Hydrogen bonded liquid crystal from nitrophenols and alkoxystilbazoles

Article abstract of DOI:10.1039/a700575j

Tree series hydrogen bonded adducts have been formed between 4-alkoxy-4'-stilbazoles and the nitrophenols 4-nitrophenol, 3-nitrophenol and 2,4-dinitrophenol.Each series is mesomorphic displaying both nematic and smectic.A phases, in sharp contrast to the

Full text of DOI:10.1039/a700575j

Hydrogen bonded liquid crystals from nitrophenols and alkoxystilbazoles
Daniel J. Price,‡b Kimberley Willis,b Tim Richardson,b Goran Ungarb and Duncan W. Bruce*a
aDepartment of Chemistry, University of Exeter, Stocker Road, Exeter, UK EX4 4QD
Fax: +44 1392 263434. Email: d.bruce@exeter.ac.uk.
bCentre for Molecular Materials, Dainton Building, University of SheÃeld, SheÃeld, UK S3 7HF
Three series of hydrogen bonded adducts have been formed between 4-alkoxy-4∞-stilbazoles† and the nitrophenols 4-nitrophenol,
3
-nitrophenol and 2,4-dinitrophenol. Each series is mesomorphic displaying both nematic and smectic A phases, in sharp contrast
to the behaviour of the individual components. The binary phase behaviour of mixtures of 3-nitrophenol and 4-octyloxystilbazole
is reported, and gives conclusive evidence for the formation of a one-to-one hydrogen bonded adduct. Electronic spectroscopy of
2
,4-dinitrophenol/decyloxystilbazole was very informative and showed that the higher energy, ionic hydrogen bonded state,
corresponding to proton transfer, was significantly populated through the mesophase, and that the mesophase provides an
additional stabilisation for this state.
phenol and 2,4-dinitrophenol. Each of these complexes will be
Introduction
referred to by the abbreviated format X-Nn; where X refers to
In the field of liquid crystals, hydrogen bonding has long been
recognised as crucial to mesophase formation in certain sys-
tems. An early, and now classic, example is that of the
alkoxybenzoic acids,1 where dimerisation of the molecules
gives enhanced structural anisotropy. Another class of hydro-
gen bonded mesogen are the alkylsilanediols,2 the aggregation
of which has been the subject of investigation.3 The silanediols
are in turn related to the large class of mesomorphic poly-
alcohols, characterised by periodicity within the mesophase
structure, and including compounds from the straight chain
a,b,y-alkanetetraols,4 to carbohydrates.5 In the latter case,
mesophase formation is driven by a microphase separation
of diÂerent parts (polar and non-polar) of the molecules.
More recently, the results of studies of complementary
heteromeric hydrogen bonded systems have been reported.6
Lehn and co-workers7 demonstrated induced mesomorphism
in ‘pseudo’ main-chain polymers and low molecular mass
materials self-assembled from complex, complementary parts.
On a similar theme but at a structurally simpler level Yu,8
GriÃn,9 ourselves10 and particularly Kato,11 and their respect-
ive co-workers, have described many mesomorphic materials
based on the adducts formed between substituted pyridines
and carboxylic acid groups.
the position of substitution on the phenol and n is the length
of the carbon chain on the stilbazole (thus, 2,4-N6 would
represent the complex between hexyloxystilbazole and 2,4-
dinitrophenol).
The proposed structures of these complexes are shown
below; variation of the alkyl chain length, n, of the stilbazole
gave three homologous series. The mesomorphism of these
materials was studied by optical microscopy and diÂerential
scanning calorimetry (DSC). The phase behaviour, transition
temperatures and associated thermal parameters are given in
Tables 1–3.
The behaviour of these adducts is very diÂerent to that
shown by the individual components. For example, the alkoxy-
stilbazoles14 show a narrow range of smectic B and crystal
smectic E phases, typically between 80 and 90 °C, while the 4-
nitro-, 3-nitro- and 2,4-dinitro-phenols simply melt at 113, 96
and 108 °C, respectively.
Alongside work based on this carboxylic acid pyridyl inter-
action, we have examined liquid crystalline systems where the
proton donor is a phenol. We demonstrated monotropic
mesomorphism in complexes of 4-cyanophenol with 4-alkoxy-
stilbazoles,12 and, more recently, enantiotropic nematic and
smectic A phases in complexes of 3-cyanophenol with 4-
alkoxystilbazoles,13 showing how structural anisotropy, and
hence mesophase stability, can be simply controlled in these
complexes. It was also clear that the phenol–stilbazole hydro-
gen bond was suÃciently strong to allow the formation of
mesophases from the complexes.
Having studied these cyanophenols, we then turned our
attention to nitrophenols as other examples of phenols bearing
polarisable groups, and three series of complexes containing
nitrophenols were prepared.
Mesomorphism of the Complexes
Hydrogen bonded adducts were formed between 4-alkoxystil-
Complexes of 4-nitrophenol with 4-alkoxystilbazole (4-Nn)
bazole and the following nitrophenols: 4-nitrophenol, 3-nitro-
This series has the phase behaviour shown in Fig. 1.
Monotropic nematic phases are seen for homologues from
methoxy (4-N1) to octyloxy (4-N8). The nonyloxy derivative
(4-N9) sees the onset of smectic behaviour, with both nonyloxy
(4-N9) and decyloxy (4-N10) derivatives showing monotropic
†
4
‡
Stilbazole=styrylpyridine; 4∞-stilbazole denotes attachment at the
-position of the pyridine ring.
Anglia, University Plain, Norwich, UK NR4 7TJ.
