Reflection colour changes in cholesteric liquid crystals after the addition and photochemical isomerization of mesogenic azobenzenes tethered to sugar alcohols

Article abstract of DOI:10.1039/b900890j

We controlled reflection colour of the glass-forming cholesteric liquid crystals over the whole visible region using the smectic liquid crystalline photochromic dopants on the basis of the different mechanism from the conventional helical twisting power c

Full text of DOI:10.1039/b900890j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Reflection colour changes in cholesteric liquid crystals after the addition and
photochemical isomerization of mesogenic azobenzenes tethered to sugar
alcohols†
a
Haruhisa Akiyama, Asuka Tanaka, Hideo Hiramatsu, Jun’ichi Nagasawa and Nobuyuki Tamaoki‡
b
b
a
a
*
*
Received 15th January 2009, Accepted 19th May 2009
First published as an Advance Article on the web 1st July 2009
DOI: 10.1039/b900890j
We controlled reflection colour of the glass-forming cholesteric liquid crystals over the whole visible
region using the smectic liquid crystalline photochromic dopants on the basis of the different
mechanism from the conventional helical twisting power change, which gave rise to full-colour
recording and display materials at a lower additive concentration than that reported previously. We
simply tethered multiple mesogenic azobenzene units to sugar alcohols to obtain effective dopants in
which the intramolecular mesogenic moieties could adopt smectic-like alignments. The new dopants
that featured linear sugar alcohol backbones exhibited increasingly stable smectic phases upon
increasing the number of mesogenic groups in the molecule; a corresponding cyclic sugar alcohol
derivative exhibited no such smectic phase. Addition of these smectic liquid crystalline dopants to the
glass-forming cholesteric liquid crystalline material induced increasingly larger pitch shifts upon
increasing the number of intramolecular mesogenic groups in the dopant. Dopants prepared from
stereoisomeric sugar alcohol backbones provided similar liquid crystalline phases and induced similar
pitch shifts after their addition to the cholesteric liquid crystals. The colour images recorded on the thin
layer of the cholesteric liquid crystalline mixture were stabilized by fixing the cholesteric alignment into
ꢀ
glassy states, which withstood heating at 70 C.
dopant when the ChLCs are composed of mixtures of nematic
liquid crystals and small amounts of chiral compounds.8 This
means that the addition of a chiral dopant enforces a twisting of
the cholesteric alignment. The proportionality constant in the
relationship between the concentration and the helical pitch is
referred to as the helical twisting power. Because this value
depends on the relative content of the liquid crystal and dopant
molecules, the change in chemical structure of a chiral photo-
responsive dopant in response to light produces a change in its
helical twisting power before and after irradiation, resulting in
a change in the reflection colour of the ChLC. Another method
to control the reflection colour is the addition of an achiral
photoresponsive dopant into the ChLC.4,6,7 Because the
concentrations and chemical structure of the chiral compounds
before and after irradiation are unchanged in this case, it is
impossible to explain the phenomenon in terms of changes to the
chiral twisting power. Indeed, the mechanism was not clear until
recently. In 2003, we found that for achiral dopants added to
glass-forming ChLCs, a relationship exists between the elonga-
tion of pitch and the increased contribution of cybotactic SmA
clusters in the ChLC phase.9 A cybotactic smectic cluster is
a dynamic and fluctuating smectic-like domain observed in
nematic phases.10 The presence of cybotactic domains can be
confirmed by the presence of layer structures in X-ray diffraction
patterns. Such clusters are observed in the transition from
nematic to smectic phases:10 upon cooling from a nematic state,
the contribution of the smectic cluster gradually increases with
an increase in the cholesteric pitch prior to the transition to the
smectic phase finally occurring. For this reason, the addition of
1. Introduction
Cholesteric liquid crystals (ChLCs) find practical application in
colour information technology.1 The formation of the coloured
state results from Bragg reflection of circularly polarized light
from the periodically twisted structure in the ChLCs.2 The
biggest advantage of this method is the ability to realize rewrit-
able colour displays because the cholesteric colour is tunable
through reversible changes in the helical pitch of the twisted
structure. It is known that the photochromic reactions of a small
amount of photochromic compounds can reversibly control the
properties of a large amount of the liquid crystals such as phase
transitions and alignment directions.3 Similarly, one well-estab-
lished approach toward reversible control over the helical pitch
employs photochemical reactions of photoresponsive additives
doped into the ChLCs.4–7 In general, the helical pitch shortens
proportionally upon increasing the concentration of a chiral
aNanotechnology Research Institute, National Institute of Advanced
Industrial Science and Technology (AIST), Central 5, 1-1-1 Higashi,
Tsukuba, Ibaraki, 305-8565, Japan. E-mail: h.akiyama@aist.go.jp; Fax:
+81-298-61-4669; Tel: +81-298-61-4418
bCollege of Industrial Technology, Nihon University, 1-2-1 Izumi-cho,
Narashino, Chiba, 275-8575, Japan
† Electronic
supplementary
information
(ESI)
available:
Characterization data; MALDI-TOF-MS and absorption spectra. See
DOI: 10.1039/b900890j
‡ Present address: Research Institute for Electronic Science, Hokkaido
University, N20, W10, Kita-ku, Sapporo, 001-0020, Japan.
