"Tuning fork"-shaped mesogens: Large hysteresis in the interdigitated layer structure in the liquid crystal phases

Article abstract of DOI:10.1039/b406177b

Compounds 1, N,N-bis{4-[4-(4-alkoxybenzoyloxy)benzyloxy] benzoyl}alkylamines, were designed and synthesized as "tuning fork"-shaped molecules, and their properties were investigated by polarized light microscopy and differential scanning calorimetry, and

Full text of DOI:10.1039/b406177b

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
‘‘Tuning fork’’-shaped mesogens: large hysteresis in the
interdigitated layer structure in the liquid crystal phases{
Takashi Kajitani,a Yuichi Miwa,b Naru Igawa,b Masaki Katoh,b Shigeo Kohmoto,b
Makoto Yamamoto,b Kentaro Yamaguchic and Keiki Kishikawa*b
aQuality Materials Science, Graduate School of Science and Technology,
Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba 263-8522, Japan
bDepartment of Materials Technology, Faculty of Engineering, Chiba University,
1-33 Yayoi-cho, Inage-ku, Chiba 263-8522, Japan. E-mail: kishikawa@faculty.chiba-u.jp
cChemical Elemental Analysis Center, Chiba University, 1-33 Yayoi-cho, Inage-ku,
Chiba 263-8522, Japan
Received 23rd April 2004, Accepted 14th June 2004
First published as an Advance Article on the web 3rd August 2004
Compounds 1, N,N-bis{4-[4-(4-alkoxybenzoyloxy)benzyloxy]benzoyl}alkylamines, were designed and synthesized
as ‘‘tuning fork’’-shaped molecules, and their properties were investigated by polarized light microscopy and
differential scanning calorimetry, and the layer distances were measured by X-ray diffraction. The observed
smectic A liquid crystal phases indicated the existence of large hysteresis in XRD. The multiple broad peaks of
the layer distances originated in the plural extent of interdigitation of the alkyl chains between the layers, and the
layer distances strongly depended on the hysteresis. Further, the chiral and racemic liquid crystalline compounds,
(R)-1k and rac-1k, showed a clear difference in the interlayer interdigitations in their mesophases.
with a covalent bond (Scheme 1 right) prevents the cancella-
1. Introduction
tion, and generates a strong molecular dipole as a ‘‘tuning
fork’’-shaped mesogen. Though the ‘‘tuning fork’’-shaped
molecules also dimerize to cancel out their polarities in the
mesophase, a large intermolecular dipole–dipole interaction is
expected between the two molecules.
In this study compounds 1 were designed and synthesized as
a ‘‘tuning fork’’-shaped molecule (Scheme 2), and the pro-
perties were investigated by polarized light microscopy and
differential scanning calorimetry (DSC), and the layer dis-
tances were measured by X-ray diffraction (XRD). As a result,
we found that the smectic A liquid crystal phases indicated
large hysteresis in XRD. The multiple broad peaks of the layer
distances originated in the plural extent of interdigitation of the
alkyl chains between the layers, and the layer distances strongly
depended on the hysteresis.
In general, smectic phases have a periodicity in one direction
to show its peculiar layer distance(s) in XRD, which is charac-
teristic of all smectic phases and in most cases does not depend
on their hysteresis because of the fluidity of liquid crystals.
However, these liquid crystal phases are highly viscous and
showed large hysteresis, even though these compounds are low
molecular weight liquid crystalline compounds.
The dipole–dipole interaction is one of the most important
interactions to generate superstructures in mesophases. Inter-
molecular dipole–dipole interactions have produced a lot of
highly ordered structures in liquid crystal phases: 1) bilayer
smectic phases,1 2) re-entrant phases,2 3) chiral phases
generated by achiral banana-shaped molecules,3 4) dimeriza-
tion of apolar molecules (anthraquinone derivatives) by their
molecular distortion,4 5) organization of bent-rod molecules
possessing a central large dipole,5 and 6) un-even parallel
association of half-disk molecules.6 Further, molecular dipoles
greatly influence the physical properties (charge mobility,7
ionic conductivity,8 fluorescence,9 electro-optic switching,10
and dipole modulation by trans–cis photoisomerization11) of
the liquid crystals. Thus, introduction of large polarity into a
molecule is one of the most important approaches for their
rapid response. In addition, the generation of novel super-
structures (ferroelectric alignment12 or helical formation13) by
controlling intermolecular dipole–dipole interactions has been
predicted theoretically in mesophases. Usually, in order to
generate a large molecular dipole, polar substituents (–CN,
–NO2, or –F) or linkage groups (–CLN–, –NLNO–, or –COO–)
are introduced in the liquid crystalline molecules.14 However,
these molecular polarities are cancelled out by having an anti-
parallel arrangement (Scheme 1 left),15 and the time-averaged
dimer behaves as a non-polar unit. In contrast, arrangement of
two polar mesogenic cores in a syn-parallel manner by linkage
2. Result and discussion
The behavior of the compounds and identification of the phases
The compounds 1a–i were synthesized by the method in our
previous paper16 (Scheme 2). The behavior of liquid crystalline
compounds 1a–i is shown in Table 1. Length of the alkyl chains
seems to be a main factor to determine the behavior of these
liquid crystalline compounds. Compounds 1a, 1c, 1d, and 1g
possessing butyl groups as the terminal chains (R1 or/and R2)
exhibited nematic phases, and very small transition enthalpies
(0.030–0.071 kcal/mol) were observed for the transitions
between the nematic and the isotropic liquid phases. At the
transition from the isotropic liquid to the nematic phase,
homeotropic textures of the nematic phases appeared simulta-
neously from all part of the isotropic sample.
Scheme 1 Arrangement of two mesogens: (left) anti-parallel organi-
zation of two polar molecules and (right) a molecule consisting of two
polar mesogenic cores in syn-parallel arrangement.
