Solvent- and phase-controlled photochirogenesis. Enantiodifferentiating photoisomerization of (Z)-cyclooctene sensitized by cyclic nigerosylnigerose-based nanosponges crosslinked by pyromellitate

Article abstract of DOI:10.1039/c4ob02390k

Cyclic nigerosylnigerose (CNN), a saucer-shaped cyclic tetrasaccharide with a shallow concave surface, was reacted with pyromellitic dianhydride in 1 : 2 and 1 : 4 ratios to give two CNN-based polymers of different degrees of crosslinking, both of which swelled upon soaking in water, acting as a 'nanosponge' (NS). These NSs evolved several phases from isotropic solution to flowing and rigid gels via suspension by gradually increasing the concentration in water. The CNN-NSs thus prepared effectively mediated the enantiodifferentiating photoisomerization of (Z)-cyclooctene (1Z) to chiral (E)-isomer (1E). The enantiomeric excess (ee) of 1E obtained was a critical function of the solvent composition and the phase evolved at different CNN-NS concentrations in water. In isotropic solution, the enantioselectivity was generally low (-4% to +6% ee) but the chiral sense of 1E was inverted by increasing the methanol content. Interestingly, the product's ee was controlled more dramatically by the phase evolved, as was the case with the cyclodextrin-based nanosponge (CD-NS) reported previously. Thus, the ee of 1E was low in solution and suspension, but suddenly leaped at the phase border of flowing gel and rigid gel to give the highest ee of 22-24%, which are much higher than those obtained with CD-NSs (6-12% ee), revealing the positive roles of the chiral void space formed upon gelation of the crosslinked saccharide polymer.

Full text of DOI:10.1039/c4ob02390k

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Solvent- and phase-controlled photochirogenesis.
Enantiodifferentiating photoisomerization of
(Z)-cyclooctene sensitized by cyclic
Cite this: DOI: 10.1039/c4ob02390k
nigerosylnigerose-based nanosponges crosslinked
by pyromellitate†
Xueqin Wei,a Wenting Liang,b Wanhua Wu,a Cheng Yang,*a Francesco Trotta,*c
Fabrizio Caldera,c Andrea Mele,d Tomoyuki Nishimotoe and Yoshihisa Inoue*f
Cyclic nigerosylnigerose (CNN), a saucer-shaped cyclic tetrasaccharide with a shallow concave surface,
was reacted with pyromellitic dianhydride in 1 : 2 and 1 : 4 ratios to give two CNN-based polymers of
different degrees of crosslinking, both of which swelled upon soaking in water, acting as a ‘nanosponge’
(NS). These NSs evolved several phases from isotropic solution to flowing and rigid gels via suspension by
gradually increasing the concentration in water. The CNN-NSs thus prepared effectively mediated the
enantiodifferentiating photoisomerization of (Z)-cyclooctene (1Z) to chiral (E)-isomer (1E). The enantio-
meric excess (ee) of 1E obtained was a critical function of the solvent composition and the phase evolved
at different CNN-NS concentrations in water. In isotropic solution, the enantioselectivity was generally
low (−4% to +6% ee) but the chiral sense of 1E was inverted by increasing the methanol content. Interest-
ingly, the product’s ee was controlled more dramatically by the phase evolved, as was the case with the
cyclodextrin-based nanosponge (CD-NS) reported previously. Thus, the ee of 1E was low in solution and
suspension, but suddenly leaped at the phase border of flowing gel and rigid gel to give the highest ee of
22–24%, which are much higher than those obtained with CD-NSs (6–12% ee), revealing the positive
roles of the chiral void space formed upon gelation of the crosslinked saccharide polymer.
Received 12th November 2014,
Accepted 22nd December 2014
DOI: 10.1039/c4ob02390k
www.rsc.org/obc
sensitizer and a prochiral substrate. Despite a considerable
number of studies devoted to chiral photosensitization since
Introduction
Catalytic photochirogenesis is one of the crucial challenging the pioneering work reported by Hammond and Cole in 1965,5
topics in current photochemistry.1–4 Catalytic asymmetric syn- to achieve high enantioselectivity in chiral photosensitization
thesis in photochemistry has been realized mostly using chiral is still a nontrivial task, as can be recognized from the limited
photosensitization, where the chirality transfer takes place in a number and type of photoreactions that afford satisfactory
diastereomeric exciplex intermediate formed between a chiral results.6–11 This is primarily due to the weak, short-lived, and
less-defined chiral environment realized in the exciplex inter-
mediates formed in conventional chiral photosensitization
aCollege of Chemistry, State Key Laboratory of Biotherapy, West China Medical
systems. For a better photochirogenic performance, supra-
School and State Key Laboratory of Polymer Materials Engineering, Sichuan
molecular photochirogenesis2 has been proposed as a new
University, 29 Wangjiang Road, Chengdu 610064, China.
