The creation of long-lasting glassy columnar discotic liquid crystals using 'dimeric' discogens

Article abstract of DOI:10.1039/a810045d

'Dimeric' discogens have been synthesised which have two triphenylene nuclei, each of which bears five β-OC₆H₁₃ substituents, which are linked through the remaining β positions by a flexible O(CH₂)(n)O polymethylene chain

Full text of DOI:10.1039/a810045d

J O U R N A L O F
The creation of long-lasting glassy columnar discotic liquid crystals
using ‘dimeric’ discogens
Neville Boden,a Richard J. Bushby,*a Andrew N. Cammidge,b Ahmed El-Mansoury,a
Philip S. Martina and Zhibao Lua
C H E M I S T R Y
aCentre for Self-Organising Molecular Systems (SOMS), University of Leeds, Leeds, UK
LS2 9JT. E-mail: richardb@chem.leeds.ac.uk
bSchool of Chemical Sciences, University of East Anglia, Norwich, UK NR4 7TJ
Received 24th December 1998, Accepted 29th March 1999
‘Dimeric’ discogens have been synthesised which have two triphenylene nuclei, each of which bears five b-OC H
6
13
substituents, which are linked through the remaining b positions by a flexible O(CH ) O polymethylene chain (n=
2 n
3–16). Calculations show that the minimum chain length for bridging between columns in the Col phase of these
h
compounds is O(CH ) O. It was found that only those dimers for which n>7 formed a Col phase. This mesophase
2 7
h
can be supercooled into a metastable glass which is relatively hard and in which the columnar order and orientation
is retained. These glasses revert to the crystalline state within a few hours but we have shown that longer-lasting
Col glasses can be engineered through introducing an extra degree of disorder into the glass by making the ‘dimers’
h
subtly unsymmetrical. These ‘dimers’ either have differently substituted triphenylene rings at the two ends of the
linking chain or an amide group within the chain itself. They give Col aligned glasses which last for months at
h
room temperature.
Most suggested applications for the columnar phases of
discotic liquid crystals1 depend on their unique alignment and
conduction2 properties but, in any real application, the ease
with which they can be processed and mechanical robustness
will also be important issues. This paper is concerned with
attempts to synthesise oligomers which combine the desirable
alignment properties of monomeric discogens with the desir-
able robustness of glassy films obtained from polymeric
discogens.
In the columnar phases of discotic liquid crystals, the
aromatic cores are stacked on top of each other and the
columns thus formed are ordered on a two-dimensional lattice.
The space between the columns is filled by mobile, fluid,
disordered alkyl chains. Provided the aromatic cores are the
same and the ‘packing fraction’ for the core and the alkyl
chains is about the same, more-or-less the same phase behav-
iour is observed regardless of whether the system is monomeric,
oligomeric or polymeric. For example, HAT6 1 shows Cr 70
Col 100 °C I3 whereas the dimer 2g shows Cr 68 Col 107 °C
h
h
I and the polymer 3 Cr 98 Col 118 °C I.4 Similarly, the low
h
angle X-ray diffraction patterns obtained for the Col phases
h
of these three are virtually indistinguishable. In other ways
there are very significant differences. Unlike their low molar
mass counterparts the polymeric systems are difficult to order
using surface effects, they give very different optical textures
(for example, Fig. 1 in ref. 5) and, whereas the Col phases of
h
monomers such as 1 show some tendency to supercool, poly-
meric systems such as 3 display very marked supercooling.5 In
supercooled materials a metastable glassy columnar state is
often formed in which the alignment, the architecture of the
columns and the disorder of the chains are preserved, but in
which the motion of the chains is ‘frozen out’. They are
mechanically stiff and relatively hard and for some applications
(for example, in the preparation of LEDs where it is necessary
to vapour deposit metal electrodes on very thin aligned films6)
this is important. In a previous paper7 we showed that ‘dimeric’
and ‘trimeric’ oligomers combine some of the desirable proper-
ties of monomers and polymers. They can be aligned by
surface interactions and supercooled to form a glassy Col
state. Compared to the polymers they are much easier to
h
also much less long-lived and the glass reverts to the crystalline
state over a period of a few hours or days.
process as ordered glassy films but, unfortunately, these are
J. Mater. Chem., 1999, 9, 1391–1402
1391

‘dimers’ or ‘twins’ have often been used as tractable models
for the polymers.14 We found that the Col phase was only
h
formed by those systems 2 with n>7. None of the simple
dimers (produced by the route shown in Scheme 1) showed
particularly long-lived glassy states. More successful strategies
for producing persistent glasses were developed which involved
the synthesis of an unsymmetrical hexyloxy/butoxy substituted
‘dimer’ (Scheme 3), and an unsymmetrical ‘amide-linked’
dimer (Scheme 4).
Modelling studies
In polymeric and oligomeric discotic liquid crystals a constraint
exists at the molecular level which does not apply to monomeric
systems. The chains linking the nuclei must be long enough
and flexible enough to bridge between the aromatic cores. This
bridging can be visualised as occurring in several different
ways; in an intracolumnar manner (intra-1,2; intra-1,3; intra-
1,4; intra-1,5; intra-1,6 etc.; Fig. 1) and/or in an intercolumnar
manner (inter-1,2; inter-1,3 etc.; Fig. 1). Each of these intro-
duces different amounts of strain into the system and this
must have some bearing on the stability of the phase. The
enthalpic cost of each type of bridging can be calculated using
force field methods. To minimise computing time, a,v-
diphenoxyalkanes 4 were initially chosen as models for the
‘dimers’. MM2 calculations15 on 4 show that (for the isolated
molecules with n>2) an equilibrium mixture exists between
an extended (all-trans) conformation 4a which minimises the
energy of the chain and a ‘stacked’ conformation 4b which
gives a less favourable chain conformation but benefits from
the attractive ring–ring van der Waals interaction. For these
isolated molecules it is this attractive ring–ring interaction
(which MM2 calculates to about 6 kcal mol−1) that dominates
the conformational equilibrium and conformer 4b is the more
stable. In the columnar phase of 2 the attractive ring–ring
van der Waals interaction between the rings is not an issue.
Each ring is always ‘sandwiched’ between two others regardless
of whether bridging is intracolumnar or intercolumnar. It is
the strain within the chain itself which we need to know. This
figure can be obtained from the calculations by factoring out
the effect of the rings (Experimental section). Table 1 gives
the MM2 calculated difference between all-trans H-(CH ) -H
2 n
chains and H-(CH ) -H chains which have been folded as
2 n
required for different types of bridging. As can be seen from
Table 1 and Fig. 4, low strain bridging in an intracolumnar
manner in the Col phase 1,2 requires n>2, 1,3 requires n>5,
h
1,4 requires n>7, 1,5 requires n>10, and 1,6 requires n>12.
Low strain bridging in an intercolumnar manner in the Col
h
The first aim of this study was to systematically investigate
the phase behaviour and glass-forming properties of discotic
oligomers.7–13 It was hoped that, by optimising the length of
the O(CH ) O chain in the ‘dimeric’ oligomers 2a–h, it would
2 n
be possible to produce a glassy state that combined the ease
of alignment of the monomers with the long life of the polymer
Col glasses. The other main aim of this investigation was to
h
investigate the effect of the length of the (CH ) chain on the
2 n
stability of the Col phase itself; a problem relevant to the
h
design of polymers and oligomers alike. In this respect we
Fig. 1 Different ways in which a chain can bridge the cores in the
were following calamitic liquid crystal science studies in which
Col phase of an oligomeric or polymeric discotic liquid crystal.
h
1392
J. Mater. Chem., 1999, 9, 1391–1402