{ Electronic supplementary information (ESI) available: colour version
of Fig. 1 and analytical data for 1c–1i. See http://www.rsc.org/
suppdata/jm/b4/b406177b/
2 6 1 2
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1
T h i s j o u r n a l i s ß T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Article Online
Table 1 Behavior of compound 1a–i
Compound
Behavioura
145 ð5:7Þ
176 ð0:045Þ
1a
CCDA
BCDC
CCDA
BCDC
Cr
N
I
129 ð{5:7Þ
176 ð{0:045Þ
99 ð2:2Þ
180 ð1:2Þ
1b
1c
1d
1e
1f
CCDA
BCDC
CCDA
BCDC
Cr
SmA
I
58 ð{2:0Þ
180 ð{1:2Þ
133 ð0:31Þ
142 ð0:071Þ
76 ð4:3Þ
CCDA
BCDC
CCDA
BCDC
Cr DCCA SmA
N
N
I
133 ð{0:29Þ
142 ð{0:060Þ
113 ð4:1Þ
147 ð0:13Þ
161 ð0:053Þ
CCDA
BCDC
CCDA
BCDC
CCDA
BCDC
Cr
SmA
I
71 ð{4:1Þ
146 ð{0:12Þ
161 ð{0:043Þ
98 ð3:1Þ
199 ð2:4Þ
CCDA
BCDC
CCDA
BCDC
Cr
SmA
I
71 ð{3:1Þ
199 ð{2:3Þ
178 ð2:7Þ
88 ð3:7Þ
CCDA
BCDC
Cr DCCA SmA
I
178 ð{2:1Þ
104 ð2:5Þ
157 ð0:19Þ
216 ð0:041Þ
1g
1h
1i
CCDA
BCDC
CCDA
BCDC
CCDA
BCDC
Cr
SmA
N
I
67 ð{2:4Þ
157 ð{0:19Þ
216 ð{0:030Þ
201 ð2:5Þ
86 ð3:8Þ
CCDA
BCDC
Cr DCCA SmA
I
201 ð{2:4Þ
210 ð2:4Þ
98 ð3:5Þ
CCDA
BCDC
Cr DCCA SmA
I
210 ð{2:3Þ
a The transition temperatures (uC) and enthalpies (in parentheses,
kcal mol21) were determined by DSC (5 uC min21) and are given
above and below the arrows. Cr, N, SmA, and I indicate crystal,
nematic, smectic A , and isotropic phases, respectively.
compound showed several broad peaks in the small angle
region (Table 2) except compounds 1f (R1 ~ n-C8H17, R2 ~
n-C12H25) and 1i (R1 ~ R2 ~ n-C12H25) which showed one
peak in each of their small angle regions corresponding to its
molecular length. Fig. 4 and Fig. 5 show the XRD patterns of
1b (R1 ~ n-C4H9, R2 ~ n-C8H17) and 1d (R1 ~ n-C8H17, R2 ~
n-C4H9), respectively.
Scheme 2 Synthesis of 1.
Compounds 1c, 1d, and 1g also exhibited smectic A phases at
the temperature range below their nematic phases. The
transition enthalpies between the nematic and the smectic A
phases were also small (0.12–0.31 kcal mol21). These small
transition enthalpies indicate a similarity between the mole-
cular aggregations before and after the transition. It is assumed
that the tuning-fork molecules exist as time-averaged dimers
with an antiparallel manner even in the isotropic and nematic
phases. In contrast, the other compounds (1b, 1e, 1f, 1h, and 1i)
only displayed smectic A phases over the wide temperature
ranges. In all these smectic A phases, typical textures of smectic
A phases, such as batonets and fan-shaped textures, appeared
from their isotropic liquid on cooling (Fig. 1). The batonets
which were parallel to either the polarizer or analyzer of the
microscope were dark, and it meant that the long axes of the
molecules were parallel with respect to the layer normal.17
From conoscopic study of the homeotropically aligned sample
in the liquid crystal phase, the uniaxiality was confirmed. These
observations in polarized light microscopy indicated that these
liquid crystal phases belonged to smectic A phases. However,
the peak shapes in DSC (Fig. 2 and Fig. 3) were different
from those of usual smectic A phases of low-weight molecules,
and a highly broad peak was observed in both of these smectic
A–isotropic phase transitions despite the high purity of the
compounds.
Estimation of the intermolecular dipole–dipole interactions by
MM2, AM1, and single crystal XRD
MM2 calculation18 of 1a indicated that the two mesogenic
cores were parallel and contacted each other in the most stable
conformation, because of the intramolecular interactions
XRD of the smectic A phases
Fig. 1 A polarized optical micrograph of 1e. Fan-shaped textures at
114 uC on cooling (6400). The directions of the crossed polarizers are
indicated by the arrows. The dark lines are parallel to the directions.
XRD of compounds 1b–1i in a glass capillary (diameter:
1.5 mm) was carried out on their smectic A phases. Every
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1
2 6 1 3

View Article Online
Fig. 2 DSC chart of 1e on cooling. Cooling rate: 5 uC min21
.
Fig. 4 XRD of 1b in a capillary method at 120 uC (liquid crystal
phase).
Fig. 5 XRD of 1d in a capillary method at 120 uC (liquid crystal
phase).
Fig. 3 DSC chart of 1f on cooling. Cooling rate: 5 uC min21
.
Table 2 The d(100) peaks in XRD data of the smectic A phasesa
b
˚
d(100)/A
˚
Molecular length/A
Entry
1
Compound
R1
R2
temp/uC
1b
n-C4H9
n-C8H17
100
120
140
160
100
110
120
130
120
130
140
100
110
90
100
150
140
150
160
140
150
160
180
190
140
150
160
170
180
34.5
34.5
35.0
35.0
48.0
46.0
45.9
44.1
35.0
34.5
32.0
38.1
37.4
49.0
48.0
46.0
41.6
40.9
42.4
46.0
45.0
45.0
45.0
41.6
50.2
50.2
48.0
46.0
49.0
30.7
32.5
32.0
31.5
42.4
40.9
40.9
40.9
27.2
27.2
27.6
33.4
33.4
29.4
27.9
27.9
28.3
38.7
37.4
36.8
39.4
35
40
35
2
3
1c
1d
n-C4H9
n-C12H25
n-C4H9
n-C8H17
4
5
1e
1f
n-C8H17
n-C8H17
n-C8H17
40
45
n-C12H25
6
7
1g
1h
n-C12H25
n-C12H25
n-C4H9
39.4
38.7
38.1
36.2
36.2
36.2
40
45
n-C8H17
37.4
38.1
41.6
37.4
39.4
38.1
8
1i
n-C12H25
n-C12H25
50
a The XRD was measured by a capillary method on the first heating. The temperature ratio was 5 uC min21 and the temperature was
maintained during each XRD measurement. b The molecular lengths were calculated by molecular modeling, based on the ‘‘tuning-fork’’
conformation.
2 6 1 4
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1

View Article Online
Scheme 3 Compound 1a’.
Fig. 6 Relative energy levels of N,N’-diformylamine in MM2 and
AM1 calculations.
Scheme 4 Dipole moments of compound 1a and 8CB.
between the two cores. In the case of compound 1a’16
(Scheme 3) which showed a nematic liquid crystalline phase,
the parallel arrangement of the two mesogenic cores was
observed in X-ray structural analysis of the single crystal. The
intramolecular CH/p interactions19 between those benzene-
20
…
˚
…
rings (CAr CAr: 3.81–3.94 A) and CLO/CLO interactions
˚
between the two esters (CLO CLO: 3.38 A) were estimated
from the interatomic distances and the spatial arrange-
ment of their moieties. Further, dimerization of the ‘‘tuning
fork’’-shaped molecules in an anti-parallel manner was also
confirmed in the crystal packing diagram. Furthermore, in
most cases (1b, 1d, 1e, 1g, 1h and 1i), XRD of the liquid crystal
phases gave the peaks of the layer distances corresponding
to their molecular lengths which were calculated based on the
‘‘tuning fork’’-shaped conformations. A large dipole moment
(12 debye) is indicated by AM1 calculation21 for the ‘‘tuning
fork’’-shaped conformation of 1a in the direction of the long
axis (Scheme 4). Time-averaged dimerization has been known
for 4’-alkyl and 4’-alkoxyl-4-cyanobiphenyls,22 and in the case
of 4-cyano-4’-octylbiphenyl (8CB in Scheme 4) the dipole
moment was only 4.8 debye in AM1 calculation. The dipole
moment of 1a (12 debye) is 2.5 times larger than that of
8CB. Accordingly, movements of the molecules in the direction
of their long axes seem to be strongly prevented by the strong
intermolecular interaction with the neighboring molecules in
the layer.
