Self-assembly and structural evolvement of polyoxometalate-anchored dendron complexes

Article abstract of DOI:10.1002/chem.201000198

A cationic dendritic molecule that has alkyl chains has been synthesized and employed to encapsulate anionic polyoxometalates through electrostatic interactions. The prepared surfactant-encapsulated polyoxometalate (SEP) complexes were used as building blocks to fabricate self-assemblies in solution and the solid state. Monodispersion, lamellar, and columnar assemblies of SEP complexes have been characterized in detail. With increasing the number of peripheral cationic dendrons on inorganic clusters, the SEPs undergo changes from globular assemblies to monodispersions in solution and from lamellar assemblies to hexagonal columnar structures in the solid state, depending on the amounts of cationic dendrons in the complexes. The structural evolvement was simulated through consideration of the size and shape of the cationic dendron and polyanionic clusters, and the experimental results are in good agreement with the interpretation of the simulations. The present research demonstrates a new kind of dendritic complex and provides a route for controlling their assembling states by simply alternating the number of cationic dendrons in the complexes.

Full text of DOI:10.1002/chem.201000198

DOI: 10.1002/chem.201000198
Self-Assembly and Structural Evolvement of Polyoxometalate-Anchored
Dendron Complexes
Yang Yang, Yizhan Wang, Haolong Li, Wen Li, and Lixin Wu*[a]
Abstract: A cationic dendritic molecule
that has alkyl chains has been synthe-
sized and employed to encapsulate
anionic polyoxometalates through elec-
trostatic interactions. The prepared sur-
factant-encapsulated polyoxometalate
(SEP) complexes were used as building
blocks to fabricate self-assemblies in
solution and the solid state. Monodis-
persion, lamellar, and columnar assem-
blies of SEP complexes have been
characterized in detail. With increasing
the number of peripheral cationic den-
drons on inorganic clusters, the SEPs
undergo changes from globular assem-
blies to monodispersions in solution
and from lamellar assemblies to hexag-
onal columnar structures in the solid
state, depending on the amounts of cat-
ionic dendrons in the complexes. The
structural evolvement was simulated
through consideration of the size and
shape of the cationic dendron and
polyanionic clusters, and the experi-
mental results are in good agreement
with the interpretation of the simula-
tions. The present research demon-
strates a new kind of dendritic complex
and provides a route for controlling
their assembling states by simply alter-
nating the number of cationic dendrons
in the complexes.
Keywords: dendrimers
dendritic complexes · polyoxomet-
alates · self-assembly · structural
evolvement
·
hybrid
A
A
Introduction
covalent bonds, such as hydrogen bond, host-guest recogni-
tion, coordination, and electrostatic interactions.[2a] Thus, the
ability to combine multicomponents in one dendritic unit
that can create new features in addition to the sum of the in-
dividual parts is of special interest.
Dendritic molecules represent a family of well-defined,
three-dimensional branched architectures and are widely
used for the construction of nanostructures and various self-
assemblies because their size, shape, flexibility, and surface
chemistry can be precisely controlled at the molecular
level.[1] Diverse self-assemblies composed of dendritic mole-
cules such as micelles, vesicles, tubes, and columnar struc-
tures have been developed through suitable molecular de-
signs. The unique properties of dendritic molecules, especial-
ly those combining functional units, exhibit potential appli-
cations in nanomaterials,[2] biomimic,[3] drug delivery,[4] and
catalysis.[5] To simplify synthetic procedures and particularly
to obtain functional dendritic systems, many components in-
cluding photochromic and luminescent groups, amino acids,
metal ions, and metal and/or semiconductor nanoparticles
have been grafted into dendritic architectures through non-
Polyoxometalates (POMs) are discrete, molecularly de-
fined inorganic metal–oxide clusters with structural variety
and interesting functional properties leading to rich applica-
tions in catalysis, optics, magnetics, and medicine.[6] More-
over, as intriguing inorganic building blocks with uniform
nanoscale size (one to several nanometers), versatile nega-
tive charges, and topologic morphologies, POMs have been
employed for the fabrication of organic–inorganic hybrid
materials.[7] Among the strategies for the modification of
POMs, a facile and simple approach currently used to fabri-
cate POM-based organic–inorganic hybrid complexes is to
replace the counterions of POMs by cationic surfactants, re-
sulting in surfactant-encapsulated POMs (SEPs).[8] An SEP
complex possessing a discrete supramolecular architecture
that has a hydrophobic shell and a hydrophilic POM core in
a well-defined composition can lead to increased stability of
inner POMs,[9] improved photophysical properties,[10] en-
hanced biological activity,[11] and can especially make POMs
soluble in weak and nonpolar organic solvents, which is very
useful for the fabrication of POM-based film devices.[8a]
Linear single- and double-chain cationic surfactants[12] have
[a] Y. Yang, Y. Wang, Dr. H. Li, Dr. W. Li, Prof. L. Wu
State Key Laboratory of Supramolecular Structure and Materials
College of Chemistry, Jilin University, Changchun 130012 (China)
Fax : (+86)431-8519-3421
E-mail: wulx@jlu.edu.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000198.
8062
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8062 – 8071

FULL PAPER
been employed to enwrap POMs, and the formed SEPs
were successfully applied in the self-assemblies. Most of
SEPs tend to assemble into lamellar structures in solution,
except in a very few cases of hexagonal or cubic structures
through using a special POM and a specially designed sur-
factant.[8a,13] It has been attempted to incorporate POMs
into dendritic architectures and the resulting complexes
were supposed to exhibit homogeneous catalysis for the ep-
oxidation of dienes.[14]
plexes in solution transformed from an aggregated state to a
monodisperse state with increasing the number of combined
cationic dendrons. Meanwhile, the complexes changed their
shapes from rodlike to disklike in the solid state, and as a
result the self-assembled structures varied from lamellar to
hexagonal columnar arrays. The proposed assembled struc-
tures were manipulated and displayed unusual agreement
with the experimental data.
In contrast to the reported results, the assembly of POMs
connected to dendritic molecules in organic solution has
been little mentioned. From a self-assembly point of view,
POMs can also be employed to adjust the self-assembly be-
havior of dendritic molecules through the formation of pre-
cisely defined complexes. This strategy reveals several
merits: The variable size and abundant shapes of POMs
match the requirement of the precise structure of dendritic
complexes and the changeable surface charges of POMs can
be used to change the amount of dendrons in one complex
architecture. In addition, binding with dendritic molecules is
favorable for the self-assembly of POMs in soft organic
Results and Discussion
Preparation and characterization of SEPs: The dendritic
molecule that has four heptyl chains, denoted as D, was syn-
thesized and characterized (see Figures S1–S3 in the Sup-
porting Information) as described in detail in the Experi-
mental Section. To evaluate the influence of the number of
cationic dendrons in the complexes on the self-assembly be-
havior in solution and the formed assembled structures in
the solid state, we chose four types of anionic POMs with
nearly the same size and morphology but with different
numbers of charges to be encapsulated. Both POM-1 and
POM-2 are Keggin-type heteropolytungstates, whereas
POM-3 is a monolacunary Keggin-type heteropolytungstate,
and POM-4 is a trilacunary Dawson-type heteropolytung-
state, the size of which is a little bigger than the Keggin-
type POMs. The preparation of the POMs and SEPs was
carried out by following the procedures described in the lit-
erature.[16,17] The main purpose of introducing alkyl chains
into the dendritic molecule is to increase the flexibility and
assembling ability of the complexes. The four corresponding
POMs are soluble in water, and in contrast all of the as-pre-
pared SEPs are immiscible in water but readily dissolve in
organic media such as cyclohexane, toluene, chloroform,
and acetone, suggesting that the POMs have been success-
fully encapsulated by dendritic molecules.
matriACHTUNGTRENNUNGces in a designed way, which is normally uncontrollable
in simply mixing systems. As a result, the dendritic shell can
have an impact on the access of small molecules to the inor-
ganic core, which may result in the special effects of en-
zymelike regioselectivity and substrate selectivity.[15] Howev-
er, to the best of our knowledge, the matching relationship
between cationic dendrons and POMs has never been in-
volved and applied to tune the existing states in solution
and the self-assembled structures in the solid state. Herein
we report the self-assembly behavior of a new class of SEP
complexes, D4SiW12O40 (SEP-D4), D5BW12O40 (SEP-D5),
D7MSiW11O39 (M=Na or K; SEP-D7), and D10Na2P2W15O56
(SEP-D10), containing different numbers of cationic dendrit-
ic molecules (D) around POM cores as shown in Scheme 1.