Table 1 Differences in energy calculated by the MM2 force field method between all-trans H-(CH ) -H chains and H-(CH ) -H chains folded in
2 n
2 n
the requisite manner for 4b, for 1,2-, 1,3-, 1,4-, 1,5- and 1,6-intrastack bridging, and 1,2- and 1,3-interstack bridging (see text)
Energy difference/kcal mol−1
n
4b
intra-1,2
intra-1,3
intra-1,4
intra-1,5
intra-1,6
inter-1,2
inter-1,3
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
(1.11)
1.00
1.57
3.91
1.97
4.27
4.48
4.83
3.47
4.73
4.05
3.43
3.73
5.41
4.76
12.47
1.12
4.03
4.74
2.79
4.8
5.49
5.55
6.16
4.99
4.28
3.6
113.7
18.04
1.9
1.25
3
3.01
4.79
3.07
6.39
3.95
4.37
5.18
5.14
107.8
15.6
4.43
3.61
4.49
2.12
3.19
3.92
5.59
6.86
7.12
6.53
4.86
6.45
6.37
6.99
6.8
111.7
27.68
3.4
2.57
2.1
1.77
3.58
4.68
4.19
5.06
6.17
109.1
15.36
4.04
4.46
4.68
2.58
4.45
4.07
107.3
59.54
5.26
2.34
3.64
3.72
3.73
6.14
6.05
142.0
47.34
11.52
1.41
3.5
phase 1,2 requires n>6 and 1,3 requires n>19. Examples of
optimised geometries are shown in Fig. 2 and 3.
Fig. 3 Examples of geometries as optimised by the force field
calculations (n=19). View from the side. 1,2-Intercolumnar bridging
(top) and 1,3-intercolumnar bridging (bottom).
Synthesis
The synthesis of the oligomers 2a–h is shown in outline in
Scheme 1. Iron() chloride mediated oxidative coupling of
3,3∞,4,4∞-tetrahexyloxybiphenyl 5a with 1-hexyloxy-2-methox-
ybenzene 6a followed by a reductive methanol workup gave
3,6,7,10,11-pentahexyloxy-2-methoxytriphenylene
7.16,17
Demethylation using lithium diphenylphosphide18 gave
3,6,7,10,11-pentahexyloxy-2-hydroxytriphenylene 8, which was
reacted with anhydrous potassium carbonate and 0.5 equiv.
of the appropriate a,v-dibromoalkane in refluxing ethanol to
give the ‘dimer’ 2. A by-product sometimes observed in these
reactions was the v-brominated product 9. These v-bromo
products proved useful intermediates in our subsequent work.
They could be obtained as the major product (yields 25–40%)
by using a 151 molar ratio of the reactants. Alternatively, they
could be made in better yield through the reaction of 3,3∞,4,4∞-
tetrahexyloxybiphenyl 5a with a 1-(v-bromoalkoxy)-2-hexy-
loxybenzene 6b,c using iron() chloride followed by a
reductive methanol workup.
The synthesis of the symmetrical butoxy substituted
oligomers shown in Scheme 2 was an essentially trivial variant
on the route to the hexyloxy substituted oligomers. Iron()
chloride mediated oxidative coupling of 3,3∞,4,4∞-tetrabutylox-
ybiphenyl 5b with 1-butoxy-2-methoxybenzene 6d followed by
a reductive methanol workup gave 3,6,7,10,11-pentabutoxy-2-
methoxytriphenylene 10. Demethylation using lithium
diphenylphosphide gave 3,6,7,10,11-pentabutoxy-2-hydroxy-
triphenylene 11, which was reacted with anhydrous potassium
carbonate and 0.5 equiv. of the appropriate a,v-dibromoal-
kane in refluxing ethanol to give the ‘dimer’ 12.
Fig. 2 Examples of geometries as optimised by the force field
calculations (n=14). Views from the side (left) and top (right) of the
column. 1,2-Intracolumnar bridging (top), 1,3-intracolumnar bridging,
1,4-intracolumnar bridging, and 1,5-intracolumnar bridging (bottom).
J. Mater. Chem., 1999, 9, 1391–1402
1393

Scheme 1 Reagents and conditions: (i) FeCl , CH Cl ; (ii) MeOH, 61% (two steps); (iii) Ph PLi, THF, 69%; (iv) Br(CH ) Br, K CO , EtOH,
2 n
3
2
2
2
2
3
25% (n=5), 34% (n=10), 40% (n=16); (v) Br(CH ) Br, K CO , EtOH, 20–68% (n=3–16).
2 n
2
3
An unsymmetrical hexyloxy/butoxy substituted ‘dimer’ 13
was synthesised as shown in Scheme 3. The reaction of
3,6,7,10,11-pentabutoxy-2-hydroxytriphenylene 11 with the
a-triphenylenyloxy-v-bromo derivative 9c gave the ‘mixed
dimer’ 13.
The synthesis of a dimer 18, in which there is an amide
group in the linking chain, is shown in Scheme 4. Ethyl 6-
(3,6,7,10,11-pentahexyloxytriphenylen-2-yloxy)hexanoate 14
was made by reacting 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 with ethyl 6-bromohexanoate. Hydrolysis and
reaction with oxalyl chloride gave the acid chloride 16
which was reacted with the amine 17 to give the amide 18.
The amine 17 was made from the bromide 9b by reaction with
sodium azide followed by a lithium aluminium hydride
reduction.
Characterisation
The phase behaviour of the new compounds was investigated
by low angle X-ray diffraction, optical polarising microscopy
and differential scanning calorimetry (DSC). The results of
these investigations and data for related systems are summar-
ised in Table 2.
Compounds 2a–c did not show liquid crystal behaviour but
2d–h gave enantiotropic liquid crystal Col phases. Low angle
h
X-ray investigations of the unoriented mesophases showed a
1394
J. Mater. Chem., 1999, 9, 1391–1402

Scheme 2 Reagents and conditions: (i) FeCl , CH Cl ; (ii) MeOH, 68% (two steps); (iii) Ph PLi, THF, 68%; (iv) Br(CH ) Br, K CO , EtOH,
2 n
3
2
2
2
2
3
46% (n=8), 34% (n=10).
single diffraction ring corresponding to the column–column
separation. Optical textures for compounds 2d–h obtained were
12b showed single reversible phase transitions at 154
and 147 °C, respectively. Compound 12b has previously been
reported by Bacher et al.8 who report a single transition at
149 °C which they identified as that of the plastic columnar
similar to those obtained for Col phases of HATN monomers
h
and samples annealed just below the Col –I transition tempera-
h
ture between glass slides quite rapidly became homeotropic.5
(Col ) to isotropic (I) type.
p
On heating, the DSC results for compounds 2d–h showed
Both of the unsymmetrical ‘dimers’ 13 and 18 gave an
endotherms for both Cr–Col and Col –I transitions (Table 2).
enantiotropic Col phase. In the case of compound 13 low
h
h
h
On cooling, only the I-Col transition and a glass transition at
angle X-ray diffraction showed a column–column separation
h
˚
ca. 35 °C were observed. At room temperature the glassy Col
of 17.7 A which compares to a value calculated using eqn. (1)
h
˚
samples formed crystallised over a period of 3–18 h and the
(Experimental section) of 17.9 A. Compounds 13 and 18
slow growth of the endotherm corresponding to the Cr–Col
transition (apparent in the DSC scan on reheating) could be
formed glasses when the Col phase was cooled to ca. 30 °C
but, unlike the glasses formed from compounds 2d–h, these
h
h
used to follow the crystallisation process. In general the longer
the linking chain the shorter was the lifetime of the glass and
the more rapidly did crystallisation occur.
glasses persist for several months at room temperature. In the
case of compound 18, a sample stored at room temperature
for three months showed 15% recrystallisation as estimated
by DSC.
The DSC scans for the butoxy-substituted ‘dimers’ 12a and
J. Mater. Chem., 1999, 9, 1391–1402
1395