Fig. 7 Three stable conformations of 1b. Tuning-fork (D-type),
asymmetric bent-rod (E-type), and symmetric bent-rod (F-type) shapes.
compact conformation, the D-type. If the packing force is not
sufficient to overwhelm the electrostatic repulsion, the molecule
may have the bent conformations, the E- and F-types. In
conformer D, both R1 and R2 directly influence the layer
distance of the smectic A phase. On the other hand, in con-
formers E and F, R2 length directly influences the layer
distance, but R1 length does not. From Table 2, it was obvious
that both R1 and R2 influenced the observed layer distances.
For example, compounds 1b and 1d which have the same
˚
molecular lengths (35 A) showed similar layer distances (1b:
˚
34.5–35.0 and 1d: 32.0–35.0 A) despite the different lengths
of the R2 groups. In the cases of 1b, 1e, and 1h, the R2 group is
the same length (n-C8H17), however, the longest layer distance
in each compound (1b: 34.5–35.0, 1e: 37.4–38.1, and 1h:
˚
41.6–46.0 A) is almost proportional to its molecular length
calculated as the tuning-fork conformation (1b: 35, 1e: 40, and
˚
1h: 45 A). Accordingly, the tuning fork conformation (D-type)
Conformation of 1 in the liquid crystal phase
was not contradictory to the layer distances observed in the
liquid crystal phases. To investigate the correlation between the
molecular length and the layer distance, each layer distance in
all the compounds was plotted against the molecular length
based on the tuning fork (D-type) and asymmetric bent
(E-type) models (Fig. 8 and Fig. 9). The solid line shows that
the layer distance equals the molecular length. The broken and
To investigate the most stable conformation of the connecting
part CO–N–CO, the relative formation energies of three N,N’-
diformylamine conformers A, B and C were calculated by
MM2 and AM1 (Fig. 6). Conformer C was the least stable of
the three conformers because of the strong electrostatic repul-
sion between the two carbonyl oxygen atoms. Conformers A
and B indicated similar values in the formation energies. Based
on the three conformations, the molecular lengths of the
conformers of 1b (D, E and F in Fig. 7) were calculated by
˚
˚
dotted lines corresponds to layer distances 5 A longer and 5 A
shorter than the molecular length, respectively. The observed
˚
layer distances fit within the range of 5 A from the molecular
˚
length in the D-type model (Fig. 8), while large differences arise
in the E-type model (Fig. 9). These results also provide further
evidence that the molecules have the tuning-fork conformation
in the liquid crystal phase.
MM2 to give 35, 55 and 53 A, respectively. If the packing force
in the liquid crystal phase is sufficient to overwhelm the
electrostatic repulsion between the two mesogenic units in the
molecule, it can be presumed that the molecule has the most
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1
2 6 1 5

View Article Online
Fig. 10 XRD of 1b in a thin layer method at 120 uC.
Fig. 8 The layer distance and the molecular length based on the
tuning-fork model (D-type). The solid, broken, and dotted lines
correspond to the layer distances equal to the molecular length, the
˚
˚
molecular length 1 5 A, and the molecular length 2 5 A, respectively,
in the tuning-fork model (D-type).
Fig. 11 XRD of 1d in a thin layer method at 120 uC.
butyl, 1d: octyl) and R2 (1b: octyl, 1d: butyl) groups in the
tuning-fork conformation.
Similarity between swallowtail compounds and 1
The broad temperature range of the smectic A phase may also
bear out the most compact conformation (D-type). To the best
of our knowledge, such a stable smectic A phase has not been
reported for bent molecules (E-type). In addition, in the case of
swallow tail molecules in the smectic A phase (Scheme 5), the
interdigitated layer distance has been reported with the layer
distance corresponding to the molecular length.23 The rod-like
molecule has three normal alkyl groups as the terminal chains,
and has an a,b-dicarbonyl as the bi-forked junction. The a,b-
dicarbonyl moiety also has three stable conformers, but they
concluded that the aliphatic double chains must be parallel to
each other in order to explain the increase of the layer distance
with the number of carbon atoms in the normal alkyl chain.
From these similarities, it was natural to conclude that our
molecules also had the tuning-fork conformation in the liquid
crystal phase.
Fig. 9 The layer distance and the molecular length based on the
asymmetric bent model (E-type). The solid, broken, and dotted lines
correspond to the layer distances equal to the molecular length, the
˚
˚
molecular length 1 5A, and the molecular length 2 5 A, respectively, in
the asymmetric bent model (E-type).
XRD of the pressed samples
From the molecular packing diagram, the conformational
calculations, the XRD experiment of the pressed samples, the
correlation between the layer distance and molecular length,
and the similarity to the swallowtail mesogen, the tuning-fork
(D-type) was most likely as the molecular conformation in the
liquid crystal phase. This indicated that the packing forces were
sufficient to overwhelm the intramolecular electrostatic repul-
sion between its two mesogenic cores.
The liquid crystalline samples were spread on a glass plate
using another glass plate to give thin layered samples which
exhibited homeotropic textures. The samples on the plates were
measured under atmospheric pressure. Both 1b and 1d showed
˚
˚
sharp [100] peaks at 2.82 (31.3 A) and 2.98 (29.6 A),
respectively (Fig. 10 and Fig. 11). It was confirmed that the
pressure in the spreading could generate the well-aligned and
compressed layer structure. The homeotropic texture indicated
that the averaged direction of the cores was parallel to the layer
normal and the layers were parallel to the plate. The decrease of
the layer distances indicated interdigitation between the alkyl
layers in the direction of the layer normal. The decreased length
˚
(4–5 A) corresponded to the length of the butyl group (gauche-
˚
form: 4.5 and all-trans-form: 5 A). It was assumed that the
decrease originated in the interdigitation between the R1 (1b:
Scheme 5 Swallowtail molecule (ref. 23).
2 6 1 6
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1

View Article Online
Fig. 12 Growth of the liquid crystal domain of 1b. XRD was
measured at 183, 171, 160, 149, 138, 126, and 114 uC on cooling (from
the bottom to the top). The arrows indicate growth of the peaks. The
cooling rate was 5 uC min21
.
