, particularly between electron rich and electron
poor groups. To explore this further binary mixtures between
compound 10, with a relatively electron rich end group, and
compound 14 with a relatively electron poor end group were
prepared and a Gibbs phase diagram was constructed, see
Fig. 8.
While the isotropic to N phase transition varies almost
linearly with respect to concentration, as shown in Fig. 8, the
SmA–N transition is non-linear and is found to be thermally
stabilised, relative to the parent material 14. This result illus-
trates that while individually electron rich or poor groups are
detrimental to the layered structure of the smectic A phase due
to electrostatic repulsion, however, this can be negated to some
extent by incorporating both electron rich and decient end
groups in binary mixtures. This is manifested in the stabilisa-
tion of the smectic A phase relative to the parent material
(1.4 ꢁC), however, given the relatively large steric bulk of the
aromatic and uoroaromatic end groups the clearing point is
much lower than the parent 11OCB.
5. When the end group(s) of a molecule is exclusively elec-
tron rich or poor this destabilises smectic phases due to
unfavourable electrostatic repulsion between groups of like
Summary and conclusion
Based on the results reported a number of important issues can charge. Hence, materials bearing two electron withdrawing
be elucidated as follows:
terminal groups – such as halogen atoms – are unlikely to
1. It is clear from the results that microphase segregation does exhibit smectic A phases, but materials with one electron
not occur for materials with chemically different end groups, and donating and one electron withdrawing group are likely to be
the claims that silicon based materials are somehow special with good candidates for the formation of smectic A phases. For
superior properties are misleading. In the smectic A phase of example, the stabilisation of the smectic A phase reported
cyanobiphenyls (and many other polar mesogens) there is exten- by Rupar et al.39 on mono-halogen terminated 4-alkoxy-40-
sive antiparallel correlation of adjacent molecules. As a conse- (alkyoxyphenyl)pyrimidines occurs because one end of the
quence, the aromatic and aliphatic sections of the molecule are molecule is polar and the other is not. Conversely, for the
segregated, while any ‘end groups’ will tend to exist close to, or halogen terminated 4-alkoxycyanobiphenlys both ends of the
within, the layer interface. This segregation is a consequence of, molecules are polar and hence stabilisation is not observed.35
rather than the impetus behind the formation of the antiparallel
correlations that exist within interdigitated smectic phases. As the
presence or absence of a bulky group does not impact upon the
References
degree of core-to-core overlap present in the interdigitated smectic
phase the view that microphase segregation is responsible for the
1 C. Tani, Novel Electro-optical Storage Effect in a Certain
Smectic Liquid Crystal, Appl. Phys. Lett., 1971, 19, 241–242.
2386 | J. Mater. Chem. C, 2015, 3, 2380–2388
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
Journal of Materials Chemistry C
2 M. Hareng and S. Le Berre, Defects – Applications To Display 19 M. V. Mitrokhin, L. Johnsen, R. Fagerberg, C. Kristiansen,
Devices Thermal Relaxation Recording Device on Smectic
Liquid Crystals, J. Phys., Colloq., 1976, C3.
3 D. Coates, W. A. Crossland, J. H. Morrissy and B. Needham,
Electrically Induced Scattering Textures in Smectic A Phases
F. J. Farrand, K. H. Holm, R. Palm, K. Netland and
V. L. Aristov, Prevention of Smectic-A LC Mixtures
Electrolytic Degradation, Mol. Cryst. Liq. Cryst., 2004, 411,
255–263.
¨
And Their Electrical Reversal, J. Phys. D: Appl. Phys., 1978, 11, 20 J. W. Goodby, I. M. Saez, S. J. Cowling, V. Gortz, M. Draper,
2025–2034.
A. W. Hall, S. Sia, G. Cosquer, S.-E. Lee and E. P. Raynes,
Transmission and Amplication of Information and
Properties in Nanostructured Liquid Crystals, Angew.
Chem., Int. Ed., 2008, 47, 2754–2787.
4 S. Khosla, K. K. Raina and H. J. Coles, Electrically induced
storage effects in smectic A phase of dyed low molar mass
liquid crystals, Curr. Appl. Phys., 2003, 3, 135–140.
