Regioselectivity control of photodimerization of liquid-crystalline cinnamoyl compounds by phase variation: Dual functionality of p-terphenyl substituent as a mesogen and a triplet sensitizer

Article abstract of DOI:10.1016/j.tet.2012.04.083

Cinnamoyl-functionalized liquid-crystalline (LC) compounds having a 4″-substituted [1,1′;4′,1″]terphenyl (p-terphenyl) as mesogens were synthesized to investigate the abilities of the mesogen to act as a triplet sensitizer. UV (365 nm) irradiation of the LC compound having 4″-cyano-p-terphenyl in the crystalline and LC phases regioselectively produced the photodimers with a head-to-tail and a head-to-head cyclobutane unit, respectively. The p-terphenyl thus played the role of triplet sensitizer because the cinnamoyl group does not absorb 365 nm light. Due to the dual functionality of the p-terphenyl as the mesogen and the triplet sensitizer, we successfully performed the photochemical and regioselective cycloaddition to obtain two different LC dimers from one LC monomer.

Full text of DOI:10.1016/j.tet.2012.04.083

Tetrahedron 68 (2012) 5513e5521
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Regioselectivity control of photodimerization of liquid-crystalline cinnamoyl
compounds by phase variation: dual functionality of p-terphenyl substituent as
a mesogen and a triplet sensitizer
,
b
a,
Hideyuki Kihara a , Yasuo Norikane , Masaru Yoshida
*
*
a Nanosystem Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan
b Electronics and Photonics Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan
a r t i c l e i n f o
a b s t r a c t
Cinnamoyl-functionalized liquid-crystalline (LC) compounds having a 400-substituted [1,10;40,100]terphenyl
(p-terphenyl) as mesogens were synthesized to investigate the abilities of the mesogen to act as a triplet
sensitizer. UV (365 nm) irradiation of the LC compound having 400-cyano-p-terphenyl in the crystalline
and LC phases regioselectively produced the photodimers with a head-to-tail and a head-to-head cy-
clobutane unit, respectively. The p-terphenyl thus played the role of triplet sensitizer because the cin-
namoyl group does not absorb 365 nm light. Due to the dual functionality of the p-terphenyl as the
mesogen and the triplet sensitizer, we successfully performed the photochemical and regioselective
cycloaddition to obtain two different LC dimers from one LC monomer.
Article history:
Received 6 February 2012
Received in revised form 20 April 2012
Accepted 23 April 2012
Available online 28 April 2012
Keywords:
Liquid crystals
Photochemistry
Sensitizers
Ó 2012 Elsevier Ltd. All rights reserved.
Cycloaddition
Regioselectivity
1. Introduction
organic compounds. In addition, the efficiency of the photo-
cycloaddition is significantly improved using triplet sensitizers.11e14
Photocycloaddition of a cinnamoyl group is one of the useful
synthetic techniques for the dimerization and cross-linking of or-
ganic compounds in order to prepare a variety of functional mate-
rials, e.g., liquid crystal alignment films, holographic recording
materials, crack healing materials, and nanogels for drug delivery.1e5
One of the most unique features of the reaction is the regio- and
stereo-selectivity at the cyclobutane unit. It has been well un-
derstood that the molecular structure of a cyclobutane unit resulting
from the [2þ2] cycloaddition by direct photo-excitation of cinnamic
acid in the crystalline phase is closely related to the crystal packing
(topochemistry).6e9 Therefore, the crystal engineering approach to
control the structure of the cyclobutane unit using the crystalline
condition has already attracted much attention.10
Meanwhile, for the practical examples for photocycloaddition of
a cinnamoyl group, indirect photo-excitation by using a triplet sen-
sitizer in the non-crystalline condition has often been employed in-
stead of the direct excitation.11e14 This is because UV light shorter
than ca. 300 nm, which is necessary for the direct excitation of
a cinnamoyl group, is likely to decompose certainparts of the starting
Actually, cinnamoyl groups attached to amorphous polymers or
liquid-crystalline (LC) polymers as their side chain have been
dimerized through indirect photo-excitation using a triplet sensitizer
additive, in order to cross-link the polymers.15e17 In some cases, the
triplet sensitizer was covalently connected to the polymer backbone
and the photodimerization of the cinnamoyl group by intramolecular
triplet energy transfer was carried out.18e20 However, to the best of
our knowledge, the photodimerization of the cinnamoyl moiety via
triplet energy transfer in a crystalline (or LC) phase has not yet been
well examined.
As another important issue, the dimerization of LC compounds,
that is, the connection of two LC monomers by a covalent bond is
one of the powerful tools for the tuning of the LC properties.21e28 LC
dimers have shown very different properties from the corre-
sponding monomers. Fig. 1 shows typical examples of the LC dimer
structures. In a linear dimer, two mesogens locate on the opposite
side relative to a connection unit, while two mesogens exist on the
same side relative to the connection unit in a U-shaped dimer.
Among the LC dimers based on the same mesogens, their LC
properties significantly vary when they have different regio- and/or
stereo-chemical geometries.29,30 Therefore, the regioselectivity
control of the LC dimer structure has been thoroughly required to
precisely tune the LC properties.
* Corresponding authors. Tel.: þ81 29 861 4442 (H.K.); tel.: þ81 29 861 9422
(M.Y.); e-mail addresses: h-kihara@aist.go.jp (H. Kihara), masaru.yoshida@aist.go.jp
(M. Yoshida).
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2012.04.083

5514
H. Kihara et al. / Tetrahedron 68 (2012) 5513e5521
365 nm,37,38 but also the mesomorphic ability of the compounds.39
We investigated the photodimerization of the LC monomers in
their crystalline and LC phases by indirect photo-excitation of the
dimerizable group via the p-terphenyl moiety and the properties of
the obtained compounds in order to clarify the dual functionality of
the p-terphenyl units as the mesogens and triplet sensitizers. We
also evaluated the effect of the auxiliary triplet sensitizer additive
on the photoreaction in the LC phase. In addition, the regiose-
lectivity of the photodimerization based on the crystalline and LC
phases and difference in the LC phases of the obtained photodimers
associated with regio-isomerism are described.
mesogen
connection
unit
connection
unit
2. Results and discussion
2.1. Syntheses and properties of the LC monomers
(a)
(b)
We used the Suzuki coupling reaction for the syntheses of the p-
terphenyl mesogens. The cinnamate monomers 1 and 2 having
cyano-p-terphenyl and nitro-p-terphenyl groups were prepared by
a similar synthetic route, as shown in Scheme 1. Due to the low
solubility of the synthetic intermediate compound, 4-(6-
bromohexyloxy)-400-N,N-dimethylamino-p-terphenyl, in common
organic solvents, we performed the Suzuki reaction as the final step
for the synthesis of 3, as shown in Scheme 2. The details of the
synthesis are described in the Experimental section.
Fig. 1. Schematic structures of (a) linear and (b) U-shaped LC dimers.
In a previous study, we successfully obtained the LC dimers by UV
irradiation of the LC monomers having a cinnamoyl group in their LC
phases in the presence of Michler’s ketone (MK: 4,40-bis(N,N-
dimethylamino)benzophenone) as a triplet sensitizer.31,32 The
cyclobutane units resulting from the [2þ2] photocycloaddition of
the cinnamoyl group exclusively had a head-to-head (HH) config-
uration, while only a trace amount of head-to-tail (HT) isomers were
observed. Consequently, the dimers adopted the U-shaped geome-
try in their LC phases. It should be noted that this regioselectivity
does not come from the LC ordering fields. It was already reported
that the photodimer with the HH cyclobutane unit was significantly
obtained by UV irradiation of an ethyl cinnamate in the molten state
because the reaction intermediates of the HH photodimer were
thermodynamically more stable than that of the HT isomer.33 If it is
possible to conduct photocycloaddition of cinnamoyl-containing LC
compounds via an intramolecular (or an intermolecular) triplet
energy transfer in the crystalline phase, the cyclobutane configu-
ration of the products will be governed by topochemistry rather
than thermodynamics. In other words, we can expect to tune the HH
or HT geometries of the LC dimer by phase variation between the
crystalline and LC states, if the molecular structures in the crystal-
line state are suitable for the HT dimer formation.