Monitoring of growth of the liquid crystal domains by XRD
Growth of the liquid crystal domains from the isotropic liquid
was monitored by XRD in the cases of 1b and 1e. On cooling,
growth of their plural peaks was observed. The small angle
region (1.50–3.00u) of XRD was measured for 1b on cooling
from 183 to 114 uC (Fig. 12). First, the two peaks at 2.25
˚
˚
(39.2 A) and 2.50 (35.3 A) grew, and then the two peaks at 2.67
˚
(33.1 A), and 2.80 (31.5 A) came out. The intensity of these
˚
peaks increased on cooling and then saturated at a certain level.
˚
In the case of 1e, the peak at 2.00 (44.1 A) appeared first at
˚
˚
183 uC, and then the peaks at 2.30 (38.4 A) and 2.42 (36.5 A)
increased on cooling from 183 to 138 uC (Fig. 13). The longest
˚
layer distances (39.2 and 44.1 A) observed in the monitoring
˚
experiments of 1b and 1e were about 4 A longer than their
˚
molecular lengths (1b: 35 and 1e: 40 A). This indicated that
Fig. 14 (a) XRD of 1e at 150 uC on heating. (b) XRD of 1e at 150 uC
on cooling.
these domains were generated as smectic Ad phases at the first
stage on cooling, in which the molecules dimerized in an anti-
˚
parallel manner with a shift (ca. 4 A) and the alkyl chains did
compounds. Then, the XRD patterns of 1e were measured at
150 uC on heating and cooling (rate: 5 uC min21) in one heat–
cool cycle (70 A 150 A 210 A 150 A 70 uC), and this
procedure was repeated three times (Fig. 14). The peak
positions and their intensities on heating are different from
those on cooling. The diffraction pattern on the first heating
not interdigitate between the layers, and then the interdigitated
layers occurred with cooling to give the peaks at the shorter
layer lengths.
In the case of 1e, after heating and maintaining the tem-
perature at 150 uC for two hours, no measurable change was
observed in the peak positions and intensities in the XRD
chart, though the temperature (150 uC) was almost the middle
of the temperature range of the smectic A phase (98–199 uC
on heating and 193–71 uC on cooling) and high enough to
remove stress in usual low molecular weight liquid crystalline
˚
(Fig. 14a) has peaks at 2.17 (40.7 A), 2.34 (37.7 A) and 2.50
˚
˚
(35.3 A), while the diffraction pattern on the first cooling
˚
(Fig. 14b) has only one peak at 2.00 (44.1 A). After the two
additional heat–cool cycles, the peak intensities decreased and
˚
the peak top converged at 1.87–1.90 (ca. 47 A) on both heating
and cooling. The smectic Ad phase without interdigitation
seems to be stable at the temperature close to the clearing point.
From these results, it is clear that the multiplicity in the layer
distances depended on the hysteresis of the sample.
Explanation for the interdigitation
In order to estimate the degree of interdigitation in the smectic
A phase, we introduced a new coefficient (Rc/c) as follows. In
the model, the long axes of the terminal chains and the rod-like
cores are parallel to the layer normal (Scheme 6). The coeffi-
cient is defined as Rc/c ~ Nchain/Ncore. In the equation, Nchain is
the number of chains crossed by a plane which is parallel to the
layer, and Ncore is the number of rod-like cores in the layer. If
the size of the plane is Splane and the alkyl chains (the number is
N’chain) in the alkyl chain layer are packed tightly, Splane/N’chain
nearly equals the cross section of an alkyl chain (~Schain). If the
rod-like cores (the number is N’core) in the core layer are packed
tightly, Splane/N’core nearly equals the cross section of a core
(~Score). When both the cores and chains are packed with
Fig. 13 Growth of the liquid crystal domain of 1e. XRD was mea-
sured at 183, 171, 160, 149, and 138 uC on cooling (from the bottom to
the top). The arrows indicate growth of the peaks. The temperature rate
was 5 uC min21
.
almost no empty space, respectively, Rc/c equals N’chain/N’core
.
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1
2 6 1 7

View Article Online
Scheme 6 Non-interdigitated and interdigitated state and Rc/c in the
case of 1-core-2-chain type compounds.
Scheme 8 Non-interdigitated and interdigitated states and Rc/c in the
cases of the 2-core-3-chain type compounds.
of the long alkyl chains seems to prevent the interdigitation
between the layers. The difference in the interdigitation might
be observed in DSC experiments of 1e and 1f. In the case of 1e,
the peak from the isotropic to the liquid crystal phase had
several small shoulders, while in the case of 1f it had a single
peak.
From monitoring of growth of the liquid crystal domains by
XRD, it was found that the longer layer distance grew at higher
temperature and the shorter distance grew at lower tempera-
ture. This can be explained by the decrease in the alkyl chain
mobility on cooling, which decreases its excluded volume and
increases the degree of the interdigitation between the layers. In
the smectic Ad phase observed in the heat–cool cycle XRD
Scheme 7 Cross sections of (a) a benzene ring and (b) an alkyl chain
˚
(a; 6.3, b; 3.6, c; 4.0 A)
Accordingly, N’chain/N’core ~ (1/N’core)/(1/N’chain) ~ (Splane
/
N’core)/(Splane/N’chain) ~ Score/Schain. We assume that the shapes
of the cross sections of a benzene ring and a alkyl chain are
an oval and a circle (Scheme 7), respectively. These areas are
2
˚
6.3 6 3.6 6 (p/4) (~Score) and 4.0 6 4.0 6 (p/4) (~Schain) A
at a rough estimate, which is calculated from their dimensions
a–c (a: width of a benzene-ring; 6.3, b: thickness of a benzene-
24
˚
ring; 3.6, and c: diameter of an alkyl chain; 4.0 A).
The obtained ratio for Score/Schain (1.4:1) indicated that the
most adequate Rc/c is approximately 1.4 in smectic A phases if
the movements of the cores and alkyl chains are not taken into
account (a temperature factor must be considered for the exact
Rc/c). For example, if one rod-like molecule has one alkyl chain
at both terminals, Rc/c is 1.0 in all levels of the alkyl chain layer
in the non-interdigitated state (Scheme 6a) and 2.0 in the
interdigitated part of the alkyl chain layer in the partially
interdigitated state (Scheme 6b). If Rc/c becomes larger than 1.4
with the interdigitation, the intermolecular distances between
the cores increase and the core–core interaction decreases. As
a consequence, the non-interdigitated state is recovered to
stabilize the layer structure. If Rc/c is much smaller than 1.4, it is
assumed that interdigitation occurs to supply the lack of space
in the alkyl chain layer.
˚
experiment of 1e (layer distance: 47 A), we propose the
dimerization as depicted in Fig. 15. In the dimer, it is assumed
that the polar connecting groups (–N–CO–, O–CO–) contact
intermolecularly to cancel out their polarities effectively. The
˚
length of the dimer was calculated to give 47 A, and it agreed
with the layer distance of the smectic Ad phase. Although the
alkyl layer of the smectic Ad phase was sparse, the excluded
volume of the alkyl chain at high temperature was large enough
to supply the lack of space. After the smectic Ad phase
appeared from the isotropic liquid on cooling, it retained its
large hysteresis.