Scheme 3 Reagents and conditions: (i) K CO , EtOH, 26%.
2
3
exchange of discs between columns [Fig. 5(b)] and this makes
it more likely that disorder will be locked into the structure
when it is cooled. This ‘problem’ is even more acute for main-
chain discotic polymers [Fig. 5(c)] and this presumably
accounts, at least in part, for the long lives of the polymer
Discussion
The failure of the ‘dimers’ 2a–c to form a Col phase is not
surprising since the short nature of the connecting chain
precludes most forms of stacking in a hexagonal array (Table 1
and Fig. 1). Furthermore modelling studies suggest that, for
compound 2a, steric crowding makes conformations in which
both rings are coplanar unlikely. Compound 2d (n=8) pos-
h
Col glasses and the slowness with which they crystallise. We
h
reasoned that the lifetime of Col glasses of ‘dimers’ would
h
also be enhanced by making the dimer unsymmetric, as shown
sesses the shortest chain for which the Col phase was observed.
Such a compound has a sufficiently long chain for low strain
in Fig. 5(d), since this introduces an additional constraint.
The unsymmetrical ‘dimers’ should align as easily as symmetric
dimers but give longer lived glasses. This proved to be the
case. Hence polarising microscopy showed that the dimers 13
and 18 can be aligned in a homeotropic manner by annealing
the sample for a few hours between glass slides just below the
h
bridging in an intracolumnar manner (1,2-, 1,3- or 1,4-) and
has one more methylene than the minimum required to bridge
in a 1,2-intercolumnar manner (Table 1 and Fig. 1). This is in
accordance with previous studies of ‘dimers’. Indeed, previous
studies of ‘dimers’ have been interpreted in terms of bridging
being ‘normally intercolumnar’.9,10 Whereas we cannot dis-
prove this claim it is difficult to understand on the basis of
the data presented in Table 1 and Fig. 4, which suggests little
energy difference between intercolumnar and intracolumnar
bridging.
Col –I transition temperature. When these aligned liquid crys-
h
tals are cooled they give aligned long-lasting Col glasses. We
h
interpret the increased lifetimes of the glasses as being a result
of their lower symmetry. In the case of dimer 18 an alternative
explanation is possible; namely that hydrogen bonding between
the amide groups is significant and makes this oligomer behave
more like a polymer.
None of the ‘dimers’ 2d–h shows phase behaviour
significantly different to that of HAT6 1 (Cr 70 Col 100 °C
I). Comparison with our previous studies of oligomers shows
h
that the dimer 2f gives a more stable mesophase (enantiotropic,
Experimental
Cr 50 Col 104 °C I) than the less symmetrical dimethoxy
h
derivative 19 (monotropic, Cr 112 I 98 Col 33 °C glassy
NMR spectra were recorded on a General Electric QE300, a
Bruker AC200 or a Bruker AM400 instrument. Chemical
shifts are relative to tetramethylsilane, coupling constants (J)
are given in Hz. Mass spectra were obtained on a VG Autospec
instrument. All peaks >20% of M+ (and less intense peaks of
particular significance) are reported. Solvents were routinely
purified according to the procedures recommended by Perrin.20
Column chromatography on silica refers to the use of Merck
silica gel 9385 Type 60 and thin layer chromatography to
Whatman AL SIL G/UV plates. Phase behaviour was investi-
gated using an Olympus BH-2 optical polarising microscope
with a Mettler FP82 HT hot stage and Perkin-Elmer 7 thermal
analysis system (cooling and heating rate, 10 °C min−1). The
glass transitions were very weak but could (usually) just be
detected by DSC. The point at which samples sandwiched
h
Col ).7 Other interesting comparisons can be made between
h
the dimers 2a–h and related dimers 20,10 21a–c11 and 2211
(Table 2).
The glassy Col states formed by dimers 2d–h are all
h
short-lasting and in general their lifetimes seem to decrease as
the length of the linking chain (n) increases (Table 2). This
led to a search for variations in the molecular design that
would give longer-lasting glasses. It is well known that the
crystallisation of glassy Col phases of monomeric discogens
h
is slowed down by the introduction of small changes in the
symmetry of the disc.19 Presumably their slow crystallisation
relates to the need to order this ‘defect’ as shown schematically
in Fig. 5(a). In the case of ‘dimers’ there is a need not only to
rotate the discs but formation of a crystal lattice also requires
1396
J. Mater. Chem., 1999, 9, 1391–1402

Scheme 4 Reagents and conditions: (i) Br(CH ) CO Et, K CO , EtOH, 37%; (ii) NaOH, H O, MeOH, 61%; (iii) ClCOCOCl, CH Cl , 75%
2 5
2
2
3
2
2 2
(crude); (iv) NaN , EtOH; (v) LiAlH , THF, 77% (two steps); (vi) THF, 16%.
3
4
between glass slides became difficult to shear was used to
provide independent confirmation. Small angle X-ray diffrac-
tion experiments were conducted with a pinhole camera con-
sisting of a Phillips generator and tubes, nickel-filtered Cu-Ka
radiation of wavelength l=0.154 nm, and a Lindemann
sample tube (0.7 mm i.d.) to plate distance of 112.8 mm. Ether
refers to diethyl ether.
1,16-Dibromohexadecane21
Phosphorous tribromide (0.69 g, 0.0025 mol) was added drop-
wise to a vigorously stirred solution of hexadecane-1,16-diol
(1 g, 3.8 mmol) in dry ether (20 cm3), and the mixture refluxed
for 5 h before being poured into water (50 cm3) and extracted
with dichloromethane (2×50 cm3). The solvent was evapor-
J. Mater. Chem., 1999, 9, 1391–1402
1397

