Separation and identification of stereoisomers of a tetrameric side- chain liquid crystalline cyclic siloxane

Article abstract of DOI:10.1039/a807003b

For the first time the four stereoisomers of a tetrameric side-chain liquid crystalline cyclic siloxane have been isolated in a pure state by high performance liquid chromatography. The structure of one of the stereoisomers was confirmed by ¹H, ¹³C and ²⁹Si NMR. The structures of the remaining three stereoisomers were assigned by using evidence from the literature and their liquid crystalline thermal properties. All four stereoisomers and the isomeric mixture exhibited calamitic SmC and SmA phases; no discotic phases were found.

Full text of DOI:10.1039/a807003b

J O U R N A L O F
Separation and identification of stereoisomers of a tetrameric side-
chain liquid crystalline cyclic siloxane
T. E. Mann,a J. Haleyb and D. Laceyb
aPETA Solutions, Perkin Elmer Ltd, Chalfont Road, Seer Green, Beaconsfield, UK HP9 2FX
bDepartment of Chemistry, Faculty of Science and the Environment, University of Hull, Hull, UK
HU6 7RX
C H E M I S T R Y
Received 8th September 1998, Accepted 7th December 1998
For the first time the four stereoisomers of a tetrameric side-chain liquid crystalline cyclic siloxane have been
isolated in a pure state by high performance liquid chromatography. The structure of one of the stereoisomers was
confirmed by 1H, 13C and 29Si NMR. The structures of the remaining three stereoisomers were assigned by using
evidence from the literature and their liquid crystalline thermal properties. All four stereoisomers and the isomeric
mixture exhibited calamitic SmC and SmA phases; no discotic phases were found.
Introduction
CH3
CH3
SiO
Pt Catalyst
Toluene
The synthesis of cyclic polysiloxanes of structure 1 has been
carried out for a number of years resulting in a wide range of
cyclic materials being prepared, some of which are now
commercially available.
SiO
H
+
n
n
CH2CH2
Mesogenic side chain
2
X = H, Me or Ph
Y = H, Me or Ph
n = 4 to 200
X
Fig. 1 The hydrosilylation reaction.
SiO
crystalline (SCLC) siloxanes of structure 2 (Fig. 1), based on
the cyclic tetramer (structure 2, n=4) and pentamer (structure
2, n=5), can also give rise to four stereoisomers. A schematic
representation of the four stereoisomers for a SCLC cyclic
tetramer is given in Fig. 2.
Y
n
1
Although
a substantial number of side-chain liquid
Most research has been undertaken on the methylphenyl-
cyclosiloxanes (structure 1, X=Me, Y=Ph) and the dimethyl-
cyclosiloxanes (structure 1, X=Y=Me) due to the fact that
they form stable cyclic polymers. Research work has also been
carried out into the different isomeric forms of these mate-
rials.1,2 When the cyclic siloxane backbone is synthesised, it is
possible to form a number of different stereoisomers for each
ring system. If we take, for example, the tetramethyltetraphe-
nylcyclosiloxane ring, it can be shown that there are four
stereoisomers formed in the preparation of the cyclic tetramer.3
These stereoisomers are formed because at each silicon atom
it is possible to have a substituent in one of two possible
spatial positions. Upon formation of the rings, as the silane
molecules interact, the spatial arrangement of the substituents
on the silicon atom governs the final arrangement when the
ring closes.
crystalline cyclic siloxanes of general structure 2 have been
prepared from the tetramethyltetrahydrogensiloxane cyclic
backbone, no attempt has ever been made to separate and
evaluate the effect of each of the individual isomers on the
thermal and physical properties of the isomeric mixture.
In this paper we detail, for the first time, the separation,
characterisation and thermal properties of the four stereo-
isomers of a SCLC cyclic tetramer.
Results and discussion
A number of structure–property correlation studies have been
carried out on cyclic SCLC siloxanes, to investigate the role
of molecular structure of (a) the polymer backbone, (b) the
After the ring has closed and formed, the arrangement of
the substituent groups is effectively fixed. The ring can undergo
the usual translational and vibrational motions, but unless the
ring is cleaved by thermal or chemical methods, the spatial
arrangement at each silicon atom will remain unchanged. It is
this ‘randomness’ of interaction of the silane molecules that
gives rise to the isomeric forms for the cyclic siloxanes.
The cyclic siloxanes that are of interest today are not
disubstituted siloxanes, but are in fact the methylhydro-
gencyclosiloxanes (structure 1, X=Me and Y=H), which can
be used in coatings, inks and adhesive technologies.4
In recent years the ability of these cyclic siloxanes to react
with an alkenic group attached to a mesogenic side chain has
led to the preparation of a wide range of side-chain liquid
crystalline polymers.5–9 This hydrosilylation reaction is shown
in Fig. 1.
=
Mesogenic Side Group
Fig. 2 Representation of the four stereoisomers for a SCLC cyclic
Like the methylphenylcyclosiloxanes, side-chain liquid
tetramer.