E-mail: yangchengyc@scu.edu.cn
strategy, wherein a robust, long-lasting, and well-defined
chiral environment is attained not only in the ground state but
bInstitute of Environmental Sciences, Shanxi University, Taiyuan 030006, China
cDepartment of ChemistryUniversity of Torino, Via P. Giuria 7, 10125 Torino, Italy.
also in the excited state. A variety of chiral supramolecular
E-mail: francesco.trotta@unito.it
dDepartment of Chemistry, Materials and Chemical Engineering “Giulio Natta”,
hosts, including cyclodextrins (CDs),12–18 chirally modified
zeolite supercages,19–21 biomolecules,22,23 and chiral hydro-
Politecnico di Milano, Piazza L. Da Vinci 32, 20133 Milano, Italy
gen-bonding templates,24–26 has been explored to achieve
modest to excellent enantioselectivities. More recently, chiral
gels27,28 and liquid crystals29 have also been employed
as an effective chiral supramolecular environment for
photochirogenesis.
eHayashibara Co., 675-1 Fujisaki, Naka-ku, Okayama 702-8006, Japan
fDepartment of Applied Chemistry, Osaka University, 2-1 Yamada-oka,
Suita 565-0871, Japan. E-mail: inoue@chem.eng.osaka-u.ac.jp
†Electronic supplementary information (ESI) available: Elemental analysis,
ATR-FTIR spectral and thermogravimetric analysis data for CNN-NSs 6 and 7.
See DOI: 10.1039/c4ob02390k
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
We have recently reported the enantiodifferentiating photo- accommodate 1Z and 2ZZ but also sensitize the geometrical iso-
isomerizations of (Z)-cyclooctene (1Z) (Scheme 1) and (Z,Z)-1,3- merization to chiral 1E and 2EZ upon photoirradiation. The
cyclooctadiene (2ZZ) sensitized by β- and γ-CD polymers cross- enantioselectivities of 1E and 2EZ obtained critically depended
linked with pyromellitate (3–5) (Scheme 2),30 which were on the phase conditions of CD-NS to afford the highest enantio-
named ‘nanosponges’ after their swelling nature upon soaking meric excesses (ee’s) at the border of the flowing and rigid gel
in water and some organic solvents. These CD-based nano- states. Such behavior had never been observed with non-poly-
sponges (CD-NSs), incorporating pyromellitate linkers, not only meric CD-based sensitizers reported earlier.31–35
A progressive aggregation mechanism has been proposed in
order to rationalize the phase-dependent photochirogenic be-
havior. The inter-particle complexation of the pyromellitate
moiety by the CD cavity exposed at the surface of CD-NS par-
ticles is considered to drive the progressive aggregations,
leading to the phase transitions, upon the gradual increase of
the CD-NS concentration. Simultaneously, the chiral environ-
ment available for the enantiodifferentiating photosensitiza-
Scheme 1 Enantiodifferentiating photoisomerization of 1Z to chiral 1E.
tion is switched from the CD cavity in isotropic media to the
void space surrounded by the exterior walls of CD in gel states
with accompanying changes in ee.30
To elucidate the role of aggregation in the supramolecular
photochirogenesis and to gain further insights into the chiral
sensitization mechanism in gel systems, we synthesized new
pyromellitate-crosslinked saccharide polymers based on cyclic
nigerosylnigerose (CNN) in this study. CNN differs from CD in
the number and connectivity of the building units, being com-
posed of only four D-glucopyranoses connected by alternating
α-(1→3)- and α-(1→6)-linkages.36–40 As a cyclic tetrasaccharide,
CNN is much smaller in ring size than CDs and does not bear
a confining hydrophobic cavity but a shallow concave which is
not sufficient to accommodate a cyclooctene molecule. There-
fore, CNN serves as an ideal saccharide building block for
investigating the effects of the aggregation of NS and the void
space derived therefrom on photochirogenic behaviors,
without the influence of cavity binding. Two crosslinked CNN
nanosponges (CNN-NSs) 6 and 7 (Scheme 2) were prepared by
crosslinking CNN with pyromellitic dianhydride (PDA) at
different CNN : PDA ratios of 1 : 2 and 1 : 4, respectively.