Current address: School of Chemical Sciences, University of East
J. Mater. Chem., 1997, 7(6), 883–891
883

Table 1 Mesomorphism, and associated enthalpies and entropies for
the 4-Nn series of hydrogen bonded adducts; monotropic transitions
are indicated by parentheses
Table 2 Mesomorphism, and associated enthalpies and entropies for
the 3-Nn series; monotropic transitions are indicated by parentheses
DH
DS
DH
DS
complex
transition
t/°C
/kJ mol−1
/J K−1 mol−1
complex
transition
Crys–I
Crys–I
t/°C
/kJ mol−1
/J K−1 mol−1
3
-N1
Crys–I
98
(97)
114
(99)
90
33.0
(1.01)
37.2
—a
89
4-N1
4-N2
4-N3
4-N4
4-N5
4-N6
4-N7
4-N8
4
-N9
118
35.2
91
(N–I)
(2.7)
96
(
N–I)
88
—a
3-N2
3-N3
3-N4
Crys–I
(N–I)
101
85
31.8
0.68
32.2
(0.4)
27.6
(0.4)
36.6
(0.41)
42.2
(0.4)
47.8
(0.42)
51.7
(0.85)
52.3
(41.9)
b
85
(
(
N–I)
N–I)
N–I)
N–I)
N–I)
N–I)
N–I)
1.9
Crys–I
(N–I)
33.2
(0.82)
35.5
(24.8)
0.93
28.5
0.54
44.6
0.7
92
(2.3)
105
(75.0)
2.6
Crys–I
100
(70)
93
98
(83)
66
(1.2)
75.5
(1.1)
103
Crys–N
(Crys∞–N)
N–I
Crys–I
(60)
89
(
(79)
82
Crys–I
3-N5
3-N6
3-N7
3-N8
Crys–N
N–I
60
86
(
(71)
97
(1.2)
115
84
1.5
Crys–I
Crys–N
N–I
79
126
1.8
(
(77)
92
(1.0)
131
88
Crys–I
Crys–N
N–I
81
41.5
0.55
41.9
—a
118
1.5
(
(77)
99
(1.2)
139
86
Crys–I
Crys–N
(Crys∞–N)
N–I
75
120
(
(80)
93
(2.4)
143
(74)
Crys–I
90
0.84
44.2
—a
2.3
(
(
(
Crys∞–N)
(79)
(70)
(82)
97
(122)
3-N9
Crys–N
(Crys∞–N)
N–I
71
129
S –N)
(70)
86
A
N–I)
(1.07)
57.4
b
(3.0)
0.77
2.2
4-N10
4-N11
4-N12
4-N13
Crys–I
155
3-N10
3-N11
3-N12
3-N13
Crys–S
A
78
51.8b
148b
(
S –N)
(84)
(86)
90
S –N
79
A
A
(
N–I)
(−0.83)c
56.3
—a
(−2.3)
N–I
90
1.11
54.1
0.22
1.25
56.5
—a
3.1
155
0.61
3.4
Crys–I
155
Crys–S
A
77
(
Crys∞–S )
(80)
89
S –N
87
A
A
S –I
2.50
36.4
—a
6.9
N–I
90
A
Crys–S
80
103
Crys–S
80
160
A
A
(
Crys∞–S )
(74)
(Crys∞–S )
(66)
A
A
S –I
91
2.98
54.9
3.50
8.2
152
9.5
S –I
92
2.92
64.6
—a
8.0
A
A
Crys–S
88
Crys–S
79
184
A
A
S –I
95
(Crys∞–S )
(70)
A
A
S –I
93
4.04
11
A
aDenotes a transition not seen in the DSC experiment. bIndicates a
second order transition. cIndicates where monotropic transitions could
not be seen or resolved as part of a heating cycle, and therefore the
cooling cycle was used.
aDenotes a transition not seen in the DSC experiment. bIndicates
combined enthalpies of transition where a baseline resolution was
not obtained.
Fig. 1 The phase behaviour of the 4-Nn series: (%) K–, (1) S –,
Fig. 2 Plot of nematic–isotropic transition temperature vs. chain
length for complexes of stilbazoles with nitro- and cyano-phenols: (%)
A
(
#) N–I
4
-Nn, (#) 3-Nn, (1) 2,4-Nn, (Z) 4-Cn, (C) 3-Cn
smectic A and nematic phases. The 4-N11 has just a mono-
tropic smectic A phase, while for the 4-N12 and 4-N13
homologues, the combination of the continued destabilisation
of the crystal phase and the increased stability of the smectic A
phase resulted in enantiotropic behaviour.
nematic phase at longer chain lengths, and in the nitro
derivatives, this is a gradual change, with intermediate chain
lengths showing both nematic and smectic A phases. In con-
trast, there is a sudden switch from nematic to smectic A in
the analogous cyano derivatives. Thus, extensive supercooling
of the nematic phase of 4-C9 never revealed a smectic phase,
yet 4-C10 showed only a smectic A structure.
This series of adducts can be compared with the analogous
complexes formed between 4-cyanophenol and alkoxystil-
bazoles,13 (4-Cn—this abbreviation refers to the analogous
complexes with 4-cyanophenol). First, in the 4-Nn series the
mesophases are more thermally stable (Fig. 2) by some
There are also similarities in the melting behaviour of both
series. After initially decreasing from the high melting methoxy
and ethoxy derivatives, an odd–even eÂect is seen in both
cases, although this is more pronounced in the 4-Nn complexes.
In both 4-Nn and 4-Cn complexes, we also noted the existence
1
0–20 °C, which we tentatively attribute to enhanced antiparal-
lel correlations induced by the nitro group in these systems.
In both series, the smectic A phase is seen to replace the
8
84
J. Mater. Chem., 1997, 7(6), 883–891

Table 3 Mesomorphism, and associated enthalpies and entropies for
the 2,4-Nn series; monotropic transitions are indicated by parentheses
showing, in addition, a smectic A phase. The 4-N12 and 4-
N13 homologues show only the smectic A phase between ca.
8
0 and 93°C. In this series, a normal odd–even eÂect is clearly
DH
DS
seen at the nematic-to-isotropic phase transition.
complex
transition
Crys–I
t/°C
/kJ mol−1
/J K−1 mol−1
For the 3-Nn series, two comparisons can be made; first,
with the structurally analogous cyano derivatives already
described,13 and secondly with the 4-Nn series to assess the
relative merits of meta and para substitution. Comparison of
the 3-Nn series with the analogous 3-cyanophenol–alkoxystil-
bazole (3-Cn) complexes revealed (Fig. 2) that the thermal
stability of mesophases for the 3-Nn complexes was again
typically 10–20 °C higher than that of their cyano counterparts.
As with the 4-Nn series, the onset of the smectic behaviour
was gradual, with intermediate derivatives being polymeso-
morphic. This is in contrast to the sudden change seen for the
2
2
2
,4-N1
,4-N2
,4-N3
150
(148)
(80)
157
(120)
(72)
139
(121)
(55)
114
(100)
(69)
114
(111)
(72)
113
(85)
101
(95)
(94)
95
43.2
102
104
117
103
127
(
(
Crys∞–I)
—a
N–I)
—a
Crys–I
44.8
(
Crys∞–I)
—a
(
N–I)
—a
Crys–I
47.8
—a
(
(
Crys∞–I)
N–I)
—a
2
,4-N4
Crys–I
41.9
(
Crys∞–I)
—a
(
N–I)
—a
3
-Cn complexes, where 3-C12 shows only a nematic phase and
2
,4-N5
Crys–I
49.3
—a
(
(
Crys∞–I)
3-C13 only a smectic A phase.