J. Mater. Chem., 1999, 9, 353–360
353

spacer group and (c) the mesogenic side chain, on the thermal
and physical properties of the polymer. However, the separa-
tion of the stereoisomers of a SCLC cyclic siloxane has never
been attempted and therefore it is not known how each
individual isomer contributes to the overall thermal properties
of the mixture. The existence of a discotic phase for these
types of materials has been predicted by molecular modelling
using the Monte Carlo technique.10,11 Further calculations
undertaken using different computer molecular modelling pro-
grammes have also identified the possibility of the existence
of a discotic-type mesophase for these cyclic siloxanes,
although many of these studies do suggest that the formation
of a calamitic mesophase is more likely.12,13 A large number
of SCLC cyclic siloxanes have been prepared incorporating a
wide range of differing mesogenic side groups and spacer
groups, but they all exhibit calamitic phases.
It is possible that the overall thermal properties of a SCLC
cyclic siloxane is being dominated by just one of the stereoiso-
mers which happens to exhibit a calamitic phase, and that one
or more of the remaining stereoisomers are forming discotic
phases. On the other hand it is also possible that the computer
modelling of these systems, which predicts discotic phases,
may be incorrect and that all the isomers are calamitic. The
only way to answer these questions is to separate the individual
isomers of a SCLC cyclic siloxane and to investigate their
phase behaviour.
the case. After many pilot studies the separation of isomer 4
was achieved with relatively little difficulty using our prepara-
tive isocratic HPLC systems. However, sample preparation of
the 3 remaining isomers proved problematic because the
removal of isomer 4 from the isomeric mixture led to a
dramatic change in the solubility of the 3 remaining isomers
in the chloroform–acetonitrile mixture, a mixture that was
used as the mobile phase in our HPLC systems. Because of
the solubility problem and the close proximity of isomer 3 to
isomers 1 and 2 in the HPLC chromatogram, it was found
necessary to change the HPLC system to a preparative gradient
HPLC system. By using a ‘cut-off’ method, isomer 3 was
separated.
The separation of the remaining two isomers was very
difficult and a number of studies using analytical HPLC had
to be carried out to obtain the best conditions for separation.
It was found necessary to use both our ‘cut-off’ method and
a column oven, set at a temperature of 30 °C, to achieve the
separation.
A detailed description of the procedures used to carry
out the separation of the stereoisomers is given in the
Experimental section.
Confirmation of structure and thermal properties of isomers
The separation of the four isomers was only the first part of
this investigation. The next step was to identify the structure
of the stereoisomers and examine their liquid crystalline
behaviour.
The separation of stereoisomers
All the following NMR spectra were carried out by Professor
J. W. Emsley at the University of Southampton. The transition
temperatures and phase identification were carried out at Hull
using DSC and optical light microscopy. When necessary X-
ray diffraction was used to identify the phases and this was
carried out by Dr R. M. Richardson at the University of
Bristol.
The separation of the individual stereoisomers of the SCLC
cyclic siloxanes of structure 3 was carried out using high
performance liquid chromatography (HPLC). Details of the
preparation of the cyclic tetramer of structure 3 and a detailed
description of the HPLC equipment and procedures used to
separate the four stereoisomers are given in the Experimental
section.
The structure of stereoisomer 3 was confirmed by 1H, 13C
and 29Si NMR spectroscopy. Stereoisomers 1, 2 and 4 all
showed similar 1H, 13C and 29Si NMR spectra due to the
symmetry of the mesogenic side groups and therefore, such
data could not be used to elucidate the structure of these three
stereoisomers. The structures of stereoisomers 1, 2 and 4 were
tentatively assigned using evidence from the literature and
their liquid crystalline behaviour.
CH3
SiO
4
Stereoisomer 3
CO2
OC8H17
(CH2)6O
The separation of this isomer was regarded as very important,
because a knowledge of the phase behaviour of this isomer
would give some indication as to the factors that influence the
phase behaviour of the bulk material. This isomer accounts
for 53% of the total composition of the isomeric mixture
(see Table 1).
The structure of the pure stereoisomer 3 was confirmed
using 1H, 13C and 29Si NMR. The NMR spectra for stereo-
isomer 3 are shown in Fig. 3 (29Si), 4 (13C) and 5 (1H), and
it is the only stereoisomer where the NMR data are
unambiguous regarding its structure.
The 29Si NMR spectrum (Fig. 3) gave a clear indication as
to the structure of stereoisomer 3 because there is only one
structure that could lead to more than one type of silicon
atom peak appearing in the NMR spectrum. A schematic
representation of this arrangement of methyl and mesogenic
side groups is shown in Fig. 6.
K1 78.4 K2 86.0 SmC 103.6 SmA 108.8 °C Iso
3
Using our HPLC assembly and data handling software, we
were able to determine the percentages of the individual
stereoisomers in the isomeric mixture and this information is
given in Table 1. These results clearly show the presence of
four stereoisomers for the SCLC cyclic siloxane studied and
that one isomer, isomer 3, constitutes over half the composition
of the isomeric mixture. Although it is possible to separate
these isomers by preparative HPLC, our earlier studies have
shown that the separation of isomers 1, 2 and 3 (especially
isomers 1 and 2) will be difficult. Indeed, this proved to be
Table 1 The percentages of the four stereoisomers in the LC isomeric
mixture
There are two silicon atoms, a and b in Fig. 6, which carry
an axially positioned methyl group but the methyl groups on
adjoining silicon atoms are cis (c) and trans (d). Thus the
silicon atoms a and b have similar environments. In the case
of the silicon atom c, although the methyl group is again
axially positioned on the silicon atom, the adjoining methyl
groups on silicon atoms a and b are both cis. This would
Isomer number
% Isomer in mixture
1
2
3
4
13
28
53
6
354
J. Mater. Chem., 1999, 9, 353–360

H3C
Si
c
CH3
Si
Si
b
CH3
O
Si
a
O
O
O
d
CH3
=
Mesogenic Side and Spacer Groups
Fig. 6 Schematic representation of the structure for stereoisomer 3.