Introduction of an asymmetric center at the N-alkyl chain
In the smectic A phases of compounds 1 except 1f and 1i, the
probability of interdigitation is explained as follows. For
example, in the case of 1e (R1 ~ R2), Rc/c is 0.75 (vv1.4) at all
levels of the alkyl chain layer in the non-interdigitated state
(Scheme 8a), and the interdigitation of the alkyl chain layers
was preferred to generate the adequate Rc/c (~1.5 (#1.4))
(Scheme 8b). In the case of R1 | R2, the interdigitated states
(Schemes 8c and d) were also preferred to achieve adequate
Rc/c in the alkyl chain layer. In the case of 1f (R1 ~ n-C8H17,
R2 ~ n-C12H25) and 1i (R1 ~ R2 ~ n-C12H25) which have
relatively long alkyl chains in this series, the flexible movement
Compounds rac-1j, rac-1k, (R)-1k and (S)-1k that possess a
1-methylheptyl group at the nitrogen atom were synthesized.
Their behavior is shown in Table 3. All these compounds
exhibited smectic A phases and the transition enthalpies of the
liquid crystal–liquid transition were larger than those of com-
pounds 1a–i. Increase of the transition enthalpies with the
substitution of R1 groups means that the difference between the
molecular aggregations of the smectic A and isotropic phases
became larger. It was assumed that the time-averaged dimeriza-
tion in the isotropic phases of 1j–1k is suppressed to some
extent by the intermolecular steric repulsion between the bulky
2 6 1 8
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1

View Article Online
Fig. 15 Dimerization of 1e in the smectic Ad phase. The polar parts of
one molecule are near those of the other molecule to cancel out those
dipoles.
Fig. 16 (a) Growth of the liquid crystal domain of rac-1k. After 30 min
heating at 150 uC, XRD was measured at 138, 132, 126, 120, 114, 108,
and 102 uC on cooling (from the bottom to the top). The arrows
R1 (1-methylheptyl) groups. The temperature range of the
liquid crystal phase of rac-1j was broader than that of the
isomer (1e) on heating, and rac-1j did not show a clear phase
transition from the smectic A to the crystal phase on cooling.
The branched N-alkyl chain (1-methylheptyl) seems to prevent
the crystallization. On the other hand, compound rac-1k
showed a clear crystal–liquid crystal phase transition on both
heating and cooling. The clearing points of chiral compounds
(R)-1k and (S)-1k were slightly lower than that of rac-1k, and
they did not show a clear transition from the liquid crystal
to the crystal phase. Accordingly, during the crystallization
indicate growth of the peaks. The temperature rate was 5 uC min21
.
(b) Growth of the liquid crystal domain of (R)-1k.After 30 min heating
at 150 uC, XRD was measured at 132, 126, 120, 114, 108, and 102 uC on
cooling (from the bottom to the top). The arrows indicate growth of the
peaks. The temperature rate was 5 uC min21
.
process, pairing of (R)-1k and (S)-1k may be necessary for
getting the stable packing structure. Growth of the liquid
crystalline domains of rac-1k and (R)-1k was monitored by
XRD on cooling. A clear difference was observed in the growth
of these peaks (Fig. 16). In the case of rac-1k, peaks at 2.20
˚
˚
˚
(40.1 A), 2.36 (37.4 A), and 2.94 (30.0 A) were observed
(Fig. 16a). On the other hand, in the case of (R)-1k, only one
˚
peak at 2.20 (40.1 A) was observed (Fig. 16b). To the best of
Table 3 Behavior of compounds 1j–k
our knowledge, this is the first example which showed a clear
difference between chiral and racemic liquid crystalline com-
pounds in interlayer interdigitations of the alkyl chains. It is
thought that the peak at 2.20 originates in the layer structure of
the optically pure molecules and the peaks at 2.36 and 2.94 are
due to that of racemic pairs of the molecules. Accordingly, in
Fig. 16a, it is assumed that the racemic domains on cooling
appeared at the higher temperature and then both the racemic
and the optically pure domains came out simultaneously. This
might indicate that spontaneous optical resolution in a fluid
liquid crystal phase25 partially occurred in this system. It was
supposed that the racemic pairs ((RS)-dimers) and chiral pairs
((RR)- and (SS)-dimers) generated by the dimerization with the
large dipole–dipole interaction were organized into their own
domains, respectively.
Compound Chirality
X
Behaviora
187 ð4:2Þ
Cr D8C1 ðC13:4AÞ SmA
I
rac-1j
rac-1k
(R)-1k
(S)-1j
racemic
racemic
R-form
S-form
–O–CO–
CCDA
BCDC
183 ð{3:9Þ
89 ð4:6Þ
130 ð4:4Þ
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
CCDA
BCDC
CCDA
BCDC
Cr
SmA
I
3. Conclusion
68 ð{3:7Þ
126 ð{4:1Þ
We could demonstrate that syn-parallel arrangement of two
mesogenic cores by linkage with a covalent bond generated
highly viscous liquid-crystal phases. It was assumed that the
large dipole–dipole interaction between the ‘‘tuning fork’’-shaped
molecules prevented free molecular movements along their
long axes. The large hysteresis in the layer distances of the
smectic phase was observed because of the high viscosity and
the small energy gap between the non-interdigitated and inter-
digitated states. From this study 1) we found that introduction
of a very strong dipole along a molecular long axis remarkably
decreased fluidity of the liquid crystal phase even in low
129 ð3:8Þ
88 ð1:4Þ
CCDA
BCDC
Cr DCCA SmA
I
I
127 ð{3:9Þ
129 ð3:8Þ
88 ð1:4Þ
CCDA
BCDC
Cr DCCA SmA
127 ð{3:9Þ
a The transition temperatures (uC) and enthalpies (in parentheses,
kcal mol21) were determined by DSC (5 uC min21) and are given
above and below the arrows. Cr, SmA, and I indicate crystal, smec-
tic A, and isotropic phases, respectively.
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1
2 6 1 9

View Article Online
molecular-weight compounds to give high viscosity in the
mesophase; 2) we could explain the probability of interdigita-
tion using Rc/c in the smectic A phase model, which was useful
to modulate the extent of the interdigitation between the alkyl
layers; 3) a clear difference between chiral and racemic liquid
crystalline compounds was observed in the growth of the liquid
crystal domains.
found: C% 76.40, H% 6.28, N% 1.40, calcd for C54H55NO8: C%
76.66, H% 6.55, N% 1.66.
