Structural elucidation of dendritic host-guest complexes by X-ray crystallography and molecular dynamics simulations

Article abstract of DOI:10.1002/chem.200700572

The multiple monovalent binding of adamantyl-urea poly(propyleneimine) dendrimers with carboxylic acid-urea guests was investigated using molecular dynamics simulations and X-ray crystallography to better understand the structure and behavior of the dynam

Full text of DOI:10.1002/chem.200700572

FULL PAPER
DOI: 10.1002/chem.200700572
Structural Elucidation of Dendritic Host–Guest Complexes by X-ray
Crystallography and Molecular Dynamics Simulations
[
a]
[b]
[a]
[c]
Theresa Chang, Koen Pieterse, Maarten A. C. Broeren, Huub Kooijman,
[
c]
[b]
[a]
Anthony L. Spek, Peter A. J. Hilbers, and E. W. Meijer*
Abstract: The multiple monovalent
binding of adamantyl–urea poly(propy-
leneimine) dendrimers with carboxylic
acid–urea guests was investigated using
molecular dynamics simulations and X-
ray crystallography to better under-
stand the structure and behavior of the
dynamic multivalent complex in solu-
tion. The results from the two methods
are consistent and suggest a preferred
molecular picture of this complicated
aggregate of multiple components. The
guest molecules can bind to the dendri-
mer in a variety of ways although most
involve hydrogen-bonding interactions
between urea groups of the dendrimer
with urea and/or carboxylic acid groups
of the guest. In addition, acid–base in-
teractions between the carboxylic acid
of the guest and the tertiary amine in
the interior of the dendritic host are
present. Our proposed structure gives
important information about the pre-
dominant dynamic interactions be-
tween the host and guest and illustrates
how they fit together and interact with
each other.
Keywords: computer chemistry
dendrimers · host–guest systems ·
molecular dynamics
structure elucidation
·
·
Introduction
tyl–urea dendrimers using noncovalent interactions resulting
in a statistical, yet adaptable, distribution of ligands at the
periphery of the dendrimer scaffold.
[5]
Multiple monovalent interactions are the simultaneous asso-
ciation of multiple ligands of one molecule or complex to
multiple binding sites on a given receptor and are important
in numerous biological processes such as cell recognition
events. Multiple interactions often have enhanced properties
that are significantly different from the properties of their
constituent monovalent ligands. An effective way of achiev-
ing a multivalent display of ligands is by attaching the li-
The adamantyl–urea dendrimers, made by reacting ada-
mantyl isocyanate with poly(propyleneimine) dendrimers,
were designed to bind to acid–urea guests via a combination
of electrostatic and hydrogen-bonding interactions
(Scheme 1). Upon mixing, proton transfer would occur from
the carboxylic acid of the guest to the tertiary amine of the
pincer and the two would be held together by electrostatic
(ion pairing) interactions. The complex would be further sta-
bilized by ureaÀurea hydrogen bonds with the guest sand-
[1–4]
gands of interest to the periphery of dendrimers.
have previously reported the functionalization of adaman-
We
wiched between the urea groups of the pincer-like moiety.
The urea moieties of the dendrimer were appended with
bulky adamantyl groups to reduce intramolecular hydrogen
bonding of urea groups within pincers, but they should bind
efficiently to guest molecules with less bulky side chains.