N–I)
—a
Comparison of the phase behaviour of the 3-Nn series with
that of the 4-Nn series shows the dramatic eÂect of structural
isomerism. The fact that the 3-Nn complexes have a higher
mesophase stability than their corresponding 4-Nn analogues
indicates the importance of structural anisotropy. In the 3-Nn
series, meta substitution counters the ‘kink’ induced in the
structure by the hydrogen bond to the phenol. This explains
2
2
,4-N6
,4-N7
Crys–I
42.8
112
99
(
S –I)
—a
A
Crys–I
36.5
—a
(
Crys∞–I)
(
S –I)
(2.53)
37.4
3.33
48.2
4.06
44.7
4.50
59.2
4.94
47.2
5.32
45.2
4.63
(7.0)
103
8.9
A
2
2
2
2
2
2
,4-N8
,4-N9
,4-N10
,4-N11
,4-N12
,4-N13
Crys–S
A
S –I
104
98
A
Crys–S
130
10.6
121
11.5
159
12.5
129
13.4
122
11.6
the greater stability of the nematic phase (T is at least 10 °C
A
NI
S –I
A
111
97
higher than in 4-Nn complexes) and the more pronounced
Crys–S
A
odd–even eÂect (Fig. 2). The occurrence of the smectic A phase
in both series is very similar appearing at about the same
temperatures for given chain lengths.
S –I
117
101
121
97
A
Crys–S
A
S –I
A
Crys–S
A
S –I
125
97
Complexes of 2,4-dinitrophenol with 4-alkoxystilbazole
A
Crys–S
(
2,4-Nn)
A
S –I
126
A
These complexes were bright yellow in contrast with all other
series previously described, which were cream coloured. The
phase behaviour is displayed in Fig. 4 and again, only nematic
and smectic A phases were observed, although this time the
phase diagram was of a diÂerent form. Thus, for short homol-
ogues (from 2,4-N1 to 2,4-N5), a monotropic nematic phase
was seen at temperatures much lower than the melting points.
Increasing the alkyl chain length saw a sudden transition to
smectic behaviour, and from the 2,4-N6 to the 2,4-N13 deriva-
tive, a smectic A phase was observed. This was stabilised with
increasing alkyl chain length, becoming enantiotropic from
aDenotes a transition not seen in the DSC experiment.
of a second, metastable crystal polymorph which typically
melted at some 10–20 °C lower than the more stable state.
Finally, it was curious to note that the stability of the nematic
phase of these complexes is higher than that of the related
3
-cyanophenol complexes, despite the more advantageous
structural anisotropy in the latter series.
Complexes of 3-nitrophenol with 4-alkoxystilbazole (3-Nn)
2
,4-N8 and reaching a thermal stability of 126°C in 2,4-N13.
The mesomorphism of the 2,4-Nn complexes can be com-
The phase behaviour of this series of complexes is shown in
Fig. 3. All derivatives with the exception of 3-N1, 3-N2 and
pared with both that shown by the 4-Nn and 3-Nn systems.
Although the thermal stability of the nematic phase does not
vary much within any given series, the stability is seen to
increase in the order 2,4-Nn<4-Nn<3-Nn (Fig. 2). This paral-
lels the increase in structural anisotropy, and is consistent with
3
-N3 display enantiotropic mesomorphism. This series shows
the largest thermal range of liquid crystal phases yet seen in
complexes based on the phenol–pyridyl interaction. The
nematic phase is seen for all homologues up to the 3-N9, after
which both the 3-N10 and 3-N11 derivatives are polymorphic
Fig. 3 The phase behaviour of the 3-Nn series: (%) K–, (1) S –,
Fig. 4 The phase behaviour of the 2,4-Nn series: (%) K–, (1) S –I,
A
A
(
#) N–I
(#) N–I
J. Mater. Chem., 1997, 7(6), 883–891
885

Table 4 Small angle X-ray scattering data
˚
˚
complex
T /°C
t/Tcl
phase
d/A
d
/Aa
d/dcalc
calc
4-N12
3-N12
2,4-N12
3-N10
2,4-N8
85
0.93
0.92
0.84
0.94
0.98
S
45.7, 4.5
43.0, 4.4
29.0
32.3
31.3
29.0
26.0
1.46
1.33
1.5
A
85
S
A
105
85
S
A
47.2, 23.2, 4.4
40.0, 4.5
N
1.37
1.54
102
S
40.0, 19.3, 11.1, 4.4
A
aMolecular components calculated using CHEM 3DTM and hydrogen bonded distance from ref. 13.
the idea that formation of the nematic phase is related to the
structural and polarisability anisotropies of the rigid core.
The stability of the smectic A phase as a function of chain
length is shown in Fig. 5. In all series, the transition tempera-
tures are seen to increase rapidly with increasing chain length
beginning to ‘level o ’ at longer homologues. The stability is
clearly much greater in the 2,4-Nn series than in both 4-Nn
and 3-Nn complexes, which are similar to one another. In
addition, we note that smectic behaviour can be induced by a
hexyloxy chain for the 2,4-Nn series, whereas nonyloxy or
decyloxy chains are needed in the 4-Nn and 3-Nn series. That
is, in these cases, much longer alkyl chains are required to
produce the microphase separation of constituent parts that is
the smectic A phase.
an interdigitated bilayer structure. However, the nematic
cannot be interdigitated as it is not layered and so we must
conclude here that we have the predicted antiparallel dimers
as illustrated in Fig. 6. On this basis, we would assume that
this arrangement existed in the S materials and was the cause
A
of the increased layer spacing there, too.
Thermodynamic Properties
From plots of entropy change as a function of chain length, n,
for a particular transition in a given series, certain trends can
be seen. In all cases crystal-to-isotropic liquid or crystal-to-
mesophase transitions can be grouped together, as diÂerences
in entropy of the transition are typically much larger than the
diÂerences found between the isotropic liquid and mesophase.
For all series we can see a general increase in DS with increasing
homologue superimposed by a small parity dependence.
However, as these values depend on factors, such as crystal
structure, which have not been evaluated, then no meaningful
comparison is possible here.
Despite the high acidity of 2,4-dinitrophenol (pK =3.96),15
a
the complex is still expected to exist predominantly in the non-
ionic hydrogen bonded form,16 on account of the small value
of DpK [the diÂerence between the pK of the conjugate acid
a
a
of the stilbazole (ca. 5–6), and the pK of the phenol]. Although
a
the complex is non-ionic (at least in the ground state17—vide
infra), the increased acidity of the phenol strengthens the
hydrogen bond, and increases its polarisability. Thus, the
combination of increased polarisation due to having two nitro
groups and increased polarisability (the hydrogen bond) that
gives an additional stabilisation to the core–core interactions,
disfavouring core–chain interactions, resulting in the predomi-
nance of the smectic A phase in the 2,4-Nn complexes.
For the nematic-to-isotropic transition, enthalpies could
only be obtained for the 4-Nn and 3-Nn series. Both series
show no general trend with values in the range 1.0 to
3.4 J K−1 mol−1. The 3-Nn series shows an odd–even eÂect,
where even homologues show larger transition temperatures
(T ) and larger entropies (DS ). That no great variation was
NI
NI
observed is in agreement with the idea that increasing the
alkyl chain length does nothing to stabilise or increase the
order in the nematic phase, and that the origin of this phase
derives primarily from the rigid core.