Results and discussion
Photosensitization in solution
In order to evaluate the chiral environment around the pyro-
mellitate linker, the circular dichroism spectra of CNN-NSs
were measured in aqueous solutions. As can be seen from
Fig. 1 and 2 (red lines), both CNN-NSs 6 and 7 exhibit the
induced Cotton effect signals at the main band of pyromelli-
tate, which are however much weaker than those observed for
CD-NSs 3–5 reported previously.30 This seems reasonable as
the shallow concave of CNN is not sufficient in size to fully
accommodate the pyromellitate linker, allowing only weak
hydrophobic interaction. Furthermore, the addition of 1Z to
an isotropic CNN-NS solution did not cause any appreciable
changes in circular dichroism, probably due to the lack of a
confining cavity and the weak interaction in the ground state.
Unlike β-CD that possesses a hydrophobic cavity to bind 1Z
Scheme 2 Chemical structures of β- and γ-CD-NSs 3–5 and CNN-NSs
6 and 7.
with a relatively high affinity,32,33 saucer-shaped CNN does not
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
in the excited state. Indeed, we have shown that the optical
yield of 1E obtained in the photosensitized isomerization of
1Z with a sensitizer-appended CNN monomer in aqueous solu-
tion is appreciably higher than those obtained with α-CD-
based sensitizers, indicating a positive role of the CNN
concave upon photosensitization.38 This previous result also
encouraged us to investigate the photoisomerization of 1Z sen-
sitized by CNN-NS in aqueous solutions, despite the appar-
ently negative result obtained upon CD spectral titration.
The enantiodifferentiating photoisomerization of 1Z sensi-
tized by CNN-NSs 6 and 7 was first performed in isotropic
aqueous solutions containing 0–80% methanol. As shown in
Table 1, the solution-phase photoisomerization of 1Z sensi-
tized by CNN-NS gave 1E in a modest yield of −4% to +6% ee.
Somewhat unexpectedly, the photosensitizations with 6 versus
7 afforded the opposite enantiomers of 1E under comparable
conditions. Thus, in aqueous solution at 0.5 °C, less
crosslinked CNN-NS 6 afforded (R)-1E in 3.9% ee, while more
crosslinked CNN-NS 7 gave antipodal (S)-1E in 4.4% ee.
Intriguingly, the enantioselectivity obtained with 6 gradually
deteriorated by raising the methanol content from 10% to 50%
and was eventually inverted in sign (affording the antipode) to
reach the highest ee of 6% in 80% methanol. It is to be noted
that 6 and 7 share the same chiral CNN monomer unit and
achiral pyromellitate linker, but differ in the degree of cross-
linking and the framework rigidity. The above results reveal
that the CNN-NS’s chiral environment for the sensitized photo-
isomerization of 1Z is readily tunable by adjusting the
monomer/linker ratio or the solvent composition. Since CNN
has only a shallow concave insufficient to fully include a
cyclooctene molecule, the enantioselectivity achieved upon
photoisomerization should rely more on the chiral environ-
ment formed by crosslinked CNN molecules, which is manipu-
latable by external factors such as temperature and solvent
properties.
Fig. 1 (a) UV-Vis and (b) circular dichroism spectra of CNN-NS 6
(0.2 mg mL−1) in aqueous solutions containing 0% (red), 10% (black),
50% (blue), and 80% methanol (green).
To elucidate the origin of the switching of product chirality
by solvent composition, the UV-Vis and circular dichroism
spectra of 6 and 7 were examined in aqueous solutions of
varying methanol contents. In pure water, both 6 and 7
1
1
showed the La and Lb bands of pyromellitate at 258 nm and
295 nm, respectively; see Fig. 1a and 2a. By increasing the
1
methanol content, the Lb band of 6 exhibited gradual batho-
chromic shifts with initial hypochromic effects up to 50%
methanol, and then a hyperchromic effect at 80% methanol,
for which discontinuous changes in solvent structure and/or
solvation mode would be responsible. Similar solvent-depen-
dent shifts were observed also for 7 (Fig. 2a). This is presum-
ably due to the change in microenvironmental polarity
experienced by the pyromellitate chromophore. Interestingly,
the circular dichroism spectra of 6 and 7 were more signifi-
cantly affected by the methanol content. Thus, 6 showed a
small positive Cotton effect peak at 241 nm and a negative one
at 259 nm in pure water. By increasing the methanol content,
the whole circular dichroism signals became more negative to
give a single negative extremum of −1.5 mdeg ellipticity at
258 nm in 80% methanol (Fig. 1b). The UV-Vis and circular
Fig. 2 (a) UV-Vis and (b) circular dichroism spectra of CNN-NS 7
(0.1 mg mL−1) in aqueous solutions containing 0% (red), 10% (black),
50% (blue), and 80% methanol (green).
have sufficient space for accommodating the 1Z molecule.