E-mail: tamaoki@es.hokudai.ac.jp; Fax: +81-11-706-9357; Tel:
+81-11-706-9356.
5956 | J. Mater. Chem., 2009, 19, 5956–5963
This journal is ª The Royal Society of Chemistry 2009

View Article Online
a dopant that can stabilize the cybotactic smectic domain in
a cholesteric phase should enlarge the cholesteric pitch. We have
used glass-forming ChLCs having the dimesogenic structure
CD8 (Scheme 1) for full-colour recording. The LC alignment of
CD8 can be quickly rearranged at the cholesteric temperature
(115–80 ꢀC). Indeed, we can control the reflection colour in this
LC temperature range within a few seconds and then fix the
image by quickly cooling the sample into the glass state. Several
kinds of photochromic additives for CD8 have been used for
photochemical tuning of the cholesteric colour.5p,6,7 One of the
most important requisites for these photochromic additives is for
one of the isomers to have the ability to induce the cybotactic
smectic domain in the host cholesteric liquid crystal, while the
other isomer exhibits no such (or little) ability. Another requisite
is that the glassy state of the mixture must exhibit thermal
stability. Among the dopants tested, we found that achiral azo-
benzene oligomers with a side-chain liquid crystal structure
possess superior properties relative to other additives in that they
have both high thermal stability and the ability to induce a large
pitch shift at low doping concentrations.7 We deduced that the
large pitch shift arose as a result of the molecular structure
stabilizing the cybotactic smectic cluster of CD8 because intra-
molecular mesogenic side chains of the oligomers can produce
a layer-like (smectic-like) structure even in a molecularly
dispersed state; they can stretch on both sides of main chain and
align in a parallel manner. In fact, these dopants themselves
exhibit an SmA phase with a layer structure in which the layer
distance matches that of the cybotactic SmA of CD8. The other
advantage of using oligomeric dopants is the good thermal
stability of the colour images. Although addition of a low-
molecular-weight compound to vitreous materials often
decreases the glass transition temperature, CD8 doped with the
oligomeric dopants maintains its glassy state up to the same glass
transition temperature as that of pure CD8. We have found,
however, that although the oligomeric dopants possess good
properties for use with ChLCs, sample preparation can be
problematic for two reasons: (i) the oligomer is generally
obtained in low yield because a chain transfer reagent, which
decreases the synthetic yield, is used for the free radical poly-
merization to reduce the molecular weight of the polymer;
(ii) strict and reproducible control over the molecular weight and
dispersity are impossible in free radial polymerizations. Because
the physical properties of low-molecular-weight polymers are
strongly influenced by their molecular weight, the second
problem is serious if oligomeric compounds are to be used
widely. In this study, we synthesized new dopants featuring
a single component molecule that can mimic the behaviour of
azobenzene side chain oligomers. Sugar alcohols present multiple
hydroxyl groups from a carbon chain backbone; their structures
can be considered as an oligomer of hydroxymethylene units. In
addition, because sugar alcohols are naturally derived, some of
them are commercially available at low cost. To realize the bunch
structure of multiple mesogenic side chains, we tethered
azobenzene-based mesogenic moieties to several kinds of sugar
alcohols. Sugar-based liquid crystalline materials prepared
through substitution of some of the hydroxyl groups in the sugar
Scheme 1 Chemical structures of the mesogenic compounds, the sugar alcohol backbones, and the cholesteric liquid crystal.
This journal is ª The Royal Society of Chemistry 2009 J. Mater. Chem., 2009, 19, 5956–5963 | 5957

View Article Online
with linear hydrophobic groups are well established; in these
materials, the liquid crystallinity originates through phase sepa-
ration of the hydrophobic and hydrophilic parts.11 In this study,
we used a sugar alcohol merely as a connector of multiple
mesogens in the accurately controllable manner. We investigated
the liquid crystalline properties of the resultant compounds and
their properties as dopants for CD8. While several kinds of
monodisperse main-chain liquid crystals12 or liquid crystalline
multipedes including dendritic molecules13 have been reported,
studies on monodisperse side-chain liquid crystalline oligomers
with linear main chains are few. We systematically investigated
the number effects of mesogenic side chains of the monodisperse
side-chain liquid crystalline oligomers from a monomer to an
octamer.
investigate the effect of the stereochemical structure of the
backbone on the cholesteric pitch. We employed D-mannitol and
inositol, which contain six hydroxyl groups each, but linear and
cyclic structures, respectively, to compare the effects of the
topological structure of the dopant on the cholesteric pitch.
Because sugar alcohols presenting eight hydroxyl groups are not
commercially available, we synthesized the xylitol dimer as an
analogue of an octamer using the method presented in Scheme 2.
We performed esterification reactions of the acid chloride
derivative R–Cl (m ¼ 10) to functionalize the hydroxyl groups of
the sugar alcohols with the mesogenic azobenzene units (Scheme
1). In addition, we synthesized another mannitol derivative
through a reaction with the azobenzene acid chloride R–Cl (m ¼
5) to investigate the effect of the molecular chain length. Not all
of the hydroxyl groups were esterified by the azobenzene deriv-
atives at room temperature, indicating that severe steric
hindrance existed around the hydroxyl groups of the sugar
alcohols. Nevertheless, reactions performed at higher tempera-
tures led to esterification of all of the hydroxyl groups with
azobenzene moieties (see Experimental).