Based on the synthetic strategy, we have designed new
cinnamoyl-functionalized LC compounds, as shown in Fig. 2. The
key to the molecular design is the 400-substituted [1,10;40,100]ter-
phenyl (p-terphenyl). The p-terphenyl group is known as a meso-
gen, which allows high isotropization temperatures and a wide
temperature range of LC phases.34,35 In addition, the p-terphenyl is
also anticipated as a triplet sensitizer to excite the cinnamoyl group,
because the existence of the relatively stable triplet-excited state of
p-terphenyl has already been reported.36 In the present study, polar
substituents, such as cyano (1), nitro (2), and N,N-dimethylamino
(3) groups, were introduced at the 400 position of the p-terphenyl
substituent to increase not only the molar extinction coefficient at
1
2
3
4
Br(CH2)6Br
K2CO3
HO
Br
Br(CH2)6O
Br
4
O
O
B
R
1
2
3
4
5
6
Br(CH2)6O
R
cat. Pd(PPh3)4, 10% Na2CO3 aq
5: R = CN
6: R = NO2
c
b
d
1
2
3
4
5
6
a
COOH
C
O
O(CH2)6O
R
e
DBU
1: R = CN
2: R = NO2
Scheme 1. Synthetic route to cinnamoyl-functionalized p-terphenyl LC monomers 1
and 2.
Br(CH2)6O
Br
4
c
b
d
1
2
3
4
a
COOH
C
O
O(CH2)6O
Br
e
DBU
7
O
O
c
b
B
N
d
1
2
3
4
5
6
a
C
O
O(CH2)6O
N
e
cat. Pd(PPh3)4, 10% Na2CO3 aq
3
Scheme 2. Synthetic route to LC monomer 3.
C O(CH2)6O
O
R
The three monomers 1, 2, and 3 have thermotropic LC phases.
Their phase transition temperatures determined by both differen-
tial scanning calorimetry (DSC) and polarizing optical microscopy
(POM) are listed in Table 1. Schlieren textures were observed for the
monomers 1 and 2 in their respective nematic temperature range. A
texture, in which focal-conic fans and homeotropic regions simul-
taneously existed, was also seen in the smectic A phase of 1. In
contrast to 1 and 2, 3 showed only an unidentified smectic phase
1: R = CN
2: R = NO2
3: R = N(CH3)2
Fig. 2. Chemical structures of cinnamoyl-functionalized LC monomers 1, 2, and 3
having p-terphenyl mesogens used in this study.

H. Kihara et al. / Tetrahedron 68 (2012) 5513e5521
5515
Table 1
Table 2
Transition temperatures and enthalpy changes for the monomers 1, 2, and 3 during
the first heating and the first cooling scans
Absorption maxima (lmax) and molar extinction coefficient (
the monomers 1, 2, and 3 in dichloromethane
3 ), and
3
at 365 nm of
Monomer
Phase transition temperaturea (ꢀC) (
Cr 140 (36.5) N 183 (0.19) I
D
H (kJ molꢁ1))
Monomer
lmax (nm) (
3
(Mꢁ1 cmꢁ1))
3
(Mꢁ1 cmꢁ1
)
365 nm
1
1
2
3
300 (38,600)
270 (34,200), 346 (21,700)
281(44,400), 324 (50,800)
500
17,800
5830
I 180 (0.42) N 134 (d)b SmA 87 (2.90) SmX 77 (22.7) Cr
Cr 158 (39.5) N 172 (0.26) I
2
3
I 169 (0.50) N 127 (36.1) Cr
Cr 145 (36.5) SmX 208 (18.5) I
I 204 (18.1) SmX
amorphous phase.11,16e20 However, it has not been clarified if such
an inter- and intramolecular triplet sensitization occurs in the low
molecular weight cinnamoyl compounds in the crystalline phase. In
order toinvestigate the sensitizing abilityof the terphenyl mesogens
in the present study, the monomers 1, 2, and 3 were irradiated with
365-nm UV light in the crystalline state. We prepared a number of
thin-layer crystalline samples of 1 and 2 by cooling their melts to
ensure reproducibility of the photoirradiation experiments. As for 3,
the powdery crystals prepared from an ethanol solution were used
for the UV irradiation, because a crystalline sample was not obtained
from its melt bycooling. Since the cinnamoyl group does not directly
absorb the UV light as already described, observation of the cyclo-
addition reaction under this condition should mean that the ter-
phenyl mesogens have the potential of being a triplet sensitizer in
order to transfer the energy to the cinnamoyl moiety.
Gel permeation chromatography (GPC) charts measured after
the UV irradiation of 1, 2, and 3 at room temperature for specific
times are depicted in Fig. 4. As shown in the charts of 1 and 2, the
peaks of the monomers at around 12.8 min decreased, while the
peaks at shorter elution times than the monomers appeared after
the UV irradiation, showing the progress of the photoreaction. We
confirmed that the early-eluted new peaks corresponded to the
photodimers not only by NMR but also by Matrix-Assisted Laser
Desorption/Ionization Time-of-Flight MS (MALDI-TOF MS). The
intensity of the dimer peak of 1 gradually increased with the in-
creasing irradiation time, while the reaction rate was much slower
than that of 2. However, the final conversions from the monomers
to the corresponding dimers were almost identical in both cases. In
contrast to 1 and 2, the UV irradiation of the crystal of 3 with
a dimethylamino substituent as the electron-donating group pro-
duced only a slight amount of the corresponding dimer, although
the monomer was certainly consumed during the photoirradiation.
We eventually found that the insoluble portions in tetrahydrofuran
(an eluent for GPC analysis) were mainly produced after the UV
irradiation of 3, indicating that an undesirable byproduct occurred
during the UV irradiation of 3. Although the reason for the differ-
ence in the photoreactivity observed for 3 is still unclear, we
eventually found that at least the two monomers 1 and 2 having the
electron-withdrawing groups successfully gave the corresponding
photodimer in the crystalline state.
a
Cr¼crystalline phase; SmX¼unidentified smectic phase; SmA¼smectic A phase;
N¼nematic phase; I¼isotropic phase.
b
Not seen by DSC.
(SmX) after melting at 145 ꢀC and changed to the isotropic phase at
208 ꢀC during the first heating. The SmX phase of 3 was found to be
a highly-ordered smectic phase by POM and X-ray diffraction (XRD)
measurements. Upon subsequent cooling and heating, 3 showed no
crystallization, but SmXeI and IeSmX transitions. The crystalline
phase of 3 was only obtained by recrystallization from the solution,
while it did not form from the melt by cooling.
The UVevis absorption spectra of the monomers 1, 2, and 3 are
shown in Fig. 3. The absorption maxima and the molar extinction
coefficient at 365 nm of the monomers are shown in Table 2. Since it
has been reported that the 4-methoxy-p-terphenyl shows an ab-
sorption maximum at 285 nm (in cyclohexane),40 the red shifts of
the spectra observed in Fig. 3 were considered to be due to the
substituent effect of the polar groups at the 400-position of the ter-
phenyl. For comparison, the structure and the absorption spectrum
of 4-(4-(hexyloxy)benzoyloxy)phenyl 4-(6-(cinnamoyloxy)hex-
yloxy)benzoate (8), which was used for the photodimerization with
MK in the previous study,31 are also shown in Fig. 3. While 8 had no
absorption at 365 nm, 1, 2, and 3 with terphenyl chromophores
showed certain absorptions in the long wavelength region. This
result clearly showed that the terphenyl mesogens are origins of the
longer absorptions and they can be selectively excited by 365-nm
UV light. It is then expected that the mesogens would be able to
sensitize the cinnamoyl group through a triplet energy transfer.
3
1
8
2
UV irr. time
UV irr. time
(h)
UV irr. time
(h)
0
0.5
1
2
6
12
(h)
8
0
0.5
1
2
6
12
0
0.5
1
2
6
12
250
300
350
400
450
500
1
Wavelength / nm
2
3
10 11 12 13 14 15 10 11 12 13 14 15
10 11 12 13 14 15
Elution time / min
Fig. 3. UVevis absorption spectra of 1, 2, 3, and 8 (for comparison) in dichloromethane
solutions (c¼0.03 mM).