5 D. J. Gardiner and H. J. Coles, Organosiloxane liquid crystals 21 C. Tschierske, Development of Structural Complexity by
for fast-switching bistable scattering devices, J. Phys. D: Appl.
Phys., 2006, 39, 4948–4955.
Liquid-Crystal Self-Assembly, Angew. Chem., Int. Ed., 2013,
52, 8828–8878.
6 N. Gheorghiu, J. L. West, A. V. Glushchenko and 22 J. Newton, H. Coles, P. Hodge and J. Hannington, Synthesis
M. Mitrokhin, Patterned eld induced polymer walls for
smectic A bistable exible displays, Appl. Phys. Lett., 2006,
88, 263511.
7 D. J. Gardiner, C. J. Davenport, J. Newton and H. J. Coles,
Electro-optic bistability in organosiloxane bimesogenic
liquid crystals, J. Appl. Phys., 2006, 99, 113517.
and properties of low-molar-mass liquid-crystalline siloxane
derivatives, J. Mater. Chem., 1994, 4, 869–874.
23 P. Kloess, J. McComb, H. J. Coles and R. Zentel, Synthesis
and properties of a new series of low-molar-mass organo-
siloxane derivatives, Ferroelectrics, 1996, 180, 233–243.
¨
24 M. Ibn-Elhaj, H. Mohwalf, M. Z. Cherkaoui and R. Zniber,
8 D. J. Gardiner and H. J. Coles, Highly anisotropic
conductivity in organosiloxane liquid crystals, J. Appl.
Phys., 2006, 39, 4948–4955.
Effect of intermolecular and interfacial interactions on the
three- and two-dimensional structure and phase behaviour
of three-block liquid crystal siloxane derivatives, Langmuir,
1998, 18, 504–516.
9 H.-Y. Chen, R. Shao, E. Korblova, W. Lee, D. Walba and
N. Clark, A bistable liquid-crystal display mode based on 25 J. R. Willmott, C. J. Davenport, J. Newton and H. J. Coles,
electrically driven smectic A layer reorientation, Appl. Phys.
Lett., 2007, 91, 163506.
10 H.-Y. Chen, R. Shao, E. Korblova, D. Walba, N. A. Clark and
Physical Properties of Siloxane Containing Oxycyanobiphenyl
Derivatives for Smectic-A Device Applications, Mol. Cryst. Liq.
Cryst., 2005, 434, 209–217.
W. Lee, Bistable SmA liquid-crystal display driven by a two- 26 L. Li, C. D. Jones, J. Magolan and R. P. Lemieux, Siloxane-
direction electric eld, J. Soc. Inf. Disp., 2008, 16, 675–681.
11 D. J. Gardiner, S. M. Morris and H. J. Coles, High-efficiency
terminated phenylpyrimidine liquid crystal hosts, J. Mater.
Chem., 2007, 17, 2313–2318.
multistable switchable glazing using smectic A liquid 27 Q. Song, D. Nonnenmacher, F. Giesselmann and
crystals, Sol. Energy Mater. Sol. Cells, 2009, 93, 301–306.
12 H.-Y. Chen and J.-S. Wu, A multistable smectic-A liquid-
crystal device with low threshold eld, J. Soc. Inf. Disp.,
2010, 18, 415–420.
13 Y. Lu, J. Guo, H. Wang and J. Wei, Flexible bistable smectic-A
liquid crystal device using photolithography and
photoinduced phase separation, Adv. Condens. Matter
Phys., 2012, DOI: 10.1155/2012/843264.
R. P. Lemieux, Tuning ‘de Vries-like’ properties in siloxane-
and carbosilane-terminated smectic liquid crystals, J.
Mater. Chem. C, 2013, 1, 343–350.
28 M. Ibn-Elhaj, A. Skoulios, D. Guillon, J. Newton, P. Hodge
and H. J. Coles, Structural characterisation of linear
dimeric and cyclic tetrameric liquid crystalline siloxane
derivatives, Liq. Cryst., 1995, 19, 373–378.
