Organic-inorganic hybrid liquid crystals derived from octameric silsesquioxanes. Effect of the peripheral groups in mesogens on the formation of liquid crystals

Article abstract of DOI:10.1039/c0jm03406a

A series of octameric silsesquioxane-based organic-inorganic hybrid compounds 7a-h was synthesised by reacting octakis(dimethylsiloxy)silsesquioxane with 4′-(undec-10-enyloxy)biphenyl-4-yl mono-, di-, or trialkoxybenzoate via platinum-catalysed hydrosilylation reaction. The chemical structures of 7a-h were characterised by ¹H, ¹³C, and ²⁹Si nuclear magnetic resonance spectroscopy, mass spectroscopy and elemental analysis. Liquid crystallinity of 7a-h was investigated by differential scanning calorimetry (DSC) and polarised optical microscopy (POM). Temperature-dependent X-ray diffraction was used to verify liquid crystal phases, revealing that 7a-d assembled to smectic C and smectic A phases. The effects of the number and length of peripheral flexible alkoxy groups in 7a-h on the formation of liquid crystals were studied using DSC and POM, which revealed that 7a-d bore one alkoxy terminal group in each mesogen exhibited a mesophase, 7e-f and 7g-h bore two and three flexible alkoxy terminal groups in each mesogen, respectively, and did not show mesomorphic behaviour. The concept of "flexible group density" (FGD), which was defined as the number of the CH₂ groups per unit volume in the periphery of the POSS molecule, was introduced to interpret the formation of liquid crystals. The results revealed that the POSS molecules with di- and tri-substituted mesogens exhibited much larger FGD than those with mono-substituted mesogens, hence the absence of mesophases for 7e-f and 7g-h. This study provides a useful archetype to predict the formation of liquid crystals when designing and synthesising liquid crystalline materials, particularly star-like or dendrimer-like molecular materials.

Full text of DOI:10.1039/c0jm03406a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2011, 21, 5248
www.rsc.org/materials
PAPER
Organic–inorganic hybrid liquid crystals derived from octameric
silsesquioxanes. Effect of the peripheral groups in mesogens on the formation
of liquid crystals
Xiaobai Wang,a Ching Mui Cho,a Wan Yong Say,ab Angeline Yan Xuan Tan,a Chaobin He,ac
b
Hardy Sze On Chan and Jianwei Xu
a
*
Received 9th October 2010, Accepted 5th January 2011
DOI: 10.1039/c0jm03406a
A series of octameric silsesquioxane-based organic–inorganic hybrid compounds 7a–h was synthesised by
reacting octakis(dimethylsiloxy)silsesquioxane with 40-(undec-10-enyloxy)biphenyl-4-yl mono-, di-, or
trialkoxybenzoate via platinum-catalysed hydrosilylation reaction. The chemical structures of 7a–h were
characterised by 1H, 13C, and 29Si nuclear magnetic resonance spectroscopy, mass spectroscopy and
elemental analysis. Liquid crystallinity of 7a–h was investigated by differential scanning calorimetry
(DSC) and polarised optical microscopy (POM). Temperature-dependent X-ray diffraction was used to
verify liquid crystal phases, revealing that 7a–d assembled to smectic C and smectic A phases. The effects
of the number and length of peripheral flexible alkoxy groups in 7a–h on the formation of liquid crystals
were studied using DSC and POM, which revealed that 7a–d bore one alkoxy terminal group in each
mesogen exhibited a mesophase, 7e–f and 7g–h bore two and three flexible alkoxy terminal groups in each
mesogen, respectively, and did not show mesomorphic behaviour. The concept of ‘‘flexible group density’’
(FGD), which was defined as the number of the CH2 groups per unit volume in the periphery of the POSS
molecule, was introduced to interpret the formation of liquid crystals. The results revealed that the POSS
molecules with di- and tri-substituted mesogens exhibited much larger FGD than those with mono-
substituted mesogens, hence the absence of mesophases for 7e–f and 7g–h. This study provides a useful
archetype to predict the formation of liquid crystals when designing and synthesising liquid crystalline
materials, particularly star-like or dendrimer-like molecular materials.
was reported in 1991,16 after which a number of POSS-based
Introduction
LCs were generated by introducing different mesogens in the
corner chains.17–23 These POSS-based LCs are generally prepared
starting from the octameric silsesquioxanes such as (Si8O12)H8
or (Si8O12)(OSi(Me)2H)8, which allow for modification with
a variety of mesogens via hydrosilylation in the presence of
a suitable platinum catalyst. The sizes and structures of these LC
materials could be well controlled by attaching different meso-
gens, revealing that various mesophases could be formed by
governing the shape of the mesogens, the spacer length between
the mesogen and the POSS core as well as the number of
substituted corner chains. For instance, the inclusion of lateral
attachments of calamitic mesogens, incomplete substitution of
the POSS core, or involvement of nematogenic mesogens in the
POSS side corner chains can lead to the formation of nematic
phases.16,22–25 By introducing chiral nematic mesogens to the
silsesquioxane core with a moderate spacer length to control the
strength of coupling (interaction) between the POSS core and
mesogens, only the chiral nematic phase could be observed.
However, mesogens, such as 40-u-alknyl-4-cyanobiphenyls26 and
bent-core mesogens,22,27 that are attached to the POSS core led to
the formation of a thermotropic smectic phase, a smectic CP or
Among the useful candidates for forming inorganic–organic
hybrid materials, polyhedral oligosilsesquioxanes (POSS),
particularly the cube-like octameric silsesquioxanes [(SiO3/2)8R8],
have attracted much interest during the past two decades.1,2
Octameric silsesquioxane has a stable, compact chemical struc-
ture and can be regarded as an inorganic nano-particle with Si–
O–Si bonds and Si corners. This stable cubic structure can be
modified by adding different types of organic functional groups
to the Si corners via suitable chemical reactions, and thus
a variety of hybrid materials can be achieved.3–15 Due to the
unique characteristics of POSS, it has been used as a core to
develop organic–inorganic hybrid liquid crystalline materials.
The first octameric silsesquioxane-containing liquid crystal (LC)
aInstitute of Materials Research and Engineering, 3 Research Link,
Singapore, 117602. E-mail: jw-xu@imre.a-star.edu.sg; Fax: +65-6872-
7528; Tel: +65-6872-7543
bDepartment of Chemistry, National University of Singapore, Singapore,
3Science Drive 3, 117543
cDepartment of Materials Science & Engineering, National University of
Singapore, 5 Engineering Drive 2, Singapore, 117576
5248 | J. Mater. Chem., 2011, 21, 5248–5257
This journal is ª The Royal Society of Chemistry 2011

View Article Online
a nematic phase. Columnar assemblies, such as rectangular and
hexagonal columnar mesophases, could be observed for lateral
mesogen-modified POSS-based LCs. Furthermore, the strength of
the coupling between the POSS core and mesogens could be
adjusted by changing the spacer length between the mesogens and
the POSS core and subsequently controlling the POSS-based LC
supramolecular assembly.28 When the spacer length between the
mesogens and the POSS core is short, for example, two CH2 units,
a ‘‘column-within-column’’ hexagonal columnar phase is formed. In
contrast, when the spacer length increases to six or ten CH2 units,
it tends to form rectangular or hexagonal columnar phases.
Flexible chains are usually linked to various mesogens, and
their functions are mostly to modulate the thermal properties,
solubility, processability and assembly architectures. However,
studies on the effect of peripheral flexible chains on the formation
of LCs are scarce. Peripheral flexible chains attached to the
mesogen are of the utmost importance in adjusting the thermal
properties, particularly phase transition temperatures, of LCs.
The number, size (length) and even shape of the flexible chains
will have an effect on the thermal properties, morphology and
assembly of LCs. Therefore, proper selection of flexible chains in
suitable numbers would be valuable in forecasting the possibility
of LC formation as well as the assembly structures. In this paper,
we report on the synthesis, structure and LC properties of a series
of octakis[(11-(40-(1-(4-alkoxy,3,5-dialkoxy or 3,4,5-trialkox-
ybenzoyloxy)biphenyl-4-yloxy)undecyl)dimethylsiloxy)]silses-
quioxanes. The spacer length between the POSS core and
mesogens is fixed to be -(CH2)11Si(Me)2O-, which is approxi-
mately equivalent to the length of thirteen methylene units, thus
minimising the coupling between the POSS core and mesogens.
