Rational design of a double-walled tetrahedron containing two different C3-symmetric ligands

Article abstract of DOI:10.1002/anie.200703789

Two is better than one: The predictability of supramolecular coordination compounds relies on an exact match between the ligands used and the steric demand of the metal center. However, the use of two different C ₃-symmetric ligands results in

Full text of DOI:10.1002/anie.200703789

Communications
DOI: 10.1002/anie.200703789
Cage Compounds
Rational Design of a Double-Walled Tetrahedron Containing Two
Different C3-Symmetric Ligands**
Iris M. Oppel (nØe Müller)* and Kirsten Föcker
The rapidly growing field of supramolecular coordination
chemistry can be divided into two classes. Polymers are
already well-known because of their possible applications as
new materials, heterogeneous catalysts, and ion exchangers.[1]
The rational design of the second class, discrete cage
molecules, can be carried out using the molecular library
method,[2] in which the stoichiometry and symmetry elements
of the product are predetermined by the molecular starting
fragments, as illustrated by tetrahedral cages. In the oldest
and most common system [M4L6], the metal centers form the
corners, which are linked by doubly bridging ligands along the
edges.[3] Building blocks that cover the faces of the cage have
been reported far less often and result in the formation of
[M4L4] or [M6L4] systems.[4,5] Nearly all examples for the last
case can be seen as truncated tetrahedra or adamantanoid
cages with wide openings at their corners, allowing guest-
exchange reactions.[5]
formed in high purity and yield. This design principle can be
used for the construction of even more complex coordination
compounds with additional ligands, such as octahedral or
trigonal-bipyramidal cages.[7,8] However, we are just starting
to understand the reactions that take place when the ligands
and metal centers do not match exactly, and a prediction of
the products is not usually possible. Herein, we report the
rational design of the first double-walled tetrahedron con-
taining a ligand–metal pair that is not perfectly matched.
Coordination cages with tetrahedral shapes have been
prepared using tris(2-hydroxybenzylidene)triaminoguanidi-
nium [H6L1]+ and tris(5-bromo-2-hydroxybenzylidene)tri-
aminoguanidinium [H6L2]+, in which the triangular faces are
linked by (CdO)2 four-membered rings (O stands for a
phenolate oxygen atom), resulting in [{(CdCl)3L1}4]8À and
[{(CdCl)3L2}4]8À, respectively.[6] The formation of a compara-
ble discrete single-walled tetrahedron containing Zn2+ was
impossible owing to the much smaller size of the Zn2+ ion
compared to the Cd2+ ion (Zn2+: 0.74 Š, Cd2+: 0.97 Š[9]).
Because of this size difference, the faces would come in closer
contact in a Zn2+ compound, which is sterically unfavorable.
On the other hand, it has been demonstrated that [H6L1]+
and [H6L2]+ are able to bind the smaller Zn2+ ions, resulting in
cationic, neutral, or anionic coordination compounds with
one or two ligands, for example [{Zn(NH3)(H2O)}3L1]+,[10]
We are interested in C3-symmetric ligands such as [H6L1]+,
[H6L2]+, and [H6L3], which cover the faces of a cage nearly
completely. We were able to construct tetrahedral coordina-
tion cages,[6] for which an exact match of the ligands to the
steric requirements of the metal centers was shown to be
essential. The supramolecular reaction then leads to products
[Zn{Zn2(H2O)3(NH3)L2}2],
and
[{{Zn(NH3)}3L2}2{m-
(OH)}3]À.[11] In these structures, the metal ions are not located
at ideal positions within the plane of the ligand but are up to
0.87(1) Š above or below this plane. Especially the last,
dimeric compound (Figure 1) suggested that it might be
possible to obtain a novel tetrahedral cage containing
Zn2+ ions instead of Cd2+. With the Zn2+ centers twisted out
of the ligand plane, it should be possible to link these ions
through the phenolate oxygen atoms to form a double-walled
tetrahedron. Moreover, the overall charge would be reduced
from 8À to 4À, as each building block carries only one
negative charge. This charge reduction should have an
additional stabilizing effect. However, the reaction of
[H6L2]Cl with even a large excess of ZnCl2 leads only to the
formation of the known dimeric compound.