Elution time / min
Elution time / min
Fig. 4. GPC charts of 1, 2, and 3 obtained by UV (365 nm) irradiation in the crystalline
phases at room temperature for various periods of time.
2.2. UV irradiation in the crystalline phases
2.2.1. Photoproduct analyses of 1, 2, and 3. It has been reported that
the aromatic nitro compounds behave as an inter- and intra-
molecular triplet sensitizer for the photodimerization of a cinna-
mate group attached to polymers as a side chain in the LC or
In order to elucidate the regioselectivity of the cyclobutane units
resulting from the UV irradiation of the crystals of 1 and 2, their 1H
NMR measurements were performed. Fig. 5 shows the 1H NMR
spectra of the monomer of 1 and the sample irradiated by UV light

5516
H. Kihara et al. / Tetrahedron 68 (2012) 5513e5521
the precursors should suggest that the p-terphenyl mesogens play
a role as a triplet sensitizer.
ε-truxillate
Ph
Ph
α-truxillate
2.2.3. Crystallographic analysis of the single crystal of 1. To clarify
the topochemistry during the photodimerization in the crystalline
state, X-ray crystallography was performed for the single crystal of
the monomer 1. The single crystal of 1 for this experiment was
COOR
Ph COOR
ROOC
ROOC
Ph
(b)
obtained by the recrystallization from
a mixed solution of
dichloromethane and hexane. The obtained structure is shown in
Fig. 6. (CCDC-844056 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.cam.a-
c.uk/data_request/cif.) The cinnamoyl group exhibited a head-to-
tail fashion to the adjacent one, and moreover, the olefinic double
bonds are very close and almost overlapped parallel to each other
(a)
ꢀ
with the distance between them of 4.06 A. This intermolecular
δ / ppm
distance meets the necessary condition for solid-state photo-
cycloaddition6e9 and totally coincident with the fact that the HT
cyclobutanes were exclusively obtained by the UV irradiation of 1 in
the crystalline phase. From the crystal structure, the p-terphenyls
are in an anti-parallel configuration, probably due to the strong
dipoleedipole interaction of the highly polarized mesogen induced
by the cyano group. Therefore, the mesogen should eventually
promote the self-assembled structure of the neighboring molecules
that is preferable for the HT dimerization in comparison to the HH
one. The crystal packing determined by the X-ray crystallography of
the single crystal of 1 seemed to be suitable for the exclusive for-
Fig. 5. 1H NMR spectra of 1: (a) before and (b) after UV (365 nm) irradiation in the
crystalline phase for 2 h. Arrows indicate peaks of the cyclobutane units.
for 2 h in the crystalline phase. The olefinic proton peaks corre-
sponding to the cinnamoyl group (two doublets at 6.45 and 7.69)
decreased, while new peaks associated with the cyclobutane units
(around 3.72, 3.96, and 4.46) appeared after the UV irradiation.
Furthermore, it was confirmed by the 2D (1He1H correlated spec-
troscopy) NMR spectrum (see Fig. S9 in Supplementary data) that
the peaks at 3.96 and 4.46 ppm had a coupling and the multiplet
peaks at around 3.72 ppm consisted of two sets of coupling peaks.
Because no coalescence of these peaks was observed in high tem-
mation of
cycloaddition. However the other
a
-truxillate, that is, syn-HT isomer by the photo-
truxillate, that is, anti-HT
-3
isomer was actually formed to give a mixture of HT dimers, as in-
dicated by the NMR measurement. This is probably due to poly-
morphism in the thin-layer crystal samples used for the
photodimerization or the local movement of the molecule during
the photoreaction process. Note that the cyclobutane regio-isomers
were occasionally obtained by the photochemical [2þ2] cycload-
dition of styryl compounds even in the crystalline phase,44,48 as
similar to the present ones.
1
perature H NMR at 100 ꢀC, we concluded that the peaks were
originated from a mixture of two HT isomers. We thus assigned the
peaks at 3.96 and 4.46 ppm to the
a-type cyclobutane, and the
peaks at around 3.72 ppm to the -type cyclobutane, based on the
3
previous literatures.41e45 These parts including two phenyl rings
were known to be truxillates and had HT configurations, as in-
dicated in the inset of Fig. 5. This result is in clear contrast to the fact
that the HH isomers were dominantly obtained upon UV irradiation
of the cinnamoyl-functionalized LC monomers in their LC phases in
our previous study.31,32 A similar result was also obtained for the 1H
NMR measurement of the UV-irradiated crystalline sample of 2,
suggesting an HT configuration of the photodimer. We thus con-
sidered that the p-terphenyl groups behaved as a triplet sensitizer
even in the crystalline state, ensuring the highly regioselective
formation of the HT dimers.
Compound 2 with a nitro group also exclusively gave the HT
dimer as described. Although the suitable single crystals of 2 were
(a)
2.2.2. Phosphorescence spectroscopy of the precursors of 1 and 2. To
investigate the triplet energy of the p-terphenyl mesogens of 1 and
2, phosphorescence spectroscopy was carried out in a frozen glass
under low temperature conditions. In this case, the corresponding
precursors 5 and 6 were used for the phosphorescence experi-
ments. (The structures of 5 and 6 are depicted in Scheme 1.) The
main reason why the precursors were used instead of the mono-
mers 1 and 2 was to take into account the possibility of phospho-
rescence quenching by an intramolecular energy transfer from the
triplet-excited mesogens to the reactive cinnamoyl group during
the measurements. Phosphorescence emissions were detected
from both of the precursors 5 and 6 in methylcyclohexane matrices
at 77 K (see Fig. S10 in Supplementary data). The triplet energies of
5 and 6 were estimated to be 245 and 232 kJ molꢁ1, respectively,
from the onsets of the corresponding phosphorescence bands
based on Kasha’s rule.46 These values are sufficiently higher than
that of the typical cinnamate compound, i.e., methyl cinnamate
(229 kJ molꢁ1).47 Although the phosphorescence spectroscopy
measurements of 1 and 2 in the crystalline state are not yet avail-
able, the present result obtained from the solution spectroscopy of
c
a
(b)
4.06 Å
4.06 Å
Fig. 6. Crystal structure of 1 obtained from X-ray crystallographic analysis: (a) unit cell
viewed along b-axis and (b) one of the HT arranged (based on the cinnamoyl group)
molecule pairs in the crystal.

H. Kihara et al. / Tetrahedron 68 (2012) 5513e5521
5517
not obtained in the present study, we assume that a similar ge-
ometry in the crystal induced by the dipoleedipole interaction
should be possible for compound 2, thus leading the regioselective
dimerization reaction.
UV irr. time
(min)
UV irr. time
(min)
UV irr. time
(min)
0
0
0
1
1
1
5
5
5
10
20
30
10
20
30
10
20
30
Based on these experimental results, we have concluded the
following (Fig. 7). The UV (365 nm) irradiation of 1 or 2 in their
crystalline states induced an excitation of the p-terphenyl meso-
gens and then the excited mesogens sensitized the cinnamoyl
group through an intramolecular (or intermolecular) triplet energy
transfer in the crystals. The triplet-excited cinnamoyl moieties
underwent the intermolecular cycloaddition reaction with the
neighboring one. Since the photodimerization regioselectively took
place, the resulting cyclobutane units exclusively had the HT con-
figuration based on the initial crystal packing. As we already de-
scribed, there has been a number of reports on the aromatic nitro
compounds as a triplet sensitizer for the photodimerization of
a cinnamate group in amorphous and LC polymers.11,16e20 To the
best of our knowledge, the present study is the first example, which
reveals that aromatic nitro and cyano compounds serve as triplet
sensitizers for the cinnamoyl group even in the highly-ordered
crystalline phases.
1
2
3
10 11 12 13 14 15 10 11 12 13 14 15
10 11 12 13 14 15
Elution time / min
Elution time / min
Elution time / min
Fig. 8. GPC charts of the UV (365 nm) irradiated samples of 1, 2, and 3 in the LC phases
at 170, 165, and 190 ꢀC, respectively, for various times.
observed upon the UV irradiation of 3, similar to the crystalline
phase described in the previous section.
