Creation of hierarchical carbon nanotube assemblies through alternative packing of complementary semi-artificial β-1,3-glucan/carbon nanotube composites

Article abstract of DOI:10.1002/chem.200701205

Much attention has been focused on exploiting novel strategies for the creation of hierarchical polymer assemblies by the control of the assembling number or the relative location among neighboring polymers. We here propose a novel strategy toward the creation of "hierarchical" single-walled carbon nanotube (SWNT) architectures by utilizing SWNT composites with cationic or anionic complementary semiartificial β-1,3-glucans as "building blocks". These β-1,3-glucans are known to wrap SWNTs helically, to create one-dimensional superstructural composites. If the cationic composite is neutralized by an anionic composite, a well ordered SWNT-based sheet structure was created. Transmission electron microscopy (TEM) observation revealed that this sheet structure is composed of highly-ordered fibrous assemblies of SWNTs. This suggests that the cationic and anionic composites are tightly packed through electrostatic interactions. Moreover, both of the final assembly structures are readily tunable by adjusting the cation/anion ratio. The self-assembling modulation of functional polymers is associated with the progress in ultimate nanotechnologies, thus enabling us to create numerous functional nanomaterials. We believe, therefore, that the present system will extend the frontier of SWNT research to assembly chemistry including "hierarchical" superstructures.

Full text of DOI:10.1002/chem.200701205

DOI: 10.1002/chem.200701205
Creation of Hierarchical Carbon Nanotube Assemblies through Alternative
Packing of Complementary Semi-Artificial b-1,3-Glucan/Carbon Nanotube
Composites
[
a, b]
[a]
[c]
[a, d]
Munenori Numata,
Kouta Sugikawa, Kenji Kaneko, and Seiji Shinkai*
Abstract: Much attention has been
focused on exploiting novel strategies
for the creation of hierarchical polymer
assemblies by the control of the assem-
one-dimensional superstructural com-
posites. If the cationic composite is
neutralized by an anionic composite, a
well ordered SWNT-based sheet struc-
ture was created. Transmission electron
microscopy (TEM) observation re-
vealed that this sheet structure is com-
posed of highly-ordered fibrous assem-
blies of SWNTs. This suggests that the
cationic and anionic composites are
tightly packed through electrostatic in-
teractions. Moreover, both of the final
assembly structures are readily tunable
by adjusting the cation/anion ratio. The
self-assembling modulation of function-
al polymers is associated with the prog-
ress in ultimate nanotechnologies, thus
enabling us to create numerous func-
tional nanomaterials. We believe,
therefore, that the present system will
extend the frontier of SWNT research
to assembly chemistry including “hier-
archical” superstructures.
ACHTREUNG
ACHTREUNG
bling number or the relative location
among neighboring polymers. We here
propose a novel strategy toward the
creation of “hierarchical” single-walled
carbon nanotube (SWNT) architectures
by utilizing SWNT composites with cat-
ionic or anionic complementary semi-
artificial b-1,3-glucans as “building
blocks”. These b-1,3-glucans are known
to wrap SWNTs helically, to create
Keywords: carbon nanotubes
·
inclusion compounds · polysaccha-
rides · self-assembly
Introduction
plications as fundamental nanomaterials. One can easily
imagine, however, that this approach is much more difficult
than their creation by using functional low molecular-weight
compounds. So far, alignment of functional polymers such
as p-conjugated polymers has been achieved through chemi-
Creation of highly-ordered assemblies by using functional
polymers as building blocks, which would lead to novel
chemical and physical properties depending on their assem-
bling modes, is of great concern, owing to their potential ap-
[1]
cal or physical techniques that utilize liquid crystal phases,
[2]
patterned surfaces, and mechanical shearing of polymer
[3]
films. These strategies have been successful in preparing
ordered polymer assemblies on a mesoscopic level. On the
other hand, tuning physicochemical properties of polymer
assemblies in nanoscale, for which individual polymers must
be manipulated during their self-assembling process, have
not yet been achieved successfully because of the lack of an
established strategy. So far, much attention has been focused
on exploiting novel strategies for the creation of hierarchical
polymer assemblies by the control of the assembling
number or the relative location among neighboring poly-
mers, and is expected to endow exceptional functionalities
in the assembled structures. The most expeditious way to
create such hierarchical polymer architectures in a bottom-
up manner is to utilize the self-assembling capabilities of
polymers, where functional polymers bearing molecular rec-
ognition groups act as building blocks in the organization
processes. Unlike the self-assembling system of small mole-
[
a] Dr. M. Numata, K. Sugikawa, Prof. S. Shinkai
Department of Chemistry and Biochemistry
Graduate School of Engineering, Kyushu University
Fukuoka 819–0395 (Japan)
Fax : (+81)92-802-2820
E-mail: seijitcm@mbox.nc.kyushu-u.ac.jp
[
b] Dr. M. Numata
Present Address: Department of Bioscience and Biotechnology
Faculty of Science and Engineering, Ritsumeikan University
Kusatsu, Shiga, 525–8577 (Japan)
[
c] Prof. K. Kaneko
Department of Material Science and Engineering
Graduate School of Engineering, Kyushu University
Fukuoka 819–0395 (Japan)
[
d] Prof. S. Shinkai
Center for Future Chemistry, Kyushu University
Fukuoka 819–0395 (Japan)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
2398
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2398 – 2404

FULL PAPER
[4]
cules, only a few attempts
have been reported for the
creation of such hierarchical
[5]
architectures from polymers.