Nevertheless, the concave is considered to be hydrophobic
enough to weakly interact with 1Z in aqueous solutions at least
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 Enantiodifferentiating photoisomerization of 1Z sensitized by CNN-NSs 6 and 7 in aqueous solutionsa
Host
Phase
Sol
Methanol content/%
T/°C
[CNN-NS]/mg mL−1
Irradiation time/min
1E/1Z ratio
eeb/%
6
0
10
0.5
0.5
25
0.5
0.5
0.2
0.2
0.2
0.2
0.2
10
15
10
10
20
30
60
60
60
60
60
60
10
10
30
30
10
20
10
5
0.03
0.05
0.01
0.02
0.02
0.02
0.07
0.01
0.15
0.07
0.15
0.01
0.07
0.06
0.13
0.04
0.03
0.05
0.05
0.02
0.01
0.06
0.06
0.05
0.07
0.01
0.01
−3.9
−2.7
−2.5
1.5
5.6
6.1
−2.3
1.0
0.1
4.5
24.3
6.1
4.4
1.0
1.0
−1.0
−3.7
−3.3
−3.6
−1.6
0.8
50
80
Suspension
Flowing gel
0
0.5
1
3
10
0.5
0.5
0
30
100
150
0.2
0.2
Gel
Sol
0
0
0.5
0.5
25
7
10
50
25
0.5
0.2
0.2
25
0.5
0.5
0.2
0.2
1
5
20
100
200
350
500
80
0
Suspension
60
60
60
60
60
60
60
0.8
0.4
5.9
7.0
21.9
1.6
Flowing gel
Gel
0
0
0.5
0.5
a Photoirradiation of 1Z (1 mM) with 6 (0.2–150 mg mL−1) or 7 (0.2–500 mg mL−1) performed at 254 nm in a quartz cell under a nitrogen
atmosphere. b Enantiomeric excess of 1E determined by chiral GC (error in ee < 0.5%); positive/negative ee values indicate the preferred
formation of (S)/(R)-enantiomers, respectively.
dichroism spectra of 7 showed similar trends in aqueous of NS surrounded by the exterior walls of CD. In this context,
methanol solutions (Fig. 2). These UV-Vis and circular dichro- the saucer-shaped CNN, lacking enough space for fully accom-
ism spectral changes indicate that increasing the methanol modating a cyclooctene molecule, is an ideal cyclic oligo-
content alters not only the microenvironmental polarity saccharide to substitute CD for proving or disproving our
around the pyromellitate chromophore but also the confor- hypothesis, and hence we prepared CNN-NSs 6 and 7 by cross-
mation of the pyromellitate linker, which inevitably affects the linking CNN with PDA under the comparable conditions
photochirogenic behavior of CNN-NS as observed.
employed for the synthesis of CD-NSs.30
Thus, the solvent-induced switching of product chirality is
The phase-transition behavior of CNN-NS was examined by
phenomenologically well correlated with the circular dichro- gradually increasing its concentration in water at ambient
ism spectral changes in aqueous solutions of different metha- temperature. As crosslinked polymers of a cyclic oligosacchar-
nol contents. Although elucidating the more detailed ide, both 6 and 7 exhibited a stepwise phase-evolution from a
mechanism of the chirality inversion, including the excited- transparent sol to a suspension, a flowing gel and finally a
state interaction of a cyclooctene substrate with a pyromellitate rigid gel, a behavior similar to CD-NS. Thus, CNN-NS 6 gave a
sensitizer, would not be readily feasible, the present methodo- homogeneous solution up to 0.4 mg mL−1 concentration,
logy may be employed as a convenient tool for obtaining both where transparent precipitates began to appear (Fig. 3a). When
enantiomers upon supramolecular photochirogenesis without the concentration was increased to 20 mg mL−1, a gel-like
preparing the antipodal host, which is often unpractical when binary phase composed of liquid and gel, or a ‘flowing gel’
naturally occurring supramolecular hosts, such as cyclodextrin state, emerged, which was eventually converted to rigid gel at a
and protein, are used.
critical gelation concentration (CGC) of 120 mg mL−1
.