1
The H NMR spectrum of 1 irradiated by the UV light at 170 ꢀC
for 30 min is depicted in Fig. 9. The peaks at 3.47 and 3.76 ppm
showed coupling, corresponding to a d-type cyclobutane as the HH
isomer (truxinate, the structure was depicted in the inset of Fig. 9).
The 1H NMR spectrum of the UV-irradiated sample of 2 also
revealed a similar NMR pattern to 1, indicating the formation of the
HH dimer. These results for the HH regioselectivity are the same as
those from our previous study on the photodimerization of the
cinnamoyl-functionalized LC monomers with MK.31,32
2. Triplet energy transfer
1. UV (365 nm) irradiation
C
O
O(CH2)6O
R
R: CN, NO2
Ph
COOR
3. Cycloaddition
O(CH2)6O
COOR
Ph
or
COOR
Ph
Ph
Ph
Ph
COOR
R
C
O
C
O
O(CH2)6O
R
δ-truxinate (enantiomer)
Fig. 7. Reaction mechanism for the formation of the HT dimers derived from 1 or 2
through triplet energy transfer from the photo-excited mesogens in the crystalline
phases.
We also confirmed that the photodimers derived from 1 and 2
showed thermotropic LC phases based on both the POM and DSC
measurements. It should be thus emphasized that the p-terphenyls
still behave as mesogens to stabilize the LC state even after the
photochemical formation of the bulky moiety in the molecular
structure, that is, the four-membered truxillate resulting from the
photocycloaddition of the cinnamoyl group. The detailed nature of
the LC phases of the photodimers derived from 1 and 2 will be
discussed in a later section.
δ / ppm
Fig. 9. 1H NMR spectra of the UV (365 nm) irradiated sample of 1 in the LC phase at
170 ꢀC for 30 min. Arrows indicate peaks of the cyclobutane unit.
We also carried out the photoirradiation of the present mono-
mers in their LC phases with MK as an auxiliary additive (2 wt %) to
examine the effect of MK with the highly efficient triplet energy
transfer ability on the photoreaction. The GPC charts of the pho-
toirradiation experiments for 1 and 2 in the presence of MK are
depicted in Fig. 10. As expected, the photodimer of 1 was efficiently
and selectively obtained, resulting in the significant suppression of
the generation of any undesirable byproduct. In contrast, the
photodimerization of 2 under MK conditions proceeded with a low
efficiency and the formation of an undesirable byproduct rather
increased when compared to the condition without MK. Although
the reason for the negative effect of MK for 2 is unclear, it can be
considered that the generation of the byproduct from the photo-
reaction of 2 in the LC phase at high temperature should mainly be
caused by the nitro group on the terphenyl part, which may induce
hydrogen abstraction under the specific photochemical conditions
as already reported.49
2.3. UV irradiation in the LC phases
As mentioned above, the HT photodimers were selectively
obtained by the UV irradiation of monomers 1 and 2 in the crys-
talline phases. In contrast, our previous studies revealed that HH
photodimers were mainly formed by UV irradiation in their LC
phases with MK.31,32 To discuss the difference in the regioselectivity
depending on the phase variation, monomers 1, 2, and 3 were ir-
radiated with UV light (365 nm) in their LC phases (at 170 ꢀC for 1,
at 165 ꢀC for 2, and at 190 ꢀC for 3). The formation of the corre-
sponding dimers was investigated by GPC measurements. The re-
sults are shown in Fig. 8. As a result, while the dimers of 1 and 2
actually appeared after the UV irradiation, a small amount of un-
desirable products with relatively high-molecular weights were
also generated in both cases, and shown to be the broad shoulders
of the peaks of the dimers. The formation of a byproduct was also

5518
H. Kihara et al. / Tetrahedron 68 (2012) 5513e5521
simultaneously observed in LC phases that appeared below the
nematic phases of both dimers, therefore, the LC phases were
identified as smectic A.50 While the isotropization temperatures (Ti)
of both dimers were higher than that of monomer 1, the degree of Ti
increase was greater for the HT dimer than that for the HH dimer. In
contrast, the temperature range of the smectic A phase of the HH
dimer was higher and wider than that of the HT dimer. It was already
reported that when the linear and U-shaped LC dimers were com-
pared based on the same mesogens, the former often shows a more
thermally stable LC phase than the latter, whereas the latter tends to
exhibit smectic phases compared to the former.29,30 Therefore, our
present results are similar to previous observations.
UV irr. time
(min)
UV irr. time
(min)
0
0
1
1
5
5
10
20
30
10
20
30
1
2
10
15
11
12
13
14
15
10
11
12
13
14
Elution time / min
Elution time / min
Fig. 10. GPC charts of the UV (365 nm) irradiated samples of 1 and 2 using MK in the
Table 3
LC phases at 170 and 165 ꢀC, respectively, for various times.
Transition temperatures and enthalpy changes for the HT and HH photodimers
derived from 1 during the first cooling and the second heating scans
Photodimer
HT
Phase transition temperaturea (ꢀC) (
D
H (kJ molꢁ1))
Although UVevis absorption spectra of 1, 2, and 8 in the solution
state are already shown in Fig. 3, we also measured absorption
spectra of the compounds in the condensed (crystalline and LC)
phases in order to demonstrate that the triplet energy transfer
takes place in those phases. As shown in Fig. S11 in Supplementary
data, all the condensed-phase samples of 1 and 2 had the absorp-
tion band including the wavelength of 365 nm, whereas 8 did not
show any absorption band at 365 nm. These results suggest that the
absorption at 365 nm of 1 and 2 originates from the terphenyl
groups but not from the cinnamoyl group even in the condensed
phases. MK, which is widely-recognized as a triplet sensitizer,
caused the dimerization of the cinnamoyl-functionalized LC com-
pounds under the condition of 365-nm irradiation in the con-
densed phase, and the same results were obtained when both 1 and
2 were irradiated with 365-nm light in the condensed phases
without MK. Therefore, it is likely that the triplet energy transfer
from the photo-excited mesogens to the cinnamoyl group of 1 and
2 took place in both the crystalline and LC phases.
We confirmed that the HH photodimer was regioselectively
obtained by the UV (365 nm) irradiation of 1 in the LC phase with or
without MK. This HH regioselectivity is in sharp contrast to that of
the photodimerization in the crystalline phase. The reason for the
drastic change in the regioselectivity is considered that the pho-
todimerization in the crystalline phase was based on top-
ochemistry whereas that in the LC phase was governed by
thermodynamics.
Cr 183 (36.1) N 208 (1.89) I
I 206 (1.93) N 183 (0.13) SmA 142 (33.2) Cr
Cr 138 (10.5) 187 SmA (1.68) N 198 (1.60) I
I 196 (1.85) N 185 (1.85) SmA 135 (10.1) Cr
HH
a
Cr¼crystalline phase; SmA¼smectic A phase; N¼nematic phase; I¼isotropic
phase.
To examine the geometric differences among the monomer, HH
and HT dimers, XRD measurements were performed for 1 and both
dimers in their smectic A phases. All the obtained XRD profiles
were found to consist of one sharp peak in a small angle region
corresponding to the smectic layer distance and a broad halo in the
wide angle region associated with the lateral packing of the
mesogens. Table 4 shows smectic layer distances (d) estimated from
the sharp peaks in the XRD profiles and the molecular lengths of
the all-trans conformation of the compounds (L) determined by
molecular modeling based on molecular mechanics. First, as for
monomer 1, d was longer than L, which indicated that in the
smectic layer, the two molecules became paired with their cyano-p-
terphenyl mesogens interdigitated with each other. It was also
suggested that the distance of the spatial overlapping of the mol-
ecules was almost the same as the length of the mesogen (about
1.5 nm), as shown in Fig. 11(a). It is well known that LC compounds
having a cyanobiphenyl or cyano-p-terphenyl mesogen often show
the LC structure with pairing of the mesogens in an anti-parallel
manner due to the strong dipole-dipole interaction.51e53 Next, as
shown in Table 4, even after the photodimerization, the d value of
the HH dimer was similar to that of 1. This result indicates that the
HH dimer adopted the U-shaped conformation (L was estimated to
be 3.08 nm by molecular modeling) and had an interdigitated
structure in the smectic A phase, as shown in Fig. 11(b). In-
terestingly, the d of the HT dimer was found to be eventually
smaller than that of 1 in contrast to the HH dimer. This result can be
explained as follows. The HT dimer adopted the linear conforma-
tion (L was estimated to be 5.25 nm) in the smectic A phase.