Many experiments showed spectroscopic evidence for the
depicted design, although a static picture can never visualize
in a reasonable way a highly dynamic multicomponent
system. Like the components of a cell or the arrangement of
surfactants in micelles and vesicles, the dendrimer–guest
complex is neither rigid nor static and the noncovalent inter-
actions holding the components together are constantly
broken and reformed. The spherical and dynamic nature of
dendrimers makes single crystal studies limited to the lower
[
a] Dr. T. Chang, Dr. M. A. C. Broeren, Prof. E. W. Meijer
Laboratory of Macromolecular and Organic Chemistry
Eindhoven University of Technology
PO Box 513, 5600 MB Eindhoven (The Netherlands)
Fax : (+31)40-245-1036
E-mail: e.w.meijer@tue.nl
[
b] Dr. K. Pieterse, Prof. P. A. J. Hilbers
Biomodeling and Bioinformatics
Eindhoven University of Technology
PO Box 513, 5600 MB Eindhoven (The Netherlands)
[
c] Dr. H. Kooijman, Prof. A. L. Spek
Bijvoet Center for Biomolecular Research
Crystal Structural Chemistry, Utrecht University
Padualaan 8, 3584 CH Utrecht (The Netherlands)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
[6]
generations. Molecular dynamics simulations of dendrim-
Chem. Eur. J. 2007, 13, 7883 – 7889
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7883

Scheme 1. Binding of an acetic acid–urea guest to a dendrimer.
ers are often focused on the overall shape, density profile
served between pincers, forming infinite hydrogen-bond
stacks in both polymorphs with urea groups rotated by 908
relative to each other. A pincer without bulky groups shows
efficient intra-pincer hydrogen bonding as illustrated in Fig-
ure 2c. The phenyl pincers are also more difficult to dissolve
in organic solvents, exemplifying the difficulty in breaking
these intra-pincer hydrogen bonds to insert a guest.
A phenyl-terminated guest molecule 3 results in a colum-
nar packing structure (Figure 3a) that is reminiscent of that
of the phenyl pincer 2. An infinite hydrogen-bonding stack
is formed with guest molecules rotated 1808 relative to the
ones above and below. Acid–urea [OÀH···O] hydrogen
[7]
and positions of the end groups. Specific supramolecular
attention in the simulation of dendrimers has been limited
[8]
to encapsulated host–guest systems. In this paper, we wish
to present an accurate representation of this dynamic host–
guest complex, based on analyses of the crystal packing
structures of model compounds and atomistic molecular dy-
namics simulations. Our proposed structure will give impor-
tant information about the predominant dynamic interac-
tions between the host and guest and illustrate how they fit
together and interact with each other.
bonds link these columns together (Figure 3b), however, no
p–p stacking interactions were observed with either pincer 2
or guest 3. The crystal structure of the adamantyl guest (Fig-
ure 3c and d) is in agreement with what was observed previ-
ously. The bulky adamantyl groups prevented urea-urea
bonds so that hydrogen bonds are formed between the urea
moieties with carboxylic acids of adjacent guest molecules
instead.
The triethylammonium salt of the guest 3 was recrystal-
lized to probe acid–base interactions present in the dendri-
mer–guest complex. In this crystal structure (Figure 4) we
observed that only half of the carboxylic acids are deproton-
ated, a common occurrence for weak acids in the solid
Results
[
9]
X-ray crystallography: Since crystals of dendritic structures
were difficult to come by, we have obtained X-ray quality
single crystals of what we thought are the relevant binding
sites of the dendrimer, the pincer-like structures with ada-
mantyl and phenyl groups (1 and 2, respectively; see Sup-
porting Information for preparation of 1), as well as guest
[10]
[11]
molecules
with the same end groups (3, 3·TEA, and 4)
shown in Figure 1. Co-crystals of pincer–guest complexes
[
12]
were also attempted but without success, suggesting that a
guest sandwiched between the arms of a pincer is not neces-
sarily a favored (stable) structure in a crystalline lattice.
The packing structures of the adamantyl pincer show that
the pincer is quite flexible (Figure 2a and b). Two poly-
morphs were found in crystals grown from the same solvent
system. These packing structures show that the adamantyl
group is bulky enough to prevent intra-pincer hydrogen
bonding between urea groups. Hydrogen bonding is ob-
[
13]
À
state,
as a strong [OÀH···O ] hydrogen bond is formed.