Mesogens with one, two or three terminal alkoxy groups are
selected to attach to the POSS core via a platinum-catalysed
hydrosilylation reaction. The effect of the length and number of
peripheral alkoxy groups in each mesogen on the formation of
LCs is explored. The phase transitions, thermal properties, and
corresponding structural changes are studied using differential
scanning calorimetry (DSC), polarised optical microscopy
(POM) and variable temperature X-ray diffraction (XRD). The
study reveals that the formation of LCs is controlled by varying
the length and the number of the terminal alkoxy groups in the
mesogens. When the mono alkoxy terminal group is attached to
each mesogen, the POSS derivatives exhibit the smectic phases.
However, when two or three alkoxy terminal groups are attached
in each mesogen, no mesophases are formed.
Results and discussion
Synthesis and structural characterisation
The synthetic routes leading to compounds 7a–h are shown in
Scheme 1. Alkenyl precursor compounds 5a–h were prepared by
reacting excess biphenyl-4,40-diol (1) with 11-bromoundec-1-ene
(2) in the presence of anhydrous K2CO3 in DMF, followed by
esterification using N,N0-dicyclohexylcarbodiimide (DCC) and
a catalytic amount of 4-dimethylaminopyridine (DMAP) accord-
ing to a similar method previously reported, with modifications.29
A hydrosilylation reaction between octakis(hydrodimethylsiloxy)
octasilsesquioxane (1) with compounds 5a–h was performed in dry
toluene at room temperature with the catalyst platinum
divinyltetramethyldisiloxane complex (Pt(dvs)) (Scheme 1).30
An excess of 5a–h was used to ensure that the hydrosilylation
was complete. Crude products were purified by silica gel
Scheme 1 Synthesis route leading to compounds 7a–h. Reaction conditions: i) K2CO3, dry DMF, 80 ꢀC, 48 h; ii) DCC, DMAP, dry CH2Cl2, reflux for
16 h in argon. iii) Pt(dvs), dry toluene, 40 ꢀC, overnight, N2.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5248–5257 | 5249

View Article Online
Table 1 Thermal behavior of 7a–ha
Transition T/ꢀC (Enthalpy/kJ molꢁ1
column chromatography to afford pure 7a–h. All intermediates
1
5a–h and final compounds 7a–h were characterised by H, 13C,
)
29Si NMR spectroscopy, and elemental analysis. The disap-
pearance of the Si–H absorption at 2142 cmꢁ1 in FT-IR spectra
1
and the signal at 4.7 ppm in H NMR spectra, in combination
POSS
Heating
Cooling
with 13C and 29Si NMR spectra, confirms exactly eight corner-
substitutions. As an example, H NMR and 29Si NMR spectra
7a
7b
7c
7d
7e
7f
Cr 124(53.7) S 220(37.5) I
Cr 122(181.3) S 233(98.8) I
Cr 127(213.3) S 232(106.4) I
Cr 119(47.5) S 211(39.8) I
Cr ꢁ8.5(13.1) I
I 213(42.9) S 115(50.1) Cr
I 227(81.2) S 106(148.1) Cr
I 227(95.9) S 112(186.1) Cr
I 199(41.8) S 111(52.6) Cr
I ꢁ14.0(17.1) Cr
1
of 7f are given in Fig. 1 and 2, respectively. Two pairs of double
doublets at d 7.57, 7.49, 7.23 and 6.95 were assigned to aromatic
protons with AB splitting systems (biphenyl part). Two singlets
at d 7.32 and 7.71 correspond to 3,5-disubstituted phenoxy
protons. The proton integration ratio of -OSi(CH3)2- at d 0.15
and -OCH2- at d 4.00 was 1 : 1, which is the same as the theo-
retical value, confirming the formation of octa-substitution in
the detection limit. In the 29Si NMR spectrum of 7f, two singlets
at d 12.75 and ꢁ108.65, which are assigned to the Si atom on the
side chains and the cage Si atom, respectively, were observed,
indicating the octameric cage structure is intact during the
synthesis. A pseudo-triplet at d 0.6 and the absence of signals at
d 2.5 both indicate that the addition of the Si–H bond to the
Cr ꢁ0.9(13.2) I
I ꢁ4.0(16.4) Cr
7g
7h
Cr 39.7(68.2) I
Cr 25(108.9) I
I ꢁ14.5(26.1) Cr
I 10(71.5) Cr
a
Cr, crystalline; S, smectic phase; I, isotropic phase. Transition
temperatures (ꢀC) were taken from second run of the DSC thermo
diagram.
associated enthalpy changes of transition are also summarised in
Table 1. Compounds 7a–d that have only one flexible alkoxy
terminal group in each mesogen displayed multiple phase tran-
sitions on heating and on cooling, indicating the formation of
mesophase. As examples, the DSC thermograms of compounds
7b and 7c are shown in Fig. 3. In Fig. 3a, it is clearly shown that
both 7c and 7d displayed mesophases upon heating and cooling.
At first glance, only two major phase transitions in the DSC
thermogram of 7c are observed on heating and on cooling.
However, a few small endothermic peaks with much lower
enthalpy changes from 0.21 to 1.17 kJ molꢁ1 than that of the
major transition, which occurred at 232 ꢀC (106.4 kJ molꢁ1), are
observed in an enlarged DSC thermogram of 7c between 120 ꢀC
and 210 ꢀC on the second heating (Fig. 3b). Similarly, a few small
corresponding exothermic peaks are also found on the second
cooling. Besides 7c, other liquid crystalline compounds, such as
7a–b and 7d, show analogous transition behaviour, although these
transitions are not as obvious as those for 7c. These negligible
phase transitions of 7c have been detected by POM, and they
roughly correspond to texture changes, as shown in Fig. 4. The
texture changes observed at 134, 145, 168, 187 and 197 ꢀC corre-
spond to the phase transitions observed in the DSC curve at 130,
b
alkene 5a–h (CH2 ]CHa-R) affords regiospecific b-addition
products, consistent with reported examples.31 7a–h are white
solids or low melting point waxy solids at room temperature
and are soluble in common organic solvents such as tetrahy-
drofuran, chloroform, etc.
Thermal behaviour and textures studied by DSC and POM
The thermal behaviours of molecules 7a–h were examined using
DSC. Their corresponding phase transition temperatures and
ꢀ
146, 165, 182 and 197 C, respectively. The enthalpy change for
negligible transitions is around 1% compared to the enthalpy
changes of major transitions. A few unusual textures, particularly,
shown in Fig. 4a and b are observed, which are assigned to
unidentified LC phases. Fig. 4c–f display a broken fan-shaped
texture, a typical smectic C (SmC) phase. In contrast, a focal conic
fan texture was observed in Fig. 4h, which is a characteristic
texture of a smectic A (SmA) phase (see below). Therefore, the
transitions with very small enthalpy changes mostly correspond to
the textures of LC phases although it is hard to completely rule out
that some of these subtle enthalpy changes are due to the intra-
molecular molecular movement of the POSS corner chains.
As shown in Table 1, 7a–h show larger enthalpy changes than
conventional LCs. Similar observations with large enthalpy
changes have been found in other POSS-based LCs.28 This is
likely due to the eight bulk mesogenic groups surrounding the
POSS core, which may enhance the intra-molecular interactions.
The identification of mesophase was studied by POM, revealing
that 7a–d exhibit SmC phases at relatively lower temperatures
and SmA phases at relatively higher temperatures. As an
example, polarised optical photographs of 7c and 7d during
Fig. 1 1H NMR spectrum of 7f in CDCl3 at room temperature.
Fig. 2 29Si NMR spectrum of 7f in CDCl3 at room temperature.
5250 | J. Mater. Chem., 2011, 21, 5248–5257
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Fig. 3 a) DSC thermograms of 7b and 7c, the 2nd run. b) Enlarged part of DSC diagram of 7c from 120 ꢀC to 250 ꢀC on heating. Small enthalpy
changes between 125–200 ꢀC show different phase transitions.
cooling are shown in Fig. 5. 7c and 7d exhibited typical focal
conic fans, corresponding to SmA phases. At the lower temper-
ature, they displayed broken fan textures, revealing the forma-
tion of SmC phases. It is worth noting that a less ordered nematic
phase was not observed for 7a–d. The octa-substitution increases
the local concentration of the mesogens upon attachment to the
POSS core, suppresses the formation of less-ordered nematic
phases and thus leads to higher-ordered phases.32 Compounds
7e–f and 7g–h have two and three flexible terminal groups in each
mesogen, respectively, but they did not exhibit mesomorphic
behaviour, as evidenced by the absence of mesophase transitions
in their DSC thermograms, which is verified by POM as well.