It was found that the existing states of the dendritic com-
The elemental analysis results suggest the chemical for-
mulae of SEP-D4, SEP-D5, SEP-D7, and SEP-D10 to be
D4SiW12O40 (6151.05), D5BW12O40 (6953.00), D7Na-
AHCNTUGERNNU(G SiW11O39) (8431.88) or D7KACHUTNGTNER(NUGN SiW11O39) (8447.99), and
D10Na2P2W15O56 (11953.76). The thermogravimetric (TG) re-
sults of SEP-D4, SEP-D5, SEP-D7, and SEP-D10 show 53.3,
60.8, 69.4, and 69.8% mass loss before 9008C, which match-
es well with the expected values, 53.8, 60.0, 68.7–68.5, and
70.4%, respectively, which are calculated from each corre-
sponding chemical formula by assuming that the organic
component has decomposed completely and that the POMs
have transformed into the corresponding oxides. Combining
the elemental analysis and TG results, we infer that the
chemical formulae of SEPs are consistent with the expected
target products. As for SEP-D7, we could not recognize the
residue metal ion as being a sodium or potassium cation
owing to the very low content in the complex and very little
difference between the two expected formulae. Actually, the
uncertainty does not affect the composition and self-assem-
bling property of the complex, therefore, further confirma-
tion was not carried out.
Scheme 1. The chemical structure of dendritic molecule (D) and schemat-
ic polyhedrons of four POMs.
Chem. Eur. J. 2010, 16, 8062 – 8071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8063

L. Wu et al.
1H NMR spectra of SEP-D4, SEP-D5, SEP-D7, and SEP-
D10 confirm the presence of cationic dendrons on the encap-
sulated clusters. In contrast to NMR signals of the dendritic
molecule alone, those of cationic dendrons in SEP-D4 show
the following changes: 1) Except for the proton peak of
tion of methanol or ethanol becomes necessary to solidify
the self-assembled structures of SEPs in chloroform. As a
substitution and a monocomponent solvent, acetone serves
the same role for the structural characterization of the as-
semblies in solution systems. To evaluate the effect of the
organic component covering on POMs for the assemblies of
SEPs, we initially investigated the self-assembling behavior
of the dendritic molecule in organic solution. TEM measure-
ments (Figure 1a) demonstrate that the dendritic ammoni-
ꢀ
ArH that is ortho to CH2OAr at the 2,6-positions, the
other proton peaks have significantly broadened and shifted
to high field, especially those near the nitrogen atom. For
example, the proton peak of the N-methyl group has re-
markably broadened and shifted toward high field by
0.16 ppm, and the proton peak of the N-methylene group
has evidently broadened and shifted to high field by
0.50 ppm. 2) The proton peak of the ArH group that is para
ꢀ
to CH2NACHTUNGTRENNUNG(CH3)3 at the 4-position, which should split into a
triplet, becomes a considerably broadened singlet and is
shifted to a higher field by 0.09 ppm. 3) The proton peak of
ꢀ
the ArH group that is ortho to CH2N
N
sitions, which should split into a doublet, becomes a highly
broadened singlet and shifts to higher field by 0.14 ppm. We
also observed similar broadening and shifting of the NMR
signals of the cationic dendrons in other SEPs. The peak
broadening and shifting clearly result from the strong elec-
trostatic interactions between the cationic dendrons and the
POMs, which restricts the mobility of the cationic dendrons
on the surface of the POMs in the organic phase, especially
for their ammonium head parts.[13a,18]
As shown in the IR spectra (Figure S6 in the Supporting
Information), the bands appearing at 2927–2929 and
2856 cmꢀ1 are assigned to CH2 antisymmetric and symmetric
stretching modes of alkyl chains, respectively. Meanwhile,
the bands at 1597–1599, 1483–1484, and 1456 cmꢀ1 can be
Figure 1. a) TEM image of dendritic molecule assemblies prepared from
acetone solution (1 mgmLꢀ1), b) XRD pattern of casting dendron assem-
blies, c) the shape of a water droplet on the surface of casting dendron
assemblies on a silica substrate, and d) schematic representation for the
reverse-bilayer structure of vesicular assemblies.
ꢀ
clearly assigned to the benzyl framework, C N scissoring,
and CH2 scissoring modes, respectively. In addition, the
bands of the W-Oc-W vibrations shift from 787–814 and
735–739 cmꢀ1 in the free POMs to 800–829 and 729–
758 cmꢀ1 in the SEPs, the vibrations of W-Ob-W change
from 879–937 cmꢀ1 in the free POMs to around 885–
um cation is not singularly molecularly dispersed but that it
is in an aggregated state in acetone, and vesicular aggrega-
tions are clearly observed with a size scale in the range of
20–30 nm. The wall thickness of the vesicular assemblies is
approximately 3.0 nm, which is accordance with the length
of two ideal dendritic molecules (LD =1.6 nm) estimated
from the data in the literature.[20] The XRD pattern further
demonstrates that the dendritic molecule forms a lamellar
structure. As seen in Figure 1b, two Bragg diffraction peaks
appear at 2.86 and 5.788, corresponding to the layer spacings
of 3.1 and 1.5 nm, respectively, which can be well indexed as
(001) and (002) reflections of a lamellar structure with a
layer spacing of 3.1 nm. Combining the length of two den-
dritic molecules, the total thickness of a single bilayer is esti-
mated to be approximately 3.2 nm. It is known that the fre-
quencies of CH2 antisymmetric and symmetric stretching
bands are sensitive to the conformation of alkyl chains: Low
wavenumbers (2915–2918 and 2846–2850 cmꢀ1) are charac-
teristic of a highly trans conformation of alkyl chains,[21]
whereas their upward shifts (2924–2929 and 2854–
2856 cmꢀ1) are indicative of the increase in gauche conform-
ers, implying that the alkyl chain conformation tends to dis-
order. On the basis of this, the observed frequencies of 2929
and 2856 cmꢀ1 of the dendritic molecules indicate the pres-
937 cmꢀ1 in the SEPs, and the W Od stretches transfer from
ꢀ
978–1001 and 945–958 cmꢀ1 in the free POMs to 972–993
and 940–951 cmꢀ1 in the SEPs. These vibration bands (as-
signed and summarized in Table S1 in the Supporting Infor-
mation) indicate that, on the one hand, the frame structures
of POMs are well kept, and on the other hand, the strong
electrostatic interaction occurs between the cationic den-
drons and the POMs.[19]
Self-assemblies of dendritic molecules and SEPs in solution:
From the previous results and related reports concerning
the assembly of surfactant-enwrapped complexes, it is
known that the solvent polarity plays an important role for
the stabilization of self-assembled structures and the crea-
tion of novel assembling morphologies.[12b] We tried several
weak polar solvents and found consistent assembling mor-
phology for the as-prepared complexes. The addition of
polar solvents can increase the stability of the assemblies in
solution so that the self-assemblies keep their structures
well even after the evaporation of solvent. Thus, the addi-
8064
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8062 – 8071

Polyoxometalate-Anchored Dendron Complexes
FULL PAPER
ence of gauche conformers and the disorder of the alkyl
chain conformation. Considering the disordered conforma-
tion of the alkyl chains and the lateral hydrophobic com-
pressing for the dendritic molecules standing in the layer, it
is reasonable that the measured layer distance (3.1 nm) is
approximate to the ideal layer spacing (3.2 nm). The static
contact angle (Figure 1c) of the dendritic molecule assem-
blies for water is approximately 868, showing a hydrophobic
surface. These results imply that the vesicular assemblies
possess a reverse-bilayer structure with the alkyl chains
facing the acetone solvent and, therefore, a schematic as-
sembling model of the dendritic molecule has been pro-
posed in Figure 1d.
