Towards technomimetic spoked wheels: Dynamic hexakis-heteroleptic coordination at a hexakis-terpyridine scaffold

Article abstract of DOI:10.1039/b718185j

The synthesis of hexakis-terpyridine 4 and an expedient approach to its dynamic hexakis-heteroleptic complexes are elaborated, the latter being readily accessible precursors for the construction of technomimetic molecular spoked wheels. The Royal Society

Full text of DOI:10.1039/b718185j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Towards technomimetic spoked wheels: dynamic hexakis-heteroleptic
coordination at a hexakis-terpyridine scaffoldw
Michael Schmittel* and Prasenjit Mal
Received (in Cambridge, UK) 26th November 2007, Accepted 21st December 2007
First published as an Advance Article on the web 9th January 2008
DOI: 10.1039/b718185j
The synthesis of hexakis-terpyridine 4 and an expedient
approach to its dynamic hexakis-heteroleptic complexes are
elaborated, the latter being readily accessible precursors for
the construction of technomimetic molecular spoked wheels.
a complete spoked wheel, thus only a covalent or coordinative
linkage between the rim elements will be required.
Technomimetic molecules are design based molecules emulat-
ing the macroscopic world on the molecular level. In this
1
context, technomimetic molecules, molecular machines, me-
2
3
tal helicates, six-leg molecules, wheelbarrows, and motors
4
5
6
have recently received considerable interest. On the other
hand, real-world spoked wheels have not found a major
7
correspondence in the nanoscopic chemical world, i.e. mole-
Fig. 1 Design of a technomimetic spoked wheel.
cular wheels with an axle unit, sturdy spokes and a circular
rim, although the concept ‘molecular wheel’ is being employed
Chart 1 depicts phenanthrolines 1,2 and terpyridines 3,4
used as building blocks for the present study. Terpyridine 3
was purchased and used as received. Synthesis of compounds 1
and 2 was achieved using established protocols elaborated in
8
in various associations in the literature. While a dazzling
multi-step approach to a covalent D6h spoked wheel has
7
ger et al., a supramolecular
recently been realised by Ho
¨
strategy forcefully profiting from suitable preorganisation of
building blocks to reduce the number of steps has not yet been
elaborated. A supramolecular approach to a D4h technomi-
14
our group. The presence of bulky alkylaryl groups at the 2,9-
positions of the respective phenanthrolines 1,2 is essential to
15
avoid formation of homoleptic complexes, thus guiding the
12,13
reaction toward heteroleptic complexes with terpyridines.
9
metic spoked wheel has been reported previously by Sanders
(
not called a spoked wheel) and more recently by Lindsey and
1
Holten (D ), but in both cases only an imperfect execution
0
3
h
of a circular rim was achieved due to the low symmetry. Other
notable approaches suffer from flexible spokes, thus generat-
1
1
ing floppy wheels. Clearly, the construction of a supramole-
cular spoked wheel poses some major challenges due to the
construction of a planar set of sturdy supporting spokes that
have to be thoroughly linked by a circular rim. This will either
require a macrocycle as a rim that is imposed on the spoke set
(
Fig. 1A), or it necessitates a directed macrocyclisation start-
ing from preoriented rim elements (Fig. 1B). Being convinced
of the latter strategy, we would like to report herein on an
approach towards the construction of a supramolecular D6h
spoke set endowed with rim elements. Our approach emanates
from a planar hexakis-terpyridine scaffold constituting the
spoke set, which is equipped in a key step with phenanthro-
lines (rim elements) that are attached to the spokes via
Chart 1 Compounds 1–4 and their cartoon representations.
1
2
the HETTAP concept
(heteroleptic terpyridine and
1
phenanthroline complex formation). For the finalisation of
3
As a key building block constituting the axle and spokes
unit, we designed the D6h symmetric hexakis-terpyridine 4.
After the first synthesis of hexaethynylbenzene (6) in 1986 by
Center of Micro and Nanochemistry and Engineering, Organische
Chemie I, Universita¨t Siegen, Adolf-Reichwein-Str. 2, D-57068
Siegen, Germany. E-mail: schmittel@chemie.uni-siegen.de;
Fax: (þ49) 271 740 3270
1
6
Vollhardt, several derivatives of this interesting family of
fully conjugated, rigid, quasi two-dimensional hydrocarbons
1
have been prepared and studied, but so far not with
7
w Electronic supplementary information (ESI) available: Experimental
procedures and spectra. See DOI: 10.1039/b718185j
0
0
00
18
2,2 : 6 ,2 -terpyridines as terminal ligands.
9
60 | Chem. Commun., 2008, 960–962
This journal is ꢀc The Royal Society of Chemistry 2008