Table 2 DSC data (onset temperatures) for the compounds described
in the text
Compound
T
/ °C
T / °C
T / °Ca
Col
I
g
2a n=3
2b n=5
2c n=7
2d n=8
2e n=9
2f n=10
2g n=12
2h n=16
9a n=5
9b n=6
9c n=10
9d n=16
12a n=8
12b n=10
13
—
—
—
81b
98b
69b
91c
92c
104c
107c
84c
68c
50b
40b
31b
154d
147d,e
98c
—
—
—
58
35 (ca. 18 h)
35 (ca. 15 h)
35 (ca. 11 h)
35 (ca. 12 h)
35 (ca. 3 h)
—
—
—
72
50
68
41
62
—
—
—
—
—
—
—
—
—
30 (>1 month)
18
89
99c
30 (>6 months)
19f
20g
21ai n=10
21bi n=12
21ci n=14
22i
112
64
—
—
—
—
98f
135c
180c
168c
147c
147c
33 (several hours)
−59 (several days)h
—
—
—
47
aFigure in parentheses is the approximate time taken for the glass to
crystallise at room temperature. bCr–I. cCol –I. dCol –I. eBacher et al.
h
p
(ref. 8) give 149 °C. fI–Col monotropic (ref. 7). gRef. 10. h‘Lifetime’
of the supercooled Col phase at room temperature. iRef. 11.
h
ated, and the crude product recrystallised from ethanol
(20 cm3) to give crude 1,16-dibromohexadecane (0.6 g, 43%)
as a white solid which was used directly in the next step; d
H
3.43 (4H, t, J 7, CH Br), 1.85 (4H, t, J 7, CH CH Br),
2
2
2
1.25–1.60 (24H, m, CH ). All of the other a,v-dibromides
2
used were commercially available.
1,3-Bis(33∞,6∞,7∞,10∞,11∞-pentahexyloxytriphenylen-2∞-
yloxy)propane 2a and related compounds 2b–2h
A stirred mixture of 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
Fig. 4 Differences in energy (kcal mol−1) calculated by the
phenylene4
8 (500 mg, 0.67 mmol), 1,3-dibromopropane
MM2 force field method between all-trans H-(CH ) -H chains and
2 n
H-(CH ) -H chains folded in the requisite manner for 1,2-intrastack
2 n
(70 mg, 0.34 mmol) and anhydrous potassium carbonate (3 g)
was refluxed in ethanol (10 cm3) for 7 days. The solution was
decanted into water (50 cm3), and extracted with dichloro-
methane (2×50 cm3), the solvent evaporated, and the product
purified by column chromatography on silica eluting with
benzene–light petroleum (7: 3). Reprecipitation from dichloro-
methane (5 cm3) with methanol (20 cm3) gave the product 2a
(0.1 g, 20%) as a white powder, mp 81 °C (Found: C, 77.45;
bridging (upper —), for 1,3-intrastack bridging (upper ---), for 1,4-
intrastack bridging (upper ,), for 1,5-intrastack bridging (upper
|{|{|{), for 1,6-intrastack bridging (upper ||{||{||{), for 1,2-interstack
bridging (lower —) and for 1,3-interstack bridging (lower ---) (see
text and Table 1).
(91 mg, 0.34 mmol) for 60 h gave 1,8-bis(3∞,6∞,7∞,10∞,11∞-penta-
hexyloxytriphenylen-2∞-yloxy)octane 2d (0.22 g, 21%) as a white
H, 9.70. C H
99 148 12
s, ArH), 7.85 (10H, s, ArH), 4.60 (4H, t,
O
requires: C, 77.70; H, 9.74%); d 8.0 (2H,
H
J
7,
powder (Found: C, 77.75; H, 9.9. C
78.05; H, 9.95%).
H
O
requires: C,
104 158 12
OCH CH CH O), 4.25 (20H, t, J 7, OCH ), 2.58 (2H, t, J 7,
2
2
2
2
2
2
2
2
OCH CH CH O), 1.78–1.98 (20H, m, OCH CH ), 1.3–15
In a similar manner 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 (500 mg, 0.67 mmol) and 1,9-dibromononane
(96 mg, 0.34 mmol) for 60 h gave 1,9-bis(3∞,6∞,7∞,10∞,11∞-penta-
hexyloxytriphenylen-2∞-yloxy)nonane 2e (0.27 g, 25%) as a
2
2
(60H, m, CH ), 0.95 (30H, t, J 7, CH ).
3
In a similar manner, 3,6,7,10,11-pentahexyloxy-2-hydroxy-
triphenylene 8 (500 mg, 0.67 mmol) and 1,5-dibromopentane
(75.8 mg, 0.33 mmol) for 55 h gave 1,5-bis(3∞,6∞,7∞,10∞,11∞-
pentahexyloxytriphenylen-2∞-yloxy)pentane 2b (0.15 g, 30%) as
white powder (Found: C, 77.9; H, 10.25. C
C, 78.11; H, 9.98%).
H
O requires:
105 160 12
a white powder (Found: C, 77.75; H, 10.05. C
H
O
In a similar manner 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 (500 mg, 0.67 mmol) and 1,10-dibromodecane
(98 mg, 0.33 mmol) for 40 h gave 1,10-bis(3∞,6∞,7∞,10∞,11∞-
pentahexyloxytriphenylen-2∞-yloxy)decane 2f (0.26 g, 24%) as
101 152 12
requires: C, 77.85; H, 9.82%); d 7.85 (12H, s, ArH), 4.22
H
(24H, t, J 7, OCH ), 1.95 (24H, m, OCH CH ), 1.35–1.55
2
2
2
(62H, m, CH ), 0.95 (30H, t, J 7, CH CH ).
2
2
3
In a similar manner 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 (500 mg, 0.67 mmol) and 1,7-dibromoheptane
(89 mg, 0.34 mmol) for 60 h gave 1,7-bis(3∞,6∞,7∞,10∞,11∞-penta-
hexyloxytriphenylen-2∞-yloxy)heptane 2c (0.22 g, 42%) as a
a white powder (Found: C, 77.90; H, 10.25. C
requires: C, 78.18; H, 10.02%).
H
O
106 162 12
In a similar manner 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 (500 mg, 0.67 mmol) and 1,10-dibromododecane
(109 mg, 0.33 mmol) for 48 h gave 1,12-bis(3∞,6∞,7∞,10∞,11∞-
pentahexyloxytriphenylen-2∞-yloxy)dodecane 2g (0.30 g, 27%)
white powder (Found: C, 77.55; H, 9.65. C
C, 77.99; H, 10.05%).
H
O requires:
103 156 12
In a similar manner 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 (500 mg, 0.67 mmol) and 1,8-dibromooctane
as a white solid (Found: C, 78.3; H, 10.4. C
C, 78.15; H, 10.1%); m/z (FAB) 1655 ([M+1]+, 5%).
H
O requires:
108 166 12
1398
J. Mater. Chem., 1999, 9, 1391–1402

cooled, filtered, the filtrate washed with water (50 cm3),
extracted with dichloromethane (2×50 cm3), the extracts con-
centrated in vacuo and the product purified by column chroma-
tography on silica eluting with dichloromethane–light
petroleum (151) to give 1-(10∞-bromodecyloxy)-2-hexyloxyb-
enzene 6c as a colourless liquid (6.56 g, 80%); d 6.88 (4H, s,
H