29Si NMR that the structure of the stereoisomer 3 is that
shown in Fig. 6; three of the mesogenic side groups are
attached to the same plane of the ring backbone, the remaining
mesogenic side group is attached to the opposite ring face.
The following transitions were found for isomer 3. The
transitions in brackets are for the isomeric mixture.
Fig. 3 The 29Si NMR spectrum for stereoisomer 3.
K 80.0 K 86.5 SmC 108.8 SmA 111.7 °C Iso
1
2
(K 78.4 K 86.0 SmC 103.6 SmA 108.8 °C Iso)
1
2
The spatial arrangement of the four mesogenic side groups
around the siloxane ring in stereoisomer 3 would cause the
cyclic tetramer to adopt the calamitic ‘bundle-type’ configur-
ation as shown in Fig. 7 (a). This would mean that the material
would be unlikely to pack in a way that would lead to the
formation of a discotic phase.
The transition temperatures for stereoisomer 3 differ only
very slightly from those of the isomeric mixture. This was not
unexpected since isomer 3 constitutes over half the isomeric
mixture. The clearing point (SmA–Iso) and the SmC–SmA
transition temperature are higher (2.9 and 5.2 °C respectively)
for stereoisomer 3 than for the isomeric mixture.
Fig. 4 The 13C NMR spectrum for stereoisomer 3.
Stereoisomers 1, 2 and 4
The 1H, 13C and 29Si NMR spectra for these three stereo-
isomers were all identical and therefore such data could not
be used to assign the configuration of stereoisomers 1, 2 and
4. To assign the configuration of these three stereoisomers, we
used both evidence from the literature and their liquid
crystalline behaviour.
From the work of Beevers and Semlyen14 on the separation
and identification of stereoisomers of cyclic tetramethyltetra-
phenylsiloxanes, they found that the amounts of the stereoiso-
mers for undiluted and solution equilibrates of this cyclic
tetramer, and the ratio of the stereoisomeric cyclics produced
by the non-equilibrium pyrolysis reaction, were very close to
the isomer ratio predicted from the random cyclisation of an
atactic polysiloxane. These results are given in Table 2. The
Fig. 5 The 1H NMR spectrum for stereoisomer 3.
mean that silicon atom c has a very slightly different environ-
ment to that experienced by silicon atoms a or b. However,
this subtle difference between the environments of silicon atom
c and silicon atoms a and b could not be resolved by 29Si
NMR and therefore all three silicon atoms appear as a single
peak in the spectrum at −20.08 ppm. In the case of the silicon
atom d the methyl group is equatorially positioned and appears
as a single peak at −20.18 ppm. The ratio of the two single
peaks is 3 (−20.08)51 (−20.18).
(c) 'Compact' structure
(a) 'Bundle-type' structure (b) 'Elongated' structure
=
=
Polysiloxane Ring
Mesogenic Side Chain
Both the 13C NMR spectrum (Fig. 4) and 1H NMR
spectrum (Fig. 5) support the strong structural evidence from
Fig. 7 Schematic representations for the different arrangements for
the stereoisomers.
J. Mater. Chem., 1999, 9, 353–360
355

Table 2 The percentages of tetramers 1 to 4 in different isomeric
Table 2 (undiluted equilibrate) the percentage ratio of tetramer
152 in the isomeric mixture is 8518, but in column 4 (prep-
aration by pyrolysis) the ratio is 16510. However, unlike
previous work on cyclic siloxanes we can now use the liquid
crystalline behaviour of our SCLC cyclic tetramers to try to
predict the configuration of stereoisomers 1 and 4. This
evidence will also substantiate that the configuration of
stereoisomer 2 is that of tetramer 3 in Fig. 8.
mixtures
% Stereoisomers in different mixtures
Equilibrate
Stereoisomers, as
defined in Fig. 8
Predicted
Undiluted
Solution
Pyrolysis
Tetramer 1
Tetramer 2
Tetramer 3
Tetramer 4
12.5
50.0
25.0
12.5
8.0
49.0
25.0
18.0
10.0
49.0
23.0
18.0
16.0
51.0
24.0
10.0
Liquid crystalline behaviour of the stereoisomers
The liquid crystalline behaviour of the four stereoisomers for
our SCLC cyclic tetramer are given in Table 3. The most
important fact that comes out of the information given in
Table 3 is that all the stereoisomers exhibit calamitic phases;
no discotic phases. Whether this will be true for all SCLC
cyclic siloxanes we cannot say, but in this case there was no
evidence of discotic phases.
assignment of the configuration of the cyclic tetramers 1–4
from Table 2 are given in Fig. 8, and these were derived by
Beevers and Semlyen.14 If we now compare these results with
those given in Table 1 for our cyclic SCLC tetramer, we can
clearly see that tetramer 2 (Table 2, Fig. 8) has the same basic
type of structure as stereoisomer 3 (Table 1, Fig. 6), and that
both contribute half the composition of the isomeric mixture.
