Conformational effect on macroscopic chirality modification of cholesteric mesophases by photochromic azobenzene dopants

Article abstract of DOI:10.1021/jp000338f

Azobenzenes having different positional substituents were dissolved in cholesteric liquid crystals to make cholesteric pitches tunable by photoisomerization. E-to-Z photoisomerization of 3,3′-disubstituted or 2,2′-dimethyl-3,3′-disubstituted azobenzenes resulted in a moderate change or even no change in the maximum wavelengths of reflectivity of the cholesteric phase, whereas azobenzene or 4,4′-disubstituted azobenzenes showed much bigger changes. The different in the conformations between the E- and Z-isomers of the azobenzenes is discussed to explain the observed characteristics. The photoinduced modification of helical pitches by photoisomerization of achiral azobenzenes was identical between a pair of enantiomeric cholesteric solvents, suggesting no preferred intrinsic handedness of the achiral azobenzenes. Experimental results with the corresponding chiral azobenzene suggested that the contribution of the conformation of the azobenzene to the effective helical twisting power is more pronounced than that of the molecular chirality arising from the asymmetric carbon at the terminal alkyl substituents. The photoinduced shortening or lengthening of the helical pitch seems to be determined crucially by the thermal characteristics of the cholesteric liquid crystals.

Full text of DOI:10.1021/jp000338f

J. Phys. Chem. B 2000, 104, 6529-6535
6529
ARTICLES
Conformational Effect on Macroscopic Chirality Modification of Cholesteric Mesophases by
Photochromic Azobenzene Dopants
Christian Ruslim and Kunihiro Ichimura*
Research Laboratory of Resources Utilization, Tokyo Institute of Technology, 4259 Nagatsuta,
Midori-ku, Yokohama 226-8503, Japan
ReceiVed: January 27, 2000; In Final Form: April 4, 2000
Azobenzenes having different positional substituents were dissolved in cholesteric liquid crystals to make
cholesteric pitches tunable by photoisomerization. E-to-Z photoisomerization of 3,3′-disubstituted or 2,2′-
dimethyl-3,3′-disubstituted azobenzenes resulted in a moderate change or even no change in the maximum
wavelengths of reflectivity of the cholesteric phase, whereas azobenzene or 4,4′-disubstituted azobenzenes
showed much bigger changes. The difference in the conformations between the E- and Z-isomers of the
azobenzenes is discussed to explain the observed characteristics. The photoinduced modification of helical
pitches by photoisomerization of achiral azobenzenes was identical between a pair of enantiomeric cholesteric
solvents, suggesting no preferred intrinsic handedness of the achiral azobenzenes. Experimental results with
the corresponding chiral azobenzene suggested that the contribution of the conformation of the azobenzene
to the effective helical twisting power is more pronounced than that of the molecular chirality arising from
the asymmetric carbon at the terminal alkyl substituents. The photoinduced shortening or lengthening of the
helical pitch seems to be determined crucially by the thermal characteristics of the cholesteric liquid crystals.
Introduction
The macroscopic chirality of cholesteric LCs is extremely
sensitive to physical stimuli such as temperature, pressure, and
electric field4 and to chemical modification of the molecules
incorporated in the systems including light-induced decomposi-
tion, isomerization, and racemization.5 This makes cholesteric
LCs significant not only for practical applications such as
cholesteric displays, optical data recording, polarizers, and
reflectors6 but also for fundamental researches.7-10 The study
of solute-solvent interactions is a main subject in the latter
case, since the effect of the molecular interactions will emerge
directly in the characteristics of the macroscopic chirality of
the cholesteric LC systems.
An intriguing subject in a variety of researches on cholesteric
liquid crystals (LCs) has been the study of their macroscopic
chirality exhibiting the handedness and the pitch of the helical
array, which is achievable when the mesophasic compounds
are intrinsically chiral (optically active) or when a small amount
of a chiral compound is added to a nematic LC. The establish-
ment of the macroscopic chirality in the latter case is a good
example of the occurrence of a solute-solvent cooperative
effect. In the limit of low concentrations, the ability of a chiral
dopant to induce helical pitch is quantitatively defined as helical
twisting power, â, according to the following equation1
Kozawaguchi and Wada7 dealt with the helical pitch of
cholesteric-cholesteric and nematic-cholesteric LC mixtures
which consist of cholesterol derivatives and concluded that
nematic-cholesteric mixtures with a small amounts of choles-
terics always give left-handed cholesteric LCs, being indepen-
dent of the handedness of the cholesterol derivatives. They
suggested that the nematic molecules lie in parallel with the
long axis of the steroid ring. This alignment diminished partially
the right-handed twist of cholesteric LCs found only when the
distance of the 3â carbon bond is relatively short (i.e., when
substituted with -OH, -Cl, -Br). The handedness of various
steroidal molecules in nematic LCs has also been reported by
Solladie et al.8 Another example of molecular interactions in
cholesteric LCs was reported by Labes, et al.,9,10 who showed
nonsymmetrical effects of enantiomeric pairs on the pitches of
some cholesteric LCs. They discussed this behavior in term of
diastereomeric interactions between the enantiomeric pairs and
the cholesteric mixtures.