X-Ray Scattering Studies
In our studies of the related 3-cyanophenol complexes,13 we
For the smectic A-to-isotropic transition, in all cases we see
a clear increase in the entropy with increasing chain length.
The data for the 4-Nn, 3-Nn and 2,4-Nn series are shown
graphically in Fig. 7. Unlike the nematic phase which has only
one type of order—defined by the order parameter and related
to the orientational distribution of molecules about the
director—the smectic A phase has two. First, it has, like the
nematic, a director and associated ‘order parameter’. Secondly,
there exists a one-dimensional periodicity giving it its layered
structure. In liquid crystals, this positional ordering varies
between two extremes. The most disordered case is a loose
association that gives layers where the periodicity is best
described by a sinusoidal distribution function. This is the case
for many thermotropic liquid crystals, especially immediately
below a nematic phase. At the other end, the lamellar structure
can be very well defined, where the layers are much more
had in one case performed X-ray scattering experiments which
clearly demonstrated that the S phase in question was inter-
A
digitated. X-Ray scattering studies were therefore carried out
in the present case to obtain similar information. The data are
collected in Table 4.
What the data clearly show is that the layer spacing in the
S
phase and the apparent molecular length in the nematic
A
phase (3-N10 only) are significantly greater than the calculated
molecular length. In the case of the S phase, this points to
A
‘
rigid’ and the periodicity approaches a series of Gaussian
functions. This is typically the case found in lyotropic phases.
The similar structures of these complexes mean that compari-
son of their entropies is valid. So, first we make the assumption
that the diÂerence between entropies of the isotropic states,
either within or between 2,4-Nn, 4-Nn and 3-Nn complexes, is
insignificant. This done, the changes in entropy of transition
can be attributed to changes in the nature of the smectic A
phases. Given that the entropies of the nematic-to-isotropic
transition are more or less independent of alkyl chain length,
no eÂect on the order parameter of the smectic A phase is
expected for an increase in chain length. Thus, lower entropies
correspond to a more sinusoidal distribution function, while
Fig. 5 The thermal stability of the smectic A phase (T
for complexes of (1) 4-Nn, (#) 3-Nn and (%) 2,4-Nn
and T
)
S I
S N
A
A
8
86
J. Mater. Chem., 1997, 7(6), 883–891

Fig. 6 The antiparallel structure adopted by the hydrogen bonded complexes
or 3-N12 derivatives. It can also be said that smectic A-to-
isotropic transitions with similar entropies will also have
similar distribution functions. Thus, since 2,4-N7 and 4-N11
are almost isoentropic, they probably have similar phase
structures. These conclusions are further supported by the X-
ray data presented in Table 4 in which it is clearly seen that it
is the two dinitrophenol complexes which show second and
third order reflections for the lamellar spacings, again suggest-
ing a more Gaussian distribution.
Binary Mixture Studies of 3-N8
Binary mixtures of various compositions were formed from
3
-nitrophenol and 4-octyloxy-4∞-stilbazole by a melt synthesis
procedure. The phase behaviour of these mixtures was studied
as a function of temperature using optical microscopy and
DSC experiments. The results are shown in Fig. 8 as a binary
phase diagram. The complexity of the phase behaviour is so
great that even with the 15 compositions studied it is not
possible to determine unambiguously the detailed structure of
the phase diagram. The lines delimiting phase boundaries,
eutectic and ‘peritectic’ like behaviour serve to highlight the
gross behaviour, and are most likely the correct interpretation
of the data.
Fig. 7 The variation in entropy of the smectic A–isotropic transition
for various homologues of the (%) 4-Nn, (1) 3-Nn and (#) 2,4-
Nn series
higher values represent a greater deviation toward a Gaussian
distribution.
Changes in the distribution function which describes the
periodicity of the layers are solely a consequence of changes
in the intermolecular interactions. By dividing molecules into
two parts (the aromatic, polarisable core and the aliphatic
chain), this system can be considered as containing only three
types of pairwise interaction. These are core–core, chain–chain
or core–chain interactions. In a rigidly layered system where
cores are separated from chains, we have only core–core and
chain–chain interactions. In a non-layered system, core–chain
interactions would also be present.
The congruent melting behaviour observed at 50% 3-nitro-
When the energy of the core–chain interaction is too
‘
costly’—that is when the diÂerence between interaction ener-
gies exceeds a certain threshold—the amount of core–chain
interactions is reduced by forming a layered structure. Further
increase in the ‘cost’ of this interaction causes a deviation from
a purely sinusoidal distribution function towards a Gaussian
one. These changes in distribution can be related to the entropy
of the phase. Thus, through the nature of the distribution
function, a larger value of DS can be related to a higher ‘cost’
of core–chain interactions. In this way we can see that changing
the alkyl chain length, in a homologous series, can only eÂect
chain–chain and core–chain interactions, and that the increase
in DS with increasing homologue, n, reflects the stabilisation
of chain–chain (or core–core) interactions above core–chain
interactions. For the 2,4-Nn series we see that each additional
carbon atom to the alkyl chain adds between 1 and
2
J K−1 mol−1 of order to the phase.
Comparison of the 2,4-Nn, 4-Nn and 3-Nn series shows the
structurally isomeric 4-Nn and 3-Nn complexes to have very
similar behaviour, whereas the addition of an extra nitro group
in the 2,4-Nn complex massively changes the order of the
phase. The smectic A phase of the 2,4-N12 derivative has an
additional 5 J K−1 mol−1 (ca. 60%) of order than the 4-N12
Fig. 8 The binary phase behaviour of mixtures of 3-nitrophenol and
4-octyloxy-4∞-stilbazole: (%) K–, (1) K+N–, (#) K+I–, (') E–, (Z)
E+N–, (2) E+I–, (C) S –, (() N–, (D) N+I–
B
J. Mater. Chem., 1997, 7(6), 883–891
887

phenol (percentages are given as mole fractions of the two
components) is conclusive evidence for the formation of a one-
to-one adduct. This, and the increased thermal stability in the
crystal and nematic phases at the 50% composition, implies
complex formation is strongly favoured over the reverse reac-
tion. It is consequently legitimate to treat the diagram as two
separate halves: from 0 to 50% 3-nitrophenol, which in reality
corresponds to a binary phase diagram of 4-octyloxy-4∞-stilba-
zole with the complex 3-N8; and from 50 to 100% 3-nitrophe-
nol, which is the phase diagram of 3-N8 with 3-nitrophenol.