The difficulty in the polymeric
system arises from how one
can introduce self-assembling
capabilities into
a polymer
backbone without losing its in-
herent functionality and from
how to assemble the polymers
through specific interpolymer
interactions without the influ-
ence of the nonspecific bun-
dling nature.
We recently demonstrated
that natural polysaccharide b-
1,3-glucans act as a one-dimen-
sional host that can include the
hydrophobic guest polymers,
such as single-walled carbon
[
6]
nanotube (SWNT),
poly-
poly(thio-
into the tubular
Figure 1. A) Proposed concept for creating the hierarchical SWNT architecture form the one-dimensional
building blocks through the electrostatic interaction. All processes including wrapping of SWNTs by curdlan
and self-assembling of the resultant composites proceed in a supramolecular manner. B) Repeating unit of nat-
ural curdlan and two kinds of semi-artificial curdlan with cationic and anionic moieties, respectively.
[7]
A
C
H
T
R
E
U
N
G
(aniline),
and
[8]
phene),
hollow constructed by the heli-
cal structure inherent to b-1,3-
[
9]
glucans, to result in the creation of water-soluble one-di-
the Experimental Section). Figure 2 shows the Vis-NIR
spectra of the aqueous solutions of the composites prepared.
The Vis-NIR spectra displays characteristic absorption
bands that we can reasonably assign to the SWNTs and veri-
fies that SWNTs are solubilized into water by wrapping
[10]
mensional nanocomposites. The system has several lines
of unique advantage: 1) when chemically modified b-1,3-glu-
cans are used as one-dimensional hosts, the exterior surface
of the resultant nanocomposites can be utilized as an inter-
action site for the construction of supramolecular architec-
tures and, 2) the strong interpolymer interactions among
guest polymers are perfectly suppressed by the wrapping
effect of b-1,3-glucans, which insulates one piece of guest
polymer to maintain its original functionality. Therefore, if a
b-1,3-glucan, which incorporates molecular recognition
groups on its exterior surface, wraps a guest polymer, the re-
sultant composite would acquire the potential to self-assem-
ble and, in turn, create a specific hierarchical architecture.
Accordingly, we synthesized two new kinds of complementa-
ry semi-artificial curdlans (one kind of b-1,3-glucans), i.e.,
+
À
them with cur-N or cur-SO . Judging from the peak sharp-
3
ness of the spectra, the SWNTs are included into the helical
+
À
structures constructed by cur-N or cur-SO3 in a 1:1 ratio.
It is well-known that a SWNT can be entrapped into the cy-
lindrical micellar structure created by SDS (sodium dodecyl
+
ammonium-modified curdlan (cur-N ) and sulfonate-modi-
fied curdlan (cur-SO ),
À
[11]
expecting that the mixture of
3
these two composites in an appropriate ratio results in the
creation of a hierarchical architecture, owing to the electro-
static interaction. To estimate the feasibility of this idea, we
chose single-walled carbon nanotube (SWNT) as a guest po-
[12]
lymer, which is now regarded as one of the most attrac-
[13]
tive future materials in nanotechnology (Figure 1).
Results and Discussion
Preparation of complementary semi-artificial curdlan/
+
Figure 2. Vis-NIR spectra of cur-N /SWNT composite (blue line), cur-
SWNT composites: Firstly, we prepared SWNT composites
À
SO
3
/SWNT composite (red line), SDS/SWNT (dotted black line), and
O, 1.0 cm cell, room temperature.
+
À
utilizing cur-N or cur-SO3 as one-dimensional hosts (see
the sheet-like structure (green line): D
2
Chem. Eur. J. 2008, 14, 2398 – 2404
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2399

S. Shinkai et al.
[
14]
sulfate). Here, one may expect that the spectral feature of
the obtained composite can be directly compared with that
of the SDS/SWNT mixture. One can see from Figure 2 that
the spectral features as well as the peak sharpness of both
These AFM images are basically very similar to those previ-
ously observed for natural b-1,3-glucan/SWNT composites.