p
-1 ) âC
(1)
Here, p and C are helical pitch length and concentration of the
dopant, respectively. The origin of this effect has been elucidated
as molecular perturbation2 and chirality transfer of the chiral
molecules to the neighboring achiral molecules through con-
formational interactions.3 When the helical pitch of a planar
oriented sample is in the range of visible light, circularly
polarized visible light will be reflected selectively. A wavelength
of the maximum reflection, λmax, is connected to the pitch and
the average refractive index n of the mixture, through
λmax ) pn
(2)
* Corresponding author. Tel: +81-45-924-5266. Fax: +81-45-924-5276.
E-mail: kichimur@res.titech.ac.jp.
10.1021/jp000338f CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/27/2000

6530 J. Phys. Chem. B, Vol. 104, No. 28, 2000
Ruslim and Ichimura
Figure 1. Chemical structures of azobenzene compounds used as
dopants.
When photochromic dopants such as azobenzene, stilbene,
indigo, and so on are dissolved in a cholesteric phase, the
photomodulation of the macroscopic chirality can be induced
by suitable light. Sackmann has showed that the photoswitching
of selective reflection of light to lead to color changes is
sometimes achieved by a cholesteric LC doped with azoben-
zene,11 but there has been no further explanations of the
microscopic phenomenon on the basis of molecular interactions.
Azobenzene appears to be very attractive as a phototrigger in
the study on structure-property relation in cholesteric media,
because of its photofatique-resistance, the simplicity of the
molecule, as well as the ease in molecular structure modification.
Recently, we have designed several novel azobenzene deriva-
tives which have distinct conformational characteristics upon
E/Z photoisomerization.12 Photoisomerization of a rod-like
E-isomer of a conventional 4,4′-disubstituted azobenzene with
UV light results in the formation of a bent-shaped Z-isomer.
However, by incorporating suitable substituents to 3- and 3′-
positions of an azobenzene or a 2,2′-dimethylazobenzene, we
have succeeded in creating novel photoisomerizable azo-
benzenes showing rod-like conformations in both E- and
Z-isomers. Such unique characteristics may serve as a good
model in exploring how the molecular conformation as well as
configuration of dopants affect the macroscopic chirality of
cholesteric solvents.
Figure 2. Reflection spectra of a right-handed (dotted line) and a left-
handed (solid line) cholesteric host, prepared by mixing chiral agent
R-1011 and S-1011 with DON-103, respectively.
TABLE 1
Y (%) at the
photostationary state of
UV light
(365 nm)
VIS light
(436 nm)
dopant
λπ-π* (nm)
ꢀZ/ꢀE × 102 a
b
1
2
3
4
5
6
4.9b
8.2
2.9
6.3
1.0
1.5
80
86
93
70
95
95
11
4.9
9.4
13
28
26
332
330
318
362
362
a The values were measured at maximum wavelength of π-π* bands
(λπ-π*) of azobenzenes by HPLC using a mixture of hexane and ethyl
acetate (80:20) as eluent. b The λπ-π* is partly superimposed on the
absorption of the LC hosts (in hexane the λπ-π* is 314 nm), so that the
ꢀZ/ꢀE is determined at π-π* bands of 354 nm.
In this paper, 3,3′-disubstituted azobenzene (1); 2,2′-dimethyl-
3,3′-disubstituted azobenzenes (2, 3); 4,4′-disubstituted azoben-
zenes (5, 6); and unsubstituted azobenzene (4) (Figure 1) were
dissolved in cholesteric LCs to form cholesteric mixtures with
light-modifiable helical pitches. The effect of E/Z photoisomer-
ization of the dopants on the helical pitch was compared and
evaluated to have an insight into the conformation effect on
macroscopic chirality, based on the calculated conformations
of the E/Z isomers of azobenzenes and the model of induced
cholesteric LCs. Thermophysical properties of the photorespon-
sive cholesteric LCs were also elucidated to support the observed
characteristics.
heated at 80 °C to obtain homogeneous LC solutions. All
mixtures showed excellent homogeneity at room temperature.