For POSS compounds 7a–d with mono-alkoxy substituted
mesogens, the alkoxy chain lengths from OC5H11 to OC12H25 in
the mesogenic side chains have a certain impact on changes in the
phase transition temperatures (Table 1), but there is no fixed
correlation between the phase transition temperatures and
lengths of the alkoxy chain. For example, on the second heating,
the crystalline–smectic transition temperatures (TCr–Ss) increased
same trend as the TCr–Ss. On the second cooling process, the
isotropic–smectic transition temperatures (TI–Ss) and smectic–
crystalline transition temperatures (TS–Crs) exhibited the same
trend as the TCr–Ss and TS–Is during the heating process. This
could be because the increase in alkoxy chain lengths (by a few
methylene units) is much less significant in contrast to the large
volume of the whole POSS molecule. As a result, there are no
significant variations for transition temperatures.
In contrast, more substitution with long alkoxy chains in
mesogens has a significant effect on the thermal behaviour of the
POSS-based molecules, and it will lead the mesophase to disap-
pear due to the depression of the order of the LC orientation.
Additional long flexible alkoxy chains in the mesogens will
impose extra freedom on the motion of these flexible terminal
chains and, thus, significantly reduce the phase transition
temperaturꢀes. For instance, the melting temperature of 7h is
about 200 C lower than that of TS–I of 7a–d. Similar observa-
tions have been reported in some star-like hydrogen-bonded
LCs.33 When one alkoxy chain is attached to the periphery of
mesogens, hydrogen bonded complexes form LCs. However,
when two or three long alkoxy chains (OC12H25) are attached to
the periphery of mesogens, hydrogen bonded LCs disappear.
ꢀ
from 213 to 227 C for OC5H12 and OC8H17, respectively, and
then slightly decreased to 119 ꢀC for OC12H25. Likewise,
smectic–isotropic transition temperatures (TS–Is) followed the
Fig. 4 Texture changes of 7c in the temperature range of 244 to 117 ꢀC on heating. All the textures observed by POM correspond to the phase
transitions in the DSC diagram.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5248–5257 | 5251

View Article Online
Table 2 Flexible group density (FGD) in the periphery of 7a–h
FGD FGD
4
3
4
3
ꢀ
ꢀ
POSS
(ꢂ10 /A )
POSS
(ꢂ10 /A )
7a
7b
7c
7d
3.16
3.05
2.92
2.65
7e
7f
7g
7h
16.5
9.30
15.3
11.3
whole POSS molecule and produce a more disordered packing
structure than 7a–d. Therefore, 7e–h behave more liquid-like and
do not exhibit any mesomorphic behaviour (Fig. 6).
Incorporation of a few flexible groups into a mesogen is
a common method to make a liquid crystal. Multiple substitution
with flexible groups in the peripheral mesogens results in the
formation of columnar or other liquid crystalline structures in
a variety of self-assemblies. For discotic-like or planar molecular
LCs, either the mesogen itself acts as the central core, or meso-
gens directly link to a planar central core34 through covalent
linkages, e.g., an ester linkage, to develop LC states as illustrated
in Fig. 7a. Development of LC state, however, usually occurs
when there is no flexible spacer between the mesogens and the
planar central core, and multiple (di- or tri-) substitution in the
mesogens will not lead to loss of LCs. In contrast, liquid crys-
talline spherical molecules derived from a three-dimensional
core, such as a cyclotriphosphazene ring or POSS core, could be
simply categorised into two typical models in which they contain
Fig. 5 Polarized optical photograph of (a) 7c at 244 ꢀC on cooling; (b) 7c
at 200 ꢀC on cooling; (c) 7d at 213 ꢀC, on cooling; (d) 7d at 175 ꢀC, on
cooling.
Substituted POSS is considered to be a spherical molecule with
a volume estimation based on the length of the corner substitu-
ents and size of the inorganic core, using the Chem 3D Pro
program. We introduced the term ‘‘flexible group density’’
(FGD), which is capable of estimating how many flexible units
per unit volume are in the periphery of the POSS molecule and
thus can act as an indicator to determine/predict the formation of
LCs. The FGD in the periphery of the POSS molecules is defined
as the ratio of the total number of carbon atoms in the flexible
chains to the periphery volume that flexible chains occupy, which
is calculated by subtracting the volume of the central core,
consisting of the POSS inorganic core, spacer and mesogenic
part, from the whole POSS molecule. The FGD is calculated in
terms of the following equation, where R and r represent the radii
of the whole POSS molecule and the central core consisting of the
POSS inorganic core, spacer and mesogenic parts, respectively.
Flexible group density in the periphery
8 ꢂ no: of carbon atoms in the alkoxy groups
¼
À
Á
4
3
p R3 ꢁ r3
As mentioned above, the POSS can be regarded as a spherical
molecule, so the value of r is equal to the sum of the radius of the
POSS inorganic core, the length of spacer and the length of
mesogen; whereas, the value of R is equal to the sum of r and the
length of the flexible chain.19 Based on the above definition, the
overall volume of a POSS molecule with mesogens substituted by
a single flexible group is approximately close to that of a POSS
molecule with mesogens substituted by two or three identical
flexible groups. Therefore, an increase in the number of flexible
groups on the mesogens will lead to an increase in FGD
(Table 2). The calculated FGD values for 7a–d ranging from 3.16
ꢂ 10ꢁ4 to 2.65 ꢂ 10ꢁ4/A are much smaller than those of 7e–f
3
ꢀ
(16.5 ꢂ 10ꢁ4 and 9.30 ꢂ 10ꢁ4/A , respectively) and 7g–h (15.3 ꢂ
3
ꢀ
3
10ꢁ4 and 11.3 ꢂ 10ꢁ4/A , respectively), suggesting that additional
ꢀ
Fig. 6 The effect of peripheral flexible groups in mesogens on the
long flexible substitutions on 7e–h result in excess flexibility to the
formation of liquid crystals.
5252 | J. Mater. Chem., 2011, 21, 5248–5257
This journal is ª The Royal Society of Chemistry 2011

View Article Online
8.05, 13.20, 19.59, 22.0 and 27.48ꢀ. A strong scattering at a low
angle of 4.75ꢀ (2q) indicates the presence of a lamellar structure.
When it was heated up from room temperature to 120 ꢀC,
a strong, sharp scattering at 4.71ꢀ (2q) remained, accompanied
by two strong diffraction peaks at 19.4ꢀ (2q) and 7.0ꢀ (2q). The
strong liquid-like broad scattering at 7.0ꢀ (2q), corresponding to
ꢀ
the d-spacing of 12.7 A, is due to the Si–O clusters in the POSS
core.16,19,21,36,37 Similar observations have been reported in some
siloxane-based liquid crystalline polymers and in some siloxane
core-based LCs.38 The first order diffraction is too weak to be
observed within the angular range of the XRD experiment. The
ꢀ
ꢀ
strong scattering at 4.71 corresponds to a d-spacing of 18.8 A,
indicating a second-order diffraction peak. The volume of 7c is
ꢀ
estimated to be 43 A by the Chem 3D Pro program. As the
ꢀ
measured d-spacing of the second-order diffraction is 19 A, the
corresponding d-spacing for the first-order diffraction should be
ꢀ
equal to 38 A, which is shorter than the modelled estimation,
indicating the presence of the interdigitated structure involving the
overlapping of the flexible chains between neighbouring layers.
To exactly identify the phase transitions, XRD graphs from
200 to 250 ꢀC with the calculated d-spacings of the second-
order diffractions are shown in Fig. 9. When the measuring
Fig. 7 The schematic representations of three types of LCs derived from
a planar molecule (a), a three-dimensional core with (b) or without
a spacer between the internal core and mesogens (c).
or do not contain spacers between the three-dimensional central
core and mesogens, as shown in Fig. 7b and c, respectively. If
there is no spacer, it will result in a strong coupling between the
central core and mesogens. Similar to the molecule with a planar
core, multiple substitution in this model causes changes only in
the architecture of molecular assembly and, thus, gives rise to
different mesomorphisms instead of the disappearance of LC.