2. Results and discussion
2.1 Synthesis and liquid crystalline properties of azobenzene
mesogens appended to sugar alcohols
To synthesize monomer, dimer, tetramer, hexamer, and octamer
analogues, we used methanol, ethylene glycol, L-threitol, meso-
erythritol, D-mannitol, scyllo-inositol, and xylitol dimer as
backbones. We used the stereoisomers L-threitol and meso-
erythritol, which possess four hydroxyl groups each, to
Table 1 lists the phase transition temperatures of these
compounds, as estimated using dynamic scanning calorimetry
(DSC) and polarizing optical microscopy (POM). The mono-
meric and dimeric azobenzene compounds, AzMe and AzEg,
respectively, do not exhibit any mesophases. We observed
Scheme 2 Synthetic method for the preparation of the xylitol dimer.
Table 1 Physical characteristics of azobenzene compounds with alcohol backbones
Alcohol
Cooling rate/ꢀC minꢁ1
Transition temperature/ꢀC
Layer distance /nm
AzMe
AzEg
AzTh
Methanol
Ethylene glycol
L-Threitol
Iso 65 Cr (Cr 72 Iso)
Iso 104 Cr
Iso 98 SmA 87 SmB 79 SmE
Iso 100 SmA 96 Cr
Iso 100 SmA 77 SmB 70 SmE
Iso 101 SmA 89 Cr
Iso 107 SmA 84 SmB
Iso 109 SmA 86 SmB
Iso 162 Cr (Cr 171 Iso)
Iso 113 SmA 85 SmB (SmB 87 SmA
114 Iso)
—
—
2
5
2
5
2
5
2
2
2
3.14 (SmB)
3.14 (SmA)
3.13 (SmB)
3.05 (SmA)
3.11 (SmB)
—
AzEr
meso-Erythritol
AzMn
D-Mannitol
AzIna
AzXyD
scyllo-Inositol
Xylitol dimer
3.18 (SmA), 3.08 (SmB)
AzMnC5
D-Mannitol
2
Iso 118 SmA 59 SmB (SmB 61 SmA
121 Iso)
4.73 (SmA), 4.71 (SmB)
a
Very small peaks were observed below melting points.
5958 | J. Mater. Chem., 2009, 19, 5956–5963
This journal is ª The Royal Society of Chemistry 2009

View Article Online
monotropic SmA, SmB, and SmE phases for the tetrasubstituted
compounds AzTh and AzEr, but the corresponding peaks in their
DSC curves were broad, partially overlapping, and changed
shape depending on the cooling rate. When we maintained these
samples at any temperature at which the materials exhibited
SmA phases, we observed (POM) a transition to a higher-order
smectic or crystal phase after 10–30 min. These gradual transi-
tions at constant temperatures indicate that these SmA phases
were thermodynamically unstable. These two tetrasubstituted
compounds exhibited similar phase behaviour regardless of the
configurations of their sugar backbones. Whereas one of our
hexasubstituted compounds (AzMnC5) exhibited SmA and SmB
phases on both heating and cooling, the other (AzMn) exhibited
monotropic SmA and SmB phases only on cooling, but the two
peaks corresponding to each transition were clearly separated
and unaffected by the cooling rate. We observed thermody-
namically stable (enantiotropic) SmA and SmB phases for the
octasubstituted compound AzXyD. Taken together, the phase
behaviour of these materials indicates a tendency for increased
stabilization of the smectic phases upon increasing the number of
azobenzene side chains. We suspect that the larger of these
compounds can form structures similar to that of side chain-type
liquid crystalline polymers, i.e., parallel alignments of extended
side chains on both sides of a backbone; such a parallel alignment
can be stabilized through lateral interactions between pairs of
mesogenic side chains and is enhanced upon increasing the
number of side chains. The layer lengths of the smectic phases of
the series of mesogenic compounds derived from R–Cl (m ¼ 10)
were 3.0–3.2 nm for both the SmA and SmB phases. In its fully
extended trans conformation, the monomeric compound AzMe
has a length of ca. 3.2 nm, which is almost equal to the smectic
layer spacing. The smectic layer distance of 4.5 nm for AzMnC5
does not match the molecular length (2.7 nm) of the model
compound, R–OMe (m ¼ 5), indicating the existence of an
interdigitated layer structure. All of our compounds possessing
four or more azobenzene moieties exhibited liquid crystalline
phases, except for the hexasubstituted cyclic compound AzIn,
which did not exhibit any mesophases. Because AzIn possesses
six azobenzene units isotropically bonded to an inositol back-
bone in a two-dimensional surface, it is unlikely that a layer
structure or a smectic phase would form. Thus, a linear backbone
presenting several mesogenic side groups is necessary to produce
the liquid crystalline material exhibiting smectic phases.
Although AzTh, AzMn, and AzMnC5 possess chiral backbones,
we did not observe chiral mesophases originating from their
molecular chirality.