The injection of cholesteric mixtures into LC cells was
performed at their isotropic phases. The cells were subjected to
irradiation with UV and visible light successively to give the
photostationary states. Absorption spectra of the azobenzene
compounds were also probed at the same time to confirm the
level of photoisomerization processes. The level of isomerization
is defined as Z-fraction, Y (%), at UV or visible light photo-
stationary state, calculated through the following equation.
1 - (At/A0)
Y )
× 100%
(3)
Results and Discussion
1 - (ꢀZ/ꢀE)
Enantiomeric chiral agents, R-1011 and S-1011, were chosen
in preparing cholesteric solvents because they exhibit a sym-
metrical effect on the induced cholesteric pitch, with R-1011
having right-handed and S-1011 having left-handed sense. The
helical twisting power â in nematic DON-103, a mixture of
cyclohexanoic acid phenyl esters, is 30.1 (µm wt %)-1. Selective
reflection spectra13 with λmax at 670 and 668 nm were confirmed
for 7.3 wt % of both chiral agents in DON-103 (Figure 2). Here,
A0 and At are π-π* absorbances of azobenzenes at initial and
photostationary states, respectively, whereas ꢀZ and ꢀE are molar
absorption coefficients of Z- and E-isomers, respectively. Using
HPLC, the ratio of ꢀZ to ꢀE in solutions can be calculated. Table
1 showed the spectroscopic features of the azobenzene com-
pounds in nematic solvents before and after irradiation with light.
It is obvious that UV light irradiation produced 70-95% of
Z-isomers, whereas visible light induced only 5-30% of
Z-isomers in cholesteric hosts. A tendency that Y values of 4,4′-
disubstituted azobenzenes (5, 6) are higher than those of the
other azobenzenes (1-4) can also be observed. The level of
the transformation depends on the nature of the compounds, as
represented by their absorption spectra, and on the excitation
wavelength. Figure 3 (a) depicts the modulation of the right-
λ
max were designed to avoid the overlapping of the spectra with
that of azobenzene dopants. These two mixtures were used as
cholesteric LC hosts and will be addressed as the right-handed
cholesteric host and the left-handed cholesteric host in the
following section.
Into the cholesteric LC hosts were dissolved approximately
2 wt % of azobenzenes, separately, and these mixtures were

Effect of E/Z Photoisomerization on Azobenzenes
J. Phys. Chem. B, Vol. 104, No. 28, 2000 6531
that of a cholesteric mixture with an azobenzene, respectively.
The proportionality constant k can be considered as a physical
quantity referring to the degree of helical pitch modulation
induced by one mole of the azobenzene dopant. Figure 4
provides comparative views on the value of k for the azoben-
zenes in both right-handed (a) and left-handed (b) cholesteric
hosts upon alternate irradiation with UV and visible light for
two cycles. Notice that k > 0 indicates that the λmax is shifted
to longer wavelength (lengthening of the pitch) with respect to
that of the cholesteric host.
It is important to notice that the helical pitch modulation
induced by E/Z photoisomerization depends significantly on the
molecular conformations of the azobenzene dopants. Essentially,
it is found here that photochemical E/Z isomerization does not
always influence the macroscopic chirality of cholesteric LCs,
as shown by dopant 3. It should be stressed here that this
phenomenon is not due to the difference in concentration ratios
between E- and Z-isomers, since during the whole photoisomer-
ization process the λmax remained unchanged in both cholesteric
hosts. A conventional 4,4′-disubstituted azobenzene, 5, showed
a much more significant effect on the pitch modulation than
azobenzene (4), 3,3′-disubstituted azobenzenes (1), and 2,2′-
dimethyl-3,3′-disubstituted azobenzenes (2, 3). As has been
known, when achiral solutes are dissolved in cholesteric LCs,
they may perturb the helical arrangement.11 The distinct
photoisomerization effect of the azobenzenes here is attributed
to the E/Z conformational interplay of the dopants with the LC
molecules arranged in the helical superstructure. On the basis
of computational and experimental studies on molecular con-
formations of geometrical isomers of azobenzenes with different
positional substituents reported before,12 the change from rod-
like conformation of E-5 to bent conformation of Z-5 after UV
light irradiation gave a remarkable impact to the molecular
arrangement in the system, leading to a large change of the
observed pitch or ∆λmax. However, in the case of 3, rod-like
conformation of its E-isomer is retained even in its Z-isomer
(see Figure 5). The elongated rod-like conformation of its
Z-isomer is energetically much more stable than other possible
conformations. Consequently, the E-to-Z transformation of 3
gave no perturbation to the helical arrangement in the cholesteric
LCs. This unique behavior of 3 was also confirmed in another
cholesteric host consisting of main components of biphenyl and
triphenyl LCs. The results were exactly the same, i.e., no change
in λmax modification at all during photoisomerization (Figure
6). 3,3′-Disubstituted azobenzene (1) and 2,2′-dimethyl-3,3′-
disubstituted azobenzene (2) showed a moderate change of pitch
as illustrated in Figure 4. The origin seems to stem from the
fact that conformations other than the rod-like one of Z-isomer
may coexist as has been discussed in our earlier paper.12 Notice
that the presence of methyl group at 2- and 2′-positions
generating a steric hindrance and the type of linkage (ester) in
3- and 3′-positions are the key points that stabilize the elongated
rod-like conformation of E-3 when compared with those of 1
or 2. Additionally the half width of the spectra in all cases
remain unchanged, indicating the invariance of the birefringence
upon photoisomerization.