For the right hand side of the diagram (50–100% 3-nitrophe-
nol) the behaviour is that of two components whose isotropic
liquid phases are completely miscible but do not form solid
solutions. Here the eutectic point is formed at about 67%
complexes in decreasing wavenumber such that 3-C8>4-
C8>3-N8>4-N8. This suggests that the hydrogen bonding is
strongest in the nitro derivatives, and that the 4-substitution
gives a stronger hydrogen bond than 3-substitution. This is
anticipated given the eÂect of these substituents upon the
proton donating ability (which can be related to pK ) of these
a
phenols. The order is primarily attributed to the stronger
electron-withdrawing properties of the nitro group over that
of the cyano group, and secondarily the fact that any 3-
substituted derivatives, unlike the 4-substituted, are not conju-
gated to the phenolic oxygen. The broad n
band at ca.
OH
2500 cm−1 suggests a ‘weak’ single minimum potential. This is
anticipated for the monosubstituted adducts on the basis of
the DpK values. However, with the increased acidity of the
a
3
-nitrophenol with a melting point at ca. 65 °C. The nematic
2,4-dinitrophenol, an unsymmetric double minimum potential
phase, is quickly destabilised becoming monotropic just above
may be expected in the 2,4-Nn complexes.
6
0% 3-nitrophenol, and biphasic behaviour is seen at this
When compared to 3-Cn, 4-Cn, the nitro-substituted adducts
show additional bands. In the 3-N8 and 4-N8 complexes a
broad band with a vibrational fine structure, centred at ca.
1800 cm−1, is observed. This is more intense in the 4-N8
derivative than the 3-N8, and is not seen at all in the analogous
cyano complexes. In 2,4-N8 two additional bands at 2116 and
2022 cm−1 are seen. Although the hydrogen bond strength in
2,4-N8 is expected to be comparable with carboxylic acid–
pyridyl adducts, no similarity in the IR spectra is observed.
Complexes of carboxylic acids and pyridyl moieties are
described as forming intermediate to strong hydrogen bonds,
and give rise to ‘type i’ spectra as defined by Hadzi,19 showing
a trio of bands, A–C at typically 2800, 2500 and 1800 cm−1.
The additional bands observed in 2,4-N8 are probably over-
tones and combinations of other stretches.
clearing transition.
On the left hand side (0–50% 3-nitrophenol) the diagram
shows the behaviour of mixtures of 4-octyloxy-4∞-stilbazole
and 3-N8. This is much more complicated than the right hand
side. It arises not only because now both of the components
are mesomorphic, but also because diÂerent phases show
diÂering degrees of miscibility. In particular while both crystal
phases are immiscible, the smectic E phase shows a limited
miscibility for a range of compositions. This is seen at the
E
+II transition, where a near flat phase boundary
stilbazole
occurs at about 77 °C for compositions 85 to 80% stilbazole.
This implies that the composition of the smectic E phase here
is between 90 and 85% stilbazole, while the isotropic phase is
relatively enriched in the complex 3-N8.
It is not surprising that the crystal smectic E phase of the
stilbazole is rather more tolerant of 3-N8 impurity than
the crystal phase. Given the more disordered nature of the
smectic E phase, such behaviour is understandable. For these
compositions the first transition on heating occurs at about
Other noticeable changes are more subtle. Generally we see
the pyridyl n stretch increase from 1590 cm−1 by ca. 10 cm−1
CN
upon complexation. Similarly, the predominantly phenolic n
CO
increases by ca. 20 cm−1 to 1250 cm−1. The magnitude and
direction of each of these changes is as expected for the change
6
6 °C, and it is likely that the corresponding eutectic composi-
in the environment for these groups.20 The c and d , ‘out-
OH
OH
tion is at about 70% stilbazole. A final point worth noting is
the overall stability of the nematic phase. The delimiting line,
representing its uppermost thermal stability is unsymmetric
across the whole phase diagram. The phase is tremendously
destabilised by the non-mesogenic 3-nitrophenol, on the other
hand, as one might expect, the mesogenic stilbazole impurity
is quite well tolerated with relatively little depreciation in the
phase stability.
of’ and ‘in-plane’ deformations of the pure phenols are not
seen in the spectra of the adducts. These vibrations are believed
to be responsible for the increase in background absorption21
below 1300 cm−1.
Variable temperature IR studies
Studies were performed on a sample of 3-N5. The sample was
heated under an argon atmosphere and spectra were recorded
every 5 °C from room temperature (23 °C), to the isotropic
phase at 140°C. Other than a general loss of spectral resolution,
two major changes were observed. The first was an unsym-
Spectroscopic Investigations
Infrared studies
metric broadening of the n band at ca. 2500 cm−1, and a
OH
shift in position of the band maximum to higher frequencies.
IR spectroscopy was used to examine the hydrogen bonding
in these complexes. Samples were run as KBr discs and were
examined between 4000 and 450 cm−1. We acknowledge the
suggestion18 that KBr may not be an inert matrix for these
hydrogen bonded adducts, as the possibility of a hydrogen
bonded interaction with the bromide ion exists. However,
identical spectra were obtained for a sample of solid 3-N5 and
this adduct in a KBr disc. We also note that the hydrogen
bonded complexes of Yu and co-workers8 prepared as Nujol
mulls gave identical spectra to those in KBr discs. We believe
that the changes in homologue and core structure of our
complexes are unlikely to cause a change in the interaction
with KBr.
This shift and broadening of the n
is consistent with a
OH
lengthening of the hydrogen bond, and an increased polaris-
ability, reflecting the increased number of possible environ-
ments of this bond. Part of this region of the spectra is
obscured by various n stretches, which are unaÂected by the
CH
changes in temperature.
The second, and most impressive change was the complete
and sudden loss of the band at ca. 1800 cm−1 upon melting.
This broad, medium-intensity band, is present in the crystal at
6
0°C, but disappears completely at 65 °C when the adduct has
melted into the nematic phase. From this observation we can
say that the band is in some way related to a lattice mode.
The most obvious changes upon complexation are observed
Solid state NMR studies
in the n
region above 1700 cm−1. The n
band in the
OH
OH
phenol at ca. 3200 cm−1 is replaced by a broad band in the
complex centred at ca. 2500 cm−1 showing some substructure.
This is indicative of the formation of a stronger hydrogen
bond in the complex than in the pure phenol. Comparison of
The timescale and sensitivity of the NMR experiment should
reveal information about the average state of the hydrogen
bonding, by changes in the chemical shifts of the atoms in the
proximity of the interaction. Preliminary solid state MAS 13C
spectra were obtained for the complex 3-N8 at ambient tem-
the spectral positions of the n stretching bands ranks the
OH
8
88
J. Mater. Chem., 1997, 7(6), 883–891

perature (23 °C) in the crystal phase, at 94 °C in the nematic
state, and at 99 °C in the isotropic liquid. No discernible
change was seen in these spectra, suggesting that in this
complex the isotropisation is not driven by a dissociation of
the complex into its constituent parts. Further, and more
detailed studies are already underway.