An example for synthesis of N,N-bis{4-[4-(4-alkoxybenzoyloxy)-
benzyloxy]benzoyl}alkylamine 1
To a 100 ml flask were added 3c (1.00 g, 1.19 mmol), ethanol
(20 ml), THF (20 ml), and 10% palladium-activated carbon
(1.00 g). The mixture was stirred at room temperature for 2 h
under an atmosphere of hydrogen. After filtrating off by celite,
the solution was concentrated in vacuo to give N,N-bis[4-(4-
hydroxybenzoyloxy)benzoyl]dodecylamine as a white solid
(0.79 g). The solid was added into a 100 ml round-bottom
flask, and 4-dodecyloxybenzoylchloride (0.77 g, 2.38 mmol),
4-(N,N-dimethylamino)pyridine (0.80 g, 6.55 mmol), and THF
(50 ml) were added. The mixture was stirred at room tem-
perature for 2 days. To the solution was added ethyl acetate
(50 ml), and the solution was washed with 1 N HCl (100 ml)
and aqueous NaHCO3 solution (300 ml) and dried over
anhydrous MgSO4. The solvent was evaporated under reduced
pressure, and a white solid was obtained as the residue. The
crude product was purified by silica gel chromatography
(chloroform) to give 1i as a white solid (0.89 g, 61%).
1b: yield 66.8%; white solid (methanol–ethyl acetate); n/cm21
(KBr) 2930, 1735, 1650, 1610, 1415, 1275, 1175, 780; dH
(500 MHz, CDCl3, TMS) 0.87 (t, 6H, J ~ 7.0 Hz), 0.97 (t, 3H,
J ~ 7.0 Hz), 1.23–1.38 (m, 22H), 1.80 (t, 6H, J ~ 6.9 Hz), 4.03
(t, 6H, J ~ 6.9 Hz), 6.96 (d, 4H, J ~ 8.6 Hz), 7.10 (d, 4H, J ~
8.9 Hz), 7.33 (d, 4H, J ~ 8.9 Hz), 7.52 (d, 4H, J ~ 8.6 Hz), 8.12
(d, 4H, J ~ 8.9 Hz), 8.20 (d, 4H, J ~ 9.2 Hz); dC (100.4 MHz,
CDCl3, TMS) 13.79, 14.09, 20.34, 22.64, 25.96, 29.05, 29.20,
29.30, 31.14, 31.78, 47.38, 68.37, 114.39, 120.83, 121.79, 122.16,
126.34, 130.21, 131.83, 132.40, 134.24, 153.47, 155.53, 163.62,
163.82, 164.27, 173.37; Elemental analysis found: C, 72.50%
H% 6.66, N% 1.30, calcd for C62H67NO12 (10.5H2O): C%
72.50, H% 6.67, N% 1.36.
4. Experimental
The spectral data of 2a, 2b, 3a, 3b, 1a, and 1e were reported in
our previous paper.16
An example for preparation of N,N-di(4-benzyloxybenzoyl)-
alkylamine 2
To a 200 ml round-bottom flask were added 4-benzyloxy-
benzoyl chloride (8.63 g, 35.0 mmol), n-dodecylamine (3.98 ml,
17.5 mmol), triethylamine (14.63 ml, 105.0 mmol), and toluene
(100 ml). The mixture was then stirred at reflux for 2 days.
After cooling, the solution was washed with water (100 ml), 1 N
HCl (100 ml), and aqueous NaHCO3 solution (300 ml), and
dried over anhydrous MgSO4. The solvent was evaporated
under reduced pressure, and a yellow solid was obtained as the
residue. The crude product was purified by silica gel chromato-
graphy eluting with chloroform to give 2c as a white solid
(3.73 g, 35.2%).
2c: mp 108.5–109.0 uC (methanol–ethyl acetate); n/cm21(KBr)
2924, 1706, 1646, 1606, 1509, 1455, 1253, 1170, 759; dH
(500 MHz, CDCl3, TMS) 0.90 (t, 3H, J ~ 6.7Hz), 1.25–1.43
(m, 18H), 1.82 (t, 2H, J ~ 6.9 Hz), 4.01 (t, 2H, J ~ 6.9 Hz),
5.00 (s, 4H), 6.98 (d, 4H, J ~ 9.0 Hz), 7.34 (t, 2H, J ~ 9.0 Hz),
7.38 (t, 4H, J ~ 9.0 Hz), 7.41 (d, 4H, J ~ 9.0 Hz), 8.03 (d, 4H,
J ~ 9.0 Hz); dC (100.4 MHz, CDCl3, TMS) 14.11, 22.71, 27.06,
29.10, 29.32, 29.36, 29.55, 29.65, 29.68, 31.96, 47.73, 70.24,
114.10, 127.01, 127.77, 128.20, 129.18, 130.68, 135.86, 161.10,
173.26; Elemental analysis found: C% 79.16, H% 7.89% N,
2.07, calcd for C40H47NO4: C% 79.30, H% 7.82, N% 2.31.
Synthesis of 1j
Synthesis of 1j was carried with the same procedure as those of
1a–i.
A typical procedure for synthesis of N,N-bis[4-(4-benzyloxy-
benzoyloxy)benzoyl]alkylamine 3
1j: white solid (methanol–ethyl acetate); n/cm21(KBr) 2956,
2925, 2855, 1739, 1511, 1272, 1064, 760; dH (500 MHz, CDCl3,
TMS) 0.88 (t, 3H, J ~ 7.1 Hz), 0.90 (t, 6H, J ~ 7.0 Hz),
1.29–1.50 (m, 28H), 1.55 (d, 3H, J ~ 7.0 Hz), 1.83 (quint, 4H,
J ~ 6.9 Hz), 1.8–1.9 (m, 1H), 2.12–2.19 (m, 1H), 4.05 (t, 4H,
J ~ 6.9 Hz), 4.89 (sext, 1H, J ~ 7.0 Hz), 6.98 (d, 4H, J ~
8.9 Hz), 7.10 (d, 4H, J ~ 8.6 Hz), 7.35 (d, 4H, J ~ 8.6 Hz), 7.52
(d, 4H, J ~ 8.9 Hz), 8.14 (d, 4H, J ~ 8.9 Hz), 8.22 (d, 4H, J ~
8.9 Hz); dC (500 MHz, CDCl3, TMS) 14.28, 14.31, 19.22, 22.83,
22.88, 26.21, 27.42, 29.31, 29.34, 29.44, 29.54, 31.98, 32.02,
35.21, 55.64, 68.63, 114.67, 121.13, 122.04, 122.39, 126.62,
130.38, 132.09, 132.65, 135.76, 153.68, 155.81, 163.88, 164.09,
164.51, 173.74; Elemental analysis found: C% 3.57, H% 6.83,
N% 1.20, calcd for C66H75NO12: C% 73.79, H% 7.04, N% 1.30.