For example, cyclotriphosphazene,35 in which each mesogen
contains only one flexible chain, yields a nematic phase. In
contrast, mesogenic moieties that consist of two or three long
flexible chains exhibit columnar mesophases. In this study, the
spacer length between the POSS core and mesogens is equivalent
to thirteen CH2 units, which is long enough to make the POSS
core and mesogens decouple. Thus, the formation of LC will
largely rely on the relative magnitude of FGD in the periphery,
multiple substitutions will lead to more flexible units surrounding
the periphery of the molecule, and, hence, the molecule is more
likely to behave as a liquid. This liquid-like behaviour, in fact,
could be evidenced by the dramatic drop in the melting
temperatures of 7e (ꢁ8.5 ꢀC) and 7g (39.7 ꢀC) compared with
ꢀ
temperature increases from 200 to 240 C, the d-spacing of the
second-order diffraction remains unchanged. When temperature
ꢀ
increases from 240 to 250 C, the d-spacing of the second-order
ꢀ
diffraction increases from 18.4 to 19.0 A. This gives strong
evidence for the formation of the tilted SmC phase. In contrast to
a SmA phase, in which the director is perpendicular to the
smectic plane, POSS molecules in a SmC phase assemble to an
arrangement approximately parallel to layers in which the
director is titled with respect to the normal layer. In terms of
d-spacing of the second-order diffraction, the tilt angle is esti-
mated to be 15ꢀ. In fact, the change in the d-spacing of the
second-order diffractions at the low angle is in good agreement
with the texture changes observed by POM, which reveal the
characteristic broken fan of a SmC phase at 200 ꢀC and conic fan
texture of a SmA phase at 244 ꢀC (Fig. 4).
Conclusions
In summary, we synthesised a series of organic–inorganic hybrid
POSS-based LCs. The chemical structures of 7a–h were charac-
terised by 1H, 13C, and 29Si NMR, mass spectroscopy and
elemental analysis. Temperature-dependent X-ray diffraction
was used to verify mesophases, revealing that 7a–d showed SmC
and SmA phases. The DSC also showed that minor exothermic
and endothermic transitions with very small enthalpy changes
were detected. These subtle transitions corresponded to the
texture changes observed by POM. Three molecular models and
the FGD concept were proposed to address the effect of the
flexible chain on the formation of LCs. The number and length of
the FGD in the periphery surrounding the mesogenic and POSS
core have a significant effect on determining the formation of
LCs, particularly when the coupling between the central core
and mesogens is minimised. The POSS-based LCs with mesogens
substituted by single flexible chains tended to form smectic
mesophases. When the mesogens were substituted by two or
three flexible chains, LCs disappeared. These observations
strongly supported the hypothesis that the FGD in the periphery
ꢀ
ꢀ
that of 7d (TCr–S: 119 C and TS–I: 211 C), which has the same
molecular weight as 7e and 7g.
Based on the above discussion, we can conclude that the
number and length of peripheral groups are important factors not
only to control the phase transition parameters and architectures
of assembly, but also to determine the formation of mesophase.
Therefore, it would be essential to select the suitable number and
length of peripheral flexible groups when designing and synthe-
sising star-like and dendrimer-like liquid crystalline materials.
X-Ray diffraction study of POSS-based LCs
Variable-temperature wide angle XRD was performed to verify
the liquid crystal phases observed. Using 7c as an example, the
X-ray diffraction data of 7c on heating are summarised in
Table 3. Fig. 8 shows the variable-temperature wide angle X-ray
diffraction (WAXD) of compound 7c. At room temperature, it is
crystalline and shows multiple diffraction peaks at 4.75, 7.09,
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5248–5257 | 5253

View Article Online
Table 3 X-ray diffraction data of 7c at different temperatures on hea-
tingb
of the POSS-based star-like molecules is a key feature for
determining the formation of LCs. This study could offer a useful
and pragmatic guideline to predict the formation of LCs when
designing and synthesising star-like, or even dendrimer-like,
liquid crystalline materials. Further studies focusing on how the
coupling between a POSS core with reduced spacer length and
mesogens, as well as the peripheral flexible groups, jointly and
interdependently affects the formation of LCs are currently
under way in our group.
2q/d-spacings
T/ꢀC
diffractions (2q)
ꢀ
d-spacing (A)
25
4.75, 7.09, 8.05,a 13.20,
19.59, 22.0, 27.48
4.71, 6.96, 19.42
4.69, 6.87, 19.17
4.63, 6.86, 18.86
4.53, 7.07, 18.54
4.79, 7.22, 17.89
4.79, 7.27, 17.27
4.79, 7.36, 17.05
4.66, 7.33, 17.00
18.59, 12.46, 10.98,
6.70, 4.53, 4.04, 3.24
18.75, 12.69, 4.57
18.83, 12.86, 4.63
19.07, 12.88, 4.70
19.50, 12.50, 4.78
18.44, 12.24, 4.96
18.44, 12.15, 5.13
18.44, 12.00, 5.20
18.96, 12.06, 5.22
120
140
160
180
200
220
240
250
Experimental section
Instrumentation
1
The H, 13C and 29Si NMR spectra were acquired on a Bruker
a
Shoulder peak. b Very weak peak.
DCR 400 MHz spectrometer in deuterated chloroform with
tetramethylsilane as an internal standard. The operating
1
frequency for H, 13C and 29Si are 400.1, 100.6 and 79.5 MHz,
respectively. The coupling constant is expressed as ‘‘J’’ with units
of Hz. Fourier transform infrared (FTIR) spectra were recorded
on a Perkin-Elmer Fourier transform infrared spectrophotom-
eter 2000 using KBr pellets. Thermal behaviors of compounds
were studied on a TA instrument DSC 2920 under a heating and
cooling rate of 5 ꢀC minꢁ1 in nitrogen. The peak temperatures of
the endo-thermo and exo-thermo were taken as the phase tran-
sition temperatures. The photomicrographs of the complexes
were taken under a Nikon OPTIPHOT2-POL polarizing optical
microscope fitted with
a hot stage using a temperature
programmer TP_93 and a TMC-6RGB1/20 color CCD camera.
The sample was placed on a glass slide, cover with a glass cover
slip and heated on the hot stage at 10 ꢀC minꢁ1 and cooled at 2 ꢀC
minꢁ1. X-ray diffraction pattern was recorded on a Bruker
GADDS X-ray diffractometer equipped with an external heating
ꢀ
element using Ka radiation wavelength (l ¼ 1.5418 A). The tube
voltage and amperage were set to 40 kV and 30 mA, respectively,
and 2q step scans of 0.02ꢀ and a counting time of 0.2 s per step
were used for all XRD measurements. The powdered samples
were packed into very thin-walled capillary tubes, heated to their
melt state and air-cooled to the room temperature before the
X-ray diffraction patterns were recorded. Electron ionization
mass (EIMS) spectra were recorded on a Micromass 7034E mass
spectrometer. Elemental analysis was conducted on a Perkin-
Elmer 240C elemental analyzer for C, H, and N determination.
Fig. 8 X-ray diffraction patterns from 25 to 250 ꢀC.
Materials
Octakis(dimethylsiloxy)silsesquioxane (6) was purchased from
Hybrid Plastics and dried in an oven at 40 ꢀC overnight prior to
use. 11-Bromo-1-undecene (2) was purchased from Alfa Aesar.
N,N-Dicyclohexylcarbodiimide and 4,40-dihydroxybiphenyl and
other chemicals were bought from Sigma-Aldrich and used as
received. Monoalkoxy, dialkoxy or trialkoxybenzoic acids were
prepared according to reported methods.33
Synthesis of compounds
40-(Undec-10-enyloxy)biphenyl-4-ol (4). 11-Bromo-1-undecene
(1.18 mL, 5.38 mmol), 4,40-biphenol (5.00 g, 26.85 mmol) and
potassium carbonate (3.71 g, 26.85 mmol) were added into dry N,
N-dimethylformamide (DMF) (100 mL). The reaction mixture
Fig. 9 X-ray diffraction patterns at the low angle region at the
temperature from 200 to 250 ꢀC.
5254 | J. Mater. Chem., 2011, 21, 5248–5257
This journal is ª The Royal Society of Chemistry 2011

View Article Online
was stirred at 80 ꢀC for 48 h and then solvent was removed under
reduced pressure. Dichloromethane was added into the solid
residue. The suspension was stirred at room temperature for 15
min and then was filtered off. The dichloromethane was removed
and the residue was recrystallized in dichloromethane to afford
white solid 3. Yield: 0.76 g, 42%.