ArH), 4.00 (4H, t, J 6.6, OCH ), 3.40 (2H, t, J 6.7, BrCH ),
2
2
1.75–1.95 (6H, m, OCH CH and BrCH CH ), 1.10–1.60
2
2
2
2
[18H, m, OCH CH (CH ) and BrCH CH (CH ) ], 0.90 [3H,
2
2
2 3 2 3
2
2
t, J 6.8, O(CH ) CH ].
2 5
3
2-(5∞-Bromopentyloxy)-3,6,7,10,11-pentahexyloxytriphenylene
9a and related compounds 9c and 9d
A stirred mixture of 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene
8 (500 mg, 0.67 mmol),4 1,5-dibromopentane
(460 mg, 2 mmol) and anhydrous potassium carbonate (7 g)
in ethanol (30 cm3) was refluxed for 17 h before being poured
into water (50 cm3) and extracted with dichloromethane
(2×50 cm3). The solvent was evaporated and the crude
product purified by column chromatography on silica eluting
with benzene–light petroleum (753) and recrystallisation from
dichloromethane–methanol
gave
2-(5∞-bromopentyloxy)-
3,6,7,10,11-pentahexyloxytriphenylene 9a (0.15 g, 25%) as a
white solid, mp 68 °C (Found: C, 71.60; H, 9.35; Br, 8.70.
C H O Br requires: C, 71.21; H, 9.12; Br, 8.94%); d 7.85
53 81 6
H
(6H, s, ArH), 4.20 (12H, t, J 9, OCH ), 3.55 (2H, t, J 7,
2
CH Br), 1.95 (12H, m, OCH CH ), 1.35–1.5 (34H, m, CH ),
2
2
2
2
0.95 (15H, t, J 7, CH CH ).
2
3
In a similar manner 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 (220 mg, 0.29 mmol) and 1,10-dibromodecane
(43 mg, 0.15 mmol) for 40 h gave 2-(10∞-bromodecyloxy)-
3,6,7,10,11-pentahexyloxytriphenylene 9c (0.1 g, 34%) as a
white solid, mp 50 °C (Found: C, 71.85; H, 9.45; Br, 7.95.
C H O Br requires: C, 72.2; H, 9.50; Br, 8.28%); d 7.85
Fig. 5 Schematic representation of ‘frozen-in disorder’ within the
glassy form of the Col phase which needs to be overcome as the
h
sample crystallises: (a) monomeric discogen with lowered symmetry
58 91 6
H
(6H, s, ArH), 4.22 (12H, t, J 7, OCH ), 3.45 (2H, t, J 7,
to the disc, (b) ‘dimeric’ discogen, (c) discogenic main-chain polymer,
and (d) unsymmetrical ‘dimeric’ discogen.
2
CH Br), 1.95 (12H, m, OCH CH ), 1.35–1.5 (44H, m, CH ),
2
2
2
2
0.95 (15H, t, J 7, CH CH ).
2
3
In a similar manner 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 (220 mg, 0.29 mmol) and 1,16-dibromohexa-
decane (280 mg, 0.72 mmol) for 30 h gave 2-(16∞-
bromohexadecyloxy)-3,6,7,10,11-pentahexyloxytriphenylene 9d
(0.12 g, 40%) as a white solid (Found: C, 73.50; H, 10.1; Br,
7.35. C H O Br requires: C, 73.32; H, 9.89; Br, 7.62%); d
In a similar manner 3,6,7,10,11-pentahexyloxy-2-hydroxy-
triphenylene
8 (500 mg, 0.67 mmol) and 1,16-dibromo-
hexadecane (130 mg, 0.34 mmol) for 60 h gave 1,16-
bis(3∞,6∞,7∞,10∞,11∞-pentahexyloxytriphenylen-2∞-yloxy)hexade-
cane 2h (0.40 g, 68%) as a white solid (Found: C, 78.35; H,
10.0. C
H
O
requires: C, 78.55; H, 10.23%).
64 103 6
H
112 174 12
7.85 (6H, s, ArH), 4.22 (12H, t, J 7, OCH ), 3.45 (2H, t, J
Transition data for compounds 2a–h are given in Table 2.
The 1H-NMR spectra of compounds 2c–h were virtually
identical to that for 2b except for the integration of the
2
7, CH Br), 1.95 (12H, m, OCH CH ), 1.35–1.5 (56H, m,
2
2
2
CH ), 0.95 (15H, t, J 7, CH CH ).
2
2
3
multiplet at d 1.35–1.55 due to the methylenes. All compounds
2a–h showed the expected parent ions in the mass spectrum.
H
2-(6∞-Bromohexyloxy)-3,6,7,10,11-pentahexyloxytriphenylene
9b
1-(6∞-Bromohexyloxy)-2-hexyloxybenzene 6b
Ferric chloride (5.0 g) was carefully added to a vigorously
stirred solution of 3,3∞,4,4∞-tetrahexyloxybiphenyl5 (1.5 g,
2.7 mmol) and 1-(6-bromohexyloxy)-2-hexyloxybenzene 6b
(2.0 g, 5.6 mmol) in dichloromethane (40 cm3). The mixture
was stirred for 2 h, carefully poured into methanol (100 cm3)
and the crude product was filtered off and purified by column
chromatography on silica eluting with dichloromethane–light
petroleum (253) to give 2-(6∞-bromohexyloxy)-3,6,7,10,11-
pentahexyloxytriphenylene 9b as a white solid (1.37 g, 56%);
A mixture of 1,6-dibromohexane (14.6 g, 60 mmol, 2 equiv.),
2-hexyloxyphenol22 (5.8 g, 30 mmol), and anhydrous potass-
ium carbonate (27.8 g, 0.2 mol) in ethanol (20 cm3) was
refluxed for 72 h. The mixture was poured onto water (50 cm3),
extracted with dichloromethane (2×50 cm3) and the organic
solvent was removed in vacuo. The crude product was purified
by chromatography on silica eluting with light petroleum–dich-
loromethane (251) to give 1-(6∞-bromohexyloxy)-2-hexyloxyb-
d
7.84 (6H, s, ArH), 3.97 (12H, t, J 7, OCH ), 3.45 (2H, t,
enzene 6b as a clear colourless oil (8.56 g, 80%); d 6.87 (4H,
H
2
H
J 7, CH Br), 1.85 (12H, m, OCH CH ), 1.35–1.7 (36H, m,
CH ), 0.97 (15H, t, J 7, CH CH ).
s, ArH), 3.97 (4H, t, J 7, OCH ), 3.45 (2H, t, J 7, CH Br),
2
2
2
2
2
1.85 (4H, m, OCH CH ), 1.3–1.7 (12H, m, CH ), 0.97 (15H,
2
2
3
2
2
2
t, J 7, CH CH ).
2
3
2-(10∞-Bromodecyloxy)-3,6,7,10,11-pentahexyloxytriphenylene
1-(10∞-Bromodecyloxy)-2-hexyloxybenzene 6c
9c
A mixture of 2-hexyloxyphenol (3.88 g, 20 mmol),22 1,10-
dibromodecane (10 g, 33 mmol, 1.6 equiv.) and anhydrous
potassium carbonate (11.1 g, 80 mmol, 4 equiv.) in ethanol
(25 cm3) was heated under reflux for 24 h. The mixture was
Ferric chloride (3.6 g, 22 mmol) was added to a stirred solution
of 3,3∞,4,4∞-tetrahexyloxybiphenyl5 (6.0 g, 11 mmol) in
dichloromethane (50 cm3) at 0 °C. After 30 min, 1-(10-bromo-
decyloxy)-2-hexyloxybenzene 6c (6.56 g, 16 mmol) and further
J. Mater. Chem., 1999, 9, 1391–1402
1399