Figure 3. Reflection spectra of right-handed cholesteric LCs containing
2.3 wt % of achiral compound (a) 5 and (b) 3 at initial (solid lines),
UV light photostationary (broken lines), and visible light photostationary
(dotted lines) states. Notice that these three spectra overlapped with
each other in (b). Inset: UV absorption spectra of the azobenzenes at
the corresponding states indicated in the figures.
handed cholesteric LC containing 2.3 wt % of 5 upon photoir-
radiation with UV and visible light, respectively. The initial
λ
max (698 nm) shifted to 662 and 688 nm after UV and visible
light exposure to give photostationary states, respectively.
Considering the level of Z-fraction after UV irradiation (in this
case 95% as indicated in the inset), the λmax can be interpreted
as the apparent effect of the Z-isomer of 5. Surprisingly, in a
keen contrast, the right-handed cholesteric LC containing 2.3
wt % of 3 did not show any changes of λmax at all, despite the
Z-fraction as large as 93% (Figure 3b). Other azobenzenes in
the same host exhibited the modulation of λmax in which the
extent depends on the structure of azobenzenes.
Quantitative interpretation of the λmax modulation using the
concept of effective twisting power â of the azobenzenes is not
very authentic, due to the question of whether these achiral
azobenzenes possess â, which we shall see in the next section.
Therefore, the following relation is used, taking into account
that, in the limit of low concentration of a dopant, the λmax of
the reflection spectra shifts linearly with the concentration c
(in molality).
If we compare the value of k only for achiral azobenzenes
(solid lines in Figure 4a and b), we observe that all azobenzene-
doped cholesteric LCs show initially λmax shifted to longer
wavelengths (k > 0 or enlargement of the pitch) with respect
to those of the cholesteric hosts, independent of the handedness
of these hosts. Besides, the k values are approximately the same
in both right-handed and left-handed hosts. This suggests that
there exists no preferred handedness of these achiral azoben-
(λmax)t ) (λmax)0 + kc
∆λmax ) [(λmax)t - (λmax)0] ) kc
(4)
Here, (λmax)0 and (λmax)t indicate λmax of a cholesteric host and

6532 J. Phys. Chem. B, Vol. 104, No. 28, 2000
Ruslim and Ichimura
Figure 4. Photoinduced alteration of wavelength of maximum reflection as defined in eq 4 for azobenzene compounds 1 (diamond), 2 (triangle),
3 (inverted triangle), 4 (circle), 5 (square), and 6 (cross) in the (a) right-handed and (b) left-handed cholesteric LCs. I, UV, and Vis of the horizontal
axis indicate initial, UV light photostationary, and visible light photostationary states, respectively.
Figure 5. The most stable conformation of E- and Z-isomers of model compounds of 3 and 5, simulated by molecular mechanics MM2 followed
by semiempirical MOPAC calculation.
zenes. Our result here is in contrast to that reported by Labes
et. al who observed the shortening of the pitch of cholesteric
mixtures doped with rod-like molecules.14 However they stated
later that this effect could not be reproduced.10 Compound 5 is
a typical rod-like molecule in its E-isomers, showing a nematic
nature; yet, no indication of the existence of handedness was
detected. The results by Kozawaguchi and Wada7 arguing that
nematic LCs behave like left-handed cholesteric in right-handed
cholesteric LCs (cholesteryl chloride, cholesteryl mercaptan, and
cholestery bromide) was understandable only in a region far
from the limitation of dilute solutions where a concentration
versus pitch curve shows an inflection point at a high nematic
LC concentration and consequently, the concept of helical
twisting power, thus guest-host interaction becomes ambiguous.