Urea–urea hydrogen bonds are not present in this crystal
structure because of the possibility of ion–dipole interac-
tions, which are usually stronger than dipole–dipole (hy-
drogen bonding) interactions. We can conclude from the
crystal structures shown above that there are many more
types of interactions available to join dendrimer and acid–
[14]
Figure 1. Structures investigated using X-ray crystallography.
7884
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7883 – 7889

Dendritic Host–Guest Complexes
FULL PAPER
Figure 2. Solid-state superstructure of a,b) two polymorphs of the adamantyl pincer and c) the phenyl pincer.
Figure 3. a) Phenyl guests form infinite stacks via urea-urea hydrogen bonding and b) the stacks are linked to other stacks via acid-urea bonds. Adaman-
tyl–urea guests form columns via c) [NÀH···O] urea acid hydrogen bonding within the column and d) [OÀH···O] acid–urea hydrogen bonding between
stacks to the carboxylic acid groups.
À
Figure 4. X-ray crystal structure of 3·TEA shows the formation of a) a protonated–unprotonated acid pair linked together with an ion–dipole [OÀH···O ]
À
interaction as well as b) [NÀH···O ] hydrogen bond. The overall packing structure is composed of the aforementioned interactions as well as ammonium
ion–urea interactions (c).
[15]
urea guests in addition to the carboxylate/ammonium and
urea–urea hydrogen-bonding interactions illustrated in
Scheme 1.
Molecular dynamics simulations: Using NAMD 2.5, mo-
lecular dynamics (MD) simulations were performed on the
3rd- and 5th-generation dendrimers shown in Figure 5 (8
Chem. Eur. J. 2007, 13, 7883 – 7889
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7885

E. W. Meijer et al.
The diaminobutane core of the
dendrimers is situated in the
center of the aggregates. Over-
all, the bulky end groups of the
adamantyl-terminated dendrim-
ers result in more crowding at
the periphery compared with
the phenyl-terminated dendrim-
ers.
The driving force for the for-
mation of each of the structures
Figure 5. Structures investigated using MD simulations.
and 32 pincer-like moieties, respectively) having adamantyl
(
D1a, D1b) or phenyl groups (D2a, D2b) at the periphery.
The dendrimers were analyzed both with and without acid–
urea guests 5. The dendrimers were simulated for 2 ns using
1
fs time steps, and atom coordinates and velocities were
sampled every picosecond. In simulations that included
acid–urea guests (8 and 32 guests for 3rd- and 5th-genera-
tion dendrimers, respectively) initial structures were gener-
ated in 1 ns pre-simulations from extended dendrimer struc-
tures (for details see Supporting Information). Guest mole-
cules were uniformly distributed on the surface of a sphere,
tightly enclosing these dendrimers. For simulations that only
involved the dendrimer, we used just the extended dendri-
mer structures as a starting point. These extended structures
quickly collapse to form a more compact structure as favor-
able van der Waals interactions and hydrogen bonds (by
means of electrostatic interactions) are formed between dif-
ferent sections of the macromolecule/complex. In each simu-
lation, a starting temperature of 500 K was used and at this
temperature noncovalent bonds could still be broken and re-
formed. The temperature was gradually decreased to 300 K
in the first nanosecond and remained at 300 K during the
last nanosecond of the simulations. As the temperature de-
creases, the conformations of the dendrimers are locked, as
the system no longer has enough energy to break hydrogen
bonding and/or acid–base interactions. The acid–base reac-
tions of these guests with the basic dendrimers were simulat-
ed as acid–base interactions (i.e., electrostatic interactions)
Figure 6. Structure of the dendrimers and dendrimer complexes after a
2 ns simulation. The dendrimers are depicted in blue (darker blue for the
interior), while the guests are colored pink. For each structure a front
and side view is shown.