[*] Dr. I. M. Oppel (nØe Müller), K. Föcker
Lehrstuhl für Analytische Chemie
NC 4/27
This result is likely due to a missing stabilization at the
corners, which can be changed by the introduction of addi-
tional side chains on the aromatic ring. As depicted in
Figure 1, the proton bound to C(4) in each dimer is in close
contact with the Br atom at C(5) in the neighboring ligand. If
this proton at C(4) is exchanged for a stabilizing group (e.g. a
methoxy group), the corner of the cage should be closed more
Ruhr-Universität Bochum
44780 Bochum (Germany)
Fax: (+49)234-321-4420
E-mail: iris.oppel@rub.de
[**] We gratefully acknowledge financial support from BayerMaterial-
Science and a doctoral fellowship from the FCI (grant 176291).
À
tightly and stabilized by weak O CH3···Br interactions.
Comparable stabilizing effects have been described.[12]
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
402
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 402 –405

Angewandte
Chemie
Figure 1. Triangular faces (top) and molecular structure (bottom) of
[{{Zn(NH3)}3L2}2{m-(OH)}3]À. The gray and white atoms in the molec-
ular structure correspond to the gray and white faces in the top
depiction.
Therefore, we synthesized tris(3,5-dibromo-2-hydroxy-4-
methoxy-benzylidene)triaminoguanidinium
chloride
([H6L3]Cl) as an additional ligand. Both [H6L2]Cl and
[H6L3]Cl were then treated with six equivalents of ZnCl2 in
a methanol–triethylamine mixture, resulting in the formation
of orange-brown crystals with a large unit cell (cubic, a =
41.829(1) Š, V= 73189(4) Š3). The diffraction pattern was
weak, but using CuKa radiation, we were able to record a
satisfactory intensity data set. The structure was solved by
direct methods and reveals the expected double-walled
Figure 2. a, b) Double-walled triangular faces, comparable to the
dimeric complex in Figure 1. c) The asymmetric unit of 1. The
connection of the double-walled units by phenolate O atoms, resulting
in a (ZnO)2 four-membered ring, is highlighted by an oval. Counter-
cations and solvent molecules have been omitted for clarity. The part
of the asymmetric unit marked by the curly bracket corresponds to the
triangular face in (b).
tetrahedron
with
the
formula
(C6H16N)4(H2O)-
[{Zn3({m-(OCH3)}3{Zn3(solvent)3L3})L2}4] (1).[13]
[L2]5À ligand, with distances of 0.12(1) to 0.15(1) Š. These
values are larger than those observed in the Cd tetrahedra
(0.02(1)–0.08(2) Š). The phenyl rings are also twisted out of
the plane of the central CN6 core, and all dihedral angles
adopt values of 29(1)8. In this way, the phenolate oxygen atom
can reach the Zn2+ center of the neighboring face. These
values are comparable to those found in the Cd cages (16(1)–
36(1)8). Because of the bridging methanolate ligands, the
Zn2+ ions bound by the outer [L3]5À ligands point inside the
dimeric building blocks, with average distances to the central
CN6 plane of 0.11(2) Š. These ligands show much less
distortion, as can be seen by the average dihedral angle
between the central CN6 core and the phenyl rings (14(2)8).
The cage is formed by four dimeric units that each contain
both ligands. Each ligand binds three Zn2+ centers linked by
three bridging methanolate ligands (Figure 2a,b). The inner
walls of the tetrahedron are formed by the four {Zn3L2} units,
and the phenolate oxygen atoms bridge the neighboring faces,
as expected. One of the {Zn3L2} units is located on the C3 axis;
as a result, the asymmetric unit contains only a third of the
cage (Figure 2c), the remaining parts of which are symmetry-
related. Each outer {Zn3(solvent)3L3} face is formed by a
ligand bound to three Zn2+ centers, which adopt trigonal-
bipyramidal coordination geometry. Two equatorial positions
are occupied by the bridging methanolate ligand and a solvent
molecule (water for the ligand located on the C3 axis and
methanol for the others). In the four inner [L2]5À units, the
central CN6 cores show the same screw direction, making the
cage chiral. In the outer [L3]5À units, the same situation is
observed, but the screw goes the opposite way. As shown in
Figure 2b, the Zn2+ centers are located below the plane of the
À
Furthermore, the expected weak O CH3···Br interactions,
three at each corner, can be observed (3.05(1) to 3.17(1) Š,
119(1)8 to 133(1)8 at the Br atoms, Figure 3).