29 E. Corsellis, D. Guillon, P. Kloess and H. Coles, Structural
characterization of mono- and di-mesogenic organosiloxanes:
The impact of siloxane content on biphenyl benzoate
systems, Liq. Cryst., 1997, 23, 235–239.
14 C.-H. Chen, V. Y. Zyryanov and W. Lee, Switching of Defect
Modes in a Photonic Structure with a Tristable Smectic-A
Liquid Crystal, Appl. Phys. Express, 2012, 5, 082003.
15 Y. Lu, J. Wei, Y. Shi, O. Jin and J. Guo, Effects of fabrication 30 C. Tschierske, Non-conventional liquid crystals
–
the
condition on the network morphology and electro-optical
characteristics of a polymer-dispersed bistable smectic A
liquid-crystal device, Liq. Cryst., 2013, 40, 581–588.
16 D. J. Gardiner and H. J. Coles, High-solubility liquid crystal
dye guest–host device, Proc. SPIE, 2007, 6587, DOI:
10.1117/12.723041.
importance of micro-segregation for self-organisation, J.
Mater. Chem., 1998, 8, 1485–1508.
31 A. J. Leadbetter, J. C. Frost, J. P. Gaughan, G. W. Gray and
A. Mosley, The structure of smectic
A phases of
compounds with cyano end groups, J. Phys., 1979, 40, 375–
380.
17 R. R. Deshmukh and A. K. Jain, The complete 32 A. C. Griffin and S. R. Vaidya, Effect of Molecular Structure
morphological, electro-optical and dielectric study of
dichroic dye-doped polymer-dispersed liquid crystal, Liq.
Cryst., 2014, 41, 960–975.
on Mesomorphism 22. Liquid Crystalline 1-Alkanol
Derivatives, Mol. Cryst. Liq. Cryst., 1989, 173, 85–88.
33 E. Terazzi, C. Bourgone, R. Welter, J.-L. Gellani, D. Goullion,
G. Rogez and B. Donnio, Single-Molecule Magnets with
Mesomorphic Lamellar Ordering, Angew. Chem., Int. Ed.,
2008, 47, 490–495.
18 H. Kataoka, N. Shigeno, M. Munakata and T. Urabe,
Reective guest–host liquid-crystal display device, US
6016178 A, 1996.
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. C, 2015, 3, 2380–2388 | 2387

View Article Online
Journal of Materials Chemistry C
Paper
34 E. Terazzi, G. Rogez, J.-L. Gallani and B. Donnio,
Supramolecular Organization and Magnetic Properties of
behaviour in liquid crystals, Chem. Commun., 2006, 4107–
4109.
Mesogen-Hybridized Mixed-Valent Manganese Single 37 D. J. Gardiner and H. J. Coles, Enhancing lifetime in a
Molecule Magnets [MnIII8MnIV4O12(Lx,y,z-CB)16(H2O)4], J. Am.
Chem. Soc., 2013, 135, 2708–2722.
bistable smectic A liquid crystal device, J. Phys. D: Appl.
Phys., 2007, 40, 977–981.
35 E. J. Davis, R. J. Mandle, B. K. Russell, P. Y. Foeller, 38 T. Itahara, Effect of the pentauorophenyloxy group on
M. S. Cook, S. J. Cowling and j. W. Goodby, Liquid- liquid crystalline behaviour, Liq. Cryst., 2007, 32, 115–118.
crystalline structure-property relationships in halogen- 39 I. Rupar, K. M. Mulligan, J. C. Roberts, D. nonnenmacher,
terminated derivatives of cyanobiphenyl, Liq. Cryst., 2014,
41, 1635–1646.
36 S. J. Cowling and J. W. Goodby, Interfacial layer interactions:
their effects on synclinic and anticlinic smectic mesophase
F. Giesselmann and R. P. Lemieux, Elucidating the
smectic-A-promoting effect of halogen end-groups in
calamitic liquid crystals, J. Mater. Chem. C, 2013, 1, 3729–
3735.
2388 | J. Mater. Chem. C, 2015, 3, 2380–2388
This journal is © The Royal Society of Chemistry 2015