1H NMR d [ppm]: 7.43(t, overlapped dd, 4H, Ar–H), 6.94
(d, 2H, J ¼ 8.4 Hz, Ar–H), 6.88 (d, 2H, J ¼ 8.4 Hz, Ar–H), 5.82
(m, 1H, HC]CH2), 5.02 (m, 2H, HC]CH2), 4.79(s, 1H, OH),
3.98(t, 2H, J ¼ 6.4 Hz, OCH2), 2.05(m, 2H, CH2HC]CH2), 1.79
(m, 2H, OCH2CH2), 1.51–1.25 (br, 12H, (CH2)6). 13C NMR
d [ppm]: 158.60, 154.86, 139.47, 134.11, 133.52, 128.19, 127.93,
115.85, 115.09, 114.37. Anal. Calc for C23H30O2: C 81.62%, H
8.93%, found: C 81.35%, H 8.69%.
40-(Undec-10-enyloxy)biphenyl-4-yl 4-(octyloxy)benzoate (5c).
1
Yield: 66%. H NMR d [ppm]: 8.08 (d, 2H, J ¼ 8.4 Hz, Ar–H),
7.51 (d, 2H, J ¼ 8.4 Hz, Ar–H), 7.43 (d, 2H, J ¼ 8.4 Hz, Ar–H),
6.90 (overlapped, dd, 4H, J ¼ 8.8 Hz, Ar–H), 5.74 (m, 1H, HC ¼
CH2), 4.89 (m, 2H, HC]CH2), 3.98 (t, 2H, J ¼ 6.4 Hz, OCH2),
3.93 (t, 2H, J ¼ 6.4 Hz, OCH2), 1.98 (m, 2H, CH2HC ¼ CH2),
1.74 (m, 4H, OCH2CH2), 1.46–1.35 (m, 4H, OCH2CH2CH2),
1.34–1.16 (m, 18H, (CH2)9), 0.83 (t, 3H, J ¼ 6.8 Hz, CH3). 13C
NMR d [ppm]: 165.36, 163.93, 159.16, 150.43, 139.54, 138.91,
133.22, 132.63, 128.44, 128.01, 122.32, 122.03, 115.23, 114.69,
114.46, 68.72, 68.51, 34.13, 32.15, 29.86, 29.77, 29.73, 29.67,
29.55, 29.47, 29.46, 29.29, 26.41, 26.345, 22.99, 14.40. EI-MS
(m/z): Found: 570.5. Calc: 570.8. Anal. Calc for C38H50O4: C
79.96%, H 8.83%, found: C 79.63%, H 8.59%
40-(Undec-10-enyloxy)biphenyl-4-yl 4-(dodecyloxy)benzoate (5d).
Yield: 56%. 1H NMR d [ppm]: 8.15 (d, 2H, J ¼ 8.8 Hz, Ar–H), 7.57
(d, 2H, J ¼ 8.4 Hz, Ar–H), 7.50 (d, 2H, J ¼ 8.4 Hz, Ar–H), 7.23 (d,
2H, J ¼ 8.8 Hz, Ar–H), 6.97 (overlapped, dd, 4H, J ¼ 8.8 Hz, Ar–
H), 5.81 (m, 1H, HC]CH2), 4.98 (m, 2H, HC]CH2), 4.04 (t, 2H, J
¼ 6.4 Hz, OCH2), 3.99 (t, 2H, J ¼ 6.4 Hz, OCH2), 2.06 (m, 2H,
CH2HC]CH2), 1.81 (m, 4H, OCH2CH2), 1.44 (m, 4H,
OCH2CH2CH2), 1.38–1.27 (m, 26H, (CH2)13), 0.89 (t, 3H, J ¼ 6.4
Hz, CH3). 13C NMR d [ppm]: 165.37, 163.94, 159.16, 150.43, 139.55,
138.92, 133.23, 132.64, 128.45, 128.02, 122.03, 121.55, 115.23,
114.70, 114.47, 68.73, 68.52, 34.14, 32.27, 30.01, 29.98, 29.94, 29.91,
29.87, 29.78, 29.74, 29.72, 29.69, 29.67, 29.47, 29.30, 26.42, 26.35,
23.03, 14.44. EI-MS (m/z): Found: 626.5. Calc.:626.9. Anal. Calc for
C42H58O4: 80.47%, H 9.33%, found: C 80.30%, H 9.04%
General synthetic procedures of 5a–h
4-n-Alkoxybenzoic acid (0.71 mmol), 40-hydroxybiphenyl unde-
cene-1-yl ether (0.20 g, 0.59 mmol) and dimethylaminopyridine
(DMAP) (0.030 g, 0.24 mmol) were dissolved in 13.0 mL of dry
dichloromethene. The reaction mixture was purged with argon gas
for 10 min before the addition of dicyclohexylcarbodiimide (DCC)
(0.19 g, 0.89 mmol) in 4.0 mL dry dichloromethene. The mixture
was then refluxed for 16 h under argon gas protection. After the
reaction is complete, the resulting suspension was removed by
filtration. The filtrate was washed with water and dried, and the
organic solvent was removed. The residue was column chroma-
tographed on silica gel using a mixture of dichloromethane and
hexane (V : V ¼ 65 : 35) as eluent to afford 5 as a white solid.
40-(Undec-10-enyloxy)biphenyl-4-yl 4-(pentyloxy)benzoate (5a).
Yield: 73%. 1H NMR d [ppm]: 8.09 (d, 2H, J ¼ 8.8 Hz, Ar–H), 7.50
(d, 2H, J ¼ 8.8 Hz, Ar–H), 7.44 (d, 2H, J ¼ 8.4 Hz, Ar–H), 7.17 (d,
2H, J ¼ 8.4, Ar–H), 6.90 (dd, 4H, J ¼ 8.8 Hz, Ar–H), 5.76 (m, 1H,
HC]CH2), 4.90 (m, 2H, HC]CH2), 3.98 (t, 2H, J ¼ 6.6 Hz
OCH2), 3.93 (t, 2H, J ¼ 6.6 OCH2) 1.98 (m, 2H, CH2HC ¼ CH2),
1.74 (m, 4H, OCH2CH2), 1.40 (m, 4H, OCH2CH2CH2), 1.38–1.19
(m, 12H, (CH2)6), 0.88(t, 3H, J ¼ 6.8, CH3). 13C NMR d [ppm]:
165.45, 163.94, 159.17, 150.43, 139.56, 138.93, 133.24, 132.65,
128.45, 128.02, 122.33, 122.04, 115.24, 114.70, 114.47, 68.72, 68.53,
34.15, 29.87, 29.78, 29.74, 29.68, 29.48, 29.30, 29.17, 28.51, 26.42,
22.78, 14.32. EI-MS (m/z): Found: 528.4. Calc.: 528.7. Anal. Calc
for C35H44O4: C 79.51%, H 8.39%, found: C 79.05%, H 8.16%.
40-(Undec-10-enyloxy)biphenyl-4-yl 3,5-di(hexyloxy)benzoate
(5e). Yield: 78%. 1H NMR d [ppm]: 7.60 (d, 2H, J ¼ 8.5 Hz, Ar–H),
7.53 (d, 2H, J ¼ 10.1 Hz, Ar–H), 7.37 (s, 2H, Ar–H), 7.27 (d, 2H, J ¼
8.6 Hz, Ar–H), 6.99 (d, 2H, J ¼ 8.7 Hz, Ar–H), 6.75 (s, 1H, Ar–H),
5.84 (m, 1H, HC]CH2), 4.98 (m, 2H, HC]CH2), 4.02 (m, 6H,
OCH2), 2.08 (m, 2H, CH2HC]CH2), 1.82 (m, 6H, OCH2CH2),
1.49–1.20 (m, 24H, (CH2)12), 0.97 (t, 6H, J ¼ 6.8 Hz, CH3). 13C
NMR (CDCl3) d [ppm]: 165.59, 160.71, 159.20, 150.29, 139.64,
139.13, 133.14, 131.62, 128.51, 128.12, 122.27, 115.23, 114.53, 108.62,
107.55, 38.83, 68.52, 34.21, 31.96, 31.92, 29.79, 29.70, 29.55, 29.34,
26.46, 26.10, 23.00, 14.43. EI-MS (m/z): Found, 642.6; Calc. 642.4
40-(Undec-10-enyloxy)biphenyl-4-yl 3,5-di(dodecyloxy)benzoate
(5f). Yield: 60%. 1H NMR d [ppm]: 7.58 (d, 2H, J ¼ 8.6 Hz, Ar–
H), 7.51 (d, 2H, J ¼ 8.6 Hz, Ar–H), 7.33 (s, 2H, Ar–H), 7.24
(d, 2H, J ¼ 8.8 Hz, Ar–H), 6.97 (d, 2H, J ¼ 8.7 Hz, Ar–H), 6.71
(s, 1H, Ar–H), 5.84 (m, 1H, HC]CH2), 4.98 (m, 2H, HC]
CH2), 4.00 (m, 6H, OCH2), 2.05 (m, 2H, CH2HC]CH2), 1.80
(m, 6H, OCH2CH2), 1.49–1.20 (m, 54H, (CH2)), 0.88 (t, 6H, J ¼
6.5 Hz, CH3). 13C NMR (CDCl3) d [ppm]: 165.51, 160.75, 159.24,
150.36, 139.57, 139.12, 133.19, 131.69, 128.47, 128.07, 122.22,
115.29, 114.47, 108.70, 107.64, 68.87, 68.56, 34.15, 32.28, 30.00,
29.95, 29.88, 29.74, 29.70, 29.58, 29.48, 29.32, 26.40, 23.03, 14.43.