mentioned above is the stabilizing electrostatic interactions
arising for the most part from hydrogen-bonding and acid–
base interactions. Hydrogen bonds between urea groups
make up a majority of these stabilizing forces because of
both i) the large number of urea groups present in either the
dendrimer or dendrimer–guest complex and ii) the strength
of these bifurcated hydrogen bonds. The hydrogen-bonding
interactions between the urea groups of the dendrimer and
guest molecules were analyzed using the atom trajectories
obtained from the molecular dynamics simulations outlined
in Table 1. The urea–urea hydrogen-bonding interactions an-
alyzed are defined as consisting of two hydrogen bonds
where the distances between the hydrogen atoms of the
donor and the oxygen atom of the acceptor are 2.5 Š or
less. Hence, the total number of hydrogen-bonding interac-
tions possible is the same as the number of urea groups.
Even though nearly all the urea groups participate in hydro-
gen-bonding interactions, only about 45% of these are bifur-
[16]
consistent with the gas phase.
As shown in Figure 6, 3rd-generation adamantyl– and
phenyl–urea dendrimers, D1a and D2a, do not exhibit a
globular structure. The groups at the periphery (adamantyl
or phenyl) tend to cluster on one side of the molecule to
maximize the number of urea–urea hydrogen-bonding inter-
actions, leaving the interior of the dendrimer as well as the
diaminobutane core exposed. Addition of guests improves
the surface coverage, resulting in a more spheroid shape,
however, still having an accessible dendrimer interior. Fifth
generation dendrimers (D1b, D2b) are more globular in
shape although patches of dendrimer interior are still ex-
posed. In some cases, even the diaminobutane core is still
exposed on these dendrimers. The addition of guests 5 to
5th-generation dendrimers gave structures with only small
portions of the dendrimer interior appearing at the surface.
7886
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7883 – 7889

Dendritic Host–Guest Complexes
FULL PAPER
Table 1. Urea–urea hydrogen-bonding distributions in dendrimer/dendri-
mer complexes.
Table 2. Number and location of acid–base interactions in dendrimer–
guest complexes.
[
b]
Dendrimer Urea
Urea linkages present
Dendrimer struc-
ture
Acid–base
Tertiary amines involved
in acid–base interactions
[
a]
structure
linkages [%]
interactions
[
a]
[b]
within
between
between
[%]
at location
dendrimer [%]
dendrimer
guests [%]
a
b
g
d
e
and guests [%]
[%]
[%]
[%]
[%]
[%]
D1a
D1a·5
D1b
D1b·532
D2a
D2a·5
D2b
43
45
48
47
42
49
46
43
100
48
100
45
100
52
100
49
–
48
–
46
–
45
–
42
–
4
–
9
–
3
–
9
D1a·5
D1b·532
D2a·5
D2b·532
8
42
36
22
38
20
6
0
20
3
37
3
60
12
63
8
–
22
–
–
57
–
8
8
0
16
74
[
a] Given as the percentage of possible acid–base interactions. [b] Subdi-
8
vision of the percentage of acid–base interactions per location of the in-
volved tertiary amines, where a–e represent each successive dendrimer
shell (a being the dendrimer core).
D2b·532
[
a] Given as the percentage of possible urea linkages. [b] Subdivision of
the percentage of urea linkages per linkage type.
number of other stabilizing interactions (including urea–
urea hydrogen bonds) in the complex.
cated out of all possible linkages in all dendrimers with or
without acid–urea guests. The remainder participate either
in longer distance hydrogen-bonding interactions with urea
or carboxylate groups or interact only with one hydrogen
bond.
The urea–urea hydrogen bond and acid–base interactions
mentioned in the pincer model only accounts for two out of
the eight possible combinations of donor/guest interactions.