This gave us strong evidence that, provided that all the Si–H
sites on the cyclic tetramethyltetrahydrogensiloxane had been
substituted by the mesogenic side chain, then the ratio of the
stereoisomers in our cyclic SCLC tetramer mixture should be
very similar to those for the cyclic methylphenylsiloxane
tetramer mixture given in Table 2. The loss of the Si–H signal
in the IR and the absence of any other peaks, except for the
presence of the four stereoisomers, in the HPLC chromatogram
for our cyclic SCLC tetramer (see Experimental section),
showed that we had 100% substitution of the Si–H by the
mesogenic side chain. Furthermore, some unpublished work
by J. A. Semlyen, University of York, showed that the cyclic
tetramethyltetrahydrogensiloxane used in this work was a
random atactic cyclic polymer similar to the cyclic methylphe-
nylsiloxane used in the work presented in Table 2. This enabled
us to predict, with some certainty, that stereoisomer 2 has the
same structure as tetramer 3, both contributing to about 25%
of the composition of their respective isomeric mixtures.
The configuration of stereoisomers 1 and 4 could not be
predicted from the information given in Table 2 since the
relative ratios of tetramers 1 and 4 in Table 2 depended on
equilibration and the mode of preparation, i.e. in column 2 of
The more ordered phase S found in stereoisomer 4 in
1
Table 3 was originally identified by optical microscopy as a
smectic B phase. X-Ray crystallography studies carried out by
Dr R. M. Richardson at the University of Bristol proved that
this was not the case. Although X-ray data confirm that there
is a transition at 102.5 °C as shown by DSC (very small peak)
and optical microscopy, the data clearly show that this is not
a smectic C to smectic B transition but merely a subtle
structure change from the normal SmC phase to a more highly
ordered smectic C phase. The DSC data also confirm this fact
in that the enthalpy for the transition is 1.58 J g−1, which is
smaller than would be expected for a smectic C to smectic B
transition. The formation of this more ordered SmC phase is
obviously due to the ability of the pure isomer to pack much
more efficiently than the other three stereoisomers. This
increase in ordering due to the packing of the mesogenic side
groups is indicative of the structures given for tetramers 1 and
3 in Fig. 8, which would mean that stereoisomers 2 and 4
must have the configuration of tetramers 1 and 3, since both
stereoisomers have high SmC and SmA thermal stabilities.
This would mean that stereoisomer 1 must have the configur-
ation of tetramer 4. In Fig. 8, if the black oblong shapes were
mesogenic side chains and not benzene rings, then the more
‘elongated’ structure of tetramer 3, compared to the more
‘compact’ structure of tetramer 1 [see Fig. 7 (b) and (c)] would
lead to this cyclic SCLC tetramer exhibiting the higher clearing
point, but the spatial arrangement of the mesogenic side chains
in tetramer 3 would also be more detrimental to the formation
of crystal (K, K and K ) phases than that given for tetramer
H3C
O
H3C
O
1
2
Si
Si
CH3
1. Since stereoisomer 2 in Table 3 has only one crystal–crystal
transition at 58.8 °C and this is considerably lower than the
two crystal–crystal transitions given for stereoisomer 4, this
led us to believe that stereoisomer 2 has the structure of
tetramer 3 and stereoisomer 4 that of tetramer 1. The alternate
spatial arrangement of the mesogenic side chains in tetramer
4 would be detrimental to the formation of crystal, SmC and
SmA phases, and this was a good reason for us to believe that
stereoisomer 1 has the structure of tetramer 4.
Si
Si
H3C
H3C
O
O
Si O
CH3
Si O
CH3
Si
Si
O
O
CH3
Tetramer 2
Tetramer 1
H3C
Si
H3C
CH3
CH3
In the case of stereoisomers 1 and 2, where the mesogenic
side groups are split evenly between the facial planes of the
siloxane ring, these arrangements of the mesogenic side chains
have manifested low crystal to SmC transitions, and this gives
rise to very wide SmC ranges being exhibited by these two
tetramers. It is interesting to note that these two stereoisomers
also give the widest SmA ranges, wider than that observed for
the isomeric mixture.
H3C
O
CH
Si
3
Si
Si
O
Si O
O
Si
Si O
CH3
Si
O
O
O
CH3
Tetramer 4
Tetramer 3
=
Conclusions
We have, for the first time, successfully separated and identified
the stereoisomers of a SCLC cyclic siloxane. In the case of
stereoisomer 3 the structure of the isomer was confirmed by
Fig. 8 The assignment of the configuration of cyclic tetramers 1–4
in Table 2.
356
J. Mater. Chem., 1999, 9, 353–360

Table 3 The thermal data for the four stereoisomers and the LC isomeric mixture
Transition temperature/°C
Isomer
K –K
K –SmC
K –S
K–SmC
S –SmC
SmC–SmA
SmA–I
SmC range
SmA range
1
2
2
2
1
1
Mix
78.4
39.2
86.0
60.0
103.6
94.1
108.8
108.8
114.0
108.8
102.1
117.5
111.7
116.1
17.6
34.1
50.0
22.3
25.8a
5.2
8.0
8.7
2.9
2.1
1
2
3
4
58.8
80.0
76.8
86.5
88.2
102.5
aThis range includes the S phase range.
1
1H, 13C and 29Si NMR but for stereoisomers 1, 2 and 4 their
structures were deduced using evidence from the literature
and their liquid crystalline behaviour. The configurations of
the four stereoisomers of the cyclic tetramer and their
corresponding transitions are given in Fig. 9.