EI-MS (m/z): Found, 810.9; Calc. 810.6
40-(Undec-10-enyloxy)biphenyl-4-yl 4-(heptyloxy)benzoate (5b).
Yield: 71%. 1H NMR d [ppm]: 8.10 (d, 2H, J ¼ 8.8 Hz, Ar–H),
7.51 (d, 2H, J ¼ 8.8 Hz, Ar–H), 7.44 (d, 2H, J ¼ 8.8 Hz, Ar–H),
7.17 (d, 2H, J ¼ 8.4 Hz, Ar–H), 6.90 (overlapped, dd, 4H, J ¼
8.8 Hz, Ar–H), 5.81 (m, 1H, HC]CH2), 4.95 (m, 2H, HC ¼
CH2), 3.98 (t, 2H, J ¼ 6.4, 6.8 Hz, OCH2), 4.02 (t, 2H, J ¼ 6.4,
6.8 Hz, OCH2), 2.03 (m, 2H, CH2HC ¼ CH2), 1.81 (m, 4H,
OCH2CH2), 1.50 (m, 4H, OCH2CH2CH2), 1.31–1.19 (m, 16H,
(CH2)8), 0.90 (t, 3H, J ¼ 6.6 Hz, CH3). 13C NMR d [ppm]:
165.36, 163.93, 159.16, 150.43, 139.54, 138.91, 133.23, 132.63,
128.44, 128.00, 122.31, 122.03, 115.23, 114.70, 114.46, 68.72,
68.51, 34.14, 32.11, 29.86, 29.77, 29.73, 29.67, 29.48, 29.46,
29.37, 29.29, 26.41, 26.31, 22.94, 14.39. EI-MS (m/z): Found:
556.4. Calc. 556.7. Anal. Calc for C37H48O4: C 79.82%, H
8.69%, found: C 79.58%, H 8.48%
40-(Undec-10-enyloxy)biphenyl-4-yl
3,4,5-tris(butyloxy)
benzoate (5g). Yield: 76%. 1H NMR d [ppm]: 7.59 (d, 2H, J ¼ 8.4
Hz, Ar–H), 7.51 (d, 2H, J ¼ 8.8 Hz, Ar–H), 7.43 (s, 2H, Ar–H),
7.23 (d, 2H, J ¼ 8.4 Hz, Ar–H), 6.97 (d, 2H, J ¼ 8.8 Hz, Ar–H),
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5248–5257 | 5255

View Article Online
5.84 (m, 1H, HC]CH2), 4.98 (m, 2H, HC]CH2), 4.07 (m, 6H,
OCH2), 4.00 (t, 2H, J ¼ 6.8 Hz, 2.05 (m, 2H, CH2HC]CH2),
1.84–1.73 (m, 8H, OCH2CH2), 1.60–1.20 (m, 20H, (CH2)5), 1.00
(t, 9H, J ¼ 6.5 Hz, CH3). 13C NMR (CDCl3) d [ppm]: 165.46,
159.23, 153.41, 150.41, 143.64, 139.56, 139.18, 133.18, 128.46,
128.06, 124.39, 122.30, 115.28, 114.47, 109.20, 73.60, 69.45,
68.55, 34.15, 32.74, 31.76, 29.87, 29.78, 29.74, 29.69, 29.48, 29.31,
26.43, 19.63, 19.52, 14.16. EI-MS (m/z): Found, 658.6; Calc.
658.4
23.05, 22.79, 18.07, 14.45, 14.34, 0.34. 29Si NMR d [ppm]: 12.33,
ꢁ108.65.
Octakis[(11-(40-(1-(4-heptyloxybenzoyloxy)biphenyl-4-yloxy)
undecyl)dimethylsiloxy)]silsesquioxane (7b)
Yield: 56% 1H NMR d [ppm]: 8.14 (d, 16H, J ¼ 8.8 Hz, Ar–H),
7.55 (d, 16H, J ¼ 8.4 Hz, Ar–H), 7.48 (d, 16H, J ¼ 8.8 Hz, Ar–
H), 7.22 (overlapped with CDCl3, 16H, Ar–H), 6.97 (overlapped
dd, 32H, Ar–H), 4.03 (t, 16H, J ¼ 0.66 Hz, OCH2), 3.96 (t, 16H,
J ¼ 6.6 Hz, OCH), 1.79 (m, 32H, OCH2CH2), 1.46 (m, 32H,
OCH2CH2CH2), 1.32–1.19 (m, 160H, (CH2)10), 0.92 (t, 24H, J ¼
40-(Undec-10-enyloxy)biphenyl-4-yl
3,4,5-tris(dodecyloxy)
benzoate (5h). Yield: 42%. 1H NMR (400 MHz, CDCl3)
d [ppm]: 7.59 (d, 2H, J ¼ 8.8 Hz, Ar–H), 7.52 (d, 2H, J ¼ 8.8
Hz, Ar–H), 7.43 (s, 2H, Ar–H), 7.23 (d, 2H, J ¼ 8.4 Hz, Ar–
H), 6.98 (d, 2H, J ¼ 8.8 Hz, Ar–H), 5.81 (m, 1H, HC]CH2),
4.95 (m, 2H, HC]CH2), 4.04 (m, 8H, OCH2), 2.04
(m, 2H, CH2HC]CH2), 1.82 (m, 8H, OCH2CH2), 1.49 (m,
8H, OCH2CH2CH2), 1.38–1.27 (m, 58H, (CH2)29), 0.88
(m, 9H, CH3). 13C NMR (CDCl3) d [ppm]: 165.44, 159.20,
153.36, 150.38, 143.56, 139.52, 139.03, 133.14, 128.43,
128.02, 124.34, 122.28, 115.24, 114.46, 109.14, 73.95, 69.72,
68.51, 34.13, 32.29, 32.27, 30.72, 30.10, 30.08, 30.07,
30.04, 30.00, 29.98, 29.92, 29.86, 29.77, 29.75, 29.73, 29.70,
29.66, 29.46, 29.29, 26.45, 26.43, 26.41, 23.04, 23.03,
14.42. EI-MS (m/z): Found: 995.0. Calc. 995.5. Anal. Calc
for C66H108O4: 79.63%, H 10.93%, found: C 79.49%, H
10.24%.
6.8 Hz, CH3), 0.60 (m, 16H, SiCH2), 0.13 (s, 48H, Si(CH3)2). 13
C
NMR d [ppm]: 165.35, 163.92, 159.18, 150.41, 138.89, 133.18,
132.64, 128.43, 127.99, 122.32, 122.05, 115.21, 114.69, 68.72,
68.53, 33.87, 32.12, 30.13, 30.11, 30.06, 29.92, 29.84, 29.77, 29.50,
29.39, 26.52, 26.32, 23.39, 22.95, 18.13, 14.40, 0.34. 29Si NMR
d [ppm]: 12.93, ꢁ108.60.
Octakis[(11-(40-(1-(4-octyloxybenzoyloxy)biphenyl-4-yloxy)
undecyl)dimethylsiloxy)]silsesquioxane (7c)
Yield: 27%. 1H NMR d [ppm]: 8.14 (d, 16H, J ¼ 8.4 Hz, Ar–H),
7.55 (d, 16H, J ¼ 8.4 Hz, Ar–H), 7.48 (d, 16H, J ¼ 8.4 Hz, Ar–
H), 7.22 (d, 16H, J ¼ 8.4 Hz, Ar–H), 6.97 (overlapped dd, 32H,
Ar–H), 4.03 (t, 16H, J ¼ 6.4 Hz, OCH2), 3.97 (t, 16H, J ¼ 6.4 Hz,
OCH2), 1.80 (m, 32H, OCH2CH2), 1.48 (m, 32H, OCH2-
CH2CH2), 1.32–1.26 (m, 176H, (CH2)11), 0.88 (t, 24H, J ¼ 6.8
Hz, CH3), 0.60 (m, 16H, SiCH2), 0.13 (s, 48H, Si(CH3)2). 13C
NMR d [ppm]: 165.35, 163.92, 159.17, 150.41, 138.89, 133.17,
132.64,128.43, 127.99, 122.32, 122.04, 115.20, 114.69, 68.72,
68.52, 33.87, 32.16, 30.13, 30.11, 30.06, 29.91, 29.84, 29.80, 29.76,
29.69, 29.57, 29.49, 26.52, 26.36, 23.39, 23.00, 18.12, 14.42, 0.34.