From the dynamic light scattering (DLS) results in
Figure 2, it can be seen that SEP-D4 and SEP-D5 are in an
aggregated state in acetone, with the hydrodynamic diame-
ters of approximately 396 and 122 nm, respectively. Contrary
Figure 3. TEM images of a) SEP-D4, b) SEP-D5, c) SEP-D7, and d) SEP-
D10 in acetone solution with a concentration of 1 mgmLꢀ1
.
dark spot with the diameter scale of approximately 3–4 nm,
which is in agreement with the exact size of the complex.
The results for SEP-D10 are similar to those of SEP-D7. The
difference of assembling properties for these two SEPs from
SEP-D4 and SEP-D5 could be reasonably explained as deriv-
ing from the change of the cationic dendron number on the
periphery of the POMs. The complexes containing less cat-
ionic dendrons have enough space to reorganize themselves
and assemble together and this is driven by hydrophobic
forces, whereas the complexes that have more cationic den-
drons are able to remain apart from each other in solution,
resulting in the monodispersed state.
To understand the self-assembled structure of the spheri-
cal assemblies of SEP-D4 and SEP-D5 further, XRD was
employed for further characterization and the results
(Figure 4) show that the lamellar structures exist in both
spherical assemblies of SEP-D4 and SEP-D5. Three diffrac-
tion peaks in the XRD pattern of SEP-D4 (Figure 4a)
appear at 2.76 (001), 5.48 (002), and 8.188 (003), suggesting
a layered structure with thickness 3.2 nm. The intensity of
the (003) reflection is a bit stronger than the (002) reflec-
Figure 2. DLS diagrams of a) SEP-D4, b) SEP-D5, c) SEP-D7, and
d) SEP-D10 in acetone solution with a concentration of 1 mgmLꢀ1
.
to what is expected, by increasing the number of cationic
dendrons around the POMs we observed that SEP-D7 and
SEP-D10 are clearly amorphous in acetone, rather than in an
aggregated state, because the observed hydrodynamic diam-
eters are 4.2 and 3.6 nm, respectively, which are very small
and approximate to the size scale of a single complex. TEM
and SEM images of SEP-D4 (Figure 3a and Figure S7a in
the Supporting Information) reveal that the complex assem-
blies have a spherical shape, and the diameter determined
from these measurements is consistent with the DLS results.
The precise structure of these spheres of SEP-D4 cannot be
observed by TEM successfully because the assemblies are
quite solid and the electron beam cannot penetrate the ob-
jects. The TEM and SEM results (Figure 3b and S7b in the
Supporting Information) of SEP-D5 demonstrate consistent
phenomena with that of SEP-D4. However, SEP-D7 and
SEP-D10 complexes do not form clear aggregations in ace-
tone, as demonstrated by their TEM images shown in Fig-
ure 3c and d, respectively, which is in accordance with the
DLS results. Interestingly, SEP-D7 appears as a separated
Figure 4. XRD patterns of the assemblies of a) SEP-D4 and b) SEP-D5
complexes prepared from acetone solution.
Chem. Eur. J. 2010, 16, 8062 – 8071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8065

L. Wu et al.
tion, indicating a symmetrical double-layered structure.[12g]
Combining the diameter of one Keggin-type POM (1.0 nm)
and the length of two cationic dendrons (3.2 nm), the ideal
diameter of one SEP-D4 complex is about 4.2 nm, which is
much larger than the measured layer distance. Because the
CH2 symmetric and antisymmetric stretching vibrations for
SEP-D4 appear at 2856 and 2929 cmꢀ1, respectively, the
shorter layer spacing should be mainly derived from the
conformation disorder of alkyl chains.[21] Note that the layer
spacing of the dendritic complex is close to that of the pre-
cursor dendron. Based on the present experimental data, it
is difficult to explain the phenomenon reliably. However, it
is possible that the rigid POM clusters support the bending
of the cationic dendrons with enough space to make the
packing looser than dendritic molecules alone, leading to a
shorter bilayer length (2.2 nm) than the measured layer dis-
tance (3.1 nm) of separated dendritic molecules. Likewise,
SEP-D5 displays a similar lamellar structure and consistent
layer spacing with the estimated results. The static contact
angles (Figure S8 in the Supporting Information) of dry
SEP-D4 and SEP-D5 castings for water are 102 and 998, re-
spectively, indicating a hydrophobic surface composed of
alkyl chains. Thus, spherical assemblies of these two com-
plexes are deduced as possessing a vesiclelike structure with
reverse bilayers, as previously fabricated by different com-
plexes.[12b] Again, in agreement with the TEM observation,
we did not find any diffractions belonging to lamellar struc-
tures in the XRD data of SEP-D7 and SEP-D10, further con-
firming that more cationic dendrons covering the POMs can
lead to a change of the assembled structure of SEPs.
Figure 5. XRD patterns of a) SEP-D4, b) SEP-D5, c) SEP-D7, and
d) SEP-D10 assemblies in the solid state.
the CH2 group emerging at 2856 and 2929 cmꢀ1,[21] which
should be the main reason for the shorter layer distance.
From these results and from the fact that strong phase seg-
regation occurs between the hydrophilic POM-1 cores and
the hydrophobic dendrons, we can propose a self-assembled
structure of SEP-D4 in the solid state (Scheme 2b), that is,
the cationic dendrons are piled on two opposite sides of in-
organic clusters and extend along the layer direction under
the packing interaction. This rodlike reorganization of the
complex in shape is, of course, favorable for tight parallel
organization. The POM-1 clusters constituting the central
layer embedded by cationic dendrons further assemble into
a lamellar structure. SEP-D5 also has a lamellar structure,
and its XRD pattern (Figure 5b) exhibits a layer spacing of
Self-assemblies of SEPs in the solid state: To investigate the
solvent effect on the self-assembled structures during the
evaporation, we examined the solid structures of SEP com-
plexes prepared through directly drying the sample solu-
tions. The XRD results in Figure 5a and b demonstrate that
the solids of SEP-D4 and SEP-D5 prepared from chloroform
possess similar lamellar structures to those from acetone so-
lution, whereas we could not observe the globular assem-
bling morphologies seen in Figure 3. Considering the insta-
bility of SEP-D4 and SEP-D5 assemblies during solvent
evaporation, which is due to the weak interaction between
complexes, it is possible that the vesicular assemblies that
occurred in chloroform collapse into lamellar structures in
the solid state, as for the case found in a previous report.[12b]
Three Bragg diffractions in the XRD pattern of SEP-D4
(Figure 5a) appear at 2.68 (001), 5.44 (002), and 8.208 (003),
supporting the assignment of a lamellar structure. However,
the layer spacing of 3.3 nm is much smaller than the ideal
layer distance but it is in accordance with the layer spacing
of the assemblies in acetone implying the unchanged trans-
fer of the assemblies. Again, having a slightly stronger (003)
reflection than a (002) reflection indicates that the solid as-
semblies possess a symmetrical double-layered structure.[12g]
Similarly, the IR spectrum of SEP-D4 reveals the disordered
conformation of the alkyl chains in SEP-D4 solid because of
the symmetric and antisymmetric stretching vibrations of
Scheme 2. Schematic representation of a) dendritic molecule D, b) the
complex SEP-D4 and its bilayer structure, in which these complex units
are accommodated in the layers without packing curvature, and c) the
complex SEP-D10 and its hexagonal columnar structure, in which the
complex adapts itself to an oblate shape to form the hexagonally packed
cylinders instead of layers.