In a dendrimer–guest complex, there are two types of hy-
drogen bond donors, urea NÀH and acid OÀH and they can
The presence or absence of guest molecules does not
change the percentage of urea–urea hydrogen-bonding inter-
actions in the macromolecule/complex. As a result of steric
issues arising from bulky adamantyl groups, urea–urea
bonds found in simulations regarding adamantyl–urea den-
drimers (with or without guests) exhibit torsion angles of
around 908 (perpendicular orientation) as was found in all
crystal structures containing adamantyl urea groups. In con-
trast, there was no preferential urea–urea torsion angle
found in simulations pertaining to phenyl dendrimers.
In dendrimer host–guest complexes, about 45% of all the
ureaÀurea bonds formed are between dendrimer and guests,
form hydrogen bonds with all four acceptors, urea C=O,
acid C=O, acid OÀH as well as the tertiary amines of the
dendrimer. Hydrogen-bonding interactions with the acid OÀ
H as acceptor are negligible for the most part. However, all
of the C=O groups of carboxylic acid groups of guest mole-
cules are involved in hydrogen-bonding interactions, the ma-
jority of them (ꢀ80%) with urea NÀH groups and the rest
with acid OÀH groups of another guest. In addition, bifur-
cated urea–urea hydrogen bonds only account for ꢀ80% of
all urea–urea hydrogen bonds in the dendrimer–guest com-
plexes.
It is noteworthy that the behavior of the 5th-generation
phenyl dendrimer complex is, in most respects, very similar
to that of the analogous adamantyl dendrimer complex. The
main difference between the two structures is the urea–urea
hydrogen bond torsion angles. The adamantyl dendrimers
and complexes exhibit strong preference for orthogonal bi-
furcated hydrogen bonds while the torsion angles of the
phenyl dendrimers and complexes appear to be randomly
distributed. The latter seems contradictory to the preferen-
tial co-planar or perpendicular orientation of urea–urea
signifying the full integration of guest molecules into these
dendrimers. Approximately 50% of urea linkages are
formed between urea groups within dendrimers, presumably
because they constitute 67% of urea groups in the complex.
Hydrogen-bonding interactions of this type between guest
molecules are small in most cases (3–9%), with a tendency
to be larger in case of the fifth-generation dendrimers.
Acid–base interactions between the tertiary amine groups
of the dendrimer and the carboxylic acids of the guests are
also important for the complexation event and are the
second stabilizing interaction holding the guest and dendri-
mer together. In the gas phase, proton transfer is only likely
between strong acids and strong bases and, hence, in these
dendrimer–guest complexes acid–base interactions should
exist as hydrogen bonds between two polarized groups
rather than an ion pair. Since tertiary amines are present in
all layers of the dendrimer (even at the core), it is conceiva-
ble that guest molecules can bind to amines in the interior
of the dendrimer as well as with those near the periphery.
However, acid–urea guests are mainly found near the pe-
riphery of the dendrimer (Table 2). The guests interact with
amines at the outer layers of the dendrimer because there
are more tertiary amines available and to maximize the
[17]
bonds,
but is the direct result of modeling hydrogen
bonds through electrostatic interactions (which lack direc-
tionality). Hence, the perpendicular orientation of the urea–
urea linkages in adamantyl-functionalized dendrimers can
solely be attributed to steric repulsions.
Discussion
Both the crystal structures and the molecular dynamics sim-
ulations clearly show that binding of the acid–urea guests to
the dendrimer via the mechanism proposed in Scheme 1 is
unlikely. The chaotic nature of the host–guest complex, how-
ever, makes it difficult to visualize what exactly is happening
Chem. Eur. J. 2007, 13, 7883 – 7889
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7887

E. W. Meijer et al.
in the dendrimer. There are many feasible ways a guest mol-
ecule can bind to the dendrimer, often via multiple interac-
tions to different parts of the dendrimer, and sometimes
even to other guests. In fact, all the secondary interactions
found in the crystal structures are also present in the MD
simulations and conceivably contribute to the overall stabili-
ty as well as complexity of the system. The examples shown
in Figure 7 exemplify the typical interactions found in a den-
drimer–guest complex.
guest molecules bound to the dendritic host. For the com-
plex, the urea groups will provide most of the stabilizing in-
teractions from participating in ureaÀurea [N-H···O], acidÀ
urea [OÀH···O] and ureaÀacid [NÀH···O] hydrogen bonds.