ferric chloride (3.6 g) were added. The mixture was stirred at
room temperature for 2 h, carefully poured into methanol
(300 cm3), cooled to 0 °C, and the crude product was filtered
off and purified by column chromatography on silica eluting
with dichloromethane–light petroleum (151) and recrystallised
from ethanol to give 2-(10∞-bromodecyloxy)-3,6,7,10,11-penta-
hexyloxytriphenylene 9c as a white solid (6.56 g, 73%), mp
50 °C (Found: C, 72.5; H, 9.75; Br, 8.3. C H BrO requires
OCH CH CH ), 0.97 (3H, t, J 7.0, OCH CH CH CH ); m/z
(EI) 180 (M+, 30%).
2
2
2
2
2
2
3
3,6,7,10,11-Pentabutoxy-2-methoxytriphenylene 10
3,3∞,4,4∞-Tetrabutoxybiphenyl 5b (6.6 g, 15 mmol), ferric
chloride (8.4 g, 50 mmol) and 1-butoxy-2-methoxybenzene 6c
(5.4 g, 30 mmol, 2 equiv.) were added to dichloromethane
(20 cm3), stirred at room temperature and the progress of the
reaction monitored by TLC. After the reaction was complete,
the mixture was carefully poured into methanol (300 cm3),
and the resultant mixture cooled to 0 °C. The crude product
was filtered off, purified by column chromatography on silica
eluting with dichloromethane–light petroleum (251), and
recrystallised from ethanol to give 3,6,7,10,11-pentabutoxy-2-
methoxytriphenylene 10 as a white solid (6.3 g, 68%); Cr 92.1
(36) Col 101 (4) Col 119.3 (6) I 116.6 (6) Col 98.3 °C
58 91
6
C, 72.25; H, 9.51; Br, 8.29%); d 7.84 (6H, s, ArH), 4.23
H
(12H, t, J 6.5, OCH ), 3.40 (2H, t, J 6.8, CH Br), 1.85 (14H,
2
2
m, OCH CH and Br CH CH ), 1.1–1.7 (42H, m, CH ), 0.93
(15H, t, J 6.8, CH CH ); m/z (FAB) 964 (M+, 100%).
2
2
2
2
2
2
3
1,2-Dibutoxybenzene23
A mixture of catechol (22 g, 0.2 mol), 1-bromobutane (68.5 g,
0.5 mol) and anhydrous potassium carbonate (70 g, 0.5 mol)
in ethanol (100 cm3) was heated under reflux for 24 h. The
reaction mixture was filtered, poured onto water (50 cm3),
extracted with dichloromethane (2×50 cm3) and concentrated
in vacuo to give 1,2-dibutoxybenzene as a colourless liquid
1
2
2
(4 J g−1) Col glass (Found: C, 75.5; H, 8.85. C H O
1
39 54 6
requires C, 75.69; H, 8.79%); d 7.81 and 7.82 (6H, m, ArH),
4.24 (10H, t, J 6.6, OCH ), 4.10 (3H, s, OCH ), 1.94 (10H,
H
2
3
m, OCH CH ), 1.62 (10H, m, OCH CH CH ), 1.05 (15H, t,
2
2
2
2
2
(41 g, 92%) (Found: C, 75.35; H, 10.0. C H O requires C,
J 7.4, OCH CH CH CH ); m/z (EI) 618 (M+, 100%).
14 22 2
75.64; H, 9.97%); d 6.88 (4H, s, ArH), 4.00 (4H, t, J 6.6,
H
2
2
2
3
3,6,7,10,11-Pentabutoxy-2-hydroxytriphenylene 11
OCH ), 1.80 (4H, m, OCH CH ), 1.52 (4H, m,
2
2
2
2
OCH CH CH ), 0.98 (6H, t, J 6.3, OCH CH CH CH ); m/z
2
2
2
2
2
3
Diphenylphosphine15 (3.35 g, 18 mmol) was added to stirred
dried tetrahydrofuran (30 cm3) under argon at 0 °C followed
by butyllithium solution in hexane (11.2 cm3, 1.6 M, 18 mmol)
over 30 min. After a further 30 min, 3,6,7,10,11-pentabutoxy-
2-methoxytriphenylene 10 (5.56 g, 9.0 mmol) was added, the
mixture heated under reflux and the progress of the reaction
monitored by TLC. Once the reaction was complete the
mixture was poured onto ice, extracted with dichloromethane
(2×100 cm3), the organic phase washed with aqueous hydro-
chloric acid (2 M, 100 cm3) and water (100 cm3) and concen-
trated in vacuo. The crude product was subjected to column
chromatography on silica eluting with dichloromethane–light
petroleum (151), and the product recrystallised from ethanol
to give 3,6,7,10,11-pentabutoxy-2-hydroxytriphenylene 11 as a
white crystalline solid (3.70 g, 68%), mp 101.5–102.5 °C
(EI) 222 (M+, 14%).
1,2-Dibutoxy-4-iodobenzene
Iodine monochloride (50 g, an excess) in chloroform (150 cm3)
was added dropwise to 1,2-dibutoxybenzene (41 g, 0.19 mol)
in chloroform (100 cm3) with stirring. The progress of the
reaction was monitored by 1H-NMR spectroscopy. Once the
reaction was complete, the liquid was decanted off, washed
with sodium metabisulfite solution (1 M, 100 cm3) and water
(100 cm3), concentrated in vacuo and the residue purified by
column chromatography on silica eluting with dichloro-
methane–light petroleum (253) to give 1,2-dibutoxy-4-iodoben-
zene as an orange oil (47 g, 71%) (Found: C, 47.65; H, 6.0.
C H IO requires C, 48.28; H, 6.08%); d 7.18 (1H, d, J 8.4,
14 21
2
H
ArH), 7.16 (1H, s, ArH), 6.60 (1H, d, J 8.4, ArH), 3.95 (4H,
t, J 6.6, ArOCH ), 1.78 (4H, m, OCH CH ), 1.50 (4H, m,
(Found: C, 74.75; H, 8.65. C H O requires C, 75.46; H,
38 52 6
8.66%); d 7.90 (1H, s, ArH), 7.83 (4H, m, ArH), 7.78 (1H,
2
2
2
H
OCH CH CH ), 0.97 (6H, t, J 6, OCH CH CH CH ); m/z
s, ArH), 5.91 (1H, s OH), 4.24 (10H, m, OCH ), 1.93 (10H,
2
2
2
2
2
2
3
2
(EI) 348 (M+, 25%).
m, OCH CH ), 1.57 (10H, m, OCH CH CH ), 1.04 (15H, t,
2
2
2
2
2
J 7.0, OCH CH CH CH ); m/z (EI), 604 (M+, 100%).
2
2
2
3
3,3∞,4,4∞-Tetrabutoxybiphenyl 5b24
1,8-Bis(3∞,6∞,7∞,10∞,11∞-pentabutyloxytriphenylen-2∞-
yloxy)octane 12a
1,2-Dibutoxy-4-iodobenzene (20 g) was added to copper
powder (20 g) and mixed intimately. The mixture was heated
carefully to about 270 °C. An exothermic reaction occurred
and the temperature rose to about 300 °C. The mixture was
cooled, extracted with dichloromethane (2×100 cm3), concen-
trated in vacuo, and ethanol was added to initiate crystallisation
to give 3,3∞,4,4∞-tetrabutoxybiphenyl 5b as a white solid (4.4 g,
A mixture of 2-hydroxy-3,6,7,10,11-pentabutyloxytriphenylene
11 (1.21 g, 2.0 mmol), 1,8-dibromooctane (0.28 g, 1.0 mmol)
and anhydrous potassium carbonate (1.0 g) in ethanol (5 cm3)
was heated under reflux for 72 h. The mixture was cooled to
0 °C, filtered, washed with water (25 cm3), and extracted with
dichloromethane (2×25 cm3), the solvent removed in vacuo,
and the residue purified by column chromatography on silica
eluting with dichloromethane and finally recrystallised from
ethanol to give 1,8-bis(3∞,6∞,7∞,10∞,11∞-pentabutoxytriphenylen-
2∞-yloxy)octane 12a as a white solid (0.60 g, 46%) (Found: C,
35%), mp 93–94 °C (Found: C, 75.85; H, 9.8. C H O
28 42 4
requires C, 75.98; H, 9.56%); d 7.07 (4H, m, ArH), 6.94 (2H,
H
d, J 8.7, ArH), 4.03 (8H, t, J 6.6, OCH ), 1.82 (8H, m,
OCH CH ), 1.53 (8H, m, OCH CH CH ), 0.98 (12H, t, J 7.4,
2
2
2
2
2
2
OCH CH CH CH ); m/z (EI) 442 ([M]+, 100%).
2
2
2
3
76.25; H, 9.10. C H
7.83 (12H, s, ArH), 4.24 (24H, t, J 6.5, OCH ), 1.93 (24H,
O
requires C, 76.21; H, 8.98%); d
84 118 12
H
1-Butoxy-2-methoxybenzene 6d25
2
m, OCH CH ), 1.62 (24H, m, OCH CH CH ), 1.04 [30H, t,
A mixture of 2-methoxyphenol (12.4 g, 0.1 mol), 1-bromo-
butane (16.4 g, 0.12 mol) and anhydrous potassium carbonate
(27.8 g, 0.2 mol) in ethanol (50 cm3) was heated under reflux
for 24 h. After the reaction had finished, the reaction mixture
was filtered, washed with water (50 cm3), extracted with
dichloromethane (2×50 cm3) and concentrated in vacuo to
give 1-butoxy-2-methoxybenzene 6d as a colourless liquid (15 g,
2
2
2 3
2
2
2
J 7.1, O(CH ) CH ].
3
1,10-Bis(3∞,6∞,7∞,10∞,11∞-pentabutyloxytriphenylen-2∞-
yloxy)decane 12b
A mixture of 2-hydroxy-3,6,7,10,11-pentabutoxytriphenylene
11 (1.05 g, 1.7 mmol), 1,10-dibromodecane (0.26 g, 0.9 mmol)
and anhydrous potassium carbonate (1.0 g) in ethanol (5 cm3)
was heated under reflux for 72 h. The mixture was cooled,
filtered, washed with water (20 cm3), and extracted with
83%) (Found: C, 73.0; H, 9.0. C H O requires C, 73.30; H,
11 16 2
8.94%); d 6.89 (4H, s, ArH), 4.02 (2H, t, J 6.8, OCH ), 3.85
H
2
(3H, s, OCH ), 1.83 (2H, m, OCH CH ), 1.51 (2H, m,
3
2
2
1400
J. Mater. Chem., 1999, 9, 1391–1402