8066
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8062 – 8071

Polyoxometalate-Anchored Dendron Complexes
FULL PAPER
3.2 nm, almost the same as that of SEP-D4. Interestingly,
through carefully dispersing a solid powder of SEP-D4 and
SEP-D5, the assembled structures can be further identified
from TEM images (Figure 6a and b) in which a clearly lay-
ered structure is found with an estimated layer spacing of
approximately 3.0 nm, which is in perfect agreement with
the value estimated from the XRD data.
tance. Summarizing these factors, we propose a packing
model for SEP-D7 and SEP-D10 in the solid state as present-
ed in Scheme 2c. The hexagonal columnar structure matches
the face-to-face stacking of the two SEPs in a disklike
shape, and the flexible alkyl chains on the periphery fill the
space between the columns. The hydrophilic regions of the
complexes interact preferably with each other and, there-
fore, segregate from the hydrophobic regions into distinct
domains. The segregation effects between hydrophilic and
hydrophobic components should be one of the main driving
forces for the formation of a columnar structure. The hexag-
onal packing modes of SEP-D7 and SEP-D10 are supported
by the TEM characterization. From the high-resolution
TEM images of SEP-D7, we clearly observed both the linear
structure (Figure 6c) and the lateral hexagonal spots (Fig-
ure 6d). The dark POM domain is clearly surrounded by the
light organic region, and the intercolumnar distance is ap-
proximately 3.4 nm from the magnified TEM image, which
is in good agreement with the lattice parameter of the a axis
derived from the XRD results. The electron diffractions of
the planform and side view (insets of Figure 6c and d) of
the assemblies are in a linear and a hexagonal fashion, re-
spectively, further confirming the assignment. Thus, we can
see that SEP-D7 and SEP-D10 complexes prefer to form hex-
agonal columnar assemblies rather than lamellar assemblies
in the solid state when they are prepared from chloroform.
So far, there are very few reports about the columnar
phases based on SEPs,[13a–c] and to the best of our knowledge
this is an unprecedented example of the hexagonal columnar
self-assembled structures formed by SEPs with small-sized
POMs.
Figure 6. TEM images of a) SEP-D4, b) SEP-D5, c) SEP-D7 viewed per-
pendicular to the columnar direction, and d) SEP-D7 viewed parallel to
the columnar direction. The insets in c) and d) are the corresponding
electron diffraction patterns.
Discussion of self-assembled structures and structural
evolvement: The above results indicate that with increasing
the number of cationic dendrons around the POMs, a struc-
tural evolvement takes place from lamellar to hexagonal
columns in the solid state. Because the employed POMs are
similar in size and morphology, the molecular geometry of
the cationic dendron and its number covering on the POMs,
as well as its array on the periphery of the SEPs, become
the crucial factors in determining the self-assembled struc-
tures and morphologies of the SEPs in solution and the
solid state. Here, we try to discuss the possible mechanism
for understanding such a structural evolvement.
Very interestingly, in contrast to the nonstructured state
in solution, we observed the self-assembled structures of
SEP-D7 and SEP-D10 in their solid states. Compared with
the lamellar assemblies of SEP-D4 and SEP-D5, the XRD
pattern of SEP-D7 in Figure 5c exhibits five diffraction
peaks at 2.82 (100), 4.96 (110), 5.72 (200), 7.648 (210), and
8.388 (300), which can be well indexed to a hexagonal col-
umnar structure with a lattice parameter of a=3.6 nm
Similar to liquid-crystal systems,[22] the orientation and
array of the cationic dendrons in SEPs have important ef-
fects on the self-assembled structures. In view of the feasible
rearrangement on the exterior of POMs, the dendrons can
adopt different orientation states depending on the free
space around the POM cores owing to the requirement for
tight assembly. Under the conditions of similar size and mor-
phology of the POMs, increasing the number or the density
of cationic dendrons leads to a larger surface occupation (or
a smaller free space) and definite changes of the shape and
packing state of SEPs accordingly. In this case, the accom-
modation of SEP complexes confined within the rod shape
for lamellar array becomes unlikely. Therefore, the SEPs
p
(2d100
/
3). Consistent with the case in which the layer
length is much smaller than the ideal diameter of SEP-D4
and SEP-D5, the length of the a axis is also much smaller
than the estimated ideal value (4.2 nm) of one monolacu-
ACHTUNGTRENNUNGnary Keggin-type POM (1.0 nm) and two cationic dendrons
(3.2 nm). The XRD pattern (Figure 5d) of SEP-D10 suggests
a similar hexagonal columnar structure with a lattice param-
eter of a=3.8 nm. Considering that CH2 symmetric and anti-
symmetric stretching vibrations appear at 2856 and
2927 cmꢀ1, respectively, the conformation of alkyl chains of
cationic D dendrons in SEP-D7 and SEP-D10 is quite disor-
dered, apparently leading to the smaller intercolumnar dis-
Chem. Eur. J. 2010, 16, 8062 – 8071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8067

L. Wu et al.
that have seven or ten cationic dendrons adopt an oblate
orientation instead of the parallel distribution that is found
in the lamellar assemblies of SEP-D4 and SEP-D5. With
such an orientation, the SEPs have to transform into a disk-
like state and then stack into columns, which further self-as-
semble parallel to each other in a regular manner to form a
hexagonal two-dimensional array. The orientation state of
the cationic dendrons can be approximately predicted by
volume calculations by filling the periphery of the SEPs.
For a convenient comprehension, we divided the POM
cluster into six regions similar to the faces of a cube, as
shown in Scheme 3a. The maximum possessive volume (V)
volumes, it is easy to find that both VD4 and VD5 are smaller
than Vl, whereas the values of VD7 and VD10 are located just
between Vl and Vh. The former two complexes should self-
assemble into lamellar structures, and owing to the space
limit for the latter two complexes, the layered structures no
longer remain and break into tight hexagonal columnar
packing structures. The larger occupation of the cationic
dendrons in SEP-D7 and SEP-D10 with respect to that in
SEP-D4 and SEP-D5 makes these two complexes favor or-
ganization into columnar assemblies, leading to the tight
hexagonal packing structures. The main driving forces can
be suitably ascribed to the phase segregation of incompati-
ble units (the hydrophilic POM core and hydrophobic den-
dritic part), the aggregation of compatible units, and the
minimization of volume.[23] In contrast to the present results,
through increasing the number of the linear surfactants such
as dioctadecyldimethylammonium bromide (DODA),[12b–d]
the yielding lamellar structure does not change at all. This is
because the DODA skeletons are greatly flexible and in-
complete for steric constraints. So, the dendritic surfactants
play a special and irreplaceable role in the structural evolve-
ment owing to their rigidity and divergent shape. It is clear
that the balance between the formation of the lamellar
structures and hexagonal columnar structures can be finely
tuned by manipulating the number of cationic dendrons on
the periphery of the SEPs.