29Si NMR d [ppm]: 12.70, ꢁ108.68.
Octakis[(11-(40-(1-(4-alkoxybenzoyloxy)biphenyl-4-yloxy)
undecyl)dimethylsiloxy)]silsesquioxane (7a–h)
General synthetic procedures of 7a–h. 40-Hydroxybiphenyl
undecene-1-yloxy-4-n-pentylbenzoate (50.0 mg, 0.095 mmol)
was dissolved in 3 mL of dry toluene followed by octakis
(hydrodimethylsiloxy)octasilsesquioxane (POSS) (11.2 mg, 0.011
mmol). The reaction mixture was purged with nitrogen gas for
30 min before Pt(dvs) (10.2 mL) was added. The reaction was
stirred at 42 ꢀC under nitrogen for 12 h. After that, toluene was
removed under reduced pressure. The crude product was
repeatedly column chromatographed on silica gel using
dichloromethane–hexane (V : V ¼ 1 : 1) and then dichloro-
methane and ethyl acetate (V : V ¼ 95 : 5) as eluent to afford
pure 7a–h. The relatively low yields for some of POSS molecules
are because of repeated column chromatography.
Octakis[(11-(40-(1-(4-dodecyloxybenzoyloxy)biphenyl-4-yloxy)
undecyl)dimethylsiloxy)]silsesquioxane (7d)
Yield: 56%. 1H NMR: d [ppm]: 8.14 (d, 16H, J ¼ 8.4 Hz, Ar–H),
7.55 (d, 16H, J ¼ 8.4 Hz, Ar–H), 7.48 (d, 16H, J ¼ 8.8 Hz, Ar–
H), 7.22 (d, 16H, J ¼ 7.2 Hz, Ar–H), 6.93 (overlapped dd, 32H,
Ar–H), 4.03 (t, 16H, J ¼ 6.8 Hz, OCH2), 3.96 (t, 16H, J ¼ 6.4 Hz,
OCH2), 1.79 (m, 32H, OCH2CH2), 1.48 (m, 32H, OCH2-
CH2CH2), 1.38–1.27 (m, 240H, (CH2)15), 0.88 (t, 24H, J ¼ 6.8
Hz, CH3), 0.60 (m, 16H, SiCH2), 0.13 (s, 48H, Si(CH3)2). 13C
NMR d [ppm]: 165.34, 163.93, 159.18, 150.42, 138.89, 133.18,
132.64, 128.43, 127.99, 122.32, 122.05, 115.21, 114.69, 68.73,
68.53, 33.84, 32.28, 30.13, 30.11, 30.08, 30.06, 30.01, 29.99, 29.94,
29.91, 29.83, 29.76, 29.73, 29.70, 29.49, 26.51, 26.35, 23.38, 23.03,
18.12, 14.44, 0.33. 29Si NMR d [ppm]: 12.75, ꢁ108.68.
Octakis[(11-(40-(1-(4-pentyloxybenzoyloxy)biphenyl-4-yloxy)
undecyl)dimethylsiloxy)]silsesquioxane (7a)
Yield: 15%. 1HNMR d [ppm]: 8.15 (d, 16H, J ¼ 8.8 Hz, Ar–H),
7.55 (d, 16H, J ¼ 8.4 Hz, Ar–H), 7.48 (d, 16H, J ¼ 8.4 Hz, Ar–
H), 7.26 (d, overlapped with CDCl3, 16H, Ar–H), 6.97 (over-
lapped dd, 32H, Ar–H), 4.11 (t, 16H, J ¼ 6.4 Hz, OCH2), 4.02
(t, 16H, J ¼ 6.4 Hz, OCH2), 1.81 (m, 32H, OCH2CH2), 1.46 (m,
32H, OCH2CH2CH2), 1.38–1.23 (m, 128H, (CH2)8), 0.95 (t, 24H,
J ¼ 7.05, CH3), 0.60 (m, 16H, SiCH2), 0.13 (s, 48H, Si(CH3)2).
13C NMR d [ppm]: 165.37, 163.91, 159.15, 150.40, 138.90, 133.18,
132.64, 128.44, 128.01, 122.33, 122.02, 115.20, 114.67, 68.70,
68.52, 33.89, 32.29, 30.05, 30.02, 29.72, 29.17, 28.51, 26.05, 23.35,
Octakis[(11-(40-(1-(3,5-dihexyloxybenzoyloxy)biphenyl-4-
yloxy)undecyl)dimethylsiloxy)]silsesquioxane (7e)
Yield: 73%. 1H NMR d [ppm]: 7.56 (d, 16H, J ¼ 8.4 Hz, Ar–H),
7.48 (d, 16H, J ¼ 8.4 Hz, Ar–H), 7.31 (s, 16H, Ar–H), 7.22(d,
16H, J ¼ 8.4 Hz, Ar–H), 6.95 (d, 16H, J ¼ 8.8 Hz, Ar–H), 6.70
(s, 8H, Ar–H), 3.98 (m, 48H, OCH2), 1.80 (m, 48H, OCH2CH2),
1.60–1.20 (m, 224H, (CH2)14), 0.62 (m, 16H, Si–CH2), 0.14
(s, 48H, Si–CH3). 13C NMR (CDCl3) d [ppm]: 165.56, 160.70,
5256 | J. Mater. Chem., 2011, 21, 5248–5257
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Chemistry, vol. 57, ed. R. West, The Netherlands, Elsevier Inc.,
2008, pp. 1–116.
2 D. B. Cordes, P. D. Lickiss and F. Rataboul, Chem. Rev., 2010, 110,
2081–2173.
3 R. Y. Kannan, H. J. Salacinski, P. E. Butler and A. M. Seifalian, Acc.
Chem. Res., 2005, 38, 879–884.
159.20, 150.27, 139.09, 133.09, 131.61, 128.48, 128.08, 122.27,
115.19, 108.60, 107.52, 68.82, 68.52, 33.94, 31.96, 30.17, 30.12,
29.96, 29.90, 29.79, 29.55, 26.56, 26.10, 23.43, 23.00, 18.15, 14.43,
0.11. 29Si NMR (CDCl3) d [ppm]: 12.78, ꢁ108.71
4 H. Liu, S. Zheng and K. Nie, Macromolecules, 2005, 38, 5088–
5097.
5 J. Miao, L. Cui, H. P. Lau, P. T. Mather and L. Zhu, Macromolecules,
2007, 40, 5460–5470.
Octakis[(11-(40-(1-(3,5-didodecyloxybenzoyloxy)biphenyl-4-
yloxy)undecyl)dimethylsiloxy)]silsesquioxane (7f)
Yield: 25%.1H NMR d [ppm]: 7.57 (d, 16H, J ¼ 8.4 Hz, Ar–H),
7.49 (d, 16H, J ¼ 8.4 Hz, Ar–H), 7.32 (s, 16H, Ar–H), 7.23
(d, 16H, J ¼ 8.4 Hz, Ar–H), 6.95 (d, 16H, J ¼ 8.8 Hz, Ar–H),
6.71 (s, 8H, Ar–H), 4.00 (m, 48H, J ¼ 6.4 Hz, OCH2), 1.80 (m,
16H, OCH2CH2), 1.60–1.20 (m, 464H, (CH2)26), 0.89 (t, 48H,
J ¼ 6.4 Hz, CH3), 0.62 (m, 16H, Si–CH2), 0.15 (s, 48H, Si–CH3).
13C NMR (CDCl3) d [ppm]: 165.49, 160.75, 159.24, 150.35,
139.10, 133.16, 131.69, 128.45, 128.04, 122.22, 115.26, 108.70,
107.62, 68.86, 68.58, 33.83, 32.29, 30.00, 29.96, 29.75, 29.59,
29.52, 26.52, 26.41, 23.37, 23.04, 18.11, 14.43, 0.07. 29Si NMR
(CDCl3) d [ppm]: 12.75, ꢁ108.65.