Therefore, the HT dimer had an intercalated structure with the two
mesogens that overlapped at each molecular end, as shown in
Fig. 11(c). The intercalation would be the main reason for the
2.4. LC properties of the photodimers of 1 and 2
From a comparison between the LC monomers and dimers, we
observed that the photodimerization significantly changes the LC
properties of the initial monomers. As for monomer 2, the photo-
dimer was obtained only in the crystalline phase, and it adopted the
HTconfiguration for the cyclobutane unit (linear dimer) as drawn in
Fig. 1(a). The HT dimer of 2 (irradiation time: 12 h) exhibited a ne-
matic phase from 201 to 190 ꢀC and then a smectic A phase to 180 ꢀC
upon cooling. The isotropicenematic phase transition temperature
significantly increased compared to the monomer 2 by 32 ꢀC. Note
that the increase of the phase transition temperature is quite rea-
sonable for the linear dimer formation.21 As already described, both
of the photodimers having HT and HH cyclobutane units were se-
lectively obtained in high conversion by the UV irradiation of 1.
Therefore, we examined the LC properties of the HT and HH dimers
of 1. As already described, the UV irradiation of 1 in the crystalline
phase for 12 h selectively gave the HT dimer with the linear dimer
structure, while the similar UV irradiation of 1 but in the LC phase
with MK for 30 min exclusively provided the HH dimer with the U-
shaped geometry as illustrated in Fig. 1(b). The phase transition
temperatures of the dimers determined by DSC and POM are listed
in Table 3. Focal-conic fans and homeotropic regions could be
Table 4
Smectic layer distances (d) and molecular lengths (L) for the monomer 1, and the HT
and HH photodimers
Compound
da (nm)
L
b (nm)
1
4.65
4.53
3.95
3.13
3.08
5.25
HH dimer
HT dimer
a
Determined by XRD in the smectic A phase.
Estimated by molecular modeling of fully extended chains.
b

H. Kihara et al. / Tetrahedron 68 (2012) 5513e5521
5519
(b)
3.08 nm
(a)
3.13 nm
COO(CH ) O
COO(CH ) O
CN
CN
COO(CH ) O
CN
NC
NC
O(CH ) OOC
NC
O(CH ) OOC
O(CH ) OOC
1.5 nm
4.65 nm
4.53 nm
3.95 nm
(c)
5.25 nm
NC
O(CH ) OOC
COO(CH ) O
NC
CN
NC
O(CH ) OOC
COO(CH ) O
CN
O(CH ) OOC
COO(CH ) O
CN
NC
O(CH ) OOC
COO(CH ) O
CN
Fig. 11. Schematic representation for self-assembled structures of the smectic A phases of (a) 1, (b) HH dimer, and (c) HT dimer derived from 1, estimated from XRD and molecular
modeling.
decreasing d value. It is thus noteworthy that we successfully
obtained both the linear and U-shaped LC dimers from the same LC
monomer by a regioselective photoreaction based on the precise
control of the reaction conditions.
regioselective photodimerization of LC cinnamoyl compounds
should be a productive method to selectively obtain the linear and
U-shaped LC dimers from a variety of LC monomers. A related study
on the tuning of the LC properties by this method is now in
progress.
3. Conclusions
4. Experimental section
4.1. General methods
In this study, three cinnamoyl-functionalized 400-substituted-p-
terphenyl LC compounds were synthesized and their photochem-
ical properties were carefully examined under several different
reaction conditions. As a result, it was successfully found that the
regioselective photodimerization of the LC compounds using the
400-substituted p-terphenyls as unprecedented dual functional
groups for the mesogens and the triplet sensitizers. Among the
cyano- (1), nitro- (2), and N,N-dimethylamino- (3) substituted LC
compounds, only 1 and 2 with electron-withdrawing groups un-
derwent photodimerization with the appropriate reaction condi-
tions. The photodimers with an HT cyclobutane configuration were
regioselectively obtained in high yield from 1 and 2 by UV light
(365 nm) irradiation in the crystalline phases, revealing that the p-
terphenyl moiety acted as a triplet sensitizer. Phosphorescence
spectroscopy of the precursors of 1 and 2 suggested that the triplet
state energies of the cyano- and nitro-substituted p-terphenyls are
high enough to transfer energy to the cinnamoyl moiety. X-ray
crystallography of the single crystal of 1 showed that the crystal
packing between the double bonds was suitable for the solid-state
photocycloaddition of the cinnamoyl group, supporting the HT di-
mer formation. In contrast, a photodimer with an HH cyclobutane
unit was exclusively produced by the UV irradiation of 1 in the LC
phase with or without Michler’s ketone (MK) as the auxiliary triplet
sensitizer, based on the thermodynamic stability of the in-
termediate of the HH dimer in the molten state at high tempera-
ture. The HT and HH photodimers from 1 and 2 showed thermally
stable nematic and a smectic A phases, despite the existence of the
bulky cyclobutane unit. Especially in case of 1, the XRD analyses of
the photodimers in their smectic A phases proved that the HH and
HT dimers adopted the interdigitated and intercalated structures,
respectively. We consider that the present results for the
The 1H, 13C, and 2D (1He1H correlated spectroscopy) NMR
spectra were recorded by a Varian Gemini 300BB (300 MHz) or
a JEOL JNM-ECX400 (400 MHz) using TMS as the internal stan-
dard. Temperature dependent 1H NMR was performed using the
JEOL JNM-ECX400 and the experimental temperature range was
from 25 to 100 ꢀC (solvent: toluene-d8). The UVevis absorption
spectra in the solutions were obtained using a Jasco V-570
spectrometer. The UVevis absorption spectra of the thin-layer
crystalline and LC samples of 1, 2, and 8 were measured by an
Otsuka electronics MCPD-7000 using a transmittance mode. The
samples were prepared between two cover glasses. The mea-
surements of the crystalline samples were performed at room
temperature. The temperatures of the LC samples (170 ꢀC for 1,
165 ꢀC for 2, 150 ꢀC for 8) were controlled by a Linkam LK-600PH
during the measurements. The phosphorescence spectra were
recorded by a Jasco FP-777 spectrofluorometer using a PMA-281
unit to delay detection of the emission light. Solutions for the
phosphorescence spectroscopy were deoxygenated by bubbling
with nitrogen for 5 min prior to the measurements. IR spectra
were recorded on a Jasco FT/IR-680 Plus. Gel permeation chro-
matography (GPC) was conducted using a Shimadzu system (an
LC-10ADvp pump unit, an SPD-10Avp UV detector (detection
wavelength: 320 nm for 1, 350 nm for 2, 330 nm for 3), a CTO-
10Avp column oven, and an SCL-10Avp controller) with Shodex
KF800D and KF805L columns, using THF as the eluent. The
thermal properties of the samples were evaluated by polarized
optical microscopy (POM) using an Olympus BH2 equipped with
a
Mettler FP82HT hot stage and by differential scanning

5520
H. Kihara et al. / Tetrahedron 68 (2012) 5513e5521
calorimetry (DSC) using a Seiko Instruments, Inc., DSC 6200. The
heating and cooling rates were 5 ꢀC minꢁ1. The powder X-ray
diffraction patterns of the molten samples were measured at the
LC temperatures using a Rigaku diffractometer (type 4037) with
an imaging plate (R-Axis IV) or a Rigaku Rint2000. The single
crystal X-ray diffraction measurement was performed at 183 K
using a Bruker CCD area-detector diffractometer (Smart APEXII)
with CHCl3 (50 ml). The combined CHCl3 was dried over MgSO4,
filtered, and evaporated under reduced pressure. The residue was
purified by recrystallization from CHCl3 to give 1 (0.14 g, 56%) as
white crystals. IR nmax (neat, cmꢁ1): 2945, 2869, 2225, 1698, 1636,
1598, 1489, 1469, 1175, 979, 811, 773. 1H NMR (300 MHz, CDCl3)
d,
ppm: 7.73 (s, 4ArH; H-3,4), 7.69 (d, J¼16.1 Hz,1 olefinic H; H-d), 7.67
(d, J¼9.1 Hz, 2ArH; H-5 or 6), 7.64 (d, J¼9.1 Hz, 2ArH; H-5 or 6), 7.56
(d, J¼8.8 Hz, 2ArH; H-2), 7.53e7.51 (m, 2ArH; H-c), 7.39e7.37 (m,
3ArH; H-a,b), 6.99 (d, J¼8.8 Hz, 2ArH; H-1), 6.45 (d, J¼16.1 Hz, 1
olefinic H; H-e), 4.24 (t, J¼6.6 Hz, 2H; COOCH2), 4.04 (t, J¼6.4 Hz,
2H; ArOCH2), 1.88e1.74 (m, 4H; COOCH2CH2CH2CH2CH2CH2O),
1.56e1.51 (m, 4H, COOCH2CH2CH2CH2CH2CH2O). 13C NMR
with Mo-Ka radiation. The single crystal of 1 suitable for the X-
ray diffraction measurements was obtained from the mixed so-
lution of dichloromethane and hexane. Matrix-assisted laser
desorption ionization time-of-flight mass spectrometry (MALDI-
TOF MS) was conducted using an Applied Biosystems Voyager-DE
Pro instrument operated in the reflector mode with 2,5-
dihydroxybenzoic acid as the matrix. Molecular modelings were
performed using a Chem3D Ultra 8.0. The MM2 calculation and
the geometry were fully optimized.