Since the CNN concave is too shallow to strongly hold the
linker moiety (i.e. pyromellitate), the host–guest complexation
is not likely to play a major role in the aggregation and phase
evolution of CNN-NS. As an alternative mechanism, we
propose that the inter-particle hydrogen-bonding interactions
between CNN moieties and the hydrophobic interaction of the
linker moiety with the CNN’s inside/outside walls are jointly
responsible for the progressive aggregation of CNN-NS nano-
Photosensitization in gels
In our previous study on the enantiodifferentiating photo-
isomerization of cyclooctenes 1Z and 2ZZ sensitized by
CD-NSs 3–5,30 we have demonstrated that the ee of 1E and 2EZ
produced is highly sensitive to the phase of CD-NS evolved in
water, which led us to a tentative conclusion that the photo-
sensitization occurs not in the CD cavity but in the void space
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 3 (a) Upper panel: Photographs of aqueous solutions of CNN-NS 6
at 0.2, 5, 40 and 120 mg mL−1 (from left to right). Lower panel: Sche-
matic drawings of the phase evolution of 6. (b) Dynamic light scattering
(DLS) intensities measured at 0.20 (red), 0.38 (blue), and 0.80 mg mL−1
(black) concentrations of 6.
particles. CNN-NS 6 showed a phase transition from solution
to suspension at a 0.4 mg mL−1 concentration, which is lower
than those of β-CD-NSs 3 and 4 but slightly higher than that of
γ-CD-NS 5, while the CGC of 6 is lower than those of 4 and 5.
These results suggest that the inter-particle interaction of
CNN-NS does not greatly differ from that of CD-NS. Taking
into account the fact that both CD-NS and CNN-NS show rela-
tively high CGC values, we deduce that the inter-particle inter-
action is relatively weak for these two nanosponges, probably
due to the shallow penetration of the pyromellitate moiety to
the CD cavity or the CNN concave, as well as the relatively weak
hydrogen-bonding interaction in water. Dynamic light scatter-
ing (DLS) studies confirmed a gradual growth of the particle
size upon increasing the concentration of CNN-NS. As illus-
trated in Fig. 3b, the DLS examination revealed that 6 forms
aggregates of ca. 90, 140 and >250 nm diameter at 0.2, 0.38
and 0.8 mg mL−1 concentrations, respectively, while visible
precipitates were formed at a 0.4 mg mL−1 concentration.
The more crosslinked CNN-NS 7 showed a similar stepwise
phase-evolution behavior in water, but the phase transitions
from a solution to a suspension, to a flowing gel and then to a
rigid gel occurred at concentrations of 0.4, 50 and 350 mg
Fig. 4 The ee profile of 1E as a function of the concentration of (a) 3,
(b) 4, (c) 5, (d) 6 and (e) 7 upon enantiodifferentiating photoisomeriza-
tion of 1Z. The blue, green, yellow and pink regions represent the sol,
suspension, flowing gel and rigid gel states, respectively.
mL−1, respectively, each of which is higher than the corres-
ponding value observed for the less crosslinked 6, suggesting
weaker inter-particle interactions for 7. In 7, there should be a
lower number of CNN units exposed on the surface, compared
to 6, which may discourage the inter-particle hydrogen-
bonding interactions, leading to the higher phase transition
concentrations. Nevertheless, CNN-NS 7 formed suspension at
lower concentrations and the suspension state was kept over a
wider range of concentrations, compared to CD-NSs 3 and 4.
In this regard, it is be to noted that the circular dichroism
spectrum of 6 does not show any essential changes in shape in
solution (0.2 mg mL−1) and in flowing gel (120 mg mL−1),
except the small negative Cotton effect at 297 nm, as can be
seen from Fig. 5. This result indicates that CNN-NS nano-
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
obtained with CD-NSs30 under comparable conditions. This
result reinforces our previous claim that not the individual
host cavity but the polymer void surrounded by the host’s
exterior walls is responsible for the chiral induction in the
supramolecular photosensitization in crosslinked CD and
CNN gels.30 In the present case, the concave shape of CNN
may provide a better chiral environment for the effective
enantiodifferentiating photoisomerization of 1Z sensitized by
a pyromellitate crosslinker located nearby. It is interesting that
CNN’s concave itself is not sufficient to tightly hold the sub-
strate molecule and indeed only a poor ee was obtained upon
enantiodifferentiating photosensitization in isotropic solu-
tion,32 whereas CNN aggregates formed in polymer gel show
much better photochirogenic performance. Thus, the orig-
inally poor chirality-transfer ability of CNN-NS was signifi-
cantly improved by supramolecular aggregation, for which the
closer contacts with the CNN’ inside and/or outside walls as
well as the well-ordered but less dynamic nature of the gel are
likely to be jointly responsible.
Fig. 5 Circular dichroism spectra of CNN-NS 6 at 0.2 mg mL−1 (red
line: a solution in a conventional quartz cell) and at 120 mg mL−1 (blue
line: a thin layer of flowing gel sandwiched with two quartz plates).
particles aggregate at higher concentrations through the inter-
particle interactions on the surface, while the chiral environ-
ment around the pyromellitate inside the particle is not sig-
nificantly altered.