[6]
Furthermore, in the present system, the amount of curdlan
wrapping on SWNTs can be estimated by using a phenol/
sulfuric-acid reaction system that indicates the curdlan/
+
À
composites, i.e., cur-N /SWNT composite and cur-SO /
SWNT composite, are similar to that of the SDS/SWNT
mixture. This indicates that individual SWNTs are actually
3
+
SWNT ratio can be estimated to be 0.33 (w/w) for cur-N /
À
SWNT and 0.46 (w/w) for cur-SO /SWNT. From these re-
3
+
À
+
À
wrapped by cur-N or cur-SO . Atomic force microscopy
sults, it is worthy to mention here that cur-N and cur-SO3
3
(
AFM) observations of the composites support this view
have the similar wrapping capability for SWNTs, resulting in
the similar helical pattern on SWNTs. The findings suggest
more quantitatively; indeed fibrous composites (1.5–1.7 nm
height) can be observed after casting them on mica surface
+
À
that cur-N /SWNT composite and cur-SO /SWNT compo-
3
(
Figure 3A,B, and S1 in the Supporting Information).
site may be used as “complementary” one-dimensional
building blocks to create higher-order hierarchical self-
assembled architectures of SWNTs.
Taking the diameter of the SWNT used into consideration,
the 1.6 nm height (average) implies that both composites
consist, mostly, of one piece of SWNT. Furthermore, it can
be seen from the magnified image (Figure 3C) that the peri-
odical helical structure is constructed on the composite sur-
face. This finding indicates that SWNTs are included in the
[6]
Self-assembly of the complementary composites: In the
present system, the final structures as well as self-assembling
+
/
behaviors are predominantly governed by the initial cur-N
+
À
À
hydrophobic cavity constructed by cur-N or cur-SO .
cur-SO3 ratio in the mixture. To create well-developed two-
or three-dimensional hierarchical architectures from these
one-dimensional composites, the cationic charges on the cur-
3
+
N /SWNT composite must be neutralized by the anionic
À
charges on cur-SO /SWNT composite. Based on this work-
3
+
ing hypothesis, we prepared first the two respective cur-N /
À
SWNT and cur-SO /SWNT composite solutions to contain
3
the same concentration of SWNT and by using the same
volume. To monitor the neutralization process, the zeta-po-
tential values were measured. The zeta-potential value of an
+
aqueous solution containing cur-N /SWNT composite was
estimated to be +48.9 mV, whereas an aqueous solution
À
containing cur-SO /SWNT composite showed À49.5 mV.
3
Both solutions were very stable without precipitate forma-
tion for several weeks at room temperature. Once these two
solutions were mixed in the same volume under the very di-
luted condition (for the experimental procedures, see the
Supporting Information), the zeta-potential value of the re-
sultant mixture showed À0.53 mV without accompanying
precipitate formation, thus indicating that the potential
charges on these composites are almost neutralized to give a
self-assembling composite through the electrostatic interac-
tion.
To confirm the type of self-assembling architectures creat-
ed during this neutralization process, we cast the resultant
aqueous mixture on a substrate and observed the morpholo-
gies by various microscopes (AFM images are shown in Fig-
ure 3D). As shown in Figure 3D, we observed a well-devel-
oped sheet structure with micrometer-scale length, which is
entirely different from the very fine fibrous structures ob-
+
À
served for individual cur-N /SWNT and cur-SO /SWNT
3
composites (Figure 3A,B). The scan profile of the sheet
structure also revealed that several thin layers were piled up
to form the sheet-like structure, the surface of which is
almost flat in the micrometer range (Figure 3E). We could
also confirm the creation of the similar sheet structure by
scanning electron microscopy (SEM) (see the Supporting In-
formation, Figure S2). The presence of SWNTs in the creat-
ed sheet structure was evidenced by Raman spectroscopic
+
3À
Figure 3. AFM images of A) cur-N /SWNT composite and B) cur-SO
/
SWNT composite, respectively. C) Magnified AFM image of (B).
D) AFM image of the sheet-like structure after mixing cur-N /SWNT
/SWNT composite. E) Height profile of the
sheet-like structure: the AFM tip was scanned along the black line. In
this AFM image, the thickness of each thin layer is estimated to be
+
À
composite and cur-SO
3
ꢀ
3.5 nm.
