Cationic shape-persistent macrocycles: The unexpected formation of a nano-size supramolecular dimer

Article abstract of DOI:10.1021/ja057999g

Shape-persistent macrocycles with a rigid pyridyl core and a flexible oligo-alkyl corona aggregate to some extent in nonpolar solvents to form large (tubular) aggregates. To increase the assembling tendency, the intraannular pyridyl groups of the macrocyc

Full text of DOI:10.1021/ja057999g

Published on Web 02/16/2006
Cationic Shape-Persistent Macrocycles: The Unexpected Formation of a
Nano-Size Supramolecular Dimer
Svetlana Klyatskaya,†,§ Nico Dingenouts,*,† Christine Rosenauer,‡ Beate Mu¨ller,‡ and Sigurd Ho¨ger*,†
Institute for Chemical Technology and Polymer Chemistry, UniVersita¨t Karlsruhe,
Engesserstr. 18, 76131 Karlsruhe, Germany, and Max Planck Institute for Polymer Research,
Ackermannweg 10, 55128 Mainz, Germany
Received November 24, 2005; E-mail: hoeger@chemie.uni-karlsruhe.de; dingenouts@polymer.uni-karlsruhe.de
1
The supramolecular organization of shape-persistent macrocycles
in one, two, or three dimensions has recently become a topic of
great interest.1 Apart from infinite aggregates, the formation of well-
defined finite-size assemblies from mesoscale building blocks is
currently an attractive goal in supramolecular chemistry.2 For
example, a dimerization has been observed for hemi-alkylated
cyclopeptides or for macrocycles with chiral helicene units in the
backbone.3-5 Furthermore, nanoscale supramolecular objects are
formed by metal-supported self-assembly of rigid rings containing
exotopic binding sites.6 One of our aims during the last years dealt
with the 1D organization of shape-persistent macrocycles by
solvophobic interactions. We could show that rigid macrocycles
that are decorated with oligo-alkyl side chains at their periphery
aggregate in solvent mixtures that contain a nonsolvent for the rigid
core.7 However, only a weak association could be observed due to
the restricted compound solubility that did not allow an increase
of the association constants by further decreasing the solvent
polarity. One way to overcome the solubility problem is to attach
polystyrene oligomers to the rigid backbone. The coil-ring-coil
block copolymers are well soluble in cyclohexane at slightly
elevated temperatures.8 Below about 35 °C, they aggregate into
supramolecular hollow cylindrical brushes.
work. A broadening of the signals in the H NMR spectrum in
cyclohexane was a first hint of an aggregation of the molecules
(see Supporting Information). The aggregate formation was further
investigated by dynamic light scattering (DLS) measurements. The
distribution of relaxation times in the autocorrelation function was
obtained by CONTIN analysis.10 It contains two maxima, one at a
hydrodynamic radius of approximately 2 nm, which corresponds
to the macrocycle 1, the other at approximately 50 nm (Figure 1a).
The fraction of the larger species in solution increases with higher
concentration. We assume that the larger objects are a mixture of
tubular aggregates of different length which are formed to minimize
the contact area between the aromatic core of the macrocycles and
the cyclohexane, as it is observed for the coil-ring-coil block
copolymers.8 However, for 1, the relative peak intensities in the
CONTIN fit clearly show that the aggregates are only a minor
fraction of the solute. To enforce the aggregation the (intraannular)
pyridyl groups of the macrocycle were alkylated by treating 1 with
ethylbromide.11 2 is well soluble in chlorinated solvents, THF, and
even in cyclohexane.
Alternatively, the solubility can be increased by variation of the
backbone of the macrocycles. It might be possible that parts of the
macrocycle that are not in plane with the rigid ring backbone (for
steric reasons) increase the solubility and at the same time maybe
support, but at least not hinder, the aggregation. Since a 1,2,3-
substitution pattern does not allow all phenyl rings to lie in one
plane, we intended to investigate the solubility behavior of such
macrocycles (such as 1) if they are decorated with extraannular
oligo-alkyl chains.9
While the 1H NMR spectrum in dichloromethane is well resolved,
the spectrum in cyclohexane is again broadened (see Supporting
Information). This observation, together with the structure of 2,
indicated an aggregation of these amphiphiles into extended tubular
aggregates that avoid a contact of the ionic parts of the molecule
with the cyclohexane. However, contrary to what we expected for
these proposed aggregates, even at a concentration of 2 wt %, the
solution is neither highly viscous nor does it exhibit any birefrin-
gence, as it is found for cyclohexane solutions of the coil-ring-
coil block copolymers. DLS measurements (1.4 wt %) gave an
unexpected result (Figure 1a). The CONTIN fit shows now only
one maximum that corresponds to a hydrodynamic radius (RH) of
approximately 3-4 nm, and no larger aggregates could be detected.