Conclusion
In this work, we have employed cyclic tetrasaccharide CNN as
a ‘pseudo’ or ‘hemi’ host that lacks a confining cavity to fully
accommodate a guest substrate. Our immediate task was to
assess and compare the photochirogenic performances of the
CNN’s shallow concave at the molecular level in isotropic solu-
tion and at the supramolecular level in aggregates. A more
crucial task was to critically examine our previous claim that
the chiral polymer void formed upon aggregation of cyclic
oligosaccharide hosts plays major roles in determining the
stereochemical outcomes of supramolecular photochirogenic
reactions.
For these purposes, we synthesized two CNN-based nano-
sponges of different degrees of crosslinking (6 and 7) and
employed them as chiral supramolecular sensitizers for med-
iating the enantiodifferentiating photoisomerization of 1Z to
1E in aqueous solutions containing 0–80% methanol. The
enantioselectivity of 1E obtained was highly sensitive to the
solvent composition and the degree of crosslinking, revealing
a switching of the product chirality, which provides us with a
convenient tool for manipulating the photochirogenic out-
comes. Both 6 and 7 showed the phase evolution from sol to
rigid gel by increasing the CNN-NS concentration in water.
Interestingly, the product’s ee was a critical function of the
phase of CNN-NS to achieve the highest enantioselectivity at
the phase border of flowing and rigid gels. This result un-
ambiguously reveals that the polymer void surrounded by the
exterior walls of the host is the source of chirality transfer
upon photosensitization with CD- and CNN-NSs. The concept
and methodology developed in this study are not restricted to
the present system, but expandable to other substrates and
photochirogenic reactions by choosing suitable sensitizing
crosslinkers and chiral building blocks.
Photoisomerization of 1Z was sensitized by 6 and 7 in
various phases prepared by changing the CNN-NS concen-
tration. The E/Z ratio obtained at a relatively early stage of
photoisomerization (Table 1) was significantly phase-depen-
dent, varying from 0.01 in rigid gel to 0.15 in suspension and
flowing gel. The particularly low E/Z ratio of 0.01 in rigid gel
may be attributed to the less-efficient photosensitization due
to the slow diffusion of the substrate to the excited sensitizer
moiety within its lifetime. The enantioselectivity of 1E pro-
duced was also a critical function of the phase of the media;
see Table 1 and Fig. 4. For instance, in the photoisomerization
of 1Z sensitized by 7 (Fig. 4e), the ee of 1E decreased from
4.4% to almost zero by increasing the CNN-NS concentration
from 0.2 to 1 mg mL−1, with an accompanying phase tran-
sition from sol to suspension, and stayed low throughout the
suspension region (0.4–50 mg mL−1), but jumped to 21.9% at
the border of the flowing and rigid gel regions. A similar
phase-dependent ee profile was observed upon sensitization
with 6 (Fig. 4d), affording an even higher ee value of 24.3%
near the border of the flowing and rigid gel regions. These ee
profiles are consistent in general with those obtained with
CD-NSs 3 and 430 (Fig. 4a and b), but significantly differ from
the ee profile obtained with γ-CD-NS 5 (Fig. 4c), which pos-
sesses a much larger CD cavity than 3 and 4 and gives a nearly
racemic product in all the anisotropic phases.30 These resem-
bling photochirogenic behaviors suggest the operation of con-
ceptually the same enantiodifferentiation mechanism in both
CNN-NS and CD-NS photosensitizations.
It is noteworthy that CNN-NSs, possessing no confining
cavity but a shallow concave, afford 1E in such good ee’s as
22–24% in gel states, which far exceed the ee values of 6–14%
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Experimental section
Acknowledgements
Instruments
This work was supported by the National Natural Science
UV-Vis and circular dichroism spectra were obtained on Foundation of China (no. 21372165 and 21321061), State Key
JASCO V-650 and JASCO J-A500-150 spectrometers, respectively. Laboratory of Polymer Materials Engineering, grant no.
DLS measurements were performed using
a Zetasizer sklpme2014-2-06, Comprehensive Training Platform of Special-
Nano S90 (Malvern Instruments, Worcestershire, UK) ized Laboratory, College of Chemistry, Sichuan University and
instrument.
Japan Society for the Promotion of Science (no. 21245011 and
26620030).
Materials
CNN was obtained from Hayashibara Co., Inc. Cyclooctene was
purchased from Tokyo Chemical Industries and used after
fractional distillation. Other chemicals were commercially
available and used without further purification.