2400
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2398 – 2404

Carbon Nanotube Assemblies
FULL PAPER
measurements; that is, the sheet structure provides charac-
thin layer. Furthermore, in other sheet structures (Fig-
ure 4E), two kinds of dark layers appear regularly; one dark
layer shows at ꢀ10 nm intervals and another dark layer
shows at ꢀ2 nm intervals. As seen in Figure 4F, the perio-
dicity of the narrow dark layer can be quantitatively esti-
mated to be 1.6 nm by utilizing the Fourier-filtered image,
the value of which is consistent with that estimated from
Figure 4C. We consider that the wide dark layers with
ꢀ10 nm intervals (Figure 4E) are ascribed to a moirØ pat-
tern created by the overlap of a few sheets. Accordingly, the
results support the view that, in the present system, a thin
layer consisting of highly-ordered fibrous assemblies with
1.6 nm intervals is a fundamental structure, which is able to
regularly stack on one another, leading to the formation of
the sheet structures. Notably, the Vis-NIR spectral features
of this sheet-like architecture are almost the same as those
À1
À1
teristic SWNT peaks at 262 cm and 1592 cm (see the
Supporting Information, Figure S3), which are assignable to
the radial breathing mode (RBM) and G-band of SWNT,
A
C
H
T
R
E
U
N
G
respectively.
TEM is a powerful tool to study how SWNTs are ar-
ranged in the obtained sheet structure. Figure 4 shows sever-
al TEM images obtained from different regions on a speci-
men supporting grid. In the lower-magnified TEM image
(
Figure 4A), we could recognize a sheet structure extending
to several micrometer length, the size of which is almost
consistent with that observed in AFM and SEM images. In
addition, it can be seen from Figure 4B that the sheet struc-
ture is composed of highly-ordered fibrous assemblies. Since
the TEM images were obtained without staining, the ob-
served contrast should arise from the presence of SWNTs.
The electron diffraction pattern (inset in Figure 4B) reveals
that the fibrous assembly has some crystalline nature, sug-
gesting that cationic and anionic composites are tightly
packed through the electrostatic interaction. The regular
alignment of the composite in the thin layer was supported
by a polarization microscope imaging method (see the Sup-
porting Information, Figure S4). This view is also supported
+
À
of cur-N /SWNT composite and cur-SO /SWNT composite
3
+
as shown in Figure 2. This fact supports the view that cur-N
and cur-SO3 act as a sheath for SWNTs, so that the direct
À
interaction among SWNTs can be suppressed even in the
sheet-like structure, making the practical application as elec-
tronic and photoactive devices possible. The EDX (energy-
dispersive X-ray spectroscopy) analysis provides direct evi-
+
+
by the magnified TEM image. In Figure 4C, cur-N /SWNT
dence that the sheet-like structure is made up of cur-N /
À
À
composite and cur-SO /SWNT composite can be recog-
SWNT composite and cur-SO /SWNT composite. In Fig-
3
3
nized as a dark layer with very regular intervals. The perio-
dicity of the dark layer is estimated to be ꢀ2 nm, which is
almost consistent with the diameter of the individual com-
posite obtained by the AFM height profile. The finding im-
ure 4D, two characteristic peaks are detected at 0.39 eV and
2.30 eV, which are assignable to nitrogen and sulfur arising
+
À
from cur-N and cur-SO , respectively. It is still unclear, at
3
present, why the 2-D sheet structure, but not the 3-D
+
À
+
plies that cur-N /SWNT composite and cur-SO /SWNT
packed structure, is formed from cur-N /SWNT and cur-
3
À
composite interact with their exterior surfaces to form the
SO /SWNT composites. However, dynamic light scattering
3
Figure 4. TEM image of A) sheet-like structure (low magnification), B) and C) magnified images of the thin layer (inset: electron diffraction pattern ob-
tained from the sheet). D) Elemental analysis of the sheet-like structure based on EDX. The spectrum was corrected from the red-square in (B).
E) Magnified TEM image of the sheet-like structure containing several thin layers. F) Fourier translation image of (D) and extracted periodical patterns.
Chem. Eur. J. 2008, 14, 2398 – 2404
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2401

S. Shinkai et al.
(
DLS) measurements revealed that the self-assembling ag-
gregates, the size of which is not as large as the sheet-like
structures observed by TEM, are formed at least in the ini-
tial aqueous solution. This self-assembling aggregate would
be further organized into the sheet-like structure during
slow evaporation of water on a substrate surface. When the
two kinds of solutions were mixed at high concentration
À1
(
above 0.79 mgmL ), black precipitate was formed. These
results suggest the view that slow-organization process of
the composites would be indispensable for the creation of
the sheet-like structure. It is worthy to emphasize through
the present work that the hierarchical SWNT architecture
can be finally created by just mixing two kinds of aqueous
+
À
solution containing cur-N /SWNT composite and cur-SO /
3
SWNT composite in the appropriate ratio.