The complete double-walled tetrahedron 1 is depicted in
Figure 4. The overall charge of 4À is compensated by four
triethylammonium ions, one of which is located inside the
Angew. Chem. Int. Ed. 2008, 47, 402 –405
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
403

Communications
central N atoms of the countercations could be found in the
Fourier difference synthesis. Nevertheless, the data could be
satisfactory corrected by employing the SQUEEZE routine
in the program Platon (see the Experimental Section),
resulting in an electron count of 4220 e per unit cell or
about 527 e per cage. This value, minus the electrons from the
three countercations, corresponds to approximately 35 water
molecules (or a comparable number of methanol molecules)
per cage; this number can be confirmed by elemental analysis.
Compound 1 is very poorly soluble in most common NMR
solvents. Nevertheless, we were able to dissolve sufficient
1
substance in hot [D6]acetone to obtain a H NMR spectrum.
A broad, poorly defined multiplet in the aromatic region was
observed, as is expected for the two different ligands. The
most important information is that only one signal set could
be observed for the triethylammonium ions, thus indicating
that under such harsh conditions the cage is not stable on the
NMR timescale.
Figure 3. Corner of the double-walled tetrahedron; the OCH3···Br
interactions of 1 are depicted in dark gray.
In conclusion, we have successfully carried out the
rational synthesis of a coordination cage with the outer
shape of a tetrahedron by employing a metal–ligand pair that
does not exhibit a perfect match. With the help of a ligand
tailored for this purpose, we were able to isolate and
characterize the first example of a double-walled coordina-
tion cage.
Experimental Section
[H6L2]Cl was prepared according to literature methods.[14]
[H6L3]Cl:
3,5-Dibromo-2-hydroxy-4-methoxybenzaldehyde
(2.00 g, 6.45 mmol) in ethanol (30 mL) was slowly added to triami-
noguanidinium chloride (303.6 mg, 2.16 mmol) dissolved in a hot
mixture of ethanol (10 mL) and water (5 mL). After adjusting to pH 3
with HCl(aq), the suspension was allowed to cool to room temperature.
The precipitate was collected, washed with diethyl ether and dried
under reduced pressure. Yield: 1.8738 g (1.84 mmol, 85.37%).
Elemental analysis calcd for C25H21N6O6Br6Cl·H2O, (1034.36):
C 29.03, H 2.24, N 8.12, found: C 29.12, H 2.42, N 8.14. 1H NMR
(400 MHz, [D6]DMSO, 228C): d = 12.09 (s, 1H; NH), 10.67 (s, 1H;
=
OH), 8.86 (s, 1H; HC N), 8.32(s, 1H; HC(6)), 3.84 ppm (s, 3H;
OCH3).
1: ZnCl2 (10.2mg, 0.0748 mmol) was combined with [H 6L2]Cl
(8.7 mg, 0.0141 mmol) and [H6L3]Cl (12.7 mg, 0.0125 mmol). A
mixture of methanol (2mL) and triethylamine (41.76 mL) was
added. After a few days, orange-brown cubes of 1 were formed in a
yield of 22.1 mg (2.3 ” 10À3 mmol, 73.6%). Elemental analysis calcd
for
C233H252N52O61Br36Zn24·20H2O·4MeOH, 1 (9691.18): C 29.37,
H 3.20, N 7.52, found: C 29.01, H 3.04, N 8.02.