Electronic spectroscopy of 2,4-N10
Unlike all of the other hydrogen bonded systems we have
examined, the 2,4-Nn complexes are golden yellow. This is
also in stark contrast to the components, the stilbazoles being
cream coloured; l (THF) 326 nm, and 2,4-dinitrophenol is
max
a pale yellow; l (THF) 295, sh 340 nm. The nature of the
Scheme 1 Proton transfer between (a) the normal state of hydrogen
bonding in the adduct and (b) the ionic hydrogen bonded form
max
hydrogen bond greatly aÂects the electronic states in each
conjugated chromophore. The large shift in the position of the
absorption maxima experienced by both parts makes these
complexes ideal for investigations by UV spectroscopy. Indeed
the results of our preliminary studies have already been
reported.17
that in the absence of more appropriate data such comparisons
are acceptable.
The fact that isosbestic behaviour is observed up to 121 °C
implies only one pair of species, and that no significant
dissociation is seen up to this point. That is, isotropisation is
driven by the structural anisotropy and properties of the
complex as a whole, and not by the dissociation into constitu-
ent parts. It is informative to plot the intensities of the key
transitions as a function of temperature and state; this is done
in Fig. 10. Thus, bands B and D are seen initially to increase
at the expense of band A. The transition C is apparently seen
to increase, as it is not deconvoluted from bands B and D.
Above 121 °C, we see the beginning of a general broadening
of the spectra, a deviation from isosbestic behaviour, and a
decrease in the intensities of all transitions. This is attributed
to the dissociation of the complex from both ionic and non-
ionic hydrogen bonded states. Dissociation from both of these
states would lead to the observed hypsochromic and batho-
chromic shifts in the spectra. This would support the proposal
from Kato22 that dissociation does not occur until the isotropic
state is reached and that, indeed, the fact that the hydrogen
bonded complex is in a mesophase (particularly in the case of
smectic phases) may serve to stabilise the hydrogen bond.
A final point to note concerns the relative populations of
the two states. As both stilbazole and stilbazolium have similar
extinction coeÃcients (30 600 and 33 000 dm3 cm mol−1,
respectively), we can say that a near 50% population is
obtained above ca. 114 °C. However, a further increase in
temperature above the clearing point sees a relative decrease
in the population of the ionic hydrogen bonded state. To
The electronic spectra of 2,4-N10—whose mesomorphism is
Crys·97·S ·117·I, was recorded every 0.33 °C at 90–126 °C.
A
Reproducible spectra were obtained once the material had
been allowed to crystallise. Fig. 9 shows a heating run from
9
0–118 °C; glass was used as a substrate giving a ‘cut-o ’
point of ca. 310 nm. Fortunately, all major transitions are well
above this value. In order to assign the spectra, it is useful to
make comparison with certain fragments. In addition to the
components (vide supra), we find that, decyloxystilbazolium
chloride is bright yellow having a l (THF) at 368 nm, and
max
potassium 2,4-dinitrophenoxide is bright orange, l (THF)
max
3
63 and 424 nm.
There are four bands in the spectra, A–D, and we have
assigned each of these to the lowest energy transitions in 4-
decyloxy-4∞-stilbazole, 4-decyloxy-4∞-stilbazolium, 2,4-dinitro-
phenol and 2,4-dinitrophenate, respectively. The observed
isosbestic behaviour at 90–121 °C shows that only two species
are present, which we identify as the non-ionic and ionic
hydrogen bonded states (Scheme 1) and which shows that an
increase in temperature populates the (excited) ionic state.
Thus, the hydrogen bond clearly has an unsymmetric double
minimum potential. That the non-ionic state is the ground
state is also supported by the low value of the DpK between
a
stilbazolium and 2,4-dinitrophenol, which is between 1 and 2.
Lindemann and Zundel16 suggest that usually a value of 4 is
required to eÂect 50% proton transfer. Although comparisons
of solution with solid state spectra, and predictions based on
aqueous acidities, must be viewed with some caution, we feel
Fig. 9 Electronic spectra of the adduct 2,4-N10 at 90–118 °C, as it
passes from the crystal through the smectic A phase and into
isotropic state
Fig. 10 The variation of the key transitions as
a
function of
temperature: (%) 348 (A), (1) 368 (B), (#) 400 (C) and (') 426 nm (D)
J. Mater. Chem., 1997, 7(6), 883–891
889

account for this deviation from a normal thermal population
of the ionic excited state, we propose that the environment of
the smectic A phase, where there is a lamellar microphase
separation of alkyl chains from the aryl cores, provides an
additional stabilisation of the ionic hydrogen bonded state, as
here the hydrogen bond is predominantly surrounded by
polarisable (aryl groups) and polarised (nitro groups) func-
tionalities. Thus, we re-emphasise the important conclusion
that significant dissociation from the hydrogen bonded com-
plexes does not occur until the isotropic state is reached and
that isotropisation is not driven by a breakdown of the
hydrogen bonded complex, indeed the mesophase may act to
stabilise the hydrogen bond.
in good yield by coupling 4-methylpyridine and the corre-
sponding 4-alkoxybenzaldehyde, followed by an acid-catalysed
elimination to convert the intermediate secondary alcohol into
the product. The procedure is described for the butoxy deriva-
tive. Analogous procedures were employed for all other homol-
ogues. All components gave satisfactory characterisation. The
behaviour of the stilbazoles was identical to that of authentic
samples.14
General sample preparation
Complexes were formed by dissolving exactly equimolar
amounts of the components in tetrahydrofuran and removing
the solvent on a rotary evaporator. The compromise to purity
that mixing brings is estimated to be not more than 1%.
Experimental
Instrumentation
Mixture preparation
The components were characterised by microanalysis, per-
formed at SheÃeld University, and by 1H and 13C NMR on a
Bruker WM250 spectrometer. Chemical shifts are quoted
relative to an internal standard, and all coupling constants are
given in Hz. 13C NMR data were obtained from J-modulated
spectra. All components gave satisfactory characterisation.
Routine IR spectroscopy was performed, unless otherwise
specified, as KBr discs on a Perkin Elmer 1600 series FTIR.
Variable temperature IR studies were performed using a
Perkin Elmer 1710 FTIR equipped with an Infrared Associates
Ltd NO 161373 liquid nitrogen cooled detector, working to a
resolution of 4 cm−1. The sample was deposited by evaporation
from a diethyl ether solution onto a silicon rod, within a
pressure cell. The cell was filled with argon to a pressure of 1
atmosphere at room temperature (23 °C) and sealed. The cell
was then equipped with a heater and thermocouple and a
temperature controlled to an accuracy of ±5°C. Attenuated
total reflectance FTIR spectra were obtained every 5 K,
through crystal, nematic and isotropic phases at
Homogeneous mixtures were formed from solutions of the
components, by removing the solvent (tetrahydrofuran) at ca.