Although a number of computer studies have predicted that
these types of cyclic siloxanes should exhibit discotic phases,
all four stereoisomers and the isomeric mixture gave calamitic
SmC and SmA phases; no discotic phases were found. It is
interesting to note that besides the ordered SmC phase exhib-
ited by stereoisomer 4, all four stereoisomers gave the same
liquid crystalline phase sequence as the isomeric mixture, i.e.
SmC–SmA–Iso. No other phases were found for either the
four stereoisomers or the isomeric mixture.
chromatography conditions12 unless stated otherwise. The
stationary phase used was Sorbsil C60 (40–60 mm).
Analytical high performance liquid chromatography
(HPLC) was carried out using a reversed phase HPLC column
(25×0.46 cm, Dynamax, Microsorb, 5 mm C18 column) and
a Spectroflow 757 UV detector (254 nm) with data handling
facilities.
The compounds outlined in Scheme 1 have relatively simple
structures and in our view the NMR, IR and MS data given
are more than adequate to confirm the structure of these
compounds. We believe that microanalysis would convey no
additional information as to the structure of our compounds
and, in this case, is not needed. 4-Octyloxyphenol is a known
compound.15
The structure of stereoisomer 4, where all four mesogenic
side groups lie on the same face of the siloxane ring, is
conducive to the formation of the SmC phase, whereas the
‘two up, two down’ arrangement of the mesogenic side chains
(pairs of mesogenic side chains are split evenly between the
two facial planes of the siloxane ring) in stereoisomer 2 is
more conducive to the formation of the SmA phase. The
alternating arrangement of the mesogenic side chains found in
stereoisomer 1 is detrimental to the formation of both the
SmC and SmA phases to such an extent that this is the only
stereoisomer where the thermal stability of either the SmC or
SmA phase is lower than that found for the isomeric mixture.
The synthetic route to the preparation of the SCLC cyclic
polymer is given in Scheme 1.
Preparation of 4-octyloxy-1-benzyloxybenzene
4-Benzyloxyphenol (10.00 g, 0.05 mol), 1-bromooctane
(9.65 g, 0.05 mol), potassium carbonate (34.00 g, 0.25 mol)
and potassium iodide (0.10 g) in dry butanone (100 ml) were
heated under reflux, with stirring, overnight. The potassium
salts were removed by filtration (Hyflo) and the solvent
removed by distillation under reduced pressure. The crude
ether was then purified by column chromatography on silica
gel, eluting with dichloromethane, and the resultant product
recrystallised (ethanol), to yield 4-octyloxy-1-benzyloxyben-
zene as a white crystalline solid (purity, 100% by HPLC).
Experimental
Yield=12.30 g (79%), mp 97–98 °C. 1H NMR (CDCl ) d 7.35
Preparation of the SCLC cyclic polysiloxane
3
(4H, m), 6.85 (2H, m), 5.00 (2H, s), 3.90 (2H, t), 1.75 (2H,
1H Nuclear magnetic resonance spectroscopy (NMR) were
carried out on a JMN GX270FT spectrometer, using deuter-
ated chloroform as a solvent and tetramethylsilane as the
internal standard. The peak types in the NMR spectra are
denoted by the following notations: singlet (s), doublet (d),
triplet (t), quartet (q) and multiplet (m).
Infrared spectra were obtained using a Perkin Elmer 487G
or a Perkin Elmer 983G spectrophotometer. Samples were
prepared as either potassium bromide discs, or, if liquid, were
run on single crystal sodium chloride discs.
m), 1.35 (10H, m), 0.90 (3H, m). IR (KBr) n
2960, 2940,
max
2920, 2860, 1510, 1250 cm−1. MS m/z 312 (M+), 200, 109,
91 (100%).
Preparation of 4-octyloxyphenol
4-Octyloxy-1-benzyloxybenzene (10.00 g, 0.03 mol), cyclohex-
ene (4.00 g, 0.05 mol) and 5% palladium on charcoal (0.2 g)
in ethanol (100 ml) and glacial acetic acid (20 ml) were heated
under reflux, with stirring, for 3 h. After filtration (Hyflo) the
solvent was removed by distillation under reduced pressure.
The crude phenol was purified by recrystallisation (ethanol)
to yield 4-octyloxyphenol as a white powder (purity, >99%
by HPLC). Yield=5.80 g (88%), mp 61.0 °C (lit.,15
Mass spectra were obtained using a Finigan Mat 1020
Automated GCMS Spectrometer. Results are quoted where
M+ represents the molecular ion and the base peak represented
by (100%).
60.0–61.0 °C). 1H NMR (CDCl ) d 7.40 (2H, m), 6.85 (2H,
Differential Scanning Calorimetry (DSC) thermograms were
obtained using a Perkin Elmer DSC 7, with a TAC 7/PC
interface and a controlled cooling accessory. The heating rate
was initially 20 °C min−1 and subsequent heating and cooling
at 10 °C min−1. Calculations were made using Perkin Elmer
PC based software. The instrument was calibrated against an
indium standard (melting temperature 156.6 °C, DH 28.45
J g−1).
3
m), 5.00 (1H, s), 3.90 (3H, t), 1.75 (2H, m), 1.35 (10H, m),
0.90 (2H, m). IR (KBr) n
3600–3250, 2960, 2930, 2920,
max
2860, 1510, 1240 cm−1. MS m/z 222 (M+), 178, 167, 149,
136, 123, 110 (100%).