Fig. 1 Reflection colour shifts (Dl/nm) of the films of CD8 containing 2
wt% of the dopants, quenched from 95 ꢀC. The values were estimated as
averages from three–five measurements. Error bars represent standard
deviations.
exhibits SmA phases, induced a large reflection colour shift. The
values of the shifts observed after the addition of the tetrasub-
stituted compounds AzTh and AzEr were almost identical, but
slightly smaller than those observed after the addition of AzMn
and AzXyD. The addition of AzEg and AzMe, which individually
exhibit no mesophases, induced small and very small shifts,
respectively. The addition of AzIn, which also exhibits no
mesophases, brought about no change in the reflection colour of
CD8, presumably because of the immiscibility between CD8 and
AzIn, which we ascertained by observing a phase separated
structure under POM. Taken together, these findings support
our hypotheses that a large cholesteric pitch shift is induced after
the addition of smectic compounds that have the potential to
stabilize smectic clusters in the cholesteric phase. In terms of their
stable smectic properties, we would predict that the pitch shift
induced by AzXyD, with its greater number of side chains, should
be superior to that of AzMn; in contrast, the pitch shifts induced
by these two compounds were quite similar. To understand this
behaviour, we should take into account the fact that AzXyD has
a larger molecular weight than AzMn and, therefore, it was
present at a lower molar concentration under conditions of
a constant weight concentration. Therefore, we deduce that the
negative effect that a lower concentration has on the pitch shift is
balanced out by the positive effect of increasing the number of
side chains, but only up to the hexamer level. The addition of
AzMnC5, with its shorter alkyl chain length, induced much less
of a colour change than did AzMn. A similar distinct dependence
of the alkyl chain length has been observed previously for other
dopants;6a,7,9 it reflects the phenomenon that large pitch shifts can
be induced only when the molecular length of the additive
matches the layer distance of the smectic cluster in the ChLC.
Indeed, the layer distances of the smectic phases of AzTh, AzEr,
AzMn, and AzXyD (3.0–3.1 nm; see Table 1) as single compo-
nents are very close to the layer length of the cybotactic
smectic cluster in CD8 (2.8 nm).9 Therefore, we must select
molecules of appropriate length for use as dopants. Our
investigation of a series of monomeric compounds revealed,
however, that not only is the molecular length important in
causing a large pitch shift, but so too is the chemical structur-
e. Among previously reported6,7,9 monomeric dopants for
CD8, we observed the largest pitch shift after the addition of
4,40-bis(nonyloxy)azobenzene (AzOC9). The pitch shift was
2.2 Effect of additives on cholesteric pitch
We estimated the reflection band shifts of the glass-forming
cholesteric liquid crystal CD8 containing 2 wt% of each of our
new compounds from the change in the reflection band position
in the absorption spectra recorded after quenching the chole-
steric alignment in these samples through rapid cooling from
ꢀ
a liquid crystalline temperature (95 C). Fig. 1 plots the differ-
ences in the reflection maximum wavelengths of CD8 in the
presence and absence of the additives; it reveals the abilities of
the dopants to induce the cholesteric pitch shift. The addition of
AzTh, AzEr, AzMn, or AzXyD, each of which individually
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 5956–5963 | 5959

View Article Online
smaller than those induced by AzTh and AzEr, but larger than
that induced by AzEg. In contrast, the only monomeric analogue
in our series of new compounds, AzMe, exhibited quite a small
shift, suggesting that the structure of the mesogenic side chain
moieties of our new dopants was not optimum. Conversely, it is
a proof that the bundling of multiple mesogenic side chains is
a useful method to obtain effective dopants—even for molecules
featuring inappropriately designed mesogenic units. Fig. 2 shows
X-ray diffraction (XRDꢀ) patterns for CD8 containing 2 wt% of
AzMn and AzMe at 95 C, where the intensity is normalized at
the maximum of the broad peaks around 17ꢀ. In both cases, the
peaks corresponding to the layer distance of 2.8 nm are observed
at 3.2ꢀ, and are considered to be derived from the cybotactic
smectic A clusters. The peak of CD8/AzMn is higher and sharper
than that of CD8/AzMe. For more detailed comparison, the
intensity and sharpness (FWHM; full width at half maximum) of
the XRD peaks of the several cholesteric mixtures are listed in
Table 2. The CD8 mixtures with AzMn, AzEr, AzEg and AzMe,
which induced large pitch shifts in that order, exhibited the sharp
and strong XRD peaks in the same order. The correspondence
relation supports the hypothesis that enhancement of the cybo-
tactic clusters upon addition of the dopants causes the cholesteric
pitch shifts. Fig. 3 presents UV–Vis absorption spectra of the
mixtures of CD8 and each dopant in the forms of their cholesteric
glassy films. The absorption peaks at ca. 350 nm and at longer
wavelengths correspond to the p–p* transition of the trans-
azobenzene moieties and to the reflection band, respectively. The
expanded regions around the p–p* absorption band of the
azobenzene moieties, presented in the inset to Fig. 3, reveals that
the spectral shapes of the CD8/AzMe and CD8/AzEg systems
were the same. The spectra of CD8/AzMn and CD8/AzXyD also
had the same shapes, but they differed from those of CD8/AzMe
and CD8/AzEg in that their absorption maxima appeared at
shorter wavelength. The spectral shapes of CD8/AzTh and
Fig. 3 UV–Vis absorption spectra of films of CD8 containing 2 wt% of the
dopants, quenched from 95 ꢀC; inset: expanded region of selected spectra.