Scheme 3. a) Supposed cubic models with two opposite facets occupied
for the lamellar structure and four adjacent facets filled for the 2D hex-
agonal columnar structure, and b) the organic shell of the SEP (the dark
ring represents the organic shell composed of cationic dendrons, r1 repre-
sents the radius of the POM, and r2 represents the radius of the SEP).
of cationic dendrons can be divided into six units, corre-
sponding to the cube faces during the solvent-driven phase
separation. For the lamellar structures, the cationic dendrons
were forced into two opposite facets, whereas for the hexag-
onal columnar structures, the dendrons were instead pressed
to occupy four adjacent facets in a rather flat disklike shape,
which enabled the most efficient interaction of the hydro-
philic regions of neighboring complexes. For the SEP form-
ing the lamellar structure, the maximum possessive volume
(Vl) of the dendrons should be two of the six units, that is,
V/3, whereas for the SEP forming the hexagonal columnar
structure, the maximum occupied volume (Vh) of the den-
drons would be four of the six units, that is, 2V/3. As a
result, the hexagonal columnar structure can form only
when the total volume of cationic dendrons (VDn) is in
agreement with Vl <VDn ꢁVh. From Scheme 3b, V can be
calculated according to Equation (1), in which r1 denotes the
radius of POM, and r2 is the radius of SEP.
Conclusion
As a whole, the dendritic molecule that has four heptyl
chains was synthesized, and four types of dendritic surfac-
tant-encapsulated polyoxometalates containing different
numbers of cationic dendrons were prepared depending on
the charges of the POMs. We found that SEP-D4 and SEP-
D5 complexes, which have a low number of cationic den-
drons, self-assemble into vesicular aggregations with lamel-
lar structures in acetone, whereas with a relatively high
number of cationic dendrons, SEP-D7 and SEP-D10 com-
plexes become monodisperse in acetone solution and exhibit
hexagonal columnar structures in the solid state. The
branched molecular structure of the cationic dendron and
the matching relationship between the cationic dendrons
and POMs plays an important role in the formation of hex-
agonal columnar structures. A mechanism of the structural
evolvement based on the orientation change of the cationic
dendrons on the periphery of SEPs as a result of their
varied number in one complex has been proposed, and the
type of self-assembled structures can be effectively predicted
through comparing the total volume of the dendrons in one
complex with their maximum possessive volumes for the la-
mellar and hexagonal columnar structures. Therefore, we
identify a facile and feasible route to control the assembled
structures in solution and in the solid state by modifying the
number of the cationic dendrons on the periphery of SEPs.
Furthermore, the above-mentioned structural evolvement
concept may provide a general pathway to construct other
V ¼ 4pðr23 ꢀ r13Þ=3
ð1Þ
According to the similar size of the POMs in the present
study, r1 is approximately 0.5 nm and r2 is equal to a/2,
which can be estimated from the lattice parameter of the
hexagonal columnar structure of SEP-D7. Thus, Equation (1)
provides the calculated V of 23.9 nm3. Accordingly, Vl is
8.0 nm3 and Vh is 16.0 nm3. The volume of the cationic den-
dron (VD) can be estimated to be 1.3 nm3 by employing a
similar dendritic molecule as that in reference [20]. There-
fore, the volumes of the organic parts of SEP-D4, SEP-D5,
SEP-D7, and SEP-D10 should be VD4 =5.2, VD5 =6.5, VD7
=
9.1, and VD10 =13.0 nm3, respectively. From these calculated
8068
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8062 – 8071

Polyoxometalate-Anchored Dendron Complexes
FULL PAPER
17.4 mmol) were mixed and stirred under reflux for 20 h in a N2 atmos-
phere. The excess methanol was removed through rotary evaporation,
and then the reaction mixture was redissolved in hot water. The insoluble
impurities were filtered out, and the filtrate was cooled to room tempera-
ture to give a white crystalline solid (yield: 6.1 g, 56.0%). 1H NMR
(CDCl3, TMS): d=3.89 (s, 3H; -COOCH3), 4.86 (s, 2H; OH-Ar), 6.56 (t,
J=2 Hz, 1H; ArH para to -COOCH3 at the 4-position), 7.09 ppm (d, J=
2 Hz, 2H; ArH ortho to -COOCH3 at the 2,6-positions).
types of self-assemblies based on POMs such as the cubic-
packed structure by further increasing the number of the pe-
ripheral cationic dendrons in one SEP. These self-assembled
structures are promising structures for use as supramolec-
ular soft templates that may contribute to unique effects on
the catalytic, optical, and magnetic properties of POMs.
Methyl 3,5-bis(heptyloxy)benzoate (2): n-Heptyl bromide (42.6 g,
237.9 mmol),
1
(10.0 g, 59.5 mmol), anhydrous K2CO3 (24.7 g,
Experimental Section
178.7 mmol), 18-crown-6 (32.3 mg), and acetone (150 mL) were mixed to-
gether, and the mixture was stirred under reflux in a N2 atmosphere for
24 h. Then, the reaction mixture was monitored by TLC analysis and
cooled to room temperature. The solid salts were filtered out, and the
solvent in the filtrate was evaporated under reduced pressure. The liquid
mixture was then purified by using silica-gel column chromatography
(eluent: 1:1 (v/v) chloroform/cyclohexane) to give a colorless liquid
(yield: 14.8 g, 68.8%). 1H NMR (CDCl3, TMS): d=0.89 (t, J=6.5 Hz,
6H; -CH3), 1.29–38 (m, 12H; -(CH2)3-), 1.41–1.47 (m, 4H; -CH2-CH3),
1.74–1.78 (m, 4H; -CH2-CH2-O-Ar), 3.89 (s, 3H; -COOCH3), 3.96 (t, J=
6.5 Hz, 4H; -CH2-O-Ar), 6.63 (t, J=2 Hz, 1H; ArH para to -COOCH3 at
the 4-position), 7.15 ppm (d, J=2 Hz, 2H; ArH ortho to -COOCH3 at
the 2,6-positions).
Materials: H4SiW12O40 (POM-1) was purchased from Sinopharm Chemi-
cal Reagent and was used as received. The other POMs, K5BW12O40
(POM-2), K6Na2SiW11O39 (POM-3), and Na12P2W15O56 (POM-4) were
freshly prepared according to the literature procedures.[16] 3,5-Dihydroxy-
benzoic acid was obtained from Alfa Aesar and 3,5-dihydroxybenzyl al-
cohol was obtained from ACROS. All of the starting compounds for syn-
thesis were used without further purification, and all of the used solvents
were analytical grade. Doubly distilled water was used in the experi-
ments. Silica gel (100–200 mesh) was used for column chromatography.
Synthesis of dendritic molecule D: The cationic dendritic quaternary am-
monium bearing four heptyl chains was synthesized step by step by using
the convergent method.[24] The detailed synthetic procedures are de-
scribed in Scheme 4. The chemical structure of the dendritic molecule
was confirmed by FTIR and 1H NMR spectroscopy, as well as MALDI-
TOF mass spectrometry.
3,5-Bis(heptyloxy)benzyl alcohol (3): A solution of 2 (2.0 g, 5.5 mmol) in
distilled THF (10 mL) was added dropwise to a stirred suspension of
LiAlH4 (0.3 g, 7.9 mmol) in distilled THF (40 mL) in an ice-water bath.
After the addition, the mixture was stirred for an additional 24 h at room
temperature. The reaction was then quenched by the successive addition
of 95% ethanol (ca. 50 mL) and water (ca. 2 mL). The insoluble impuri-
Methyl 3,5-dihydroxybenzoate (1): 3,5-Dihydroxybenzoic acid (10.0 g,
65.0 mmol), methanol (150 mL), and p-toluene sulfonic acid (3.0 g,
Scheme 4. Synthetic route of the dendritic molecule D.
Chem. Eur. J. 2010, 16, 8062 – 8071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8069

L. Wu et al.
ties were filtered out. The residual solution was extracted with chloro-
form (3ꢁ50 mL), and the combined chloroform solution was then dried
and evaporated under reduced pressure to dryness to give a yellow
liquid. The crude product was purified by using silica-gel column chroma-
tography (eluent: chloroform) to give a light-yellow liquid (yield: 1.7 g,
(0.09 g) under stirring, and its initial molar ratio to POM-1 was controlled
at 3.8:1. The organic phase was separated, washed with water (3ꢁ30 mL)
and dried over anhydrous sodium sulfate. SEP-D4 was obtained by evap-
orating the chloroform to dryness. The product was stored in a vacuum
desiccator until the weight remained constant (yield: 0.14 g, 82.4%).