60°C, then heating the mixture while stirring with a magnetic
follower, at up to ca. 120 °C, where only a single isotropic
phase is present, and then cooling quickly.
1-(4-Butoxyphenyl)-2-(4-pyridyl)ethanol. A solution of diiso-
propylamine (3.757 g, 37.2 mmol), in dry tetrahydrofuran
(200 cm3; freshly distilled from sodium benzophenone ketyl),
was cooled under a nitrogen atmosphere to −78°C.
Butyllithium (23.5 cm3, 1.6 mol dm−3 solution in hexanes,
37.6 mmol) was added dropwise over a period of 40 min. The
reaction was allowed to stir for a further 30 min while a
temperature of −78 °C was maintained. After this time a
solution of 4-methylpyridine (3.459 g, 37.2 mmol) in tetra-
hydrofuran (30 cm3) was added dropwise over 1 h. The reaction
mixture, which became bright orange, indicating the formation
of the carbanion, was stirred for a further 40 min. Subsequent
dropwise addition of 4-butoxybenzaldehyde (6.0 g, 33.7 mmol)
in tetrahydrofuran (30 cm3) over 30 min was accompanied by
a colour change of the reaction mixture from orange to pale
yellow. The mixture was stirred for 30 min and then allowed
to warm to room temp. over 3 h. The excess pyridyl carbanion
was quenched with water (20 cm3), upon which the reaction
became instantly colourless, the mixture was then neutralised
with dilute aqueous hydrochloric acid. All of the solvents were
removed under reduced pressure on a rotary evaporator. The
solid was partitioned between dichloromethane (200 cm3) and
water (3×200 cm3) and washed once with saturated brine
solution (1×200 cm3). The organic phase was dried over
anhydrous magnesium sulfate, filtered and the solvent removed
on a rotary evaporator, giving the crude product as an oil
which crystallised upon standing. Recrystallisation from hep-
tane gave the product (yield 4.95 g, 18.3 mmol, 54%). mp 87 °C
(found: C, 75.1; H, 7.8; N, 5.2. C H NO requires: C, 75.3;
4
000–200 cm−1.
Tetrahydrofuran (THF) was refluxed under nitrogen over
lumps of sodium and benzophenone until the solution became
purple and was freshly distilled immediately prior to use.
Solution state electronic spectra were recorded on a Phillips
PU8720 UV–VIS scanning spectrophotometer. The variable
temperature ‘solid’ state electronic spectroscopy was performed
in transmission, with a Photal Otsuka Electronics Deuterium
Lamp, MC-962A and Spectromultichannel Photo Detector,
MCPD-100. The sample was heated on a Mettler hot stage
with an FP90 control processor. Heating rates of 1 and
4
K min−1 were employed and spectra were taken every 0.33
or 1 K. Samples were prepared by placing a saturated solution
of the material in tetrahydrofuran onto a glass microscope
slide, covering with a coverslip and allowing the solvent to
evaporate at ca. 60 °C.
1
7 21
2
The mesomorphism was characterised by heated stage polar-
ising optical microscopy, using a Zeiss Labpol microscope
fitted with a Linkam TH600 hot stage and a PR600 tempera-
ture controller. Enthalpies of transitions were recorded by a
Perkin Elmer DSC 7 instrument, at heating and cooling rates
of between 5 and 10 K min−1 were employed.
H, 7.8; N, 5.2%); n
cm−1 (KBr) n
3202 br vs; d
H
max
O–H
(250 MHz; CDCl ) 0.97 (3H, H , t, 3J 9.4), 1.40–1.58 (2H,
3
a
ac
H , m), 1.68–1.82 (2H, H , m), 2.93 (1H, H , dd, J 16.9, 6.9),
b
c
j
3.03 (1H, H , dd, J 16.9, 10.0), ca. 3.0 (1H, H , br s), 3.94 (2H,
j
n
H , t, 3J 8.1), 4.85 (1H, H , dd, J 10.0, 6.9), 6.84 (2H, H ,
d
cd
i
f
AA∞XX∞, J 10.6), 7.05 (2H, H , AA∞XX∞, J 7.5), 7.20 (2H, H ,
l
g
X-Ray diÂraction patterns were recorded with an Image
Plate area detector (MAR Research) using graphite-monochro-
matised Cu-Ka pinhole collimated radiation. The samples, in
glass capillaries, were held in a temperature-controlled cell.
The beam path was flushed with helium.
AA∞XX∞, J 10.6) and 8.35 (2H, H , AA∞XX∞, J 7.5); d (63 MHz;
m
C
d
CDCl ) 13.9 C , 19.6 C , 31.3 C , 45.1 C , 67.7 C , 74.0 C ,
3
a
b
c
j
i
114.4 C , 125.0 C , 127.1 C , 135.6 C , 147.8 C , 149.3 C and
f
158.8 C .
e
l
g
h
k
m
Components
3
-Nitrophenol (BDH) was crystallised from water. 2,4-
Dinitrophenol (BDH) was dried under high vacuum and 4-
nitrophenol (Hopkin and Williams Ltd) was used without
further purification. 4-Alkoxy-4∞-stilbazoles were synthesised
8
90
J. Mater. Chem., 1997, 7(6), 883–891

trans-4-Butoxy-4∞-stilbazole
{trans-4-[2-(4-butoxyphenyl)
J. W. Goodby and G. W. Gray, J. Chem. Soc., Faraday Trans. 1,
1
982, 713.
ethenyl]pyridine}. To a solution of 1-(4-butoxyphenyl)-2-(4-
pyridyl)ethanol (4.921 g, 18.15 mmol) in toluene (300 cm3) was
added toluene-p-sulfonic acid (10.021 g, 56.26 mmol) and pyri-
dine (4.430 g, 56.08 mmol). The reaction mixture was heated
under reflux overnight and a Dean–Stark apparatus was used
to remove the water produced by azeotropic distillation.
After cooling, a solution of potassium hydroxide (3.614 g,
4
5
F. Henrich, S. Diele and C. Tschierske, L iq. Cryst., 1994, 17, 827.
For reviews of this subject see: G. A. JeÂery and L. M. Wingert,
L iq. Cryst., 1992, 12, 179; G. A. JeÂery, Acc. Chem. Res., 1986,
19, 168.
6
7
C. Paleos and D. Tsiourvas, Angew. Chem., Int. Ed. Engl., 1995,
34, 1696.
M. Kotera, J.-M. Lehn and J.-P. Vigneron, J. Chem. Soc., Chem.
Commun., 1994, 197; M.-J. Brienne, J. Gabard, J.-M. Lehn and
I. Stibor, J. Chem. Soc., Chem. Commun., 1993, 420.
6
4.42 mmol) in a mixture of water (200 cm3) and ethanol
(
5 cm3) was added, and the reaction stirred for a further 12 h.