To a 500 ml flask were added 2c (5.00 g, 8.25 mmol), ethanol
(80 ml), THF (80 ml), and 10% palladium-activated carbon
(1.50 g). The mixture was stirred at room temperature for 2 h
under an atmosphere of hydrogen. After filtrating off by celite,
the solution was concentrated in vacuo to give N,N-di(4-
hydroxybenzoyl)dodecylamine as a white solid (3.51 g). The
solid was added to a 100 ml round-bottom flask, and
4-benzyloxybenzoylchloride (5.09 g, 20.6 mmol), 4-(N,N-
dimethylamino)pyridine (2.00 g, 16.4 mmol), and THF
(50 ml) were added. The mixture was stirred at room tempera-
ture for 27 h. To the solution was added ethyl acetate (50 ml),
and the solution was washed with 1 N HCl (100 ml) and
aqueous NaHCO3 solution (300 ml) and dried over anhydrous
MgSO4. The solvent was evaporated under reduced pressure,
and a white solid was obtained as the residue. The crude
product was purified by silica gel chromatography (chloro-
form) to give 3c as a white solid (2.92 g, 42.0%).
3c: mp 82.5–83.0 uC (methanol–ethyl acetate); n/cm21 (KBr)
2924, 1717, 1689, 1646, 1604, 1510, 1456, 1258, 1165, 758; dH
(500 MHz, CDCl3, TMS) 0.89 (t, 3H, J ~ 7.0 Hz), 1.27–1.45
(m, 18H), 1.83 (t, 2H, J ~ 7.0 Hz), 4.04 (t, 2H, J ~ 7.0 Hz),
5.16 (s, 4H), 7.05 (d, 4H, J ~ 8.9 Hz), 7.11 (d, 4H, J ~ 8.9 Hz),
7.38 (t, 2H, J ~ 7.0 Hz), 7.41 (t, 4H, J ~ 7.4 Hz), 7.45 (d, 4H,
J ~ 7.3 Hz), 7.54 (d, 4H, J ~ 8.6 Hz), 8.11 (d, 4H, J ~ 9.2 Hz);
dC (125.65 MHz, CDCl3, TMS) 14.12, 22.70, 27.10, 29.07,
29.32, 29.37, 29.54, 29.61, 29.65, 31.94, 47.73, 70.26, 114.79,
121.62, 121.84, 127.51, 128.34, 128.76, 130.21, 132.42, 134.05,
136.08, 153.77, 163.25, 164.00, 173.48; Elemental analysis
Optical resolution of 2-aminooctane
In a 300 ml round-bottom flask (S)-mandelic acid (13.4 g,
88.1 mmol) was dissolved in ethanol (88.1 ml), and then ethyl
acetate(88.1ml)andracemic2-aminooctane(14.8 ml, 88.1mmol)
were added. The solution was heated to dissolve the white
crystals at 75 uC in a hot water bath. The solution was allowed
to stand overnight. The crystals (13.2 g, 29% de) were separated
by filtering with suction. The recrystallization procedure was
repeated two more times to give a pure diastereomeric salt (2nd
crop: 7.0 g (80% de), 3rd crop: 5.2 g (98% de)).
In a 100 ml round bottom flask, the salt (0.407 g, 2.46 mmol)
was suspended in diethyl ether (20 ml), and a solution of KOH
(3.04 g, 54.2 mmol) in water (30 ml) was added. The solution
was stirred for 10 min at room temperature. The organic layer
2 6 2 0
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1

View Article Online
3
Y. Matsunaga and S. Miyamoto, Mol. Cryst. Liq. Cryst., 1993,
237, 311; T. Akutagawa, Y. Matsunaga and K. Yasuhara, Liq.
Cryst., 1994, 17, 659; T. Niori, F. Sekine, J. Watanabe,
T. Furukawa and H. Takezoe, J. Mater. Chem., 1996, 6, 1231;
D. R. Link, G. Natale, R. Shao, J. E. Maclennan, N. A. Korblova
and D. M. Walba, Science, 1997, 278, 1924; G. Heppke and
D. Moro, Science, 1998, 279, 1872; D. Shen, A. Pegenau, S. Diele,
I. Wirth and C. Tschierske, J. Am. Chem. Soc., 2000, 122, 1593;
D. M. Walba, E. Ko¨rblava, R. Shao, J. E. Maclennan,
D. R. Link, M. A. Glaser and N. A. Clark, Science, 2000, 288,
2181–2184.
S. Norvez, F.-G. Tournilhac, P. Bassoul and P. Herson, Chem.
Mater., 2001, 13, 2552.
I. A. Levitsky, K. Kishikawa, S. H. Eichhorn and T. M. Swager,
J. Am. Chem. Soc., 2000, 122, 2474; K. Kishikawa, M. C. Harris
and T. M. Swager, Chem. Mater., 1999, 11, 867.
was separated and dried over anhydrous MgSO4. The ether
solution was carefully concentrated at room temperature using
a rotary evaporator. The residual oil was put in a glass tube
oven and the distillation gave clear oil (2-aminooctane, 0.113 g,
37%). (S)-2-aminooctane (98% ee): [a]D19 19.12u (0.0113 g ml21
,
benzene), cell length: 100 mm, temperature: 19.2 uC, integra-
tion time: 60 s. (Ref. 26: (R)-2-octylamine, [a]D 27.13u
(0.0121 g ml21, benzene)).
Determination of the enantiomeric excess. The obtained
2-aminooctane was reacted with 3,5-dinitrobenzoylchloride
in the presence of triethylamine in THF at room temperature
to give the corresponding amide. The enantiomeric excess of
the amide was determined by HPLC using a chiral column
packed in our laboratory (column: 4.6 mm 6 500 mm, solvent:
hexane–isopropanol (19:1), flow rate: 2.0 ml min21, detection:
254 nm, retention time of (R)-isomer: 54.3 min, retention time
of (S)-isomer: 64.0 min).27
4
5
6
7
K. Kishikawa, S. Furusawa, T. Yamaki, S. Kohmoto,
M. Yamamoto and K. Yamaguchi, J. Am. Chem. Soc., 2002,
124, 1597.