Small-angle X-ray scattering experiments (SAXS) in cyclohexane
support that result, showing only particles with a small radius of
gyration (RG) of 1.75 nm (as expected smaller than RH) at all
concentrations (see Supporting Information). Moreover, the ob-
1 is well soluble in chlorinated solvents, aromatic solvents, THF,
and also cyclohexane, as we speculated at the beginning of our
† University of Karlsruhe.
§ On leave from the Institute of Chemical Kinetics and Combustion SB RAS,
Institutskaya 3, Novosibirsk 630090, Russia.
‡ Max Planck Institute.
9
3150
J. AM. CHEM. SOC. 2006, 128, 3150-3151
10.1021/ja057999g CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
extended aggregates (as it is observed for 1). However, the ion-
triggered aggregation process induces a conformational locking of
the macrocycles in the boat conformation that self-blocks the top
and bottom of the dimer, hindering a further aggregation after the
dimerization. As far as we know, this behavior has not been
described before for shape-persistent macrocycles.
In summary, we have shown by detailed DLS and X-ray
scattering experiments that nanometer-size dicationic shape-
persistent macrocycles in nonpolar solvents do not form infinite
tubular aggregates, as might be expected from their chemical
structure, but dimerize to defined supermolecules. The exclusive
dimerization is not programmed by synthetic efforts but is a result
of the ability of the macrocycles to adopt during the aggregation a
self-complementary shape that blocks the top and bottom of the
supermolecule, thus preventing a further aggregation. It has to be
marked that dimerization occurs even at the lowest investigated
concentration. Currently, we are investigating the concentration and
counterion dependence of the aggregation process further as well
as the behavior of this and other charged macrocycles in different
environments.
Figure 1. (a) CONTIN fit of DLS-derived distribution of 1 in cyclohexane
(1.5 wt %, dashed gray line) and of 2 in cyclohexane (1.4 wt %, solid
line). (b) Experimental scattering intensity of 2 in cyclohexane (O) compared
to theoretical scattering of different models: (dashed black line) tube due
to the packing of 20 macrocycles; (dashed gray line) tube due to the packing
of 100 macrocycles; (gray line) sphere (2.2 nm) (for a scattering vector q
< 1.0 nm-1 the curve superimposes with the scattering curve of the disk);
(black line) disk with a diameter of 4.5 nm and a height of 2.35 nm.
Acknowledgment. Financial support by the Deutsche Fors-
chungsgemeinschaft is gratefully acknowledged. We gratefully
acknowledge beamtime and local support by the ESRF, 6 rue Jules
Horowitz, BP 220, F-38043 Grenoble Cedex, France.
Figure 2. Proposed structure of the supramolecular dimer (only the rigid
backbone of the rings is displayed): the extraannular p-(tert-butyl)phenyl
groups block the top and bottom of the dimer and prevent an aggregation
of the supermolecules.
Supporting Information Available: Synthesis and characterization
of all new compounds as well as information about the X-ray scattering
experiments. This material is available free of charge via the Internet
at http://pubs.acs.org.
served scattering data (O) can be clearly distinguished from spheres
and tubular objects of different length. The observed data correspond
well to the calculated scattering profile of a disk with a diameter
of 4.5 nm and a height of 2.35 nm. By assuming a compound
density of 1.14 g/cm3,12 the molecular weight of the scattering object
is calculated to be 10 000 g/mol.
References
(1) For recent reviews on shape-persistent macrocycles, mainly phenylene
ethynylene-based, see: (a) Moore, J. S. Acc. Chem. Res. 1997, 30, 402-
413. (b) Ho¨ger, S. J. Polym. Sci. Part A: Polym. Chem. 1999, 37, 2685-
2698. (c) Haley, M. M.; Pak, J. J.; Brand, S. C. Top. Curr. Chem. 1999,
201, 81-130. (d) Grave, C.; Schlu¨ter, A. D. Eur. J. Org. Chem. 2002,
3075-3098. (e) Zhao, D.; Moore, J. S. Chem. Commun. 2003, 807-818.
(f) Yamaguchi, Y.; Yoshida, Z. Chem.sEur. J. 2003, 9, 5430-5440. (g)
Ho¨ger, S. Chem.sEur. J. 2004, 10, 1320-1329.
(2) (a) Moore, J. S. Curr. Opin. Colloid Interface Sci. 1999, 4, 108-116. (b)
Schmittel, M.; Kalsani, V. Top. Curr. Chem. 2005, 245, 1-53.