Preparation of 4-octyloxyphenyl 4∞-(hex-5-enyloxyphenyl)
benzoate
Optical microscopy was performed using an Olympus BH2
polarising microscope, fitted with a Mettler FP52 hot stage
and a Mettler FP5 controller. Samples were prepared as thin
films between a glass slide and a glass cover slip.
4-(Hex-5-enyloxy)benzoic acid (3.00 g, 10 mmol), 4-octyloxy-
phenol (2.70 g, 10 mmol) and dimethylaminopyridine (0.25 g,
1.2 mmol) were dissolved in dry dichloromethane (100 ml)
under anhydrous conditions. N∞,N-Dicyclohexylcarbodiimide
(2.54 g, 10 mmol) in dry dichloromethane (15 ml) was added
Column chromatography was carried out under flash
J. Mater. Chem., 1999, 9, 353–360
357

5.00 (1H, m), 4.05 (2H, t), 3.95 (2H, t), 2.15 (2H, m), 1.80
(4H, m), 1.6 (2H, m), 1.50–1.25 (10H, m), 0.90 (3H, t). IR
(KBr) n
3000–2840, 1720, 1600, 1500, 1260, 760 cm−1. MS
H3C
Si
H3C
max
m/z 424 (M+), 312, 109, 91 (100%), 71, 57.
CH3
Si
O
Si
O
Si
Preparation of SCLC cyclic tetramer
O
O
CH3
The cyclic tetramer backbone (0.25 g), 4-octyloxyphenyl 4∞-
(hex-5-enyloxyphenyl)benzoate (1.8 g) and Speiers catalyst
(4–6 drops of a prepared solution) were dissolved in dry
toluene (15 cm3), and heated with stirring at 60 °C under
anhydrous conditions for 24 h. The progress of the reaction
was monitored using IR spectroscopy by examining the spectra
in the region of 2155 cm−1 (Si–H fingerprint absorption). The
disappearance of this band indicates completion of the
hydrosilylation reaction.
Upon completion any undissolved material was filtered off
and the toluene was removed by distillation under reduced
pressure. The SCLC cyclic tetramer was reprecipitated by
dissolving the tetramer in the minimum amount of dry
dichloromethane and then excess dry methanol was added (at
least four times the volume of the dichloromethane). The
resultant suspension was then centrifuged at 4500 rpm for
30 min, the liquid was then decanted off. The process was
repeated until TLC analysis (silica gel, 951 petroleum fraction
(bp 40–60 °C)–diethyl ether) showed that the monomer had
been removed.
K
1 39.2 K2 60.0 SmC 94.1 SmA 102.1 °C Iso
Stereoisomer 1
CH3
Si
H3C
Si
CH3
Si
O
Si
O
O
O
CH3
K1 58.8 SmC 108.8 1 SmA 117.5 °C Iso
Stereoisomer 2
The SCLC cyclic tetramer was then redissolved in dry
dichloromethane and filtered through a disposable syringe
filter (PTFE membrane 0.45 mm) to remove any particulates.
The solvent was carefully removed by distillation under
reduced pressure and the tetramer dried in vacuo. 1H NMR
H3C
Si
CH3
Si
H3C
O
Si
O
Si
(CDCl ) d 8.25 (2H, d), 7.15 (2H, d), 6.90 (2H, d), 6.85 (2H,
O
O
3
CH3
d), 4.05 (2H, t), 3.95 (2H, t), 2.15 (2H, m), 1.80 (4H, m),
1.65 (2H, m), 1.10–1.50 (14H, m), 0.90 (3H, t), 0.05 (12H,
m). IR (KBr) n
1560, 1280, 760 cm−1.
3000–2840, 1950, 1875, 1850, 1720, 1605,
max
K1 80.0 K2 86.5 SmC 108.8 SmA 111.7 °C Iso
A sample of the SCLC cyclic tetramer was sent to ICI
Wilton for MALDI-TOF. The MALDI-TOF mass spec-
trometer was operated in the reflectron mode at a voltage of
29 kV. The matrix employed was anthracene-1,8,9-triol–silver
trifluoroacetate in THF. The MALDI-TOF gave a [M+Ag]+
peak at 2046.4 which clearly confirms the structure of the
cyclic tetramer.
Stereoisomer 3
H3C
Si
The SmA and SmC phases exhibited by the SCLC cyclic
tetramer were confirmed by optical light microscopy and X-
ray diffraction.
Si
H3C
O
Si
O
Si
O
O
CH3
CH3
Preparation of Speiers catalyst
K1 76.8 K2 88.2 S1 102.5 SmC 114.0 SmA 116.1 °C Iso
Hydrogen hexachloroplatinate() (0.10 g) was dissolved in
dry propan-2-ol (1 ml). Dry toluene (9 ml) was added with
shaking and the solution was stored in the dark at 4 °C.
Stereoisomer 4
=
Mesogenic Side Chain
Preparation of 4-(hex-5-enyloxy)benzoic acid
Fig. 9 Structure and transition temperatures for stereoisomers 1 to 4.
The preparation of 4-(hex-5-enyloxy)benzoic acid was similar
to that outlined in ref. 16. The cooled reaction mixture was
acidified with concentrated hydrochloric acid (10 ml) and the
crude product filtered off, washed with water and recrystallised
(50% aqueous ethanol) to yield 4-(hex-5-enyloxy)benzoic acid
as a white crystalline powder (purity, >99% by HPLC).