The adamantyl substituted urea groups will give bifurcated
ureaÀurea hydrogen bonds with the urea groups aligned or-
thogonal to each other, whereas a less bulky group will lead
to a more random conformation. When the dendrimer or
guest molecules contain phenyl groups, they might partici-
pate in p-p stacking interac-
tions. Proton transfer might
occur between the acid groups
of the guests and the tertiary
amines of the dendrimer, de-
pending on the solventꢁs ability
to stabilize the newly formed
ion pair.
Although it cannot do justice
to the three dimensional, intri-
cate and dynamic nature of the
dendrimer–guest
complexes,
Figure 8 summarizes the most
Figure 7. Simulation snapshot of dendrimer–guest interactions (a–c) in a complex (middle) of D1b with 32 important interaction types
acid–urea guests 5.
found in
a dendrimer–guest
complex in a two dimensional
Figure 7a shows that guest molecules sometimes do bind
to the pincer-like moiety in the dendrimer as designed, sand-
wiched between two arms of a pincer, although not exactly
in the way shown in Scheme 1. In this example, a guest mol-
ecule forms urea–urea hydrogen bonds with one arm of a
pincer, while the carbonyl group of the carboxylic acid
moiety binds to the urea group of the other arm. It also
binds, via an acid–base interaction, to a tertiary amine of a
neighboring branch.
Guest molecules can interact favorably with other
guest(s) in addition to its links with the dendrimer. In Fig-
ure 7b, the central guest molecule has urea–urea hydrogen
bonds with two other guests. Furthermore, its carboxylic
acid group has two additional interactions: an acid–base in-
teraction with a tertiary amine and a hydrogen bond with
the acidic proton of the guest below.
Often, a hydrogen-bond stack found in many of the crys-
tal structures composed of urea moieties can be found in the
dendrimer–guest complex as shown in Figure 7c. Here we
see a somewhat curved urea stack, composed of alternating
guests and pincer arms, each time the molecule rotates rela-
tive to the urea stack as a result of either steric reasons or
reorientation of the molecule to optimize interactions with
another part of the dendrimer.
Considering the results found in the crystal structures and
MD simulations, what will the dendrimer–guest complexes
look like in solution? Of course, the structure of the com-
plex will be dependent on the polarity of the solvent, but
will be less condensed than the crystal or simulated struc-
tures as a result of competitive van der Waals and electro-
static interactions with the solvent molecules. While obvi-
ously, the association constant will determine the number of
Figure 8. Proposed dendrimer complex structure in solution, showing the
predominant interaction types between dendrimer and guest.
representation. Moreover, the structure presented here,
based on X-ray of models and simulations, is in full agree-
ment with earlier spectroscopic and mass spectrometry data
as well as light scattering.
Conclusion
The dendrimer–guest systems studied here are dynamic and
these host–guest aggregates cannot be easily represented
7888
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7883 – 7889

Dendritic Host–Guest Complexes
FULL PAPER
with a two- or even three-dimensional static picture. We
have used crystal structures of model compounds and mo-
lecular dynamics simulations to try to understand how
guests occupy the “binding sites” in adamantyl–urea den-
drimers. We have found that there is not one clear way that
guest molecules bind to the dendrimer although they do
share some common characteristics and usually form multi-
ple interactions with different parts of the dendrimer. We
have proposed a preferred structure that is in agreement
with all reported spectroscopic data. Since, dendrimers are
gaining more and more interest as carriers for (bioactive)
guests, the presented studies is of relevance for all dendritic
guest–host systems and other multiple monovalent interact-
ing systems so often observed in Nature and nanotechnolog-
ical applications.