CD8/AzEr were also identical, with their signals positioned
between those of the other two types of spectra. We conclude that
three kinds of spectra existed, with the absorption maximum
shifting to a shorter wavelength upon increasing the number of
side chains. Because the position of the absorption maximum can
reflect the conditions surrounding chromophores, these spectra
suggest that the units in the dimer and monomer were sur-
rounded by CD8 molecules only; i.e., the two azobenzene
moieties of the dimer were extended to different directions. In the
tetramers, hexamer, and octamer, the azobenzene side chains
existed in a parallel alignment and were extended across both
sides of the linear backbones, interacting with each other even in
the molecularly dispersed media. This concept supports our
hypothesis that such dopants have high abilities to stabilize
smectic clusters and induce large pitch shifts. Therefore, fine-
tuning the number of side chains is an important aspect of
designing such systems. Indeed, these new compounds are the
most effective dopants ever reported for CD8: the reflection
colour shifts induced by AzXyD and AzMn are larger than those
induced by the most effective oligomeric (Lp8Az8) and mono-
meric (AzOC9) compounds.5p,6,7,9
Fig. 2 X-Ray diffraction patterns for CD8 containing 2 wt% of AzMe
(a) and AzMn (b) at 95 ꢀC.
2.3 Photoinduced pitch shift and photoimaging
Table 2 Half-width and relative intensity of small-angle X-ray diffrac-
tion peaks at 3.2ꢀ
Azobenzene is famous as a photochromic compound that
isomerizes reversibly from E to Z upon irradiation with UV light
and from Z to E upon irradiation with visible light. Adding
certain types of azobenzene derivatives as E isomers to CD8
often induces a large pitch shift; photochromic reactions to the
corresponding Z isomers lead the system to almost recover to its
original state.5p,6,7 We investigated the photoresponse of our new
dopants in cholesteric mixtures. Fig. 4a displays the
FWHM/deg
Relative Intensity
CD8/AzMe
CD8/AzEg
CD8/AzEr
CD8/AzMn
5.3
4.9
4.4
3.8
1.00
1.16
1.23
1.41
5960 | J. Mater. Chem., 2009, 19, 5956–5963
This journal is ª The Royal Society of Chemistry 2009

View Article Online
films to estimate the thermal stability of their images. Fig. 5
displays photographs of images recorded in films obtained from
a mixture of CD8 and AzMn before and after heating for 30 min
at various temperatures. The colour image that formed in the
cholesteric glassy film remained unchanged when heating at
temperatures up to 70 ꢀC. We observed similar thermal stabilities
for the images recorded in mixtures of CD8 and AzTh, AzEr, and
AzXyD; these stabilities are comparable to those observed for
mixtures of CD8 and oligomeric dopants. The thermal stabꢀility
of the mixture of CD8 and AzEg was low (upper limit: 40 C),
which can be explained by considering the severe impurity effect
of the low-molecular-weight dopant, which lowered the glass
transition temperature of the matrix.
Fig. 4 a) Photochemical change in the reflection colour shifts (Dl/nm) of
CD8 doped with 2 wt% of AzXyD, AzMn, AzEr, or AzOC9 before and
after UV irradiation (365 nm) at different energies; dark gray bars
represent negative values. b) Reflection colour shifts of the new dopants
and AzOC9 upon irradiation with UV light at 100 mJ cmꢁ2
.
3. Conclusion
Although oligomeric mixtures of achiral dopants are effective at
controlling the cholesteric reflection colour, in this study we
synthesized single-component analogues of these oligomers.
Fig. 6 provides schematic representations of these new dopants
and summarizes the results we obtained. Because we controlled
the number of intramolecular mesogenic side chains, we could
determine the optimal number required to induce the largest
pitch shift. In addition, because we prepared isomers presenting
the same number of mesogenic side chains, we could elucidate the
effects of the configuration and topological structure (linear or
cyclic) of the main chains. We found that using a linear backbone
was necessary to induce a large pitch shift, but differences in the
configuration of the main chain had little effect. The dopants
possessing linear backbones exhibited stable smectic phases that
increased in stability upon increasing the number of side chains;
the ability to induce a pitch shift increased accordingly. These
results support our recently proposed mechanism: that the
enlarged cholesteric pitch induced by an achiral dopant origi-
nates from the stabilizing effect of the dopant on the cybotactic
smectic cluster in the cholesteric phase. Using our newly
synthesized dopants, we recorded full-colour images that
exhibited high thermal stability.