Note that they actually observed the lengthening of the pitch
when a nematic LC is doped into left-handed cholesteryl oleate,
which then may also lead to the interpretation that the nematic
is right-handed. Therefore, we come to the conclusion that
achiral molecules seem not to possess preferred intrinsic
handedness. This behavior becomes more evident when being
compared to that of the chiral azobenzene, 6, and after
elucidating the effect of the thermal properties of the cholesteric
LCs, which are discussed below.
Figure 6. Reflection spectra of a cholesteric host S-1011/RO-571 (bold
line), the same host mixed with 2.2 wt % of 3 before irradiation (solid
line), after irradiation with UV light (broken line), and visible light
(dotted line) toward the photostationary states. The latter three spectra
overlapped with each other because of the photoinert of the helical
pitch.
left-handed sense was dissolved in the right-handed host, the
When a chiral azobenzene, 6 (dotted lines in Figure 4), having
λmax before irradiation shifted to even longer wavelength (k )

Effect of E/Z Photoisomerization on Azobenzenes
J. Phys. Chem. B, Vol. 104, No. 28, 2000 6533
0.91) compared to that induced by the achiral compound 5 (k
) 0.47) (see Figure 4a), indicating the role of the opposite
molecular chirality of 6 with respect to that of the host. However,
when 6 was doped into the left-handed host (dotted line in Figure
4b), no shift to shorter wavelength (shortening of the pitch)
occurred, which is beyond our expectation. A shift also to longer
wavelength, as seen in the case of dissolving in the right-handed
host, was indeed observed. However, it is important to notice
that the shift induced by 6 in the left-handed host (k ) 0.10) is
much smaller than that in the right-handed host (k ) 0.91). This
means that essentially there exists an intrinsic shift to a shorter
wavelength when 6 is introduced to the left-handed host due to
the microscopic chirality derived from asymmetric carbon at
the alkyl substituents of 6. However, this shift was compensated
by the shift to longer wavelength attributed to the effect of the
conformation of the azobenzene unit, as confirmed in the case
of the corresponding 4,4′-disubstituted achiral azobenzenes 5.
Here, the shift induced by the conformation of the azobenzene
6 seems to be larger than that by its molecular chirality, so that
totally we observed a shift of λmax to longer wavelength in the
left-handed host. These results indicate that the effective â of
the chiral azobenzenes 6 is the contribution of at least two
components, molecular chirality and molecular conformation.
This supports the validity of the results reported by Labes et.
al, who dealt with more complicated chiral molecules doped in
cholesteric LCs.9 Thus, by comparing the behavior of achiral 5
and chiral 6 in both right- and left-handed hosts (Figure 4a and
b), it is also indirectly confirmed that there is no intrinsic
handedness in achiral azobenzenes.
Figure 7. Thermal behavior of the helical pitch of the right-handed
cholesteric host (circle), the cholesteric LC consisting of cholestryl
chloride and cholesteryl nonanoate at a ratio of 35:65 by weight (square)
and the induced cholesteric LC formed by doping 5.8 wt % of chiral
6 in nematic DON-103 (triangle).
With the increase of the fraction of Z-isomer upon irradiation
with UV light, λmax shifted to shorter wavelengths for all
mixtures in the above experiments, as though the â of Z-isomers
were stronger than E-isomers. However, this appears to be a
deceitful interpretation, since first there is no handedness in these
achiral azobenzenes, thus no effective â. And second, Z-isomers
of the azobenzenes (except for some 3,3′-disubstituted azoben-
zenes) generally disorder or even destroy LC phases.12,15,16 In
fact, Sackmann observed the opposite behavior, i.e., the shift
of λmax to longer wavelength upon E-to-Z photoisomerization,
using unsubstituted azobenzene in cholesteric LCs.10 An un-
substituted azobenzene (4) was also used in our systems and
the behavior, as seen in Figure 4, was opposite to that of
Sackmann. After a careful study, we found that this behavior
seems to depend crucially on the signs of thermal characteristics
of the helical pitch, definable as thermal coefficient R,
Figure 8. UV light-induced lengthening of the helical pitch of a
cholesteric LC consisting of 5.8 wt % of 6 in a nematic DON-103.
an anomalous temperature behavior close to the smectic-
cholesteric transition temperature was observed, due to the
formation of smectic clusters.17 Our cases deal only with normal
cholesteric states. When chiral compound 6 is dissolved in a
nematic LC (DON-103), a cholesteric mixture with a large
helical pitch is induced. In this case compound 6 acts as both
a chiral agent and a photoresponsive dopant. This mixture
exhibited (dp/dT) > 0 (see also Figure 7) and hence E-to-Z
photoisomerization resulted in lengthening of the pitch from 6
to 25 µm for the dopant concentration of 5.8 wt % (Figure 8).