(3) Ghadiri, M. R.; Kobayashi, K.; Granja, J. R.; Chadha, R. K.; McRee, D.
E. Angew. Chem., Int. Ed. Engl. 1995, 34, 93-95.
(4) Nakamura, K.; Okubo, H.; Yamaguchi, M. Org. Lett. 2001, 3, 1097-1099.
(5) A dimeric sandwich-like structure has been proposed for the complexes
between Cs+-induced tubular self-assembly of Schiff-base macrocycles
but not investigated in detail: Gallant, A. J.; MacLachlan, M. J. Angew.
Chem., Int. Ed. 2003, 42, 5307-5310.
(6) Kalsani, V.; Ammon, H.; Ja¨ckel, F.; Rabe, J. P.; Schmittel, M. Chem.s
Eur. J. 2004, 10, 5481-5492.
(7) Ho¨ger, S.; Bonrad, K.; Mourran, A.; Beginn, U.; Mo¨ller, M. J. Am. Chem.
Soc. 2001, 123, 5651-5659.
(8) (a) Rosselli, S.; Ramminger, A.-D.; Wagner, T.; Silier, B.; Wiegand, S.;
Ha¨ussler, W.; Lieser, G.; Scheumann, V.; Ho¨ger, S. Angew. Chem., Int.
Ed. 2001, 40, 3138-3141. (b) Rosselli, S.; Ramminger, A.-D.; Wagner,
T.; Lieser, G.; Ho¨ger, S. Chem.sEur. J. 2003, 9, 3481-3491.
(9) Ho¨ger, S.; Rosselli, S.; Ramminger, A.-D.; Enkelmann, V. Org. Lett. 2002,
4, 4269-4272.
(10) Provencer, S. W. Makromol. Chem. 1979, 180, 201-209.
(11) For some pyridyl macrocycles see ref 1f and additionally, see: (a) Tobe,
Y.; Nagano, A.; Kawabata, K.; Sonoda, M.; Naemura, K. Org. Lett. 2000,
2, 3265-3268. (b) Campbell, K.; McDonald, R.; Branda, N. R.;
Tykwinski, R. R. Org. Lett. 2001, 3, 1045-1048. However, we are not
aware of any reports about quarternized structures.
(12) Data from ref 8b and: Fischer, M.; Lieser, G.; Rapp, A.; Schnell, I.;
Mamdouh, W.; De Feyter, S.; De Schryver, F. C.; Ho¨ger, S. J. Am. Chem.
Soc. 2004, 126, 214-222.
(13) MM calculations have shown that the boat and chair conformations of
shape-persistent phenylene ethynylene macrocycles of this type are
energetically slightly favored over the planar conformation (ref 7). A
computational analysis explaining at a molecular level the preference of
2 for dimerization over oligomerization is in progress.
(14) (a) Cram, D. J.; Cram, J. M. Container Molecules and Their Guests; Royal
Society of Chemistry: Cambridge, 1994. (b) Ungaro, R.; Pochini, A.;
Andreetti, G. D.; Domiano, P. J. Chem. Soc., Perkin Trans. 2 1985,
197-201.
The results of the X-ray scattering experiments and DLS
experiments undoubtedly show that the quarternization of the
slightly aggregating macrocycles 1 does not lead to the formation
of extended tubular structures in the nonpolar solvent, as it is was
expected. Rather, the DLS and X-ray scattering experiments indicate
the presence of well-defined dimers of the amphiphile 2 in the
cyclohexane solution. This is also confirmed by the size of the disk
described above. A simple computer model of 2 (stiff cycle, random
walk side chains) gives an idea of the size of one macrocycle: Its
scattering can be described by a disk with a diameter of 4.3 nm
and a thickness of 1.1 nm, that is, half the disk thickness that was
extracted from the experimental scattering data. It should be noted
that within the investigated concentration regime (0.005-1.4 wt
%) no large aggregates could be detected, but dimerization occurs
completely even at the lowest concentration.
Although on first glance an infinite aggregation should be
preferred to shield the charged parts of the macrocycle from the
nonpolar cyclohexane environment, the exclusive dimer formation
can be explained with a boat conformation of the macrocycles in
which the pyridinium groups point into the dimer center that
accommodates the four bromide ions (Figure 2).13
The proposed structure resembles sandwich-type complexes of
crown ethers or inclusion complexes of calixarenes.14 As pointed
out before, well-defined aggregates of shape-persistent macrocycles
are only scarcely observed.3-5 In the cases reported before, the
structure and functionality of the single macrocycle determines the
preference of dimerization over infinite aggregation. In the example
described here, the macrocycle is in principle capable of forming
JA057999G
9
J. AM. CHEM. SOC. VOL. 128, NO. 10, 2006 3151