photoinduced shifts of CD8 doped with 2 wt% of AzXyD, AzMn,
AzEr, and AzOC9 before and after UV irradiation (365 nm). The
reflection band shifted to shorter wavelength upon increasing the
irradiation energy; it was accompanied by a decrease in the
absorption at 360 nm arising from E–Z isomerization (ESI†
Fig. S2). The photoinduced shifts in reflection colour were
almost complete after irradiation at 17 mJ cmꢁ2, regardless of the
nature of the dopant. Isomerization still proceeded when irra-
diated at light energies of up to 100 mJ cmꢁ2. Fig. 4b reveals that
the reflection colour shifts induced by our new dopants and by
AzOC9 upon irradiation with UV light at 100 mJ cmꢁ2 were all
slightly negative, except for that induced by AzXyD, which was
slightly positive. Nevertheless, in all cases these values were
almost all equal to zero, which corresponded to a blue reflection
colour. This phenomenon can been explained by considering that
the E isomer of azobenzene can behave as a mesogenic core that
affects the cholesteric pitch, whereas the bent structure of the Z
isomer cannot exist as a mesogenic core that interacts with host
LC molecules. For full-colour recording, the wavelength of the
reflection band should exist in the infrared region (>700 nm)
prior to irradiation to ensure a colourless state. If this condition
is satisfied, irradiation for a shorter period of time can also give
rise to green or red reflections. Therefore, we expected that we
could control the cholesteric colour through photochemical
reactions using some of our new dopants that induce large pitch
shifts. To realize a colourless state for full-colour recording, we
adopted a cholesteric mixture including 3 wt% of each dopant.
We recorded colour images in a 1.8 ꢂ 1.8 cm cell using a mask-
less irradiation device that generated UV light of different
intensity for each 1024 ꢂ 764 pixel. After fixing the recorded
images in the glassy LC through rapid cooling, we heated the
4. Experimental
4.1 Materials
The preparation of CD8 has been reported previously.14
Azobenzene alkyl acids were synthesized using a method similar
to that described in the literature.15
Fig. 5 Colour pictures, prepared using the maskless irradiation system,
recorded in a thin film of CD8 doped with 3 wt% of AzMn. (a) Before
heating; (b, c) after heating for 30 min at b) 70, and c) 80 ꢀC.
Fig. 6 Schematic representations of the new dopants and a summary of
the results.
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 5956–5963 | 5961

View Article Online
2,3:4,5:20,30:40,50-Tetra-O-isopropylidene-1,10-oxybis(1-deoxyxy-
litol). 1,2:3,4-Di-O-isopropylidenexylitol16 (1.27 g) and sodium
hydride (65% suspension in oil, 0.4 g) were dissolved in
dry tetrahydrofuran (THF, 50 mL) at room temperature
7.29 (d, J ¼ 8.6 Hz, 16H, ArH), 7.78 (d, J ¼ 8.2 Hz, 16H, ArH),
7.87 (d, J ¼ 8.8 Hz, 16H, ArH). MALDI-TOF-MS: calcd. for
[M + Na]+ and [M + K]+, 3897.6 and 3913.6, respectively; found
3896.8 and 3913.7, respectively. Elemental analysis. Found: C,
75.20; H, 8.81; N, 5.73. Calcd. for C240H338N16O25: C, 74.92; H,
8.86; N, 5.82%.
and then
a solution of 1:2,3:4-di-O-isopropylidene-5-O-tri-
fluoromethanesulfonylxylitol17 (1.2 eq) in dry THF (25 mL) was
added slowly. The mixture was stirred for 12 h at room temper-
ature. The solution was diluted with EtOAc and washed with
brine. After evaporating the solvent, the crude compound was
purified through column chromatography (SiO2; CH2Cl2/EtOAc,
4:1) to yield a transparent liquid (1.04 g, 42.5%). 1H NMR
(CDCl3, d): 1.39 (s, 6H, CH3), 1.43 (s, 18H, CH3), 3.64–3.70
(m, 4H, CH2O), 3.81–3.93 (m, 4H, CH2), 4.00–4.22 (m, 6H, CH).
MALDI-TOF-MS: m/z calcd. for [M + Na]+ and [M + K]+, 469.24
and 485.21 respectively; found 469.61, 485.58 respectively.
4.2 LC cell preparation
CD8 was mixed at 125 ꢀC with 2 or 3 wt% of the dopant and 5 mm
diameter epoxy resin spheres (<0.1 wt%; Nippon Shokubai). The
molten mixture was placed between two glass slides on a hot
stage (Mettler FP(82) heated at 125 ꢀC. The isotropic phase
transformed into the LC phase upon cooling to 95 ꢀC; the sample
was then dropped into iced water. This rapid cooling technique
froze the LC alignment into the glassy state.
1,10-Oxybis(1-deoxyxylitol). 1:2,3:4-Di-O-isopropylidene-5-O-
trifluoromethanesulfonylxylitol (1.04 g) was suspended in 2 M
HCl (50 mL) and then MeOH was added to form a homogenous
4.3 Physical measurements
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
reflector mode; samples were prepared through mixing MeOH
and CH2Cl2 solutions of the compounds with a-cyano-4-
hydroxycinnamic acid. Electrospray ionization time-of-flight
mass spectrometry (ESI-TOF MS) was conducted using an
ꢀ
solution. The mixture was stirred at 90 C for 6 h and then the
solvent was evaporated under vacuum to yield a sticky liquid
1
(0.67 g). H NMR (CD3OD, d): 3.53–3.70 (m, 10H, CH, CH2),
3.73–3.79 (m, 2H, CH), 3.85–3.93 (m, 2H, CH2). ESI-TOF-MS:
calcd. for [M + Na]+ m/z 309.12; found 309.1.