1H NMR (CDCl3, TMS): d=0.85 (t, J=6.5 Hz, 12H; -CH3), 1.25–1.40
(m, 32H; -(CH2)4-CH3), 1.68–1.73 (m, 8H; -CH2-CH2-O-Ar), 3.16 (s, 9H;
(CH3)3N), 3.89 (t, J=6 Hz, 8H; -CH2-O-Ar), 4.39 (s, 2H; -CH2-NACHTUNGTRENNUNG(CH3)3),
4.92 (s, 4H; Ar-CH2-O-Ar), 6.34 (s, 2H; ArH para to -CH2OAr at the 4-
position), 6.56 (s, 4H; ArH ortho to -CH2OAr at the 2,6-positions), 6.62
1
92.1%). H NMR (CDCl3, TMS): d=0.89 (t, J=6.5 Hz, 6H; -CH3), 1.30–
1.34 (m, 12H; -(CH2)3-), 1.42–1.45 (m, 4H; -CH2-CH3), 1.75–1.77 (m,
4H; -CH2-CH2-O-Ar), 3.93 (t, J=6 Hz, 4H; -CH2-O-Ar), 4.62 (s, 2H;
-CH2-OH), 6.37 (s, 1H; ArH para to -CH2OH at the 4-position),
6.50 ppm (s, 2H; ArH ortho to -CH2OH at the 2,6-positions).
(s, 1H; ArH para to -CH2N
ACHTUGNERTN(NUNG CH3)3 at the 4-position), 6.69 ppm (s, 2H;
3,5-Bis(heptyloxy)benzyl bromide (4): Compound 3 (1.5 g, 4.4 mmol) was
dissolved in dry benzene (50 mL) and was cooled to 0–58C in an ice-
water bath. PBr3 (0.2 mL, 2.1 mmol) was added dropwise with stirring.
The mixture was stirred for another 6 h at room temperature, and doubly
distilled water (30 mL) was added to quench the reaction. The benzene
in the reaction solution was evaporated under vacuum, and the mixture
was extracted with dichloromethane (3ꢁ30 mL). The combined extracts
were dried with anhydrous sodium sulfate and evaporated to dryness.
The crude product was purified by using silica-gel column chromatogra-
phy (eluent: 1:1 (v/v) dichloroform/cyclohexane) to give a light-yellow
thick liquid product (yield: 1.7 g, 95.5%). 1H NMR (CDCl3, TMS): d=
0.89 (t, 6H; -CH3), 1.30–1.34 (m, 12H; -(CH2)3-), 1.42–1.45 (m, 4H;
-CH2-CH3), 1.75–1.77 (m, 4H; -CH2-CH2-O-Ar), 3.92 (t, 4H; -CH2-O-
Ar), 4.41 (s, 2H; -CH2-Br), 6.37 (s, 1H; ArH para to -CH2Br at the 4-po-
sition), 6.51 ppm (s, 2H; ArH ortho to CH2Br at the 2,6-positions).
ArH ortho to -CH2N(CH3)3 at the 2,6-positions); IR (KBr): n˜ =3435,
AHCTUNGTRENNUNG
2954, 2929, 2856, 1597, 1484, 1456, 1377, 1167, 1059, 972, 924, 885,
800 cmꢀ1; elemental analysis (%) calcd for SEP-D4 (C208H336N4O64W12Si,
6151.05): C 40.61, H 5.51, N 0.91; found: C 41.30, H 5.59, N 0.92 (corre-
sponding to the chemical formula: D4SiW12O40 (6151.05)); the TG mea-
surement displays a 53.3% (w/w) mass loss between 40 and 6008C.
SEP-D5: The complex was prepared by following a similar procedure as
for SEP-D4, but by using POM-2 instead of POM-1. The initial molar
ratio of dendritic molecule D to POM-2 was kept at 4.5:1 (yield: 0.03 g,
76.9%). 1H NMR (CDCl3, TMS): d=0.86 (t, J=6.5 Hz, 12H; -CH3),
1.26–1.39 (m, 32H; -(CH2)4-CH3), 1.70 (s, 8H; -CH2-CH2-O-Ar), 3.17 (s,
9H; (CH3)3N), 3.89 (s, 8H; -CH2-O-Ar), 4.45 (s, 2H; -CH2-NACHTUNGTRENNUNG(CH3)3),
4.93 (s, 4H; Ar-CH2-O-Ar), 6.35 (s, 2H; ArH para to -CH2OAr at the 4-
position), 6.56 (s, 4H; ArH ortho to -CH2OAr at the 2,6-positions), 6.61
(s, 1H; ArH para to -CH2N
ACHTUGNERTN(NUNG CH3)3 at the 4-position), 6.77 ppm (s, 2H;
3ꢀ,5ꢀ-BisACHTUNGTRENNUNG(3,5-bis(heptyloxy)benzyloxy)benzyl alcohol (5): Compound 4
ArH ortho to -CH2N(CH3)3 at the 2,6-positions); IR (KBr): n˜ =3440,
AHCTUNGTRENNUNG
(1.7 g, 4.3 mmol), 3,5-dihydroxybenzyl alcohol (0.25 g, 1.8 mmol), anhy-
drous K2CO3 (1.0 g, 7.2 mmol), and catalytic amount of 18-crown-6 were
dissolved in acetone (50 mL) and the mixture was stirred under reflux in
a N2 atmosphere for 24 h. The reaction was monitored by TLC analysis.
The reaction mixture was filtered and the excess solvent was evaporated.
The liquid mixture was then purified by using silica-gel column chroma-
tography (eluent: 10:1 (v/v) cyclohexane/ethyl acetate) to give a light-
yellow thick liquid (yield: 1.0 g, 72.1%). 1H NMR ([D6]DMSO, TMS):
d=0.86 (t, J=6.5 Hz, 12H; -CH3), 1.24–1.41 (m, 32H; -(CH2)4-CH3),
1.65–1.70 (m, 8H; -CH2-CH2-O-Ar), 3.92 (t, 8H; J=6.5 Hz, -CH2-O-Ar),
4.41 (d, J=6 Hz, 2H; -CH2-OH), 4.97 (s, 4H; Ar-CH2-O-Ar), 5.18 (t, J=
6 Hz, 1H; -OH), 6.39 (t, J=2 Hz, 2H; ArH para to -CH2OAr at the 4-
position), 6.49 (t, 1H; ArH para to -CH2OH at the 4-position), 6.54–
6.55 ppm (m, 4H and 2H; ArH ortho to -CH2OAr at the 2,6-positions
and ArH ortho to -CH2OH at the 2,6-positions).
2954, 2929, 2856, 1599, 1483, 1456, 1379, 1165, 1059, 993, 951, 903,
829 cmꢀ1; elemental analysis (%) calcd for SEP-D5 (C260H420N5O70W12B,
6953.00): C 44.91, H 6.09, N 1.01; found: C 45.72, H 5.85, N 0.64 (corre-
sponding to the chemical formula: D5BW12O40 (6953.00)); the TG mea-
surement displays a 60.8% (w/w) mass loss between 45 and 6008C.