Attempt toward the creation of SWNT architectures with
controlled size and assembling number: It is generally ac-
cepted in the molecular assembling systems that the mixture
of the oppositely-charged small molecules tends to result in
nonspecific irregular assemblies through electrostatic inter-
actions, but when either cationic or anionic polymer exists
excessively, specific regular structures with the well-control-
[4]
led size and assembling number can be created. This con-
cept may be applicable to the present polymer assembling
system; the size of the hierarchical architecture as well as
the assembly composite number would be controllable by
+
À
just changing the [cur-N /SWNT]/[cur-SO /SWNT] feed
3
ratio. To test this intriguing idea, we mixed an aqueous cur-
+
N /SWNT composite solution with an excess amount of an
À
aqueous cur-SO /SWNT composite solution, adjusting the
3
+
3À
Figure 5. TEM images of A) cur-N /SWNT composite and B) cur-SO
/
+
À
[
cur-N /SWNT]/[cur-SO /SWNT] ratio to 1:5. Upon mixing
these solutions, it was confirmed that there was no precipi-
tate formation. The DLS analysis of the mixture revealed
3
SWNT composite. C) Fibrous bundle structure containing the limited
number of SWNTs ([cur-N /SWNT]/[cur-SO /SWNT]=1/5). D) Magni-
fied TEM image of (C). E) Larger bundle structure ([cur-N /SWNT]/
[cur-SO3 /SWNT]=1/3). F) Magnified TEM image of (E) (inset: extract-
+
À
3
+
À
+
À
that cur-N /SWNT composite and cur-SO /SWNT compo-
3
ed periodical patterns obtained along the red line).
site form a self-assembly structure, the size of which is
larger than those of individual composites. Furthermore, the
zeta-potential value of the resultant aqueous solution indi-
cates that the obtained assembly structure has the highly
to 1:3, the diameter of the bundle structure became larger,
accompanied by the increase in the zeta-potential from
À33.2 mV to À17.8 mV (Figure 5E,F). In Figure 5F, one can
see that the bundle structure is also composed of highly-or-
dered fibrous assemblies as seen in Figure 4C. The periodici-
ty of the layer is estimated to be ꢀ2 nm, which is almost
consistent with that observed for the sheet-like structure.
These results substantiate our hypothesis that the self-as-
sembling hierarchical architecture is predictable and control-
+
anionic surface (À33.2 mV), suggesting that cur-N /SWNT
À
composite is surrounded by cur-SO /SWNT composite to
3
give a bundle structure. In such a case, one may imagine
that electrostatic repulsion among the created bundle assem-
blies would suppress undesired further aggregate formation.
The TEM images indeed support this view (Figure 5C,D)
and the many fibrous SWNTs bundles (10–20 nm diameters)
can be recognized, which are larger than those of individual
composites shown in Figure 5A,B. Taking the narrow distri-
bution in diameter and the anionic surface into consider-
ation, one- or two-pieces of cur-N /SWNT composite would
be surrounded by several cur-SO /SWNT composites. To
further extend these findings, we prepared the mixtures
under various [cur-N /SWNT]/[cur-SO /SWNT] ratios. In
+
lable by tuning the ratio of cur-N /SWNT composite and
À
cur-SO /SWNT composite.
3
+
À
Conclusion
3
+
À
In summary, we have demonstrated that the creation of hier-
archical SWNT architectures can be easily achieved through
the electrostatic interactions between cur-N /SWNT compo-
sites and cur-SO /SWNT composites. The resultant hier-
archical architectures as well as the self-assembling process-
3
+
À
the [cur-N /SWNT]/[cur-SO /SWNT] ratios from 1:5 to
:10, the mixture gave the fibrous bundle structure similar
to those in Figure 5C,D. On the other hand, when the [cur-
N /SWNT]/[cur-SO /SWNT] ratio was increased from 1:5
3
+
1
À
3
+
À
3
2402
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2398 – 2404

Carbon Nanotube Assemblies
FULL PAPER
+
es are strongly affected by the initial [cur-N /SWNT]/[cur-
258C)d=138.06, 132.91, 105.2, 86.09, 75.96, 73.78, 71.23, 62.47, 55.35,
À
51.78 ppm.