(100 MHz, CDCl3) d, ppm: 167.17, 159.12, 145.32, 144.75, 141.31,
137.31, 134.53, 132.72, 132.55, 130.33, 128.97, 128.16, 127.64, 127.57,
127.37, 119.08, 118.31, 115.02, 110.86, 68.02, 64.64, 29.28, 28.81,
25.92. MS (MALDI-TOF) C34H31NO3 [MþH]þ: 502.2; found: 501.7.
Anal. Calcd for C34H31NO3: C; 81.41, H; 6.23, N; 2.79%. Found: C;
81.43, H; 6.17, N; 2.71%.
4.2. Synthesis
4.2.1. 4-Bromo-40-(6-bromohexyloxy)biphenyl (4). A mixture of 4-
bromo-40-hydroxybiphenyl (3.0 g, 12 mmol), 1,6-dibromohexane
(15 g, 60 mmol), K2CO3 (2.5 g), and acetone (50 ml) was refluxed
with stirring for 6 h. After cooling, the mixture was filtered, and the
filtrate was evaporated under reduced pressure. The residue was
purified by silica gel column chromatography (eluent: hexane/ethyl
acetate¼10:1 gradient to 5:1) and recrystallization from hexane to
give 4 (3.7 g, 75%) as white crystals. Mp 192 ꢀC. IR nmax (neat, cmꢁ1):
2939, 2863, 1604, 1473, 1245, 995, 809. 1H NMR (300 MHz, CDCl3)
4.2.4. 4-(6-Bromohexyloxy)-400-nitro-p-terphenyl (6). From the re-
action of 4-nitrophenyl-boronic acid pinacol ester (1.0 g, 4.0 mmol)
and 4 (1.5 g, 3.7 mmol), 6 was obtained as yellow crystals using the
same procedure as that for 5. Yield 0.81 g (52%). Mp 153 ꢀC (ne-
matic 208 ꢀC isotropic). IR nmax (neat, cmꢁ1): 2942, 2867,1739,1592,
1505, 1473, 1333, 1245, 997, 819. 1H NMR (300 MHz, CDCl3)
d, ppm:
8.31 (d, J¼8.9 Hz, 2ArH; H-6), 7.78 (d, J¼8.9 Hz, 2ArH; H-5), 7.69 (s,
4ArH; H-3,4), 7.58 (d, J¼8.8 Hz, 2ArH; H-2), 7.00 (d, J¼8.8 Hz, 2ArH;
H-1), 4.03 (t, J¼6.4 Hz, 2H; ArOCH2), 3,44 (t, J¼7.3 Hz, 2H; BrCH2),
1.94e1.82 (m, 4H; BrCH2CH2CH2CH2CH2CH2O), 1.56e1.51(m, 4H,
d
, ppm: 7.52 (d, J¼8.6 Hz, 2ArH; H-4), 7.47 (d, J¼8.8 Hz, 2ArH; H-2),
7.40 (d, J¼8.6 Hz, 2ArH; H-3), 6.95 (d, J¼8.8 Hz, 2ArH; H-1), 4.00
(t, J¼6.4 Hz, 2H; OCH2), 3.43 (t, J¼6.8 Hz, 2H; BrCH2), 1.93e1.80
(m, 4H; BrCH2CH2CH2CH2CH2CH2O), 1.55e1.50 (m, 4H;
BrCH2CH2CH2CH2CH2CH2O). 13C NMR (100 MHz, CDCl3)
d, ppm:
159.01, 147.13, 146.93, 141.42, 136.76, 132.36, 128.06, 127.69, 127.47,
127.23, 124.13, 114.89, 67.83, 33.81, 32.67, 29.08, 27.92, 25.31. MS
(MALDI-TOF) C24H24BrNO3 [MþH]þ: 454.1; found: 453.6.
BrCH2CH2CH2CH2CH2CH2O). 13C NMR (100 MHz, CDCl3)
d, ppm:
158.82, 139.69, 132.28, 131.72, 128.21, 127.90, 120.68, 114.82, 67.78,
33.80, 32.66, 29.07, 27.90, 25.29. MS (MALDI-TOF) C18H20Br2O
[MþH]þ: 411.0; found: 410.2.
4.2.5. 4-(6-Cinnamoyloxyhexyloxy)-400-nitro-p-terphenyl (2). From
the reaction of E-cinnamic acid (0.25 g, 1.7 mmol) and 6 (0.70 g,
1.5 mmol), 2 was obtained as yellow crystals by the same procedure
used for 1. Yield 0.57 g (71%). IR nmax (neat, cmꢁ1): 2950, 2868, 1698,
1634, 1594, 1513, 1473, 1339, 1176, 977, 815, 755. 1H NMR (300 MHz,
4.2.2. 4-(6-Bromohexyloxy)-400-cyano-p-terphenyl (5). To a mixture
of 4 (1.6 g, 4.0 mmol), Pd(PPh3)4 (0.46 g, 0.40 mmol), a 10% aqueous
solution of Na2CO3 (10 ml), and THF (10 ml) was added THF (10 ml)
containing the 4-cyano-phenyl-boronic acid pinacol ester (1.0 g,
4.4 mmol). The mixture was refluxed with stirring for 18 h. After
cooling, the mixture was poured into water, and extracted three
times with CHCl3 (50 ml). The combined CHCl3 was dried over
MgSO4, filtered, and evaporated under reduced pressure. The res-
idue was purified by silica gel column chromatography (eluent:
hexane/CHCl3¼1:1) and recrystallization from a mixture of toluene
and ethanol to afford 5 (1.1 g, 56%) as white crystals. Mp 136 ꢀC
(nematic 236 ꢀC isotropic). IR nmax (neat, cmꢁ1): 2939, 2864, 2234,
CDCl3)
d
, ppm: 8.31 (d, J¼8.8 Hz, 2ArH; H-6), 7.78 (d, J¼8.8 Hz, 2ArH;
H-5), 7.69 (d, J¼16.0 Hz, 1 olefinic H; H-d), 7.68 (s, 4ArH; H-3,4), 7.57
(d, J¼8.7 Hz, 2ArH; H-2), 7.55e7.51 (m, 2ArH; H-c), 7.39e7.37 (m,
3ArH; H-a,b), 7.00 (d, J¼8.7 Hz, 2ArH; H-1), 6.45 (d, J¼16.0 Hz, 1
olefinic H; H-e), 4.24 (t, J¼6.6 Hz, 2H; COOCH2), 4.03 (t, J¼6.4 Hz, 2H;
ArOCH2), 1.88e1.75 (m, 4H; COOCH2CH2CH2CH2CH2CH2O),1.56e1.51
(m, 4H; COOCH2CH2CH2CH2CH2CH2O). 13C NMR (100 MHz, CDCl3)
d,
ppm: 167.16, 159.18, 147.27, 147.07, 144.75, 141.57, 136.88, 134.53,
132.46, 130.34, 128.97, 128.18, 128.15,127.81, 127.59, 127.39, 124.26,
118.31, 115.04, 68.03, 64.64, 29.28, 28.81, 25.92. MS (MALDI-TOF)
C34H31NO3 [MþH]þ: 522.2; found: 521.7. Anal. Calcd for C34H31NO3:
C; 75.99, H; 5.99, N; 2.69%. Found: C; 75.79, H; 5.91, N; 2.60%.