Yield=11.40 g (95%), transition temperatures K 101 N 140 °C
and the mixture allowed to stir overnight, at room temperature.
Any undissolved material was filtered off and the solvent
removed by distillation under reduced pressure. The crude
ester was then purified by flash column chromatography (silica
gel), eluting with a 251 mixture of dichloromethane–petroleum
fraction (bp 40–60 °C) to yield 4-octyloxyphenyl 4∞-(hex-5-
enyloxyphenyl)benzoate as a white powder (purity, 100% by
HPLC). Yield=3.20 g (78%), transition temperatures K 25
Iso (lit.,16 K 101 N 140 °C Iso). 1H NMR (CDCl ) d 8.10
3
(2H, d), 6.95 (2H, d), 5.85 (1H, m), 5.05 (2H, t), 4.05 (2H,
t), 2.15 (2H, m), 1.85 (2H, m), 1.60 (2H, m). IR (KBr) n
max
SmC 44.6 SmA 74.6 °C Iso. 1H NMR (CDCl ) d 8.15 (2H,
2950, 1690, 1610, 1260 cm−1. MS m/z 220 (M+), 203, 138,
3
d), 7.10 (2H, d), 6.95 (2H, d), 6.90 (2H, d), 5.85 (2H, m),
82, 55 (100%).
358
J. Mater. Chem., 1999, 9, 353–360

CH2O
OH
1.1
CH2O
OC8H17
HO
CO2H
1.5
1.2
CH2=CH(CH2)4O
CO2H
HO
OC8H17
1.3
CH2=CH(CH2)4O
CO2
OC8H17
1.4
CH3
SiO
4
(CH2)6O
CO2
OC8H17
Polymer 3
Scheme 1 Reagents and conditions: 1.1 1-Bromooctane, potassium carbonate, potassium iodide, butanone, 1.2 Cyclohexene, 5% palladium on
charcoal, ethanol, glacial acetic acid; 1.3 DCC, DMAP, dichloromethane; 1.4 Pt catalyst, toluene; 1.5 6-Bromohex-1-ene, sodium hydroxide,
water, ethanol.
Separation of the stereoisomers of the SCLC cyclic tetramer by
HPLC
was programmed to identify individual peaks and set to collect
individual fractions in sequence. Each sample peak was
collected into a separate sample bottle.
The above procedure was repeated many times to yield
sufficient sample to allow separation of the remaining isomers.
The HPLC chromatogram shown in Fig. 10 was obtained
for the stereoisomer 4.
The following text details the use of HPLC in the separation
of these stereoisomers.
Separation of stereoisomer 4
For the separation of stereoisomer 4 the following isocratic
preparative HPLC system was used: Kontron HPLC pump
420, Gilson HPLC injection pump, Model 303, fitted with a
10-S pump head; Gilson automatic fraction collector 201 with
201–202 controller set for peak detection mode; Gilson
Holochrome UV detector set at 254 nm; Dupont Instruments
Zorbax ODS column (21.2 mm×25 cm).
A standard solution of the SCLC cyclic tetramer was
prepared by dissolving the tetramer (2.0 g) in chloroform
(4 ml). Acetonitrile (6 ml) was then added dropwise to ensure
that the polymer did not precipitate out of solution. A stock
solution of chloroform–acetonitrile (40560) was then prepared
and used as the mobile phase. A metered volume of the
standard solution (1.68 ml) was then injected onto the column,
and the sample eluted at 5 ml min−1. The fraction collector
Fig. 10 The HPLC chromatogram of impure stereoisomer 4.
J. Mater. Chem., 1999, 9, 353–360
359

Purification of the stereoisomer 4
Separation of the remaining two stereoisomers 1 and 2
To ensure the complete separation of stereoisomer 4, it was
necessary to further purify the above sample by using the
following preparative gradient HPLC system: Merck/Hitachi
L-6200A intelligent pump; Merck/Hitachi L-4000 UV detector,
set at 254 nm; Merck/Hitachi D-6000 interface; Merck/Hitachi
T-6300 column thermostat oven; Merck/Hitachi PC based
data handling system; Dupont Zorbax ODS column
(21.2 mm×25 cm); Gilson Automatic fraction collector 201
with 201–202 controller set for manual peak collection mode.
The stereoisomer 4 (81.3 mg) from the previous separation
was dissolved in chloroform (2 ml). Acetonitrile (3 ml) was
then added slowly to ensure that the isomer did not precipitate
out of solution. A known volume of sample (1 ml) was injected
onto the column and the sample was eluted with the 50550
chloroform–acetonitrile mixture for 2 min at a flow rate of
5 ml min−1, at 30 °C. The mobile phase composition was then
changed, using a gradient method, to chloroform–acetonitrile
(35565) over a period of 1 min. The process was then repeated
a further 4 times until all the sample had been separated to
yield pure stereoisomer 4. The solvent was then removed by
distillation under reduced pressure and the isomer was then
dried in vacuo to yield stereoisomer 4 as a white solid.
Analytical HPLC showed that the purity of stereoisomer 4
was 100%.