Notes and references
1 Y. Inoue and V. Ramamurthy, Chiral Photochemistry,
Dekker, New York, 2004.
CNN-NSs 6 and 7 were synthesized and purified according
to essentially the same procedures as that employed for the
preparation of CD-NSs 3–5.30 Thus, 2.0 g of CNN (dried at
100 °C in an oven, until a constant weight was reached) were
dissolved in 8 mL of DMSO with continuous stirring, to which
were added 2 mL of triethylamine and, 5 min later, 1.345 g
2 C. Yang and Y. Inoue, Chem. Soc. Rev., 2014, 43, 4123.
3 A. G. Griesbeck and U. J. Meierhenrich, Angew. Chem., Int.
Ed., 2002, 41, 3147.
4 C. Müller and T. Bach, Aust. J. Chem., 2008, 61, 557.
5 G. S. Hammond and R. S. Cole, J. Am. Chem. Soc., 1965, 87,
3256.
(for 6) or 2.691
g
(for 7) of pyromellitic dianhydride.
6 Y. Inoue, T. Yokoyama, N. Yamasaki and A. Tai, Nature,
1989, 341, 225.
7 Y. Inoue, H. Ikeda, M. Kaneda, T. Sumimura,
S. R. L. Everitt and T. Wada, J. Am. Chem. Soc., 2000, 122,
406.
The mixture was allowed to react at room temperature,
while the gelation began in a few seconds after the addition
of the crosslinker. After 24 h, the crosslinking appeared to
be complete, affording a monolithic block, which was ground
in a mortar, washed with deionized water and rinsed with
8 A. Bauer, F. Westkaemper, S. Grimme and T. Bach, Nature,
2005, 436, 1139.
acetone in
a Buchner funnel with an aspirator. The
polymer residue was air-dried and extracted with acetone for
9 R. Brimioulle and T. Bach, Science, 2013, 342, 840.
14 h in a Soxhlet apparatus to give CNN-NS 6 (4.3 g, 97% yield) 10 J. Sivaguru, A. Natarajan, L. S. Kaanumalle, J. Shailaja,
or 7 (5.8 g, 96% yield) as a white powder. These samples,
which were practically insoluble (but swelled) in water and
S. Uppili, A. Joy and V. Ramamurthy, Acc. Chem. Res., 2003,
36, 509.
many organic solvents, gave the satisfactory elemental ana- 11 H. Saito, T. Mori, T. Wada and Y. Inoue, J. Am. Chem. Soc.,
lyses, ATR-FTIR spectra, and thermogravimetric analyses
2004, 126, 1900.
shown in ESI.†
12 C. Yang, Chin. Chem. Lett., 2013, 24, 437.
13 (a) Y. Inoue, S. F. Dong, K. Yamamoto, L.-H. Tong,
H. Tsuneishi, T. Hakushi and A. Tai, J. Am. Chem. Soc.,
1995, 113, 2793; (b) C. Yang, T. Mori, Y. Origane, Y. H. Ko,
N. Selvapalam, K. Kim and Y. Inoue, J. Am. Chem. Soc.,
2008, 130, 8574.
Photolysis
A sample solution containing 1Z and CNN-NS in a quartz cell
was placed on a UNISOKU CoolSpek USP-203CD cryostat,
purged with nitrogen gas, and irradiated at a given tempera- 14 C. Yang, C. Ke, W. Liang, G. Fukuhara, T. Mori, Y. Liu and
ture at 254 nm with a 20 W low pressure mercury lamp Y. Inoue, J. Am. Chem. Soc., 2011, 133, 13786.
(UVL20PH-6). No appreciable phase change or transition was 15 J. Yao, Z. Yan, W. Wu, C. Yang, M. Nishijima, G. Fukuhara,
observed after irradiation. The photolyzed sample was added T. Mori and Y. Inoue, J. Am. Chem. Soc., 2014, 136, 6916.
to 10% aqueous KOH solution (2 mL), and the resulting 16 Q. Wang, C. Yang, C. Ke, G. Fukuhara, T. Mori, Y. Liu and
mixture was extracted with pentane (1 mL). An aliquot of the Y. Inoue, Chem. Commun., 2011, 47, 6849.
pentane extract was analyzed by gas chromatography (GC) on a 17 G.-H. Liao, L. Luo, H.-X. Xu, X.-L. Wu, L. Lei, C.-H. Tung
Shimadzu CBP-20 (PEG) column for the E/Z ratio. To deter- and L.-Z. Wu, J. Org. Chem., 2008, 73, 7345.
mine the ee value, an aqueous silver nitrate solution (2 mL) 18 L. Luo, S.-F. Cheng, B. Chen, C.-H. Tung and L.-Z. Wu,
was added to the organic extract at 0 °C to give a stable Ag+
Langmuir, 2010, 26, 782.
complex with 1E. The aqueous solution of the [Ag+ 1E] 19 A. Joy, S. Uppili, M. R. Netherton, J. R. Scheffer and
complex was washed twice with pentane (1 mL) and added to a V. Ramamurthy, J. Am. Chem. Soc., 2000, 122, 728.