SO /SWNT] ratio; that is, upon mixing the same concentra-
3
3À
Synthesis of cur-SO : to a mixed solvent containing MeOH (7 mL) and
tion of composites, the well-developed sheet structure is
formed, whereas an excess amount of one composite over
the other results in the bundle structure with the uniform di-
ameter. The fibrous bundle structures have some charge on
their exterior surface, so they are utilizable as building
blocks for further organization. As curdlan can entrap a va-
water (7 mL), propargyl bromide (2.0 mL) and Na
added. After stirring for 7 h at 658C, MeOH (120 mL) was added to the
reaction mixture. An excess amount of Na SO was removed by filtration
2 3
SO (2.9 g) were
2
3
followed by concentration to give a yellow solution. The resultant MeOH
solution was washed with acetone to give 1-sulfo-2,3-propyne sodium salt
as yellow precipitate (96%). The click reaction with 6-azido-6-deoxycur-
dlan was carried out according to the same procedure as described
above.
[10l]
riety of guest polymers in an induced-fit manner,
the
present concept is more broadly applicable to the design of
hierarchical architectures from other functional polymers if
they are appropriately included in b-1,3-glucan polysaccha-
rides. We believe, therefore, that considering the serious dif-
ficulties in the creation of hierarchical architectures from
synthetic polymers, the present system can open new paths
to accelerate development of the polymer assembly systems
and can extend the frontier of polymer-based functional
nanomaterials.
3À
1
Cur-SO : M
w
=34300, M
w
/M
n
=1.4. H NMR (600 MHz, 10 mm NaOD/
1
D O, 60 8C) d=8.14 (br, 1H, triazole-H), 4.88(ibr, 1H, H ), 4.69(ibr, 2H,
2
5
6
3
CH
H, H ), 3.30 ppm (ibr, 1H, H ); C NMR (125 MHz, 10 mm NaOD/
O, 25 8C) d=129.74, 104.75, 85.58, 75.98, 72.47, 71.54, 53.58,
0.43 ppm.
2
S), 4.31(ibr, 1H, H ), 3.85(ibr, 2H, H ), 3.62(ibr, 1H, H 9, 3.43(ibr,
2
4
13
1
D
2
5
+
3À
Preparation of cur-N /SWNT and cur-SO /SWNT composites: SWNTs
1.3 mg) were mixed with the aqueous solution containing cur-N or cur-
+
(
SO
À
À1
3
(1.0 mL, 5.0 mgmL ) and dispersed by sonication for 50 min using
a probe type sonicator with the sample immersed in a water bath. To
remove an excess amount of curdlan, the obtained homogeneous black
solution was subjected to gel-column chromatography (Sephadex, G-100,
eluted with water).
Experimental Section
Confirmation of the composite purity: From the molecular weight of
+
À
used curdlan, the fiber length of cur-N and cur-SO
3
can be estimated
General: Vis-NIR spectroscopic studies were performed by using a SHI-
to be less than 0.2 mm, whereas that of SWNT is estimated to be more
than 0.2 mm. Taking this fact into consideration, the purity of the ob-
tained composite solution was confirmed according to the following pro-
cedure: The composite solution was subjected to filtration by using a
PTFE membrane filter (pore size: 0.2 mm), expecting that the uncom-
MADZU
UV-3100
spectrophotometer.
Raman
spectra
of
c-SWNTs/polysaccharide composites were obtained by using a JASCO
NRS-2000 laser Raman spectrometer (Ar laser, 514 nm). DLS data and
zeta-potential values were obtained by using Sysmex Zetasizer Nano.
Transmission electron microscopy (TEM) and high-resolution TEM
+
À
+
plexed cur-N or cur-SO
3
would separate into the filtrate. Since cur-N
À
(
HRTEM) images were acquired by using a JEOL TEM-2010 (accelerate
and cur-SO
3
have a characteristic absorption band at around 230 nm
voltage 120 kV) and a TECNAI-20, FEI (accelerate voltage 200 kV), re-
spectively. The c-SWNTs/s-SPG solution was placed on a copper TEM
grid with a holey carbon support film. The TEM grid was dried under re-
duced pressure for 6 h before TEM observation. Energy dispersive X-ray
spectroscopy (EDX) spectra and EDX line scan profiles were obtained
using a TECNAI-20, FEI. Atomic force microscopy (AFM) images were
acquired in air using a NanoScope IIIa (tapping mode). The sample was
cast on mica and dried for 6 h under reduced pressure before AFM
that can be ascribed to the triazole unit, the existence of uncomplexed
+
À
cur-N or cur-SO
3
in the filtrate can be easily detected by UV-vis spec-
tra. Consequently, the UV-vis spectra thus obtained revealed that the
+
composite solution did not contain any uncomplexed cur-N and cur-
À
SO
3
, suggesting that the gel-column chromatography effectively works
as a purification tool in the present system.