dichloromethane (2×20 cm3), the solvent removed in vacuo,
and the residue purified by column chromatography on
silica eluting with dichloromethane–light petroleum (251) and
finally recrystallised from ethanol to give 1,10-
bis(3∞,6∞,7∞,10∞,11∞-pentabutoxytriphenylen-2∞-yloxy)decane 12b
as a white solid (0.40 g, 34%) (Found: C, 76.35; H, 9.20.
(12H, t, J 7, O CH CH ), 1.5–1.35 (47H, m, CH ), 0.85 (15H,
2
2
2
t, J 7, CH CH ).
2
3
6-(3∞,6∞,7∞,10∞,11∞-Pentahexyloxytriphenylen-2∞-yloxy)hexanoyl
chloride 16
A solution of oxalyl chloride (0.5 g) in dry dichloromethane
C H
86 122 12
ArH), 4.24 (24H, t, J 6.5, OCH ), 1.93 (24H, m, OCH CH ),
O
requires C, 76.64; H, 9.12%); d 7.84 (12H, s,
H
(5 cm3) was slowly added to
a stirred solution of 6-
2
2
2 3
2
3
(3∞,6∞,7∞,10∞,11∞-pentahexyloxytriphenylen-2∞-yloxy)hexanoic
acid 15 (700 mg, 0.78 mmol) in dry dichloromethane (20 cm3)
under nitrogen and the mixture left at room temperature for
24 h. The solvent was then evaporated in vacuo. The product
16 was used in its crude state for the next step.
1.35–1.70 (32H, m, CH ), 1.05 [30H, t, J 7.3, O(CH ) CH ];
m/z (FAB) 1346 (M+, 100%).
2
1-(3∞,6∞,7∞,10∞,11∞-Pentabutoxytriphenylen-2∞-yloxy)-10-
(3◊,6◊,7◊,10◊,11◊-pentahexyloxytriphenylen-2◊-yloxy)decane 13
2-(6∞-Aminohexyloxy)-3,6,7,10,11-pentahexyloxytriphenylene
17
A mixture of 2-(10-bromodecyloxy)-3,6,7,10,11-pentahexylox-
ytriphenylene 9c (0.30 g, 0.3 mmol), 3,6,7,10,11-pentabutoxy-
2-hydroxytriphenylene 11 (0.15 g, 0.25 mmol) and anhydrous
potassium carbonate (1.0 g) in ethanol (5 cm3) was heated
under reflux for 72 h. The mixture was cooled, filtered, washed
with water (50 cm3), and extracted with dichloromethane
(2×50 cm3), the solvent removed in vacuo, and the residue
purified by column chromatography on silica eluting with
dichloromethane–light petroleum (251) and finally recrystal-
lised from ethanol to give 1-(3∞,6∞,7∞,10∞,11∞-pentabutoxytriphen-
ylen-2∞-yloxy)-10-(3◊,6◊,7◊,10◊,11◊-pentahexyloxytriphenylen-
2◊-yloxy)decane 13 as a white solid (0.10 g, 26%) (Found: C,
A stirred mixture of 2-(6∞-bromohexyloxy)-3,6,7,10,11-penta-
hexyloxytriphenylene 9b (1.35 g, 1.5 mmol) and sodium azide
(500 mg) in ethanol (50 cm3) was refluxed for 24 h. The
mixture was poured into water (50 cm3), extracted with
dichloromethane (2×50 cm3) and the solvent evaporated in
vacuo. The crude azide was dissolved in dry tetrahydrofuran
(50 cm3), lithium aluminium hydride (5 equiv.) carefully
added, and the mixture stirred for 1 h. Dilute aqueous hydro-
chloric acid (20 cm3, 2 M) was carefully added and the mixture
extracted with dichloromethane (2×50 cm3). The organic
solvent was evaporated in vacuo and the crude residue purified
by column chromatography on silica eluting with methanol–
dichloromethane (159) to give 2-(6∞-aminohexyloxy)-
3,6,7,10,11-pentahexyloxytriphenylene 17 (950 mg, 77%) as a
white solid which quite rapidly turned brown on exposure to
the air and hence was used directly for the next step; Cr 52
Col 185 °C I; d 7.78 (6H, s, ArH), 5.4–6.0 (2H, br s, NH ),
77.6; H, 9.9. C H
(12H, s, ArH), 4.23 and 4.24 (24H, 2×t, J 0.5, OCH ),
1.85–2.05 (24H, m, OCH CH ), 1.25–1.70 (52H, m, CH ),
1.04 [15H, t, J 7.4, O(CH ) CH ], 0.93 [15H, t, J 6.7,
O
requires C, 77.48; H, 9.62%); d 7.83
96 142 12
H
2
2
2
2
2 3
3
O(CH ) CH ]; m/z (FAB) 1487, ([M+1]+, 60%).
2 5
3
Ethyl 6-(3∞,6∞,7∞,10∞,11∞-pentahexyloxytriphenylen-2∞-
yloxy)hexanoate 14
H
2
2
4.20 (12H, t, J 7, ArOCH R), 3.03 (2H, m, CH NH ),
2
2
1.20–1.95 (48H, m, CH ), 0.97 (15H, t, J 7, CH ).
A stirred mixture of 3,6,7,10,11-pentahexyloxy-2-hydroxytri-
phenylene 8 (2.78 g, 3.7 mmol),4 ethyl 6-bromohexanoate
(1.06 g, 4.6 mmol) and anhydrous potassium carbonate
(1.17 g) in ethanol (50 cm3) was refluxed for 36 h. The cooled
solution was decanted into water (100 cm3), the mixture
extracted with dichloromethane (2×100 cm3), the solvent was
removed in vacuo, and the product purified by column chroma-
tography on silica eluting with light petroleum–dichlorome-
thane (352). Recrystallisation from ethanol (20 cm3) gave
ethyl 6-(3∞,6∞,7∞,10∞,11∞-pentahexyloxytriphenylen-2∞-yloxy)hex-
2
3
6,6∞-Bis(3◊,6◊,7◊,10◊,11◊-pentahexyloxytriphenylen-2◊-yloxy)-
N-hexylhexanamide 18
6-(3∞,6∞,7∞,10∞,11∞-Pentahexyloxytriphenylen-2∞-yloxy)hexanoyl
chloride 16 (525 mg, 0.6 mmol) in dry tetrahydrofuran
(10 cm3) was added to
a
stirred solution of 2-(6∞-
17
aminohexyloxy)-3,6,7,10,11-pentahexyloxytriphenylene
(930 mg, 1.2 mmol) in dry tetrahydrofuran (10 cm3). After
1 h at room temperature the mixture was poured into water
(50 cm3), extracted with dichloromethane (2×50 cm3) and the
organic solvent removed in vacuo. The crude product was
purified by column chromatography on silica eluting with
dichloromethane to give the amide 18 (160 mg, 16%) as a
white solid; Cr 88.5 Col 99.1 I 94.5 Col ~30 °C glass (Found:
anoate 14 (1.21 g, 37%) as a white solid; Cr 36 Col 53 °C I
h
(Found: C, 75.7; H, 9.80. C H O requires: C, 75.85; H,
56 86 8
9.70%); d 7.85 (6H, s, ArH), 4.20 (12H, t, J 7, OCH ), 4.15
H
2
(2H, q, J 7, OCH CH ), 2.35 (2H, t, J 7, CH CH CO), 1.95
2
3
2
2
(12H, t, J 7, O CH CH ), 1.75 (2H, m, CH CH CO), 1.5–1.35
(47H, m, CH ), 1.20 (3H, t, J 7, OCH CH ), 0.85 (15H, t, J
2
2
2
2
C, 77.3; H, 9.6; N, 0.71. C
H
O N requires: C, 77.0; H,
2
2
3
108 165 13
7, CH CH ).
9.8; N, 0.83%); d 7.78 (12H, s, ArH), 4.25 (24H, t, J 7,
2
3
H
ArOCH R), 3.30 (2H, m, CH NHCO), 2.25 (2H, t, J 7,
2
2
CH CONH), 1.20–1.95 (48H, m, CH ), 0.97 (15H, t, J 7,
CH ); m/z (FAB) 1684 (M+, 40%).
6-(3∞,6∞,7∞,10∞,11∞-Pentahexyloxytriphenylen-2∞-yloxy)hexanoic
acid 15
2
3
2
A vigorously stirred mixture of ethyl 6-(3,6,7,10,11-pentahexy-
loxytriphenylen-2-yloxy)hexanoate 14 (1.27 g, 1.43 mmol) and
sodium hydroxide (1 pellet) in water (10 cm3) and methanol
(20 cm3) was refluxed for 24 h. The cooled mixture was
carefully poured into dilute aqueous hydrochloric acid (50 cm3;
2 M), extracted with dichloromethane (2×50 cm3), and the
solvent evaporated. The crude product was purified by column
chromatography on silica eluting with dichloromethane–ethyl
acetate (151) and was recrystallised from ether (10 cm3) at
0 °C to give 6-(3∞,6∞,7∞,10∞,11∞-pentahexyloxytriphenylen-2∞-ylox-
Force field calculations
The MM2 force field method was used throughout.15 To
calculate strain energies for intracolumnar bridging (Table 1)
we needed to know the separation between aromatic cores
within the columns. In 1,2-intrastack bridging it was assumed
that the aryl ring–oxygen bonds were parallel and 3.56 A
apart, in the case of 1,3-intrastack bridging 7.12 A apart, in
the case of 1,4-intrastack bridging 10.68 A apart, in the case
of 1,5-intrastack bridging 14.24 A apart, and in the case of
1,6-intrastack bridging 17.80 A apart. For calculations on
˚
˚
˚
˚
˚
y)hexanoic acid 15 (750 mg, 61%) as a white solid; Cr 60 Col
h
72 °C I (Found: C, 75.35; H, 9.6. C H O requires: C, 75.5;
54 84 8
intercolumn bridging we also need to know the separation
between the columns ( y). This was estimated using eqn. (1)
which was obtained by fitting X-ray diffraction data for the
H, 9.56%); d 7.85 (6H, s, ArH), 4.20 (12H, t, J 7, OCH ),
H
2
2.75 (1H, s, CO H), 2.35 (2H, t, J 7, CH CH COOH), 1.95
2
2
2
J. Mater. Chem., 1999, 9, 1391–1402
1401