C. W. Struijk, A. B. Sieval, J. J. E. Dakhorst, M. van Dijk,
P. Kimkes, R. B. M. Koehorst, H. Donker, T. J. Schaafsma,
S. J. Picken, A. M. van de Craats, J. M. Warman, H. Zuilhof and
E. J. R. Sudho¨lter, J. Am. Chem. Soc., 2000, 122, 11057.
T. Ohtake, M. Ogasawara, K. Ito-Akita, N. Nishina, S. Ujiie,
H. Ohno and T. Kato, Chem. Mater., 2000, 12, 782.
A.-J. Attias, C. Cavalli, B. Donnio, D. Guillon, P. Hapiot and
J. Maltheˆte, Chem. Mater., 2002, 14, 375; A. Rego, S. Kumar and
H. Ringsdorf, Chem. Mater., 1996, 8, 1402.
Synthesis of 1k
Synthesis of 1k was carried according to our reported
procedure.16
8
9
1k: n/cm21(KBr) 2924, 2855, 1727, 1685, 1645, 1606, 1511,
1456; dH (500 MHz, CDCl3, TMS) 0.86 (t, 3H, J ~ 6.9 Hz),
0.89 (t, 6H, J ~ 7.0 Hz), 1.25–1.40 (m, 24H), 1.47 (quint, 4H,
J ~ 7.1 Hz), 1.51 (d, 3H, J ~ 7.2 Hz), 1.75–1.85 (m, 1H), 1.81
(quint, 4H, J ~ 7.1 Hz), 2.10–2.15 (m, 1H), 2.75–2.82 (m, 8H),
4.03 (t, 4H, J ~ 7.1 Hz), 4.80 (sext, 1H, J ~ 7.2 Hz), 6.94 (d,
4H, J ~ 8.1 Hz), 6.96 (d, 4H, J ~ 9.0 Hz), 7.06 (s, 8H), 7.34 (d,
4H, J ~ 8.1 Hz), 8.12 (d, 4H, J ~ 9.0 Hz); dC (125.65 MHz,
CDCl3, TMS) 14.07, 14.10, 18.99, 22.58, 22.65, 25.99, 27.16,
29.10, 29.22, 29.32, 31.77, 31.80, 35.01, 36.78, 37.58, 55.10,
68.35, 114.30, 121.62, 128.25, 128.36, 128.93, 129.31, 132.24,
135.99, 138.34, 145.65, 149.38, 163.54, 165.01, 174.39.
rac-1k: white solid; Elemental analysis found: C% 78.01; H%
8.12; N% 1.44, calcd for C68H83NO8: C% 78.35; H% 8.03; N%
1.34.
10 C. Nuckolls, R. Shao, W.-G. Jang, N. A. Clark, D. M. Walba and
T. Katz, Chem. Mater., 2002, 14, 773.
11 L. Dinescu and R. P. Lemieux, J. Am. Chem. Soc., 1997, 119, 8111.
12 W. L. MacMillan, Phys. Rev., 1973, A8, 1921.
13 A. G. Khachaturyan, J. Phys. Chem. Solids, 1975, 36, 1055.
14 J. W. Goodby, in Handbook of Liquid Crystals, ed. D. Demus,
J. Goodby, G. W. Gray, H.-W. Spiess and V. Vill, Wiley-VCH,
1998, vol. 2A, ch.V.
15 Y. Matsunaga, L. Hikosaka, K. Hosono, N. Ikeda, T. Sakatani,
K. Sekiba, K. Takachi, T. Takahashi and Y. Uemura, Mol. Cryst.
Liq. Cryst., 2001, 363, 51; M. Yayloyan, L. S. Bezhanova and
E. B. Abrahamyan, Ferroelectrics, 2000, 245, 147.
16 K. Kishikawa, Y. Miwa, T. Kurosaki, S. Kohmoto, M. Yamamoto
and K. Yamaguchi, Chem. Mater., 2001, 13, 2468.
17 G. W. Gray and J. W. G. Goodby, in Smectic Liquid Crystals,
Textures and Structures, Leonard Hill, London, 1984, ch. I.
18 MM2 calculation: molecular modeling of compounds 1 and 2 was
carried out by Chem3D (Cambridge software corporation).
19 M. Nishio, M. Hirota and Y. Umezawa, The CH/p Interaction.
Evidence, Nature, and Consequences, Wiley-VCH, New York,
1998; M. Nishio, Y. Umezawa, M. Hirota and Y. Takeuchi,
Tetrahedron, 1995, 51, 8665; S. Paliwal, S. Geib and C. S. Wilcox,
J. Am. Chem. Soc., 1994, 116, 4497.
(R)-1k: white solid; Elemental analysis found: C% 78.20; H%
8.09; N% 1.39, calcd for C68H83NO8: C% 78.35; H% 8.03; N%
1.34.
(S)-1k: white solid; Elemental analysis Found: C% 78.01;
H% 7.99; N% 1.51, Calcd for C68H83NO8: C% 78.35; H% 8.03;
N% 1.34.
Acknowledgements
20 K. Kishikawa, S. Tsubokura, S. Kohmoto, M. Yamamoto and
K. Yamaguchi, J. Org. Chem., 1999, 64, 7568; K. Kishikawa,
C. Iwashima, S. Kohmoto, K. Yamaguchi and M. Yamamoto,
J. Chem. Soc., Perkin Trans. 1, 2000, 2217–2221.
21 AM1 calculation was carried out by using WinMOPAC Ver.3
(Fujitsu, Ltd.) software. J. J. P. Stewart, J. Comput. Chem., 1989,
10, 209–220 and 221–264.
22 A. J. Leadbetter, R. M. Richardson and C. N. Collings, J. Phys.,
1975, 36, 37; J. E. Lydon and C. J. Coakley, J. Phys., 1975, 36, 45;
D. A. Dunmur, M. R. Manterfield, W. H. Miller and
J. K. Dunleavy, Mol. Cryst. Liq. Cryst., 1978, 45, 127.
23 S. Diel, S. Manke, W. Weißflog and D. Demus, Liq. Cryst., 1989,
4, 301.
24 The sizes of a benzene ring and an alkyl chain were calculated
based on van der Waals radii. A. Bondi, J. Phys. Chem., 1964, 68,
443.
25 Y. Takanishi, H. Takezoe, Y. Suzuki, I. Kobayashi, T. Yajima,
M. Terada and K. Mikami, Angew. Chem., Int. Ed., 1999, 38, 2354.
26 O. Mitsunobu, M. Wada and T. Sano, J. Am. Chem. Soc., 1972,
94, 697.
27 K. Kishikawa, Y. Takada, K. Kawashima, S. Kohmoto,
M. Yamamoto and K. Yamada, Tetrahedron: Asymmetry, 1996,
7, 1733.
This work was supported by Iketani Science and Technology
Foundation and a Grant-in-Aid for Scientific Research (C)
from the Japan Society for the Promotion of Science (JSPS)
(13640571). We thank Chiba University Radioisotope
Research Center for measurement of powder X-ray diffraction
of the liquid crystals.
References
1
Y. Lansac, M. A. Glaser and N. A. Clark, Phys. Rev. E, 2001, 64,
51703/1; A. Mathis, M. Galin, J. C. Galin, B. Heinrich and
C. G. Bazuin, Liq. Cryst., 1999, 26, 973; Y. E. Shapiro and
A. Y. Boytsov, Mol. Cryst. Liq. Cryst., 1997, 299, 517; S. Taki
and S. Kai, Jpn. J. Appl. Phys., 2001, 40, 4187.
2
U. Stanislaw, R. Dabrowski, B. Gestblom and A. Kocot, Liq.
Cryst., 2000, 27, 1675; J. S. Walker and J. L. Mace, Phys. Lett. A,
1986, 115, 281; P. E. Cladis, Phys. Rev. Lett., 1975, 35, 48;
F. Hardouin, G. Sigaud, M. F. Achard and H. Gasparoux, Phys.
Lett., 1979, 71A, 347; N. H. Tinh, C. Destrade, J. Maltheˆte and
J. Jacques, Mol. Cryst. Liq. Cryst., 1982, 72, 195.
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 2 6 1 2 – 2 6 2 1
2 6 2 1