1738, 1602, 1473, 1244, 995, 812. 1H NMR (300 MHz, CDCl3)
d, ppm:
7.73 (s, 4ArH; H-3,4). 7.68 (d, J¼9.0 Hz, 2ArH; H-5 or 6), 7.64 (d,
J¼9.0 Hz, 2ArH; H-5 or 6), 7.57 (d, J¼8.7 Hz, 2ArH; H-2), 6.99 (d,
J¼8.7 Hz, 2ArH; H-1), 4.02 (t, J¼6.4 Hz, 2H; OCH2), 3.44 (t, J¼6.8 Hz,
2H; BrCH2), 1.94e1.81 (m, 4H; BrCH2CH2CH2CH2CH2CH2O),
1.56e1.51 (m, 4H, BrCH2CH2CH2CH2CH2CH2O). 13C NMR (100 MHz,
4.2.6. 4-Bromo-40-(6-cinnamoyloxyhexyloxy)biphenyl (7). From the
reaction of E-cinnamic acid (0.79 g, 5.3 mmol) and 4 (2.0 g,
4.9 mmol), 7 was obtained as white crystals by the same pro-
cedure used for 1. Yield 1.6 g (69%). Mp 71 ꢀC. IR nmax (neat, cmꢁ1):
2939, 2863, 1709, 1635, 1604, 1473, 1245, 995, 810. 1H NMR
CDCl3) d, ppm: 158.95, 145.17, 141.15, 137.20, 132.59, 132.45, 128.04,
127.51, 127.44, 127.24, 118.96, 114.87, 110.72, 67.82, 33.81, 32.67,
29.09, 27.92, 25.31. MS (MALDI-TOF) C25H24BrNO [MþH]þ: 433.5;
found: 434.1.
(300 MHz, CDCl3)
d
, ppm: 7.68 (d, J¼16.0 Hz, 1 olefinic H; H-d),
7.54e7.51 (m, 2ArH; H-c), 7.52 (d, J¼8.6 Hz, 2ArH; H-4), 7.46 (d,
J¼8.8 Hz, 2ArH; H-2), 7.40 (d, J¼8.6 Hz, 2ArH; H-3), 7.39e7.36 (m,
3ArH; H-a,b), 6.96 (d, J¼8.8 Hz, 2ArH; H-1), 6.44 (d, J¼16.0 Hz, 1
olefinic H; H-e), 4.23 (t, J¼6.6 Hz, 2H; COOCH2), 4.01 (t, J¼6.4 Hz,
2H; ArOCH2), 1.86e1.71 (m, 4H; COOCH2CH2CH2CH2CH2CH2O),
1.55e1.48 (m, 4H; COOCH2CH2CH2CH2CH2CH2O). 13C NMR
4.2.3. 4-(6-Cinnamoyloxyhexyloxy)-400-cyano-p-terphenyl
(1). E-
Cinnamic acid (94 mg, 0.63 mmol) and DBU (0.10 g, 0.67 mmol)
were dissolved in dry dimethyl sulfoxide (DMSO; 15 ml). After
stirring at 50 ꢀC for 30 min, 5 (0.22 g, 0.51 mmol) was added and
the mixture was left to stir for 3 h at 50 ꢀC. After cooling, the
mixture was poured into water (150 ml) and extracted three times
(100 MHz, CDCl3)
d, ppm: 167.05, 158.86, 144.62, 139.72, 134.39,
132.27, 131.73, 130.21, 128.84, 128.03, 127.90, 118.16, 114.85, 67.87,

H. Kihara et al. / Tetrahedron 68 (2012) 5513e5521
5521
64.51, 29.14, 28.67, 25.78. MS (MALDI-TOF) C27H27BrO3 [MþNa]þ:
References and notes
501.1; found: 501.4.
1. Ichimura, K.; Akita, Y.; Akiyama, H.; Kudo, K.; Hayashi, Y. Macromolecules 1997,
30, 903e911.
2. Hayashi, Y.; Suzuki, S.; Ichimura, K. J. Mater. Chem. 2001, 11, 1805e1811.
3. Nagata, A.; Sakaguchi, T.; Ichihashi, T.; Miya, M.; Ohta, K. Macromol. Rapid
Commun. 1997, 18, 191e196.
4. Chung, C.-M.; Roh, Y.-S.; Cho, S.-Y.; Kim, J.-G. Chem. Mater. 2004, 16, 3982e3984.
5. Ding, J.; Zhuang, X.; Xiao, C.; Cheng, Y.; Zhao, L.; He, C.; Tang, Z.; Chen, X. J.
Mater. Chem. 2011, 21, 11383e11391.
6. Cohen, M. D.; Schmidt, G. M. J.; Sonntag, F. I. J. Chem. Soc. 1964, 2000e2013.
7. Schmidt, G. M. J. Pure Appl. Chem. 1971, 27, 647e678.
8. Ramamurthy, V.; Venkatesan, K. Chem. Rev. 1987, 87, 433e481.
9. Abdelmoty, I.; Buchholz, V.; Di, L.; Guo, C. Y.; Kowitz, K.; Enkelmann, V.;
Wegner, G.; Foxman, B. M. Cryst. Growth Des. 2005, 5, 2210e2217.
10. Huang, C. H.; Bassani, D. M. Eur. J. Org. Chem. 2005, 4041e4050.
11. Robertson, E. M.; Van Deusen, W. P.;Minsk,L. M. J. Appl. Polym. Sci.1958, 2, 308e311.
12. Curme, H. G.; Natale, C. C.; Kelley, D. J. J. Phys. Chem. 1967, 71, 767e770.
13. Furumi, S.; Otomo, A.; Yokoyama, S.; Mashiko, S. Polymer 2009, 50, 2944e2952.
14. Furumi, S.; Ichimura, K. Phys. Chem. Chem. Phys. 2011, 13, 4919e4927.
15. Tsuda, M. J. Polym. Sci., Part A 1964, 2, 2907e2916.
4.2.7. 4-(6-Cinnamoyloxyhexyloxy)-400-N,N-dimethyamino-p-ter-
phenyl (3). 4-N,N-Dimethylamino-phenyl-boronic acid (0.30 g,
1.8 mmol), 7 (0.80 g, 1.7 mmol), and Pd(PPh3)4 (64 mg, 0.055 mmol)
were dissolved in a mixture of dimethoxyethane (DME; 30 ml) and
EtOH (15 ml). A 10 wt % aqNa2CO3 (10 ml)solutionwas added and the
mixture was refluxed with stirring for 10 h. After cooling, the mixture
was poured into water (100 ml) and the precipitate was collected by
filtration. The crude product was purified by silica gel column chro-
matography (eluent: CH2Cl2) and recrystallization from EtOH to af-
ford 3 (0.44 g, 51%) as pale yellow crystals. IR nmax (neat, cmꢁ1): 2939,
2865, 2796,1706,1636,1606,1541,1494,1470, 1177, 807, 767. 1H NMR
(300 MHz, CDCl3)
d
, ppm: 7.69 (d, J¼16.0 Hz,1 olefinic H; H-d). 7.62 (d,
J¼8.8 Hz, 2ArH; H-3 or 4), 7.58 (d, J¼8.8 Hz, 2ArH; H-3 or 4), 7.54e7.51
(m, 2ArH; H-c), 7.54 (d, J¼8.8 Hz, 4ArH; H-2 and 5), 7.39e7.37 (m,
3ArH; H-a,b), 6.97 (d, J¼8.8 Hz, 2H; H-1), 6.82 (d, J¼8.8 Hz, 2H; H-6),
6.45 (d, J¼16.0 Hz, 1 olefinic H; H-e), 4.23 (t, J¼6.6 Hz, 2H;
COOCH2), 4.02 (t, J¼6.4 Hz, 2H; ArOCH2), 3.00 (s, 6H; N(CH3)2),
1.87e1.74 (m, 4H; COOCH2CH2CH2CH2CH2CH2O), 1.55e1.50 (m, 4H;
16. Tsuda, M. Bull. Chem. Soc. Jpn. 1969, 42, 905e908.
17. Kawatsuki, N.; Tachibana, T.; Kamada, K. Adv. Mater. 2005, 17, 1886e1890.
18. Nishikubo, T.; Iizawa, T.; Yamada, M.; Tsuchiya, K. J. Polym. Sci., Part A 1983, 21,
2025e2045.