The method used for the separation of the remaining two
isomers was identical to that outlined previously. The sample
was prepared in a similar manner to that outlined for the
separation of stereoisomer 3, but the injection volume was
reduced to 0.5 ml. The eluent system that was used in the
separation of the remaining two isomers was a mixture of
chloroform–acetonitrile (30570), with
temperature of 30 °C.
a
column oven
Again the multiple injection method was used in conjunction
with our ‘cut-off’ method to achieve the separation of the two
remaining stereoisomers in a pure state.
The fractions containing pure stereoisomer 1 were combined
and the solvent removed by distillation under reduced pressure
to afford pure stereoisomer 1 as a white solid (100% pure by
analytical HPLC). This procedure was also carried out to
obtain a pure sample of stereoisomer 2 (white solid, 100%
pure by analytical HPLC).
Acknowledgements
The authors would like to thank Professor J. W. Emsley
(University of Southampton) for carrying out the 1H, 13C and
29Si NMR work, Professor J. W. Emsley and Dr D. F. Ewing
(University of Hull) for their help in the interpretation of the
NMR spectra and Dr R. M. Richardson (University of
Bristol) for the X-ray diffraction work. We would also like to
thank the Defence Research Agency, Malvern, for sponsoring
this research project. This paper is published by the kind
permission of HMSO.
The solvent in fractions containing stereoisomers 1, 2 and
3 was removed by distillation under reduced pressure, to yield
a white solid which was used in the following section.
Separation of stereoisomers 1, 2 and 3
The preparative gradient HPLC system used in the purification
of stereoisomer 4 was now used in the separation of stereo-
isomers 1, 2 and 3.
References
1
2
3
4
5
6
C. W. Young, P. C. Servais, C. C. Currie and M. J. Hunter, J. Am.
Chem. Soc., 1948, 70, 3758.
J. Homer, A. W. Jarvie, A. Holt and H. J. Hickton, J. Chem. Soc.
(B), 1967, 67.
H. J. Hickton, A. Holt, J. Homer and A. W. Jarvie, J. Chem. Soc.
(C ) , 1966, 149.
M. Owen, in Siloxane Polymers, ed. S. J. Clarson and
J. A. Semlyen, PTR Prentice Hall, New Jersey, 1993.
F.-H. Kreuzer, M. Gawhary, R. Winkler and H. Finkelmann, EP
0060335 (1981).
Separation of stereoisomer 3
A standard stock solution of the SCLC cyclic tetramer was
prepared by dissolving the mixture of the three remaining
isomers (0.01 g) in chloroform (2 ml). Acetonitrile (3 ml) was
then added dropwise to ensure that the isomer did not
precipitate out of solution.
R. D. C. Richards, W. D. Hawthorne, J. S. Hill, M. S. White,
D. Lacey, J. A. Semlyen, G. W. Gray and T. C. Kendrick,
J. Chem. Soc., Chem. Commun., 1990, 95.
Between injections, the sample solution and the syringe were
stored in the column oven at 30 °C. This procedure was
adopted to prevent the precipitation of isomers on standing.
The column was then eluted with a mixture of chloroform–
acetonitrile (40560). A known volume of the sample solution
(1 ml) was then injected onto the column, and the sample
eluted at a flow rate of 4 ml min−1 at 30 °C. The sample peaks
eluted from the column were collected manually and samples
were collected allowing for a ‘cut-off’ sample between the
major isomers. This method was used so that it was possible
to isolate stereoisomer 3 in a pure state as well as concentrating
the sample containing the remaining two isomers. The above
procedure was repeated many times to yield sufficient sample
of stereoisomer 3.
The solvent was then removed, under reduced pressure, and
the residue dried in vacuo to yield stereoisomer 3 as a white
solid. Analytical HPLC showed that the purity of stereoisomer
3 was 100%.
The solvent in fractions containing stereoisomers 1 and 2
was distilled off, under reduced pressure, to yield a white solid
which was used in the separation of stereoisomers 1 and 2.
7
8
F.-H. Kreuzer, D. Andrejewski, W. Haas, N. Haberle, G. Riepl
and P. Spes, Mol. Cryst. Liq. Cryst., 1991, 199, 345.
K. D. Gresham, C. M. McHugh, T. J. Bunning, R. L. Crane,
H. E. Klei and E. T. Samulski, J. Polym. Sci. A: Polym. Chem.,
1994, 32, 2039.
I. Sledzinska, J. Soltysiak and W. Stanczyk, J. Inorg. Organomet.
Polym., 1994, 4, 199.
9
10 D. R. R. Everitt, C. M. Care and R. M. Wood, Mol. Cryst. Liq.
Cryst., 1987, 153, 55.
11 D. R. R. Everitt, C. M. Care and R. M. Wood, Mol. Cryst. Liq.
Cryst., 1991, 201, 41.
12 J. Pinsl, C. Brauchle and F-H. Kreuzer, J. Mol. Elec., 1987, 3, 9.
13 E. P. Socci, B. L. Farmer, T. J. Bunning, R. Pachter and
W. W. Adams, Liq. Cryst., 1993, 13, 811.
14 M. S. Beevers and J. A. Semlyen, Polymer, 1971, 12, 378.
15 E. Klarman, L. W. Gatyas and V. A. Shternov, J. Am. Chem.
Soc., 1932, 54, 298.
16 D. Lacey, H. N. Beattie, G. R. Mitchell and J. A. Pople, J. Mater.
Chem., 1998, 8, 53.
¨
Paper 8/07003B
360
J. Mater. Chem., 1999, 9, 353–360