X-ray analysis: Intensity data for 1 were collected on an Oxford
Diffraction Xcalibur2CCD (Cu radiation, w scan). The data were
Ka
corrected for Lorentz, polarization, and absorption (Gauss method)
effects. The structure was solved by direct methods (SHELXS-97)[15]
and refined using a full-matrix least-squares procedure (SHELXL-
97).[16] All hydrogen atoms were placed at geometrically estimated
positions. Owing to the large number of disordered countercations
and solvent molecules outside the cage and the resulting low data-to-
parameter ratio, we decided to correct the X-ray data for their
influence employing the SQUEEZE routine in PLATON.[17] Never-
theless, the triethylammonium ion as well as the water molecule
inside the cage were located in the difference synthesis and refined
isotropically. They are both disordered but could be modeled with the
bond lengths fixed to literature values. CCDC-657925 contains the
supplementary crystallographic data for this paper. These data can be
Figure 4. a) The same asymmetric unit as in Figure 2c, but with the
same color code as in Figure 4b. b) Crystal structure of 1. Protons (for
(b)), countercations, and solvent molecules have been omitted for
clarity.
cage. It is disordered with an additional water molecule that is
À
fixed to the NH group by a hydrogen bond (d(N H···O) =
1.57(8) Š). Owing to the large degree of disorder outside the
3
cage (about 24650 Š or 34% of the overall cell volume is
occupied by countercations and solvent molecules), only the
404
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 402 –405

Angewandte
Chemie
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
Nieuwenhuyzen, S. L. James, Chem. Eur. J. 2006, 12, 2448, and
references therein.
[5] Recent papers on [M6L4] systems: a) Y. Nishioka, T. Yamaguchi,
M. Yoshizawa, M. Fujita, J. Am. Chem. Soc. 2007, 129, 7000;
b) T. Furusawa, M. Kawano, M. Fujita, Angew. Chem. 2007, 119,
5819; Angew. Chem. Int. Ed. 2007, 46, 5717; c) K. Nakabayashi,
M. Kawano, T. Kato, K. Furukawa, S. Ohkoshi, T. Hozumi, M.
Fujita, Chem. Asian J. 2007, 2, 164; d) Y. Kobayashi, M. Kawano,
M. Fujita, Chem. Commun. 2006, 4377; e) V. Maurizot, M.
Yoshizawa, M. Kawano, M. Fujita, Dalton Trans. 2006, 2 750; f) S.
Tashiro, M. Fujita, Bull. Chem. Soc. Jpn. 2006, 79, 833, and
references therein.
[6] a) I. M. Müller, D. Möller, C. A. Schalley, Angew. Chem. 2005,
117, 485; Angew. Chem. Int. Ed. 2005, 44, 480; b) I. M. Müller, R.
Robson, F. Separovic, Angew. Chem. 2001, 113, 4519; Angew.
Chem. Int. Ed. 2001, 40, 4385.
[7] I. M. Müller, S. Spillmann, H. Franck, R. Pietschnig, Chem. Eur.
J. 2004, 10, 2207.
Received: August 17, 2007
Published online: November 13, 2007
Keywords: cage compounds · coordination chemistry ·
.
self-assembly · supramolecular chemistry · zinc
[1] For recent reviews, see: a) A. L. Garay, A. Pichon, S. L. James,
Chem. Soc. Rev. 2007, 36, 846; b) K. A. Williams, A. J. Boydston,
C. W. Bielawski, Chem. Soc. Rev.2007, 36, 729; c) K. Biradha, M.
Sarkar, L. Rajput, Chem. Commun. 2006, 4169; d) S. Kitagawa,
S. Noro, T. Nakamura, Chem. Commun. 2006, 701; e) Y. Zhou,
M. Hong, X. Wu, Chem. Commun. 2006, 135; f) K. Ding, Z.
Wang, X. Wang, Y. Liang, X. Wang, Chem. Eur. J. 2006, 12, 5188;
g) T. Uemura, S. Horike, S. Kitagawa, Chem. Asian J. 2006, 1, 36;
h) C. Chen, B. Kang, C. Su, Aust. J. Chem. 2006, 59, 3.
[8] I. M. Müller, D. Möller, Angew. Chem. 2005, 117, 3029; Angew.
Chem. Int. Ed. 2005, 44, 2969.