SEP-D7: The complex was prepared by following a similar procedure as
for SEP-D4, but by using POM-3 instead of POM-1. The initial molar
ratio of dendritic molecule D to POM-3 was controlled at 6.5:1 (yield:
0.10 g, 78.1%). 1H NMR (CDCl3, TMS): d=0.85 (t, J=6.5 Hz, 12H;
-CH3), 1.25–1.36 (m, 32H; -(CH2)4-CH3), 1.68 (s, 8H; -CH2-CH2-O-Ar),
3.14 (s, 9H; (CH3)3N), 3.84 (s, 8H; -CH2-O-Ar), 4.49 (s, 2H; -CH2-N-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
3ꢀ,5ꢀ-BisACHTUNGTRENNUNG(3,5-bis(heptyloxy)benzyloxy)benzyl bromide (6): This compound
AHCTUNGTRENNUNG
was prepared from compound 5 by using a reaction procedure similar to
the preparation of compound 4. The crude product was directly used in
the following step without purification.
;
(C364H588N7O81W11SiNa or C364H588N7O81W11SiK, 8431.88 or 8447.99): C
51.85 or 51.75, H 7.03 or 7.02, N 1.16 or 1.16; found: C 51.52, H 7.00, N
0.58 (corresponding to the chemical formula: D7Na
or D7K(SiW11O39) (8447.99)); the TG measurement displays a 69.4% (w/
w) mass loss between 45 and 6008C.
N-3ꢀ,5ꢀ-bisACHTUNGTRENNUNG(3,5-bis(heptyloxy)benzyloxy)benzyl-N,N,N-trimethylammoni-
ACHTNUGNERT(NUGN SiW11O39) (8431.88)
um bromide (D): An aqueous solution of trimethylamine (33%, 0.7 g,
3.9 mmol) and crude product 6 (0.5 g) were dissolved in ethanol (50 mL),
and the mixture was stirred under reflux for 24 h with monitoring by
TLC analysis. The excess solvent and most of trimethylamine were
evaporated under vacuum. The thick liquid mixture was then purified by
using silica-gel column chromatography (eluent: 20:1 (v/v) chloroform/
methanol) to give a ceraceous solid (yield: 0.4 g, 69.1% over two steps).
1H NMR (CDCl3, TMS): d=0.89 (t, J=6.5 Hz, 12H; -CH3), 1.28–1.46
(m, 32H; -(CH2)4-CH3), 1.73–1.79 (m, 8H; -CH2-CH2-O-Ar), 3.32 (s, 9H;
(CH3)3N), 3.94 (t, J=6.5 Hz, 8H; -CH2-O-Ar), 4.89 (s, 2H; -CH2-N-
AHCTUNGTRENNUNG
SEP-D10: The complex was prepared by following a similar procedure as
for SEP-D4, but by using POM-4 instead of POM-1. The initial molar
ratio of dendritic molecule D to POM-4 was controlled at 8:1 (yield:
0.08 g, 88.9%). 1H NMR (CDCl3, TMS): d=0.86 (t, J=6.5 Hz, 12H;
-CH3), 1.26–1.38 (m, 32H; -(CH2)4-CH3), 1.70 (s, 8H; -CH2-CH2-O-Ar),
3.22 (s, 9H; (CH3)3N), 3.87 (s, 8H; -CH2-O-Ar), 4.49 (s, 2H; -CH2-N-
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
-CH2OAr at the 2,6-positions), 6.71 (t, 1H; ArH para to -CH2N
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
the 4-position), 6.83 ppm (d, J=2 Hz, 2H; ArH ortho to -CH2NAHCTUNGTRENNUNG
(KBr): n˜ =3435, 2954, 2927, 2856, 1599, 1484, 1456, 1379, 1167, 1122,
1080, 1059, 972, 937, 889, 816, 758 cmꢀ1; elemental analysis (%) calcd for
SEP-D10 (C520H840N10O116W15P2Na2, 11953.76): C 52.25, H 7.08, N 1.17;
found: C 52.00, H 6.37, N 1.00 (corresponding to the chemical formula:
D10Na2P2W15O56 (11953.76)); the TG measurement displays a 69.8% (w/
w) mass loss between 45 and 9008C.
the 2,6-positions); MALDI-TOF MS: m/z: 819.4 [M+]; IR (KBr): n˜ =
3423, 2954, 2929, 2856, 1597, 1488, 1462, 1379, 1167, 1059 cmꢀ1
.
Preparation of SEPs: All SEP complexes were prepared at room temper-
ature according to the previously reported procedures,[17] and were con-
firmed by 1H NMR and IR spectroscopy, elemental analysis, and TG
measurements.
Sample preparation for measurements: For the identification of the self-
assembled structures of dendritic molecule D and SEPs from acetone so-
lution, all of the sample films for XRD measurement were prepared
SEP-D4:
A solution of dendritic molecule D (0.1 g) in chloroform
(50 mL) was added dropwise into an aqueous solution (50 mL) of POM-1
8070
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8062 – 8071

Polyoxometalate-Anchored Dendron Complexes
FULL PAPER
through direct casting of the dendritic molecule D or SEP solutions
(0.5 mL, 1 mgmLꢀ1) onto silica substrates at ambient temperature. For
measurements of the self-assembled structures in the solid states, owing
to the poor solubility in ethanol all of the as-prepared SEPs were sus-
pended in ethanol by undergoing ultrasonic treatment. Taking SEP-D7 as
an example, the sample (1.0 mg) was added to ethanol (2.0 mL) and the
mixture was sonicated for 30 min. The sample film for XRD measure-
ment was prepared by directly casting 1 mL of the above mixture disper-
sion onto a silica substrate at ambient temperature. For identifying the
existing states of SEPs in acetone solution, all of the specimens for TEM
measurement were prepared through dropping some of the SEP solutions
(1 mgmLꢀ1) onto the carbon-coated copper grids, which were then dried
through suction of the excess sample solutions with filter paper at room
temperature. For the self-assembled structures in the solid state, the SEP
samples for TEM measurement were prepared in the same way as the
samples for XRD measurement. Taking SEP-D7 as an example, SEP-D7
(1.0 mg) was added to ethanol (2.0 mL) and the mixture was sonicated
for 30 min. Several drops of the dispersion were placed onto the carbon-
coated copper grid and dried through suction of the excess sample solu-
tion with filter paper.
Measurements: 1H NMR spectra were recorded on a Bruker Avance 500
instrument by using TMS as an internal reference. IR spectral measure-
ments on pressed KBr pellets were performed on a Bruker IFS66v FT-IR
spectrometer equipped with a DGTS detector (32 scans) with a resolu-
tion of 4 cmꢀ1. Element analysis (C, H, N) was carried out on a Flash
EA1112 analyzer from ThermoQuest Italia S.P.A. TG thermograms were
conducted on a Perkin–Elmer Diamond TG/DTA instrument with a
heating rate of 108Cminꢀ1 under flowing air. MALDI-TOF mass spectra
were recorded on a LDI-1700 mass spectrometer (Linear Scientific) with
chloroform as the solvent. XRD patterns were collected on a Rigaku X-
ray diffractometer (D/max rA, using CuKa radiation at 1.542 ꢂ), and the
data were recorded from 0.5 to 108. TEM images were obtained with a
Hitachi H8100 electron microscope with accelerating voltage of 200 kV
without staining. SEM measurements were performed on a JEOL JSM-
6700F field-emission scanning electron microscope. DLS measurements
were performed by using a Zetasizer NanoZS (Malvern Instruments).
The contact angles were carried out on a Drop Shape Analysis System
DSA20MK2 KRꢃSS (Edward Keller) at 258C. The amount of one drop-
let of water that was used for the measurement was approximately 3 mL.
[7] a) W. Qi, Y. Wang, W. Li, L. Wu, Chem. Eur. J. 2010, 16, 1068–
1078; b) H. Li, W. Qi, W. Li, H. Sun, W. Bu, L. Wu, Adv. Mater.
2005, 17, 2688–2692; c) W. Qi, H. Li, L. Wu, Adv. Mater. 2007, 19,
1983–1987; d) Y. Han, Y. Xiao, Z. Zhang, B. Liu, P. Zheng, S. He,
W. Wang, Macromolecules 2009, 42, 6543–6548; e) W. Bu, H. Li, H.
Sun, S. Yin, L. Wu, J. Am. Chem. Soc. 2005, 127, 8016–8017; f) H.