[6] a) G. Magro, P. Marchand, R.-M. Sebastiµn, C. Guyard-Duhayon,
A.-M. Caminade, J.-P. Majoral, Eur. J. Org. Chem. 2005, 1340–1347;
b) J. W. Kriesel, S. Konig, M. A. Freitas, A. G. Marshall, J. A. Leary,
T. D. Tilley, J. Am. Chem. Soc. 1998, 120, 12207–12215; c) A. W.
Bosman, M. J. Bruining, H. Kooijman, A. L. Spek, R. A. J. Janssen,
E. W. Meijer, J. Am. Chem. Soc. 1998, 120, 8547–8548.
[
7] a) P. Brocorens, R. Lazzaroni, J.-L. BrØdas, J. Phys. Chem. B 2005,
09, 19897–19907; b) T. C¸ a gˇ in, G. Wang, R. Martin, N. Breen,
1
W. A. Goddard III, Nanotechnology 2000, 11, 77–84; c) L. Cavallo,
F. Fraternali, Chem. Eur. J. 1998, 4, 927–934; d) C. B. Gorman, J. C.
Smith, Polymer 2000, 41, 675–683; e) M. Han, P. Chen, X. Yang, Po-
lymer 2005, 46, 3481–3488; f) K. J. Haxton, D. J. Cole-Hamilton,
R. E. Morris, Dalton Trans. 2004, 1665–1669; g) C. Jana, G. Jaya-
murugan, R. Ganapathy, P. K. Maiti, N. Jayaraman, A. K. Sood, J.
Chem. Phys. 2006, 124, 204719; h) H. Lee, J. R. Baker Jr. , R. G.
Larson, J. Phys. Chem. B 2006, 110, 4014–4019; i) P. K. Maiti, T.
C¸ a gˇ in, G. Wang, W. A. Goddard III, Macromolecules 2004, 37,
6
236–6254; j) P. K. Maiti, T. C¸ a gˇ in, S.-T. Lin, W. A. Goddard III,
Macromolecules 2005, 38, 979–991; k) W. Ortiz, A. E. Roitberg,
J. L. Krause, J. Phys. Chem. B 2004, 108, 8218–8225; l) S. Pricl, M.
Fermeglia, M. Ferrone, A. Asquini, Carbon 2003, 41, 2269–2283;
m) R. Scherrenberg, B. Coussens, P. van Vliet, G. Edouard, J. Brack-
man, E. de Brabander, K. Mortensen, Macromolecules 1998, 31,
456–461; n) N. Zacharopoulos, I. G. Economou, Macromolecules
2002, 35, 1814–1821.
Acknowledgements
This work was supported in part by the Council for the Chemical Scien-
ces of the Netherlands Organization for Scientific Research (CW-NWO)
and in part by the EC Sixth Framework Programme (as part of the
STREP Biomach, contract #505487-1).
[8] a) P. Miklis, T. C¸ aˇ in, W. A. Goddard III, J. Am. Chem. Soc. 1997,
119, 7458–7462; b) G. Teobaldi, F. Zerbetto, J. Am. Chem. Soc.
2
003, 125, 7388–7393.
[
9] CCDC-648377–648382 contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.cam.
ac.uk/data_request/cif.
[
1] For some reviews on supramolecular chemistry of dendrimers see:
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) P. J. Gittins, L. J.
Twyman, Supramol. Chem. 2003, 15, 5–23; c) M. W. P. L. Baars,
E. W. Meijer, Top. Curr. Chem. 2000, 210, 131–182; d) G. R. New-
kome, E. He, C. N. Moorefield, Chem. Rev. 1999, 99, 1689–1746;
e) F. Zeng, S. C. Zimmerman, Chem. Rev. 1997, 97, 1681–1712.
2] Dendrimers can be made using supramolecular interactions: a) S. C.