Again, this supports satisfactorily the validity of the phenomenon
mentioned above. Additionally, it has to be stressed that, if a
chiral azobenzene is introduced to cholesteric LCs, whether
E-to-Z photoisomerization will lengthen or shorten the helical
pitch depends competitively on the sign of the thermal coef-
ficient, the handedness, and the magnitude of the effective â of
the chiral dopant.
1 dp
p dT
R )
(5)
(
)
where p is the pitch length and T is the temperature. The sign
of R is determined only by the sign of (dp/dT). In the right-
handed and left-handed hosts used here (Figure 4), we found
(dp/dT) < 0. A mixture of cholesteryl chloride and cholesteryl
nonanoate (35:65 by weight), which was used by Sackmann,
was also prepared to determine its R. We found that this mixture
showed (dp/dT) > 0, contrary to our cholesteric hosts (Figure
7). This means that in a cholesteric mixture with (dp/dT) > 0,
E-to-Z photoisomerization will end up with shifting of λmax to
longer wavelength (lengthening of the pitch) and via versa. This
phenomenon can be rationalized by taking into account that the
microscopic molecular disordering induced by bent Z-isomer
of an azobenzene should give analogous impact to the system
(which appears as pitch shifting in this case) with thermally
induced disordering of molecular arrangement. However, in
some cholesteric mixtures having pretransition to smectic states,
Conclusion
Photosensitive cholesteric LCs have been achieved by dis-
solving 3,3′-, 2,2′-dimethyl-3,3′- and 4,4′-disubstituted azoben-
zenes in cholesteric LCs. It was found that upon photoirradiation
the helical pitch was modified markedly in the case of a 4,4′-
disubstituted azobenzene because of the drastic molecular
conformation change between the E- and the Z-isomers. On the
other hand, a 3,3′-disubstituted azobenzene or 2,2′-dimethyl-
3,3′-disubstituted azobenzene showed moderate modification or
no modification of the pitches at all, depending on the
substituents. A particular phenomenon was observed with

6534 J. Phys. Chem. B, Vol. 104, No. 28, 2000
Ruslim and Ichimura
compound 3, i.e., photoinert of the helical pitch during E/Z
photoisomerization. This compound retained its rod-like shape
in both E- and Z-isomers as has been confirmed by means of
molecular mechanics and molecular orbital calculation. There-
fore, E/Z transformation has almost no perturbation to the
adjacent molecular arrangement.
The achiral azobenzenes showed no preferred handedness in
cholesteric LCs within the limit of low dopant concentration.
By comparing the behavior of an achiral azobenzene and an
analogous chiral azobenzene, it is reasonable to conclude that
the effective â of the chiral azobenzene can be considered as
the contribution of at least two components, i.e., the molecular
conformation of the isomers and the molecular chirality derived
from asymmetric carbons.
The shrinking or lengthening of the helical pitch upon E-to-Z
photoisomerization of achiral azobenzenes seems to depend on
the thermal nature of the cholesteric LCs. LC phase-destructive
Z-isomer of a conventional 4,4′-substituted azobenzene will
induce a shortening of the helical pitch if (dp/dT) of the
cholesteric host has a negative sign and vice versa. However,
this is not always the case in 2,2′-dimethyl-3,3′-disubstituted
azobenzenes since their geometrical isomers, depending on the
substituents, may possess similar conformation. In the case
where two types of chiral compounds, one of which is a chiral
azobenzene, exist in the cholesteric mixtures, the behavior of
the helical pitch photomodulation depends then on the competi-
tive strength of the â and handedness of both chiral compounds.
The information obtainable from the behavior of the macro-
scopic chirality in azobenzene containing cholesteric LCs here
may serve as a comprehensive model in the interpretation of
conformational molecular interaction in LC systems, which is
of great importance in molecular modeling to achieve optimum
tailored properties.