6 L. Zhang, Q. Yang, H. Yang, J. Liu, H. Xin, B. Mezari,
P. C. M. M. Magusin, H. C. L. Abbenhuis, R. A. van Santen and
C. Li, J. Mater. Chem., 2008, 18, 450–457.
7 K. Y. Mya, Y. Wang, 1 L. Shen, J. Xu, Y. Wu, X. Lu and C. He,
J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 4602–4616.
8 J. Xu, X. Li, C. M. Cho, C. L. Toh, L. Shen, K. Y. Mya, X. Lu and
C. He, J. Mater. Chem., 2009, 19, 4740–4745.
9 J. Huang, C. He, X. Liu, J. Xu, C. S. S. Tay and S. Y. Chow, Polymer,
2005, 46, 7018–7027.
10 S. S. Nobre, C. D. S. Brites, R. A. S. Ferreira, V. Z. Bermudez,
C. Carcel, J. J. E. Moreau, J. Rocha, M. C. M. Wong and
L. D. Carlos, J. Mater. Chem., 2008, 18, 4172–4182.
11 A. Strachota, P. Whelan, J. Kriz, J. Brus, M. Urbanova, M. Slouf and
L. Matejka, Polymer, 2007, 48, 3041–3058.
12 M. J. Abad, L. Barral, D. P. Fasce and R. J. J. Williams,
Macromolecules, 2003, 36, 3128–3135.
Octakis[(11-(40-(1-(3,4,5-tributyloxybenzoyloxy)biphenyl-4-
yloxy)undecyl)dimethylsiloxy)]silsesquioxane (7g)
13 I. A. Zucchi, M. J. Galante, R. J. J. Williams, E. Franchini, J. Galy
and J. F. Gerard, Macromolecules, 2007, 40, 1274–1282.
14 E. A. Drylie, C. D. Andrews, M. A. Hearshaw, C. Jimenez-
Rodriguez, A. Slawin, D. J. Cole-Hamilton and R. E. Morris,
Polyhedron, 2006, 25, 853–858.
15 L. Cui, J. P. Collet, G. Xu and L. Zhu, Chem. Mater., 2006, 18, 3503–
3512.
16 F.-H. Kreuzer, R. Maurer and P. Spes, Makromol. Chem. Macromol.
Symp., 1991, 50, 215–228.
17 I. M. Saez and P. Styring, Adv. Mater., 1996, 8, 1001–1005.
18 R. M. Laine, C. Zhang, A. Sellinger and L. Viculis, Appl. Organomet.
Chem., 1998, 12, 715–723.
19 C. X. Zhang, T. J. Bunning and R. M. Laine, Chem. Mater., 2001, 13,
3653–3662.
20 M. C. Gravel, C. Zhang, M. Dinderman and R. M. Laine, Appl.
Organomet. Chem., 1999, 13, 329–336.
21 I. M. Saez and J. W. Goodby, Liq. Cryst., 1999, 26, 1101–1105.
Yield: 57%. 1H NMR d [ppm]: 7.58 (d, 16H, J ¼ 8.4 Hz, Ar–H),
7.49 (d, 16H, J ¼ 8.4 Hz, Ar–H), 7.42 (s, 16H, Ar–H), 7.22
(d, 16H, J ¼ 8.4 Hz, Ar–H), 6.95 (d, 16H, J ¼ 8.8 Hz, Ar–H),
4.06–3.9 (m, 48H, OCH2), 1.80 (m, 64H, OCH2CH2), 1.60–1.20
(m, 224H, (CH2)14), 0.98 (m, 72H, CH3), 0.60 (m, 16H, Si–CH2),
0.14 (s, 48H, Si–CH3). 13C NMR (CDCl3) d [ppm]: 165.43,
159.25, 153.41, 150.43, 143.67, 139.04, 133.16, 131.20, 129.19,
128.44, 128.03, 124.40, 122.30, 115.26, 109.24, 73.60, 72.16,
69.47, 68.58, 33.87, 32.75, 31.77, 30.82, 30.12, 29.85, 29.33, 26.54,
23.40, 19.63, 19.52, 18.15, 14.35, 14.15, 0.09. 29Si NMR (CDCl3)
d [ppm]: 13.03, ꢁ108.51
€
22 R. Elsaber, G. H. Mehl, J. W. Goodby and D. J. Photinos, Chem.
Commun., 2000, 851–852.
Octakis[(11-(40-(1-(3,4,5-tridodecyloxybenzoyloxy)biphenyl-4-
yloxy)undecyl)dimethylsiloxy)]silsesquioxane (7h)
23 M. O’Neill and S. M. Kelly, Adv. Mater., 2003, 15, 1135–1146.
24 I. M. Saez, J. W. Goodby and R. M. Richardson, Chem.–Eur. J.,
2001, 7, 2758–2764.
Yield: 45%. 1H NMR d [ppm]: 7.56 (d, 16H, J ¼ 8.4 Hz, Ar–H),
7.48 (d, 16H, J ¼ 8.8 Hz, Ar–H), 7.41 (s, 16H, Ar–H), 7.22
(d, 16H, J ¼ 8.4 Hz, Ar–H), 6.94 (d, 16H, J ¼ 8.8 Hz, Ar–H),
4.03 (m, 64H, OCH2), 1.81 (m, 64H, OCH2CH2), 1.52–1.41
(m, 64H, OCH2CH2CH2), 1.41–1.16 (m, 496H, (CH2)31), 0.88
(m, 72H, CH3), 0.61 (m, 16H, SiCH2), 0.13 (s, 48H, Si(CH3)2).
13C NMR d [ppm]: 153.37, 128.44, 128.02, 122.30, 115.22, 109.13,
69.74, 33.88, 32.30, 32.28, 30.74, 30.11, 30.10, 30.05, 30.01, 29.99,
29.94, 29.76, 29.74, 29.71, 26.46, 23.39, 23.04, 23.03, 18.12, 14.43,
0.34. 29Si NMR (CDCl3) d [ppm]: 13.08, ꢁ108.59
25 I. M. Saez and J. W. Goodby, J. Mater. Chem., 2001, 11, 2845–2851.
26 C. X. Zhang and R. M. Laine, J. Am. Chem. Soc., 2000, 122, 6979–
6988.
27 C. Keith, G. Dantlgraber, R. A. Reddy, U. Baumeister, M. Prehm,
H. Hahn, H. Lang and C. Tschierske, J. Mater. Chem., 2007, 17,
3796–3805.
28 J. Miao and L. Zhu, J. Phys. Chem. B, 2010, 114, 1897–1887.
29 C. Ornelas, D. Mery, E. Cloutet, J. R. Aranzaes and D. Astruc, J. Am.
Chem. Soc., 2008, 130, 1495–1506.
30 G. H. Mehl and I. M. Saez, Appl. Organomet. Chem., 1999, 13, 261–272.
31 Q. Pan, X. Chen, X. Fan, Z. Shen and Q. Zhou, J. Mater. Chem.,
2008, 18, 3481–3488.
32 K. D. Gresham, C. M. McHugh, T. J. Bunning, R. L. Crane,
H. E. Klei and E. T. Samulski, J. Polym. Sci., Part A: Polym.
Chem., 1994, 32, 2039–47.
33 J. Xu, T. C. Ling and C. He, J. Polym. Sci., Part A: Polym. Chem.,
2008, 46, 4691–4703.
Acknowledgements
This work was funded by the Institute of Materials Research and
Engineering (IMRE), A*STAR (Agency for Science, Technology
and Research), Singapore.
34 H. C. Holst, T. Pakula and H. Meier, Tetrahedron, 2004, 60, 6765–6775.
35 J. Barbera, M. Bardaji, J. Jimenez, A. Laguna, M. P. Martinez,
L. Oriol, J. L. Serrano and I. Zaragozano, J. Am. Chem. Soc., 2005,
127, 8994–9002.
36 G. H. Mehl and J. W. Goodby, Angew. Chem., Int. Ed. Engl., 1996,
35, 2641–2643.
37 G. H. Mehl, A. J. Thornton and J. W. Goodby, Mol. Cryst. Liq.
Cryst., 1999, 332, 455–461.
38 M. F. Achard, S. Lecommandoux and F. Hardouin, Liq. Cryst., 1995,
19, 581–587.
References and notes
1 P. D. Lickiss and F. Rataboul, Fully Condensed Polyhedral
Oligosilsesquioxanes (POSS): From Synthesis to Application, in:
A. F. Hilland and M. J. Fink, Advances in Organometallic
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 5248–5257 | 5257