Zimmerman, F. Zeng, D. E. C. Reichert, S. V. Kolotuchin, Science
[10] Crystals of 1 were grown by vapor diffusion of isopropyl ether into
a solution of the acetate salt of the pincer in methanol and acetone
and 2 by slow evaporation/cooling of a heated methanolic solution
(wet) containing the pincer.
[
[11] X-ray quality single crystals were grown by vapor diffusion of iso-
propyl ether into a solution of 3 in methanol. The triethyl ammoni-
um salt 3·TEA crystals were obtained by the vapor diffusion of iso-
propyl ether into an ethanol/acetone solution of the salt. The ada-
mantyl guest 4 was recrystallized by vapor diffusion of isopropyl
ether into a solution of 4 in a mixture of ethanol and chloroform.
[12] Crystals of 1 were obtained in our attempt to grow co-crystals of the
pincer with acetic acid.
1
996, 271, 1095–1098; b) K. C.-F. Leung, F. Aricó, S. J. Cantrill, J. F.
Stoddart, J. Am. Chem. Soc. 2005, 127, 5808–5810; c) W. T. S. Huck,
L. J. Prins, R. H. Fokkens, N. M. M. Nibbering, F. C. J. M. van Veg-
gel, D. N. Reinhoudt, J. Am. Chem. Soc. 1998, 120, 6240–6246;
d) H. W. Gibson, N. Yamaguchi, L. Hamilton, J. W. Jones, J. Am.
Chem. Soc. 2002, 124, 4653–4665.
[
[
3] a) S. C. Zimmerman, Y. Wang, P. Bharathi, J. S. Moore, J. Am.
Chem. Soc. 1998, 120, 2172–2173; b) A. M. Naylor, W. A. I. God-
dard, G. E. Kiefer, D. A. Tomalia, J. Am. Chem. Soc. 1989, 111,
[13] a) A. Ballabh, D. R. Trivedi, P. Dastidar, E. Suresh, CrystEngComm
2002, 4, 135–142; b) B. L. Chen, S.-C. Ng, Y.-L. Feng, S.-X. Liu,
Polyhedron 1998, 17, 4237–4247.
[14] The strength of dipole–dipole interactions is proportional to the in-
verse cube of the distance whereas that of ion–dipole is proportional
to the inverse square of the distance.
[15] L. KalØ, R. Steel, M. Bhandarkar, R. Brunner, A. Gursoy, N. Kra-
wetz, J. Phillips, A. Shinozaki, K. Varadarajan, K. Schulten, J.
Comp. Physiol. 1999, 151, 283–312.
2
1
339–2341; c) P. Walliman, P. Seiler, F. Diedrich, Helv. Chim. Acta
996, 79, 779–788.
4] a) N. Tomioka, D. Takasu, T. Takahashi, T. Aida, Angew. Chem.
998, 110, 1611–1614; Angew. Chem. Int. Ed. 1998, 37, 1531–1534;
b) S. Jockusch, N. J. Turro, D. A. Tomalia, Macromolecules 1995, 28,
416–7418; c) B. Alonso, I. Cuadrado, M. Moran, J. Losada, Chem.
1
7
Mater. 1995, 7, 1440–1442; d) Y. B. Lim, T. Kim, J. W. Lee, S. M.
Kim, H. J. Kim, K. Kim, J. S. Park, Bioconjugate Chem. 2002, 13,
[16] S. Alavi, D. L. Thompson, J. Phys. Chem. A 2004, 108, 8801–8809.
[17] H.-Y. Liao, S.-Y. Chu, New J. Chem. 2003, 27, 421–424.
1
181–1185.
[
5] D. Banejee, M. A. C. Broeren, M. H. P. van Genderen, E. W. Meijer,
Received: April 13, 2007
P. L. Rinaldi, Macromolecules 2004, 37, 8313–8318.
Published online: July 5, 2007
Chem. Eur. J. 2007, 13, 7883 – 7889
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7889