Phenol/sulfuric-acid reaction: Firstly, calibration curve was created by
using mannose as a standard saccharide. An aqueous solution containing
A
C
H
T
R
E
U
N
G
observation.
5
% (w/v) phenol was then added to 200 mL of the mannose solutions. Im-
Materials: Curdlan used here was purchased from Tokyo Kasei Kougyou.
SDS was purchased from Wako Pure Chemical Industries. Single-walled
carbon nanotubes (SWNTs) produced by means of the HiPco (high-pres-
sure decomposition of carbon monoxide) process and were obtained
from Carbon Nanotechnologies (lot no: po304). Chemical modification
on native curdlan was carried out basically according to the preceding
mediately after addition of 1.0 mL of sulfuric acid to the mixture, the re-
sultant solution assumed an intense yellow color. To complete the colora-
tion reaction, the obtained solution was kept for 40 min at room temper-
ature. The absorption maxima at 490 nm were plotted as function of sev-
eral mannose concentrations. The composite solutions containing un-
known amount of curdlans were treated with phenol followed by sulfuric
acid according to the same procedure. From the calibration chart, the
amount of curdlan wrapping on SWNT was estimated.
[
11]
paper reported by us. Here, 6-azido-6-deoxycurdlan was synthesized as
+
À
a common intermediate for cur-N and cur-SO . The synthetic route is
3
shown in Scheme S1 (see the Supporting Information).
+
Synthesis of cur-N : Propargyl bromide (9.1 mmol) was dissolved in
5
0 mL of dry THF and cooled in an ice bath. To the resultant THF solu-
tion, trimethylamine (32 mmol) was added and the mixture was stirred
for 24 h. After removing solvent, the residue was washed with THF sev-
eral times followed by drying to give 1-trimethylammonium-2,3-propyne
chloride as yellow powder (yield 91%).
Acknowledgement
We thank Dr. M. Takeuchi for fruitful discussions and Ms. M. Fujita for
AFM measurements. This work was partially supported by the Japan
6
-Azido-6-deoxycurdlan (100 mg) was dissolved in 100 mL of DMSO. To
the resultant solution, 1-trimethylammonium-2,3-propyne chloride
250 mg), copper(II) bromide (10.0 mg), ascorbic acid (50 mg), and pro-
AHCTREUNG
(
pylamine (5.2 mL) were added and the mixture was stirred for 60 h at
room temperature. The reaction product was subjected to dialysis against
distilled water (MWCO 8000) followed by lyophilization and washing
ACHTREUNG
(no.18655048).
+
with MeOH to give the ammonium modified curdlan (cur-N ) in 96%
+
1
yield. Cur-N
6
:
M
w
=32000,
M
w
/M
n
=1.6. H NMR (600 MHz,
D
2
O,
[1] a) H. Goto, K. Akagi, H. Shirakawa, Synth. Met. 1997, 84, 373;
b) T. M. Long, T. M. Swager, J. Am. Chem. Soc. 2002, 124, 3826;
c) K. P. Fritz, G. D. Scholes, J. Phys. Chem. B 2004, 108, 10141.
[2] H. Xin, T. Woolley, J. Am. Chem. Soc. 2003, 125, 8710.
1
08C) d= 8.43 (br, 1H, triazole-H), 4.91(ibr, 1H, H ), 4.71(ibr, 2H,
3
6
5
6
CH
H
2
N), 3.85(ibr, 2H, H and H ), 3.66(ibr, 2H, H and H ), 3.40(ibr, 2H,
and H ), 3.18 ppm (br, 9H, NCH ); C NMR(i125 MHz, D O,
3 2
2
4
13
Chem. Eur. J. 2008, 14, 2398 – 2404
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2403

S. Shinkai et al.
[
3] N. Akima, Y. Iwasa, S. Brown, A. M. Barbour, J. Cao, J. L. Musfeldt,
H. Matsui, N. Toyota, M. Shiraishi, H. Shimoda, O. Zhou, Adv.
Mater. 2006, 18, 1166.
4] J.-M. Lehn, Supramoleculer Chemistry, VCH, New York, 1995.
5] a) Y. Kubo, Y. Kitada, R. Wakabayashi, T. Kishida, M. Ayabe, K.
Kaneko, M. Takeuchi, S. Shinkai, Angew. Chem. 2006, 118, 1578;
Angew. Chem. Int. Ed. 2006, 45, 1548; b) M. Ikeda, Y. Furusho, K.