28% aqueous ammonia solution at 0 °C to liberate 1E. Finally, 20 J. Sivaguru, T. Shichi and V. Ramamurthy, Org. Lett., 2002,
the liberated cyclooctene was extracted with pentane (0.3 mL), 4, 4221.
and the pentane extract was analyzed by GC on a Supelco 21 J. Sivaguru, S. Jockusch, N. J. Turro and V. Ramamurthy,
β-DEX 225 column.
Photochem. Photobiol. Sci., 2003, 2, 1101.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
22 S. Asaoka, T. Wada and Y. Inoue, J. Am. Chem. Soc., 2003, 32 W. Liang, C. Yang, M. Nishijima, G. Fukuhara, T. Mori,
125, 3008.
A. Mele, F. Castiglione, F. Caldera, F. Trotta and Y. Inoue,
Beilstein J. Org. Chem., 2012, 8, 1305.
33 R. Lu, C. Yang, Y. Cao, Z. Wang, T. Wada, W. Jiao, T. Mori
and Y. Inoue, J. Org. Chem., 2008, 7695.
34 R. Lu, C. Yang, Y. Cao, Z. Wang, T. Wada, W. Jiao, T. Mori
and Y. Inoue, Chem. Commun., 2008, 374.
35 G. Fukuhara, T. Mori, T. Wada and Y. Inoue, Chem.
Commun., 2005, 4199.
23 M. Nishijima, T. Wada, T. Mori, T. C. S. Pace, C. Bohne and
Y. Inoue, J. Am. Chem. Soc., 2007, 129, 3478.
24 J. Mizoguchi, Y. Kawanami, T. Wada, K. Kodama, K. Anzai,
T. Yanagi and Y. Inoue, Org. Lett., 2006, 8, 6051.
25 T. Bach, H. Bergmann, B. Grosch and K. Harms, J. Am.
Chem. Soc., 2002, 124, 7982.
26 C. Müller, A. Bauer, M. M. Maturi, M. C. Cuquerella,
M. A. Miranda and T. Bach, J. Am. Chem. Soc., 2011, 133, 16689. 36 P. Biely, G. L. Cote and A. Burgess-Cassler, Eur. J. Biochem.,
27 A. Dawn, T. Shiraki, S. Haraguchi, H. Sato, K. Sada and
1994, 226, 633.
S. Shinkai, Chem. – Eur. J., 2010, 16, 3676.
37 G. L. Cote and P. Biely, Eur. J. Biochem., 1994, 226, 641.
28 M. Shirakawa, N. Fujita, T. Tani, K. Kaneko and S. Shinkai, 38 T. Nishimoto, H. Aga, K. Mukai, T. Hashimoto,
Chem. Commun., 2005, 4149.
H. Watanabe, M. Kubota, S. Fukuda, M. Kurimoto and
29 Y. Ishida, Y. Kai, S.-y. Kato, A. Misawa, S. Amano,
Y. Tsujisaka, Biosci. Biotech. Biochem., 2002, 66, 1806.
Y. Matsuoka and K. Saigo, Angew. Chem., Int. Ed., 2008, 47, 39 H. Aga, T. Nishimoto, M. Kuniyoshi, K. Maruta,
8241.
H. Yamashita, T. Higashiyama, T. Nakada, M. Kubota,
S. Fukuda, M. Kurimoto and Y. Tsujisaka;, J. Biosci. Bioeng.,
2003, 95, 215.
30 W. Liang, C. Yang, D. Zhou, H. Haneoka, M. Nishijima,
G. Fukuhara, T. Mori, F. Castiglione, A. Mele, F. Caldera,
F. Trotta and Y. Inoue, Chem. Commun., 2013, 49, 3510.
31 C. Yang, T. Mori, T. Wada and Y. Inoue, New J. Chem., 2007,
31, 697.
40 C. Yang, W. Liang, M. Nishijima, G. Fukuhara, T. Mori,
H. Hiramatsu, Y. Dan-oh, K. Tsujimoto and Y. Inoue, Chir-
ality, 2012, 24, 921.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015