[2] For recent reviews, see: a) J. A. Thomas, Chem. Soc. Rev. 2007,
36, 856; b) M. D. Pluth, K. N. Raymond, Chem. Soc. Rev. 2007,
36, 161; c) J.-M. Lehn, Chem. Soc. Rev. 2007, 36, 151; d) A. W.
Kleij, J. N. H. Reek, Chem. Eur. J. 2006, 12, 4218; e) S. P. Argent,
H. Adams, T. Riis-Johannessen, J. C. Jeffery, L. P. Harding, O.
Mamula, M. D. Ward, Inorg. Chem. 2006, 45, 3905; f) C. H. M.
Amijs, G. P. M. van Klink, G. van Koten, Dalton Trans. 2006,
308.
[9] R. C. Weast, Handbook of Chemistry and Physics, CRC, Boca
Raton, 1988, F-105.
[10] I. M. Müller, R. Robson, Angew. Chem. 2000, 112, 4527; Angew.
Chem. Int. Ed. 2000, 39, 4357.
[11] I. M. Müller, D. Möller, K. Föcker, Chem. Eur. J. 2005, 11, 3318.
[12] OCH3···Br distances of approximately 3 Š can be observed in
more than 70 cases in the CCDC data base. For comparable
interactions, see: a) G. R. Desiraju, Chem. Commun. 2005, 2995;
[3] Recent papers on [M4L6] systems: a) M. D. Pluth, R. G. Berg-
man, K. N. Raymond, Science 2007, 316, 85; b) B. E. F. Tiede-
mann, K. N. Raymond, Angew. Chem. 2007, 119, 5064; Angew.
Chem. Int. Ed. 2007, 46, 4976; c) D. H. Leung, R. G. Bergman,
K. N. Raymond, J. Am. Chem. Soc. 2007, 129, 2746; d) R. Frantz,
C. S. Grange, N. K. Al-Rasbi, M. D. Ward, J. Lacour, Chem.
Commun. 2007, 1459; e) D. Fiedler, R. G. Bergman, K. N.
Raymond, Angew. Chem. 2006, 118, 759; Angew. Chem. Int.
Ed. 2006, 45, 745; f) U. N. Andersen, G. Seeber, D. Fiedler, K. N.
Raymond, D. Lin, D. Harris, J. Am. Soc. Mass Spectrom. 2006,
17, 292; g) S. P. Argent, H. Adams, T. Riis-Johannessen, J. C.
Jeffery, L. P. Harding, W. Clegg, R. W. Harrington, M. D. Ward,
Dalton Trans. 2006, 4996, and references therein.
b) G. R. Desiraju, Acc. Chem. Res. 2002, 35, 565.
3
¯
[13] 0.30 ” 0.31 ” 0.31 mm , cubic, space group Pa3, a = 41.829(1) Š,
V= 73189(4) Š3, 1calcd = 1.670 gcmÀ3
,
2qmax = 121.728, l =
1.54178 Š, T= 100 K, 321316 measured reflections, 18492
independent reflections (Rint = 0.127), 7521 observed reflections
(I > 2s(I)), m = 6.825 mmÀ1, numerical absorption correction,
T
min = 0.0823, Tmax = 0.1343, 761 parameters, R1(I>2s(I)) =
0.0583, wR2(all data) = 0.1457, max/min residual electron density
À3
0.934/À1.021 eŠ
.
[14] I. M. Müller, D. Möller, Eur. J. Inorg. Chem. 2005, 257.
[15] G. M. Sheldrick, SHELXS-97, Program for Crystal Structure
Solution, University Göttingen, 1997.
[16] G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
Refinement, University Göttingen, 1997.
[17] A. L. Spek, Acta Crystallogr. Sect. A 1990, 46, C34.
[4] Recent papers on [M4L4] systems: a) M. Albrecht, S. Burk, R.
Stoffel, A. Lüchow, R. Fröhlich, M. Kogej, C. A. Schalley, Eur. J.
Inorg. Chem. 2007, 1361; b) M Albrecht, R. Fröhlich, Bull.
Chem. Soc. Jpn. 2007, 80, 797; c) J. Zhang, P. W. Miller, M.
Angew. Chem. Int. Ed. 2008, 47, 402 –405
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
405