8
9
L. J. Yu, J. M. Wu and S. L. Wu, Mol. Cryst. L iq. Cryst., 1991, 198,
407; L. J. Yu and J. L. Pan, L iq. Cryst., 1993, 14, 829; L. J. Yu, L iq.
Cryst., 1993, 14, 1303.
The phases were separated and the organic layer washed with
water (5×200 cm3) and saturated brine (1×200 cm3). The
toluene solution was then dried over anhydrous magnesium
sulfate, filtered and the solvent removed under reduced press-
ure. Crystallisation from hot acetone gave the product, trans-
C. M. Lee, C. P. Jarwala and A. C. GriÃn, Polymer, 1994, 35, 4550.
D. J. Price, PhD thesis, SheÃeld University, 1995; H. Kihara,
T. Kato, T. Uryu, S. Ujiie, K. Iimura, U. Kumar, J. M. J. Fr e´ chet,
D. W. Bruce and D. J. Price, L iq. Cryst., 1996, 21, 25; D. J. Price,
H. Adams and D. W. Bruce, Mol. Cryst. L iq. Cryst., 1996, 289, 127;
1
0
4
9
-butoxy-4∞-stilbazole (yield 3.371 g, 11.21 mmol, 62%). mp
5 °C (lit.,14 95 °C) (found: C, 80.34; H, 7.56; N, 5.57. C H ON
K. Willis, J. E. Luckhurst, D. J. Price, J. M. J. Frechet, H. Kihara,
´
2
1 19
requires: C, 80.60; H, 7.56; N, 5.53%); d (250 MHz; CDCl )
T. Kato, G. Ungar and D. W. Bruce, L iq. Cryst., 1996, 21, 585.
H
3
11 T. Kato and J. M. J. Fr e´ chet, J. Am. Chem. Soc., 1989, 111, 8533;
T. Kato and J. M. J. Fr e´ chet, Macromolecules, 1989, 22, 3818;
U. Kumar, T. Kato and J. M. J. Fr e´ chet, J. Am. Chem. Soc., 1992,
114, 6630; T. Kato, H. Kihara, T. Uryu, A. Fuijishima and
J. M. J. Fr e´ chet, Macromolecules, 1992, 25, 6836; T. Kato,
0
3
.97 (3H, H , t, 3J 7.5), 1.49 (2H, H , m), 1.77 (2H, H , m),
a
ab
b
c
.98 (2H, H , t, 3J 6.4), 6.77 (1H, H , d, 3J 16.2), 6.83 (2H,
d
dc
j
ij
H , AA∞XX∞, J 8.4), 7.15 (1H, H , d, 3J 16.2), 7.22 (2H, H ,
g
i
ij
l
AA∞XX∞, J 6.1), 7.38 (2H, H , AA∞XX∞, J 8.4) and 8.46 (2H,
f
H , AA∞XX∞, J 6.1); d (63 MHz; CDCl ) 13.9 C , 19.2 C ,
H. Kihara, U. Kumar, T. Uryu and J. M. J. Frechet, Angew. Chem.,
´
m
C
3
a
b
3
1.3 C , 67.8 C , 114.8 C , 120.6 C , 123.6 C , 128.4 C , 128.7
Int. Ed. Engl., 1994, 33, 1644; T. Kato, P. G. Wilson, A. Fuijishima
and J. M. J. Fr e´ chet, Chem. L ett., 1990, 2003; T. Kato,
J. M. J. Fr e´ chet, P. G. Wilson, T. Saito, T. Uryu, A. Fuijishima,
C. Sin and F. Kaneuch, Chem. Mater., 1993, 5, 1094; T. Kato,
A. Fuijishima and J. M. J. Fr e´ chet, Chem. L ett., 1990, 912;
M. Fukumassa, T. Kato, T. Uryu and J. M. J. Fr e´ chet, Chem. L ett.,
c
d
f
l
j
g
C , 132.8 C , 145.0 C , 150.1 C and 159.8 C .
h
i
k
m
e
1
993, 65; T. Kato, H. Kihara, T. Uryu, S. Ujiie, K. Iimura,
J. M. J. Fr e´ chet and U. Kumar, Ferroelectrics, 1994, 148, 1303.
12
13
14
D. W. Bruce and D. J. Price, Adv. Mater. Opt. Electron., 1994,
4
, 273.
H. Adams, D. W. Bruce, D. J. Price, G. Ungar and K. Willis,
J. Mater. Chem., 1995, 5, 2195.
We would like to thank the EPSRC for a studentship to
D. J. P., Dr David Apperly (EPSRC Solid State NMR service,
Durham) for his invaluable assistance with solid state MAS
NMR and Dr Tony Haynes (SheÃeld) for his assistance with
variable-temperature infrared studies. We wish to thank
Professor G u¨ nter Lattermann (Bayreuth) for suggesting the
potential benefits of a nitro group in these systems.
D. W. Bruce, D. A. Dunmur, E. Lalinde, P. M. Maitlis and
P. Styring, L iq. Cryst., 1988, 3, 385.
15 CRC Handbook of Chemistry and Physics, ed. D. R. Lide, 75th edn.,
CRC Press Inc., Boca Raton, Florida.
1
6
R. Lindemann and G. Zundel, J. Chem. Soc., Faraday Trans. 2,
1
972, 68, 979; R. Lindemann and G. Zundel, J. Chem. Soc., Faraday
T rans. 2, 1977, 73, 788.
1
7
D. J. Price, T. Richardson and D. W. Bruce, J. Chem. Soc., Chem.
Commun., 1995, 1911.
18
19
20
K. R. Seddon, personal communication, 1995.
D. Hadzi, Pure Appl. Chem., 1965, 11, 435.
D. W. Brown, A. J. Floyd and M. Sainsbury, Organic Spectroscopy,
Wiley, Bath, 1988.
References
1
2
D. Vorl a¨ nder, Z. Phys. Chem., 1923, 105, 211; G. M. Bennett and
B. Jones, J. Chem. Soc., 1939, 420.
2
1
S. L. Johnson and K. A. Rumon, J. Phys. Chem., 1965, 69, 74.
T. Kato, personal communication.
C. Eaborn, J. Chem. Soc., 1952, 2840; A. Polishchuk,
T. V. Timofeeva, N. N. Makarova, M. Yu. Antipin and Yu. T.
Struchov, L iq. Cryst., 1991, 9, 433.
2
2
3
J. D. Bunning, J. E. Lydon, C. Eaborn, P. M. Jackson,
Paper 7/00575J; Received 24th January, 1997
J. Mater. Chem., 1997, 7(6), 883–891
891