Esterification. The reaction procedure for the synthesis of
AzXyD is described below as a typical example. The other
compounds were synthesized in a similar manner. The purifica-
tion methods, synthetic yields and physical data of other
compounds, AzMe, AzEg, AzEr, AzTh, AzMn, AzMnC5 and
AzIn, are in the ESI.†
1
Applied Biosystems Mariner instrument. H NMR spectra of
samples dissolved in CDCl3 were recorded using a Varian
Gemini-300 BB spectrometer. Absorption spectra were measured
using a Hewlett–Packard UV–Vis spectrophotometer (Agilent
84(53). DSC measurements were conducted using a Seiko
DSC Exstar 6000 instrument operated at heating rates of 5 and
2
ꢀC/min. Photoimaging and UV irradiation were performed
AzXyD. Thionyl chloride (1.2 mL) was added to a solution of
11-[4-(4-hexylphenylazo)phenoxy]undecanoic acid (1.00 g,
2.14 mmol) in dry CH2Cl2 (3.5 mL). After stirring the solution
for 1 h, the solvents were evaporated under vacuum. The red
residue was dissolved in CH2Cl2 (6 mL) and then this solution
was added dropwise to a solution of the xylitol dimer (0.051 g,
0.17 mmol) in pyridine (4 mL). After heating under reflux for 24
h under a dry nitrogen atmosphere, the solution was diluted with
CH2Cl2 and washed several times with water. The organic phase
was dried (MgSO4) and then the solvent was evaporated. The
residue was dissolved in a small amount of CH2Cl2, placed in
a flask and sealed with a plug. The flask was heated at 60 ꢀC for
12 h to ensure complete thermal isomerization of all the azo-
benzene units to their E isomers. After evaporating the solvent,
the residue was purified twice by column chromatography
through SiO2, eluting once with CH2Cl2/hexane (2:1) and then
with CH2Cl2 alone under red light. The hepta, hexa, and lower-
order substituted compounds generated as byproducts were
thoroughly removed. The product was obtained as an orange
solid (0.2 g, 30%). 1H NMR (CDCl3, d): 0.88 (t, J ¼ 6.3 Hz, 24H,
CH3), 1.30 (s, 124H, CH2), 1.44 (br s, 16H, CH2C2H4OAr), 1.61
(m, 32H, CH2CH2Ar, CH2CH2COO), 1.78 (q, J ¼ 7.0 Hz, 16H,
CH2CH2OAr), 2.31 (br s, 17H, CH2COO), 2.65 (t, J ¼ 7.6
Hz,16H, CH2Ar), 3.40–3.60 (m, 4H, H-1a, H-1b, H-10a, H-10b),
3.99 (t, J ¼ 6.2 Hz, 18H, OCH2, H-5a, H-50a), 4.28–4.40 (m, 2H,
H-5b, H-50b), 5.10–5.20 (m, 2H, H-2, H-20), 5.25–5.36 (m, 2H,
H-4), 5.38–5.50 (m, 2H, H-3), 6.94 (d, J ¼ 8.8 Hz, 16H, ArH),
using a maskless UV exposure instrument (Interwave) and
a modified DLP projector system that generated pixelated
365 nm light (1024 ꢂ 764 dots). The intensity of the light at each
dot was modulated using a digital micromirror device; the image
data were provided by a JPEG file from a personal computer.
Small-angle X-ray diffractometry was conducted as reported
earlier;7 the X-ray exposure time was 15 min.
References
1 N. Tamaoki, Adv. Mater., 2001, 13, 1135, and references cited
therein.
2 (a) P. G. de Gennes, J. Prost, The Physics of Liquid Crystals,
Clarendon Press, Oxford, UK, 1993, Section 6.1.
p 219; (b)
D. Demus, J. Goodby, G. W. Gray, H.-W. Spiess, V. Vill,
Handbook of Liquid Crystals, Vol. 2A, Wiley-VCH, Weinheim,
1998, (Ch. IV 2.2.1), p 335.
3 (a) K. Ichimura, Chem. Rev., 2000, 100, 1847, and references cited
therein; (b) T. Ikeda, J. Mater. Chem., 2003, 13, 2037, and
references cited therein; (c) C. K. Prasad, G. G. Nair and G. Hegde,
Adv. Mater., 2005, 17, 2086.
4 (a) E. Sackmann, J. Am. Chem. Soc., 1971, 93, 7088; (b) S. Z. Janicki
and G. B. Schuster, J. Am. Chem. Soc., 1995, 117, 8524.
5 (a) B. L. Feringa, N. P. M. Huck and H. A. van Dorenl, J. Am. Chem.
Soc., 1995, 117, 9929; (b) R. A. van Delden, M. B. van Gelder,
N. P. M. Huck and B. L. Feringa, Adv. Funct. Mater., 2003, 13,
319; (c) A. Yu. Bobrovsky, N. I. Boiko and V. P. Shibaev, Adv.