B. Glusen, A. Greiner, A. Kettner, R. Sander, V. Stumpflen,
V. Tsukruk and J. H. Wendorff, Adv. Mater., 1997, 9, 48;
N. Boden, R. J. Bushby, J. Clements and B. Movaghar, J. Appl.
Phys., 1998, 83, 3207; N. Boden, R. J. Bushby, J. Clements,
K. Donovan, B. Movaghar and T. Kreouzis, Phys. Rev. B, 1998,
58, 3063.
HATN series26 to a second order polynomial,
y=7.50+0.388c−0.00139c
(1)
where y is in angstroms and c is the total number of carbons
in the side chains per aromatic core (6N in HATN, [5N+n/2]
for 2). Hence, for the oligomer 2c, n=7, c=33.5 and in the
3
4
C. Destrade, M. C. Mondon and J. Malthete, J. Phys. Colloq. C3,
1979, 40, 17.
˚
Col phase y would be 18.94 A. For calculations of the
h
N. Boden, R. J. Bushby and A. N. Cammidge, J. Chem. Soc.,
Chem. Commun., 1994, 465; J. Am. Chem. Soc., 1995, 117, 924.
N. Boden, R. J. Bushby and Z. Lu, Liq. Cryst., 1998, 25, 47.
T. Christ, B. Glusen, A. Greiner, A. Kettner, R. Sander,
V. Stumpflen, V. Tsukruk and J. H. Wendorff, Adv. Mater., 1997,
9, 48.
N. Boden, R. J. Bushby, A. N. Cammidge and P. S. Martin,
J. Mater. Chem., 1995, 5, 1857.
A. Bacher, I. Bleyl, C. H. Erdelen, D. Haarer, W. Paulis and
H.-W. Schmidt, Adv. Mater., 1997, 9, 1031.
amounts of strain involved in 1,2-interstack bridging the
constraint applied was that the aryl ring–oxygen bonds were
co-linear and parallel with the oxygens ( y−9.85) A apart and
5
6
˚
in the case of 1,3-intrastack bridging the oxygens
˚
(1.732y−9.85) A apart. Obtaining the energy minimum for
7
8
9
extended all-trans conformations (analogous to 4a) is a trivial
exercise but, for the folded chains (analogous to 4b) it is
difficult to be sure that the absolute minimum energy structure
has been located. This is a common problem in all cases of
‘folding’ of polymethylene chains since the energy surfaces are
characterised by a multitude of local energy minima. For each
data point in Table 1, at least three independent minimisations
starting from different initial conformations were performed
and the best minimum generated was taken as (at least close
to) the absolute minimum. To abstract the chain component
of the strain energy for each optimised geometry the PhO
groups were each replaced by H and (without changing any
other atomic positions) the energy of the chain recalculated.
Hence Table 1 gives the MM2 calculated difference between
all-trans H-(CH ) -H chains and H-(CH ) -H chains which
C. P. Lillya and Y. L. N. Murthy, Mol. Cryst. Liq. Cryst., 1985,
2, 121.
10 S. Zamir, R. Poupko, Z. Luz, B. Huser, C. Boeffel and
H. Zimmerman, J. Am. Chem. Soc., 1994, 116, 1973; D. Adam,
P. Schuhmacher, J. Simmerer, L. Haussling, W. Paulis,
K. Siemensmeyer, K.-H. Etzbach, H. Ringsdorf and D. Haarer,
Adv. Mater., 1995, 7, 276.
11 W. Kranig, B. Huser, H. W. Spiess, W. Krender, H. Ringsdorf
and H. Zimmermann, Adv. Mater., 1990, 2, 36.
12 B. Kohne, P. Marquardt, K. Praefke, P. Psaras, W. Stephan and
K. Turgay, Chimia, 1986, 40, 360; V. Percec, C. G. Cho, C. Pugh
and D. Tomazos, Macromolecules, 1992, 25, 1164; G. C. Bryant,
M. J. Cook, S. D. Haslam, R. M. Richardson, T. G. Ryan and
A. J. Thorne, J. Mater. Chem., 1994, 4, 209; A. Takaragi,
M. Sugiura, M. Minoda, T. Miyamoto and J. Watanabe,
Macromol. Chem. Phys., 1997, 8, 2583.
2 n
2 n
have been folded as required for the different types of bridging.
13 S. Zamir, E. J. Wachtel, H. Zimmerman, S. Dai, N. Speilberg,
R. Poupko and Z. Luz, Liq. Cryst., 1997, 23, 689.
14 J. W. Emsley, G. R. Luckhurst and G. N. Shilstone, Mol. Cryst.
Liq. Cryst. Lett., 1984, 102, 223.
Acknowledgements
We thank the Overseas Research Students Awards Scheme
(O.R.S.) and the University of Leeds (Tetley and Lupton
Scheme) for awards to Z.B.L. and the EPSRC for the award
of a Research Studentship to P.S.M.
15 N. L. Allinger, J. Am. Chem. Soc., 1997, 99, 8127; U. Burket and
N. L. Allinger, Molecular Mechanics, A.C.S. Monograph, 1982.
16 N. Boden, R. J. Bushby, A. N. Cammidge and G. Headdock, UK
Pat. 94118288, 11th June 1993; Synthesis, 1995, 31; J. Mater.
Chem., 1995, 5, 2275.
17 R. E. Ireland and D. M. Warlba, Org. Synth., 1998, Col. Vol. 6,
References
567.
18 M. Werth, S. H. Vallerien and H. W. Spiess, Liq. Cryst., 1991,
10, 759.
19 Purification of Laboratory Chemicals, ed. D. D. Perrin,
W. L. F. Armarego and D. R. Perrin, 2nd edn., Pergamon Press,
Oxford, 1980.
20 P. Chuit and J. Hauser, Helv. Chim. Acta, 1929, 12, 850.
21 E. Klarman, L. W. Gates and V. A. Shternov, J. Am. Chem. Soc.,
1932, 54, 1204.
22 K. H. Slotta and W. Franke, Chem. Ber., 1930, 63, 678.
23 S. Kumar and M. Manicham, Chem. Commun., 1998, 1427.
24 J. Allan and R. Robinson, J. Chem. Soc., 1926, 376.
25 C. R. Safinya, N. A. Clark, K. S. Liang, W. A. Varady and
L. A. Chiang, Mol. Cryst. Liq. Cryst., 1985, 123, 205.
1
N. Boden, R. Bissell, J. Clements and B. Movaghar, Curr. Sci.,
1996, 6, 599; N. Boden and B. Movaghar, Handbook of Liquid
Crystals, vol. 2B, ed. D. Demus, J. W. Goodby, G. W. Gray,
H. W. Spiess and V. Vill, VCH, Weinheim, 1998, p. 781.
2
N. Boden, R. J. Bushby, J. Clements, M. V. Jesudason,
P. F. Knowles and G. Williams, Chem. Phys. Lett., 1988, 152, 94;
N. Boden, R. J. Bushby and J. Clements, J. Chem. Phys., 1993,
98, 5920; D. Adam, F. Closs, T. Frey, D. Funhoff, D. Haarer,
H. Ringsdorf, P. Schuhmacher and K. Siemensmeyer, Phys. Rev.
Lett., 1993, 70, 457; N. Boden, R. C. Borner, R. J. Bushby and
J. C. Clements, J. Am. Chem. Soc., 1994, 116, 10807; N. Boden,
R. J. Bushby, J. Clements, B. Movaghar, K. J. Donovan and
T. Kreouzis, Phys. Rev. B, 1995, 52, 13274; M. Funahashi and
J. Hanna, Jpn. J. Appl. Phys., 1996, 35, L703; Appl. Phys. Lett.,
1997, 71, 602; Phys. Rev. Lett., 1997, 78, 2184; T. Christ,
Paper 8/10045D
1402
J. Mater. Chem., 1999, 9, 1391–1402