Zhang, X. Lin, Y. Yan, L. Wu, Chem. Commun. 2006, 4575–4577.
[8] a) D. G. Kurth, P. Lehmann, D. Volkmer, H. Cçlfen, M. J. Koop, A.
Mꢄller, A. Du Chesne, Chem. Eur. J. 2000, 6, 385–393; b) W. Bu, H.
Fan, L. Wu, X. Hou, C. Hu, G. Zhang, X. Zhang, Langmuir 2002,
18, 6398–6403.
[9] S. Nlate, D. Astruc, R. Neumann, Adv. Synth. Catal. 2004, 346,
1445–1448.
[10] W. Bu, L. Wu, X. Zhang, A.-C. Tang, J. Phys. Chem. B 2003, 107,
13425–13431.
[11] R. Carr, I. A. Weinstock, A. Sivaprasadarao, A. Mꢄller, A. Aksi-
mentiev, Nano Lett. 2008, 8, 3916–3921.
[12] a) M. Nyman, M. A. Rodriguez, T. M. Anderson, D. Ingersoll, Cryst.
Growth Des. 2009, 9, 3590–3597; b) H. Li, H. Sun, W. Qi, M. Xu, L.
Wu, Angew. Chem. 2007, 119, 1322–1325; Angew. Chem. Int. Ed.
2007, 46, 1300–1303; c) H. Sun, H. Li, W. Bu, M. Xu, L. Wu, J.
Phys. Chem. B 2006, 110, 24847–24854; d) Y. Wang, W. Li, L. Wu,
Langmuir 2009, 25, 13194–13200; e) Y. Yan, B. Li, W. Li, H. Li, L.
Wu, Soft Matter 2009, 5, 4047–4053; f) W. Li, W. Bu, H. Li, L. Wu,
M. Li, Chem. Commun. 2005, 3785–3787; g) W. Li, S. Yin, J. Wang,
L. Wu, Chem. Mater. 2008, 20, 514–522; h) S. Yin, W. Li, J. Wang, L.
Wu, J. Phys. Chem. B 2008, 112, 3983–3988; i) T. Akutagawa, R.
Jin, R. Tunashima, S. Noro, L. Cronin, T. Nakamura, Langmuir
2008, 24, 231–238.
[13] a) S. Polarz, B. Smarsly, M. Antonietti, ChemPhysChem 2001, 2,
457–461; b) T. Zhang, C. Spitz, M. Antonietti, C. F. J. Faul, Chem.
Eur. J. 2005, 11, 1001–1009; c) X. Lin, Y. Wang, L. Wu, Langmuir
2009, 25, 6081–6087; d) D. Volkmer, A. Du Chesne, D. G. Kurth, H.
Schnablegger, P. Lehmann, M. J. Koop, A. Mꢄller, J. Am. Chem.
Soc. 2000, 122, 1995–1998.
[14] D. Volkmer, B. Bredenkçtter, J. Tellenbrçker, P. Kçgerler, D. G.
Kurth, P. Lehmann, H. Schnablegger, D. Schwahn, M. Piepenbrink,
B. Krebs, J. Am. Chem. Soc. 2002, 124, 10489–10496.
[15] a) R. Brodbeck, T. Tçnsing, D. Andrae, D. Volkmer, J. Phys. Chem.
B 2008, 112, 5153–5162; b) S. Nlate, L. Plault, D. Astruc, Chem.
Eur. J. 2006, 12, 903–914.
[16] a) I. A. Weinstock, J. J. Cowan, E. M. G. Barbuzzi, H. Zeng, C. L.
Hill, J. Am. Chem. Soc. 1999, 121, 4608–4617; b) N. Haraguchi, Y.
Okaue, T. Isobe, Y. Matsuda, Inorg. Chem. 1994, 33, 1015–1020;
c) B. J. Hornstein, R. G. Finke, Inorg. Chem. 2002, 41, 2720–2730.
[17] a) C. F. J. Faul, M. Antonietti, Adv. Mater. 2003, 15, 673–683; b) H.
Li, P. Li, Y. Yang, W. Qi, H. Sun, L. Wu, Macromol. Rapid
Commun. 2008, 29, 431–436; c) H. Li, W. Qi, H. Sun, P. Li, Y. Yang,
L. Wu, Dyes Pigm. 2008, 79, 105–110.
[18] D. G. Kurth, P. Lehmann, D. Volkmer, A. Mꢄller, D. Schwahn, J.
Chem. Soc. Dalton Trans. 2000, 3989–3998.
[19] J. Niu, X. You, C. Duan, H. Fun, Z. Zhou, Inorg. Chem. 1996, 35,
4211–4217.
Acknowledgements
This work was financially supported by the National Basic Research Pro-
gram of China (2007CB808003), the National Natural Science Founda-
tion of China (20973082, 20921003, 20703019), and MOST. We thank the
111 Project (B06009) for supporting the visit of Prof. I. Kim at Pusan Na-
tional University who provided fruitful discussion, as well as the Open
Project of State Key Laboratory of Polymer Physics and Chemistry of
the CAS.
[20] P. B. Rheiner, D. Seebach, Chem. Eur. J. 1999, 5, 3221–3236.
[21] a) R. A. MacPhail, H. L. Strauss, R. G. Snyder, C. A. Elliger, J.
Phys. Chem. 1984, 88, 334–341; b) Q. Wang, B. Zhao, X. Zhang, J.
Shen, Y. Ozaki, Langmuir 2002, 18, 9845–9852.
[1] D. Schrçder, M. Diefenbach, T. M. Klapçtke, H. Schwarz, Angew.
Chem. 1990, 102, 119–157; Angew. Chem. Int. Ed. Engl. 1990, 29,
138–175.
[22] J. W. Goodby, I. M. Saez, S. J. Cowling, V. Gçrtz, M. Draper, A. W.
Hall, S. Sia, G. Cosquer, S. E. Lee, E. P. Raynes, Angew. Chem.
2008, 120, 2794–2828; Angew. Chem. Int. Ed. 2008, 47, 2754–2787.
[23] a) C. Tschierske, J. Mater. Chem. 1998, 8, 1485–1508; b) W. Li, J.
Zhang, B. Li, M. Zhang, L. Wu, Chem. Commun. 2009, 5269–5271.
[24] Z. Bo, X. Zhang, X. Yi, M. Yang, J. Shen, Y. Rehn, S. Xi, Polym.
Bull. 1997, 38, 257–264.
[2] a) D. K. Smith, A. R. Hirst, C. S. Love, J. G. Hardy, S. V. Brignell, B.
Huang, Prog. Polym. Sci. 2005, 30, 220–293; b) B. Donnio, S. Bua-
thong, I. Bury, D. Guillon, Chem. Soc. Rev. 2007, 36, 1495–1513;
c) Special issue “Dendrimers and Nanoscience”: D. Astruc, C. R.
Chim. 2003, 6, 709–711.
[3] J. Kofoed, J. L. Reymond, Curr. Opin. Chem. Biol. 2005, 9, 656–664.
[4] U. Boas, P. M. H. Heegaard, Chem. Soc. Rev. 2004, 33, 43–63.
[5] D. Astruc, F. Chardac, Chem. Rev. 2001, 101, 2991–3023.
[6] a) M. T. Pope, A. Mꢄller, Angew. Chem. 1991, 103, 56–70; Angew.
Chem. Int. Ed. Engl. 1991, 30, 34–48; b) Special issue on “Polyoxo-
metalates”: Chem. Rev. 1998, 98, 1–390.
Received: January 25, 2010
Published online: June 16, 2010
Chem. Eur. J. 2010, 16, 8062 – 8071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8071