(S,S)-4,4′-bis-(2-methylbutyloxy)azobenzene (6). To a 40
mL dehydrated DMF were introduced 4,4′-dihydroxy azoben-
zene (0.80 g; 3.73 mmol) and potassium carbonate (1.54 g, 11.2
mmol). The suspension was heated at 80 °C and stirred for 30
min before (S)-2-methylbutyl-p-toluenesulfonate (2.71 g; 11.2
mmol) was added dropwise. Three equimolar amounts of the
compound were added here in order to obtain also (S)-4-(2-
methylbutyloxy)-4′-hydroxy azobenzene as another product
which is used in other experiments. The products were separated
through column chromatography (silica gel, ethyl acetate/hexane
1:10). The product, 6, was purified by recrystallization from
ethyl acetate to give yellow crystals (mp 109-110 °C) in a 26%
1
yield (from 4,4′-dihydroxy azobenzene). H NMR (200 MHz,
CDCl3) δ 0.97 (t, J ) 7.4 Hz, 6 H), 1.04 (d, J ) 6.9 Hz, 6 H),
1.22-1.36 (m, 2 H), 1.50-1.66 (m, 2 H), 1.82-1.98 (m, 2 H),
3.77-3.94 (m, 4 H), 7.00 (d, J ) 9.2 Hz, 4 H), 7.86 (d, J )
9.2 Hz, 4 H). Anal. Calcd. for C22H30N2O2 (%): C, 74.52; H,
8.55; N, 7.90. Found: C, 74.21; H, 8.45; N, 7.89.
Sample Preparation. In nematic LCs, DON-103 and RO-
571, were dissolved 7.3 wt % of enantiomeric chiral agents,
R-1011 (Merck) and S-1011, separately to form right-handed
and left-handed cholesteric solvents, respectively. Approximately
2 wt % of azobenzene or substituted azobenzenes was dissolved
in the mixtures. LC cells with Grandjean textures were prepared
by sandwiching the LC solutions between plates treated with
uniaxially rubbed PVA thin films. The thickness of the cells
was adjusted by colloidal silica and was approximately 5 µm.
Photoirradiation and Cholesteric Pitch Measurement. An
LC cell was irradiated with UV and visible light sorted out from
high-pressure Hg lamp (USH-500D) passing through a combi-
nation of glass filters, UV35/UVD35 and Y43/V44 (Toshiba),
respectively. Irradiation was performed toward the photosta-
tionary states at about 23 °C. The reflection spectra of short-
pitch cholesteric mixtures were recorded on a diode array
spectrometer (HP8452A). In the case of large-pitch cholesteric
LCs, the pitch was determined by Grandjean-Cano lines in
wedge cells.18 The handedness of the cholesteric mixtures was
confirmed by contact method.19 Temperature dependence of the
cholesteric pitch was measured using a Mettler FP800 hot stage
set under an optical microscope.
Experimental Section
Materials and Characterization. (S)-2-Methylbutanol was
purchased from Kanto Chemical and p-toluenesulfonyl chloride
was from Tokyo Chemical Industry. All reagents were used
without further purification. The synthesis of 4,4′-dihydroxy
azobenzene and compounds 1-5 have been reported previously.12b
Chiral agents, R1011 and S1011 (Merck) were used as received.
DON-103 (TNI ) 74.2 °C), a nematic LC consisting of mixtures
of cyclohexanoic acid phenyl esters, and RO-571 (TNI ) 64.1
°C), a nematic LC consisting of mixtures of cyano biphenyl
and cyano triphenyl compounds, were kindly donated from
LODIC. The chemical structures of products were characterized
by 1H spectra, recorded on a Bruker AC-200 NMR spectrometer
with TMS as an internal standard, and elemental analysis. Phase
transition temperatures were determined by a DSC 22C (Seiko
Densi Kogyo) and a polarized optical microscope Olympus
BH-2 equipped with a Mettler FP800 hot stage.
References and Notes
(1) (a) Sackmann, E.; Meiboom, S.; Snyder, L. C.; Meixner, A. E.;
Dietz, R. E. J. Am. Chem. Soc. 1968, 90, 3567. (b) Baessler, M.; Laronge,
T. M.; Labes, M. M. J. Chem. Phys. 1969, 51, 3213.
(2) Gottarelli, G.; Samori, B.; Stremmenos, C. Chem. Phys. Lett. 1976,
40, 308.
(3) Gottarelli, G.; Hibert, M.; Samori, B.; Solladie, G.; Spada, G. P.;
Zimmermann, R. J. Am. Chem. Soc. 1983, 105, 7318.
(4) Coles, H. In Handbook of Liquid Crystals; Demus, D., Goodby,
J., Gray, G. W., Spiess, H.-W., Vill, V., Eds.; Wiley-VCH: Weinheim,
1998; Vol. 2A, Chapter 4, p 382.
(5) (a) Solladie, G.; Zimmermann, R. Angew. Chem., Int. Ed. Engl.