Mater., 1999, 11, 1025; (d) P. van de Witte, E. E. Neuteboom,
M. Brehmer and J. Lub, J. Appl. Phys, 1999, 85, 7517; (e)
T. Sagisaka and Y. Yokoyama, Bull. Chem. Soc. Jpn., 2000, 73,
191; (f) H.-K. Lee, K. Doi, H. Harada, O. Tsutsumi, A. Kanazawa,
5962 | J. Mater. Chem., 2009, 19, 5956–5963
This journal is ª The Royal Society of Chemistry 2009

View Article Online
T. Shiono and T. Ikeda, J. Phys. Chem., 2000, 104, 7023; (g)
C. Ruslim and K. Ichimura, Adv. Mater., 2001, 13, 37; (h) J. Lub,
W. P. M. Nijssen, R. T. Wegh, J. P. A. Vogel and A. Ferrer, Adv.
Funct. Mater., 2005, 15, 1961; (i) T. Yoshioka, T. Ogata,
T. Nonaka, M. Moritsugu, S.-N. Kim and S. Kurihara, Adv.
Mater., 2005, 17, 1226; (j) T. Mitsuoka, H. Sato, J. Yoshida,
A. Yamagishi and Y. Einaga, Chem. Mater., 2006, 18, 3442; (k)
U. A. Hrozhyk, S. V. Serak, N. V. Tabiryan and T. J. Bunning,
Adv. Mater., 2007, 19, 3244; (l) Q. Li, L. Green, N. Venkataraman,
I. Shiyanovskaya, A. Khan, A. Urbas and J. W. Doane, J. Am.
Chem. Soc., 2007, 129, 12908; (m) M. Mathews and N. Tamaoki,
J. Am. Chem. Soc., 2008, 130, 11409; (n) S. Pieraccini, S. Masiero,
G. P. Spada and G. Gottarelli, Chem. Commun., 2003, 598–599; (o)
S. Pieraccini, G. Gottarelli, R. Labruto, S. Masiero, O. Pandoli and
G. P. Spada, Chem. Eur. J., 2004, 10, 5632; (p) N. Tamaoki,
Y. Aoki, M. Moriyama and M. Kidowaki, Chem. Mater., 2003, 15,
719.
6 (a) N. Tamaoki, S. Song, M. Moriyama and H. Matsuda, Adv.
Mater., 2000, 12, 94; (b) M. Moriyama, S. Song, H. Matsuda and
N. Tamaoki, J. Mater. Chem., 2001, 11, 1003; (c) R. Davis,
V. A. Mallia, S. Das and N. Tamaoki, Adv. Funct. Mater., 2004, 14,
743.
7 H. Akiyama, V. A. Mallia and N. Tamaoki, Adv. Funct. Mater., 2006,
16, 477.
8 P. G. de Gennes, J. Prost, The Physics of Liquid Crystals, Clarendon
Press, Oxford, UK, 1993, Section 6.2. p. 239.
9 M. Kidowaki, M. Moriyama, M. Wada and N. Tamaoki, J. Phys.
Chem. B, 2003, 107, 12054.
10 D. Demus, J. Goodby, G. W. Gray, H.-W. Spiess, V. Vill, Handbook
of Liquid Crystals, Vol. 1, Wiley-VCH, Weinheim, 1998, (Ch. VII
6.3.2), p. 380; Ch. VIII 3.6.2, p. 650.
11 (a) G. A. Jefferey and L. M. Wingert, Liq. Cryst., 1992, 12, 179, and
references cited therein; (b) H. A. van Doren, E. Smits, J. M. Pestman,
J. B. F. N. Engberts and R. M. Kellogg, Chem. Soc. Rev., 2000, 29,
183, and references cited therein; (c) S. Das, N. Gopinathan,
S. Abraham, N. Jayaraman and M. K. Singh, Adv. Funct. Mater.,
2008, 18, 1632.
12 (a) C. T. Imrie and P. A. Henderson, Curr. Opin. Colloid Interface
Sci., 2002, 7, 298, and references cited therein; (b) I. Nishiyama,
J. Yamamoto, J. W. Goodby and H. Yokoyama, J. Mater. Chem.,
2003, 13, 2429; (c) C. V. Yelamaggad, I. S. Shashikala and Q. Li,
Chem. Mater., 2007, 19, 6561.
13 (a) I. M. Saez and J. W. Goodby, J. Mater. Chem., 2005, 15, 26, and
references cited therein; (b) Y. Shimizu, A. Kurobe, H. Monobe,
N. Terasawa, K. Kiyohara andK. Uchida, Chem. Commun., 2003, 1676.
14 N. Tamaoki, A. V. Parfenov, A. Masaki and H. Matsuda, Adv.
Mater., 1997, 9, 1102.
15 T. Seki, T. Fukuchi and K. Ichimura, Bull. Chem. Soc. Jpn., 1998, 71,
2807.
16 R. M. Hann, A. T. Ness and C. S. Hudson, J. Am. Chem. Soc., 1944,
66, 73.
17 S. Solyon, K. Szilagyi and L. Toldy, Liebigs Ann. Chem., 1987, 153.
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 5956–5963 | 5963