Okoshi, S. Tanahara, K. Maeda, S. Nishino, T. Mori, E. Yashima,
Angew. Chem. 2006, 118, 6641; Angew. Chem. Int. Ed. 2006, 45,
Kato, M. Tanaka, Org. Lett. 2006, 8, 235. For self-assembled cyclo-
dextrin nanotubes, see: g) A. Harada, J. Li, M. Kamachi, Nature
1993, 364, 516; h) M. Ceccato, P. L. Nostro, C. Rossi, C. Bonechi, A.
Donati, P. Baglioni, J. Phys. Chem. B 1997, 101, 5094; i) K. Yoshida,
T. Shimomura, K. Ito, R. Hayakawa, Langmuir 1999, 15, 910; j) T.
Shimomura, T. Akai, T. Abe, K. Ito, J. Chem. Phys. 2002, 116, 1753;
k) H. Okumura, Y. Kawaguchi, A. Harada, Macromol. Rapid
Commun. 2002, 23, 781; l) Our recent review, see: K. Sakurai, K.
Uezu, M. Numata, T. Hasegawa, C. Li, K. Kaneko, S. Shinkai,
Chem. Commun. 2005, 4383.
[
[
6
491; c) K. K. Ewert, H. M. Evans, A. Zidovska, N. F. Bouxsein, A.
Ahmad, C. R. Safinya, J. Am. Chem. Soc. 2006, 128, 3998.
6] a) M. Numata, M. Asai, K. Kaneko, T. Hasegawa, N. Fujita, Y.
Kitada, K. Sakura, S. Shinkai, Chem. Lett. 2004, 33, 232; b) M.
Numata, M. Asai, K. Kaneko, A.-H. Bae, T. Hasegawa, K. Sakurai,
S. Shinkai, J. Am. Chem. Soc. 2005, 127, 5875.
7] M. Numata, T. Hasegawa, T. Fujisawa, K. Sakura, S. Shinkai, Org.
Lett. 2004, 6, 4447.
8] C. Li, M. Numata, A.-H. Bae, K. Sakurai, S. Shinkai, J. Am. Chem.
Soc. 2005, 127, 4548.
[11] T. Hasegawa, M. Umeda, M. Numata, C. Li, A.-H. Bae, T. Fujisawa,
S. Haraguchi, K. Sakurai, S. Shinkai, Carbohydr. Res. 2006, 341, 35.
[12] a) S. Iijima, Nature 1991, 354, 56; b) S. Iijima, T. Ichihashi, Nature
1993, 363, 603.
[13] For self-assembly of carbon nanotubes, see: a) H. Shimoda, S. J. Oh,
H. Z. Geng, R. J. Walker, X. B. Zhang, L. E. McNeil, O. Zhou, Adv.
Mater. 2002, 14, 899; b) R. Duggal, F. Hussain, M. Pasquali, Adv.
Mater. 2006, 18, 29; c) H. Ko, V. V. Tsukruk, Nano Lett. 2006, 6,
1443; d) S. Badaire, C. Zakri, M. Maugey, A. Derre, J. N. Barisci, G.
Wallance, P. Poulin, Adv. Mater. 2005, 17, 1673; e) A. Ishibashi, Y.
Yamaguchi, H. Murakami, N. Nakashima, Chem. Phys. Lett. 2006,
419, 574.
[
[
[
[
9] T. Yanaki, T. Norisuye, H. Fujita, Macromolecules 1980, 13, 1462.
[
10] Amylose can also act as a helical host and form inclusion complexes
with various functional guests, see: a) L. S. Choi, O.-K. Kim, Macro-
molecules 1998, 31, 9406; b) O.-K. Kim, J. Je, J. W. Baldwin, S. Kooi,
P. E. Pehrsson, L. J. Buckley, J. Am. Chem. Soc. 2003, 125, 4426; c) J.
Kadokawa, Y. Kaneko, H. Tagaya, K. Chiba, Chem. Commun. 2001,
[14] M. J. OꢁConnell, S. M. Bachilo, C. B. Huffman, V. C. Moore, M. S.
Strano, E. H. Haroz, K. L. Rialon, P. J. Boul, W. H. Noon, C. Kit-
trell, J. Ma, R. H. Hauge, R. B. Weisman, R. E. Smalley, Science,
2002, 297, 593.
4
2
2
59; d) T. Kida, T. Minabe, S. Okabe, M. Akashi, Chem. Commun.
007, 1559; e) T. Sanji, N. Kato, M. Kato, M. Tanaka, Angew. Chem.
005, 117, 7467; Angew. Chem. Int. Ed. 2005, 44, 7301; f) T. Sanji, N.
Received: August 2, 2007
Published online: January 17, 2008
2404
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2398 – 2404