1984, 23, 348. (b) Lemieux, R. P.; Schuster, G. B. J. Org. Chem. 1993, 58,
100. (c) Heppke, G.; Marschall, H.; Nu¨rnberg, P.; Oestreicher, F.;
Scherowsky, G. Chem. Ber. 1981, 114, 2501. (d) Bobrovsky, A. Y.; Boiko,
N. I.; Shibaev, V. P. Liq. Cryst. 1998, 25, 679. (e) Huck, N. P. M.; Janger,
W. F.; de Lange, B.; Feringa, B. L. Science 1996, 273, 1686.
(6) See, for instance: (a) Dyer, D. J.; Scho¨der, U. P.; Chang, K. P.;
Twieg, R. J. Chem. Mater. 1997, 9, 1665. (b) Tamaoki, N.; Parfenov, A.
V.; Masaki, A.; Matsuda, H. AdV. Mater. 1997, 9, 1102. (c) Tamaoki, N.;
Song, S.; Moriyama, M.; Matsuda, H. AdV. Mater. 2000, 12, 94. (d) Hikmet,
R. A. M.; Kemperman, H. Nature 1998, 392, 476. (e) Van de Vitte, P.;
Neuteboom, E. E.; Brehmer, M.; Lub, J. J. Appl. Phys. 1999, 85, 7517.
(7) Kozawaguchi, H.; Wada, M. Jpn. J. Appl. Phys. 1975, 14, 651.
(8) Hibert, M.; Solladie, G. Mol. Cryst. Liq. Cryst. 1981, 64, 211.
(9) Yarovoy, Y. K.; Labes, M. M. Mol. Cryst. Liq. Cryst. 1995, 270,
101.
(S)-2-Methylbutyl-p-toluenesulfonate. p-Toluenesulfonyl
chloride (3.00 g; 34.0 mmol), DMAP (2.90 g; 23.7 mmol), and
triethylamine (8.00 g; 79.2 mmol) were dissolved in 20 mL of
dichloromethane. The solution was stirred and cooled in an ice-
bath, while (S)-2-methylbutanol (3.00 g; 34.0 mmol) diluted in
4 mL of dichloromethane was added dropwise. After 30 min,
the reaction was let warm to room temperature and stirred for
another 2 h. The product was purified by column chromatog-
raphy (silica, ethyl acetate/hexane 1:1) to give a product as a
1
light-yellow liquid in a yield of 95%. H NMR (200 MHz,
CDCl3) δ 0.83 (t, J ) 7.3 Hz, 3 H), 0.88 (d, J ) 6.6 Hz, 3 H),
1.04-1.26 (m, 1 H), 1.29-1.46 (m, 1 H), 1.63-1.79 (m, 1 H),
2.45 (s, 3 H), 3.74-3.93 (m, 2 H), 7.35 (d, J ) 8.0 Hz, 2 H),
7.79 (d, J ) 8.0 Hz, 2 H).
(10) Yarovoy, Y. K.; Patel, V.; Labes, M. M. Mol. Cryst. Liq. Cryst.
1998, 319, 101.

Effect of E/Z Photoisomerization on Azobenzenes
J. Phys. Chem. B, Vol. 104, No. 28, 2000 6535
(11) Sackmann, E. J. Am. Chem. Soc. 1971, 93, 7088.
(12) (a) Ruslim, C.; Ichimura, K. Chem. Lett. 1998, 789. (b) Ruslim,
C.; Ichimura, K. J. Mater. Chem. 1999, 9, 673.
(13) Selective reflection spectra indicated here were recorded as
transmitted spectra of the cholesteric mixtures arranged in Grandjean
textures.
(16) Fischer, B.; Thieme, C.; Fischer, T. M.; Kremer, F.; Oge, T.; Zentel,
R. Liq. Cryst. 1997, 22, 65.
(17) Voss, J.; Sackmann, E. Z. Naturforsch. 1973, 28a, 1496.
(18) Cano, R. Bull. Soc. Franc. Mineral. 1968, 91, 20.
(19) (a) Isaert, N.; Soulestin, B.; Malthete, J. Mol. Cryst. Liq. Cryst.
1976, 37, 321. (b) Gray, G. W.; McDonald, D. G. Mol. Cryst. Liq. Cryst.
Lett. 1977, 34, 211.
(14) Labes, M. M.; Shang, W. J. Am. Chem. Soc. 1991, 113, 2773.
(15) Sasaki, T.; Ikeda, T.; Ichimura, K. Macromolecules 1992, 25, 3807.