19. Nishikubo, T.;Iizawa, T.;Takahashi, E.; Nono, F. Macromolecules 1985,18, 2131e2135.
20. Kawatsuki, N.; Kato, K.; Shiraku, T.; Tachibana, T.; Ono, H. Macromolecules 2006,
39, 3245e3251.
21. Imrie, C. T.; Luckhurst, G. R. In Handbook of Liquid Crystals; Demus, D., Goodby, J.,
Gray, G. W., Spiess, H.-W., Vill, V., Eds.; Wiley-VCH: Weinheim, New York, NY,
1998; pp 801e833.
COOCH2CH2CH2CH2CH2CH2O). 13C NMR (100 MHz, CDCl3)
d, ppm:
167.08, 158.51, 149.95, 144.63, 139.49, 138.43, 134.44, 133.34, 130.22,
128.87, 128.76, 128.06, 127.88, 127.54, 126.90, 126.50, 118.23, 114.78,
112.80, 67.88, 64.57, 40.61, 29.22, 28.72, 25.84. MS (MALDI-TOF)
C35H37NO3 [MþH]þ: 520.3; found: 519.9. Anal. Calcd for C35H37NO3:
C; 80.89, H; 7.18, N; 2.70%. Found: C; 80.96, H; 7.17, N; 2.60%.
22. Emsley, J. W.; Luckhurst, G. R.; Shilstone, G. N.; Sage, I. Mol. Cryst. Liq. Cryst.
1984, 102, 223e233.
23. Attard, G. S.; Date, R. W.; Imrie, C. T.; Luckhurst, G. R.; Roskilly, S. J.; Seddon, J.
M.; Taylor, L. Liq. Cryst. 1994, 16, 529e581.
24. Imrie, C. T.; Henderson, P. A. Curr. Opin. Colloid Interface Sci. 2002, 7, 298e311.
25. Imrie, C. T.; Henderson, P. A. Chem. Soc. Rev. 2007, 36, 2096e2124.
26. Yoshizawa, A.; Chiba, S.; Ogasawara, F. Liq. Cryst. 2007, 34, 373e379.
27. Yoshizawa, A.; Kasai, H.; Ogasawara, F.; Nagashima, Y.; Kawaguchi, T. Liq. Cryst.
2007, 34, 547e553.
4.3. Photoirradiation in the crystalline phases
As for 1 and 2, 2 mg of the monomer was placed between cover
glasses (Matsunami Glass Ind., Ltd., Japan) and heated to the iso-
tropization temperature using the Linkam hot stage. The molten
sample was then cooled to room temperature at the rate of
5 ꢀC minꢁ1 and a thin-layer crystalline sample was obtained. The
resulting sample was irradiated by monochromatic 365-nm UV
light (5 mW cmꢁ2) from a 500-W high-pressure mercury lamp
through optical filters at room temperature. After completion of the
UV irradiation, the cover glasses were carefully separated and the
irradiated sample was analyzed by several techniques. As for the
monomer 3, the powdery sample obtained by recrystallization
from the EtOH solution was used for the UV irradiation in the
crystalline phase. The sample (2 mg) between two cover glasses
was spread as flat as possible and irradiated with UV light using the
same procedure for 1 and 2, except for not melting.
28. Yoshizawa, A.; Uehara, N.; Kurauchi, M.; Yamaguchi, A. Liq. Cryst. 2007, 34,
1121e1128.
29. Yoshizawa, A.; Kawaguchi, T. Liq. Cryst. 2007, 34, 177e181.
30. Yoshizawa, A.; Kinbara, H.; Narumi, T.; Yamaguchi, A.; Dewa, H. Liq. Cryst. 2005,
32, 1175e1181.
31. Kihara, H.; Tamaoki, N. Liq. Cryst. 2007, 34, 1337e1347.
32. Kihara, H.; Tamaoki, N. Macromol. Chem. Phys. 2008, 209, 1424e1431.
33. Egerton, P. L.; Hyde, E. M.; Trigg, J.; Payne, A.; Beynon, P.; Mijovic, M. V.; Reiser,
A. J. Am. Chem. Soc. 1981, 103, 3859e3863.
34. Smith, G. W. Mol. Cryst. Liq. Cryst. 1978, 49, 27e31.
35. Kiryanov, A. A.; Sampson, P.; Seed, A. J. J. Mater. Chem. 2001, 11, 3068e3077.
36. Cai, X. C.; Sakamoto, M.; Hara, M.; Tojo, S.; Kawai, K.; Endo, M.; Fujitsuka, M.;
Majima, T. J. Phys. Chem. A 2004, 108, 9361e9364.
37. Matsen, F. A. J. Am. Chem. Soc. 1950, 72, 5243e5248.
38. Lim, E. C. Spectrochim. Acta 1963, 19, 1967e1978.
39. Takeda, H.; Sakurai, Y.; Takenaka, S.; Miyake, H.; Doi, T.; Kusabayashi, S.; Takagi,
T. J. Chem. Soc., Faraday Trans. 1990, 86, 3429e3435.
40. Lees, W. A.; Burawoy, A. Tetrahedron 1963, 19, 419e438.
41. Benefrai, Da; Green, B. S. Tetrahedron 1974, 30, 2357e2364.
42. Green, B. S.; Heller, L. J. Org. Chem. 1974, 39, 196e201.
43. Lewis, F. D.; Quillen, S. L.; Hale, P. D.; Oxman, J. D. J. Am. Chem. Soc. 1988, 110,
1261e1267.
44. Koivukorpi, J.; Kolehmainen, E. J. Mol. Struct. 2009, 930, 116e120.
45. Hesse, M.; Meier, H.; Zeeh, B. In Spectroscopic Methods in Organic Chemistry, 2nd
Japanese ed.; Nomura, M., Ed.; Trans., Kagakudoujin: Kyoto, 2010; p 114.
46. Kasha, M. Discuss. Faraday Soc. 1950, 9, 14e19.
47. Montalti, M.; Murov, S. L. Handbook of Photochemistry, 3rd ed.; CRC/Taylor &
Francis: Boca Raton, FL, 2006.
48. Vedernikov, A.; Kuz’mina, L.; Sazonov, S.; Lobova, N.; Loginov, P.; Churakov, A.;
Strelenko, Y.; Howard, J.; Alfimov, M.; Gromov, S. Russ. Chem. Bull. 2007, 56,
1860e1883.
Acknowledgements
We thank Ms. Midori Goto of IBEC in AIST for the X-ray crys-
tallographic analyses. We also thank Mr. Toshi Aoi for the assistance
of temperature dependent NMR measurement and Dr. Masaru
Aoyagi for FTIR measurement. This work was supported by the
fundamental research fund of AIST.
49. Turro, N. J. Modern Molecular Photochemistry; Benjamin/Cummings: Menlo
Park, CA, 1978.
50. Dierking, I. Textures of Liquid Crystals; Wiley-VCH: Weinheim, 2003.
51. Lansac, Y.; Glaser, M. A.; Clark, N. A. Phys. Rev. E 2001, 64.
52. Ghanadzadeh, A.; Beevers, M. S. J. Mol. Liq. 2002, 100, 47e57.
53. Pelaez, J.; Wilson, M. Phys. Chem. Chem. Phys. 2007, 9, 2968e2975.
Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2012.04.083.