View Article Online
Fig. 2 Synthesis of 4.
In Fig. 2, the synthesis of 4 is described. The first step
involved the preparation of hexakis(trimethylsilylethynyl)ben-
zene (5) via a Sonogashira coupling. Deprotection of 5 using
1
6
K CO in methanol led to hexaethynylbenzene (6). Since 6
2
3
was found to be very unstable, it was not purified for the
ensuing step. The final transformation employed a Sonoga-
Fig. 3 Cartoon illustrations of C1–C3 and computed structures of C2
and C3 (hydrogens are eliminated for clarity).
1
3
shira coupling between 6 and 7 in the presence of Pd(0) and
CuI as catalyst. For this step exclusion of oxygen was crucial,
because traces thereof in the reaction mixture led to a noto-
rious and inseparable mixture of products.
6
1
in the mass region m/z ¼ 200 to 2000, i.e. [Cu
6
(2)
3
(4)]
51
(
m/z 913.4 Da), [Cu
4
6
(2)
3 6
(4)(PF )]
(m/z 1125.4 Da),
31
1
[Cu
6
(2)
3
6
(4)(PF )
2
]
6 3 6 3
(m/z 1442.5 Da) and [Cu (2) (4)(PF ) ]
Compound 4 was characterised using standard organic
1 13
techniques. As can be seen from the H and C NMR spectra
(m/z 1971.7 Da). Notably, signals of the linear ligand 2 or of
terpyridine 4 were not detectable. The experimental isotopic
distributions (grey lines, for 4þ and 3þ charged species) were
in good agreement with the calculated distributions (black
lines). Similar results were also obtained for the assembly C2
(
ESIw), the methyl groups appeared as a single signal at 2.16
1 13
(
H) and 21.0 ( C) ppm. In addition, the clean ESI-MS
spectra of 4 and the experimental isotopic distribution for
1
1
the protonated 4 þ 3(H ) (m/z 746.4 Da) and 4 þ 2(H ) (m/z
(
ESIw), although it should be much more labile due to the
presence of six mononuclear binding sites.
1
119.1 Da) were in full agreement with the suggested structure.
Since it is well known that terpyridines can form bishomo-
1
As can be seen from the H NMR spectra, the mesityl
1
9
leptic complexes in the presence of various metal ions, it was
a challenging task to elaborate a hexakis-heteroleptic complex
while avoiding any homoleptic ones. All initial efforts to react
protons in C1–C3 experienced a distinctive and diagnostic
high field shift to ca. 6 ppm that is notably different from that
in the free ligands (e.g. 1 in C2 exhibits a shift of 6.22 ppm as
compared to 6.96 ppm for the free ligand 1, Fig. 5). Such a
shift is typical for heteroleptic complex formation. Similar
results were obtained for C3 (6.24/6.15 ppm), but now the two
mesityl units (in vs. out) have different environments in
21
1
and 4 with Zn led to the formation of polymeric aggre-
2
0
gates as evidenced by a highly insoluble precipitate with only
negligible amounts of the desired hexakis-heteroleptic complex
being formed. Therefore, we turned our attention to the
1
2
thermodynamically less stable copper(I) complexes, also
1
prepared via the HETTAP method, as use of copper(I)
3
accordance with the overall structure (Fig. 3). Also the hex-
1
akis-terpyridine 4 experienced diagnostic shifts in the
H
avoids formation of bishomoleptic octahedral terpyridine
1
complexes. Thus, we were successful in preparing the Cu
NMR. In the free ligand, the methyl protons appeared as a
single signal at 2.16 ppm (in [D ]dichloromethane, 25 1C, 400
MHz), while, upon complexation, the signal was shifted down-
2
complex C1 which is a model for capping a spoke of ligand 4
with 1 as the rim element. Analogously, when phenanthrolines
field, now emerging at 2.28 (in C2) and 2.42/2.22 ppm (in C3)
1
or 2 were combined with terpyridine 4 in the presence of the
(
ESIw). This downfield shift is readily explained by the immer-
appropriate amount of [Cu(MeCN) ]PF in dichloromethane,
6
4
sion of the methyl groups of 4 into the deshielding region of
the mesityl rings of 1 and 2. Likewise, the shifts of the methyls
the dark red compounds C2 and C3 formed in virtually
quantitative yield (Fig. 3). These assemblies were analysed
1
13
by H NMR, C NMR, IR, UV/Vis, ESI-MS and elemental
analysis without the need for purification. They were found to
be dynamic in nature, stable for several months and soluble in
a variety of solvents like dichloromethane, acetonitrile, nitro-
methane, DMF, DMSO, etc.
Since it has been impossible so far to obtain single crystals,
we have utilised spectroscopic techniques like ESI-MS,
UV/Vis, NMR etc. to confirm the supramolecular structures.
ESI-MS is one of the most important methods to elucidate the
composition of metallo-supramolecular structures in solu-
2
1
tion. In our study, ESI-MS analysis provided conclusive
evidence for the clean formation of the expected assemblies
C1–C3. Fig. 4 displays the ESI-MS spectrum of C3 as a
representative example, showing the presence of four signals
Fig. 4 ESI-MS spectra of C3.
This journal is ꢀc The Royal Society of Chemistry 2008
Chem. Commun., 2008, 960–962 | 961

View Article Online
spoke set. Secondly, as a key step towards the construction
of a spoked wheel we have achieved the selective formation of
two multicomponent assemblies exhibiting dynamic hexakis-
heteroleptic complexation. Thereby, rim elements have been
attached onto the spokes, now being in place for a directed
macrocyclisation. For final covalent linkages between the rim
elements, however, we will need to introduce alkyne (for
oxidative linkage) or alkene (for metathesis) groups in the
0
appropriate positions of 1 (3,8-positions) and/or 2 (8,8 -posi-
tions). While this functionalisation is straightforward for 1, the
rotational flexibility in C2 may also allow for an undesired
inter ‘supramolecular’ coupling. Modification of 2 therefore
may be more promising. Work along these lines is ongoing and
hopefully will demonstrate the success of our strategy for the
preparation of molecular spoked wheels as new technomimetic
molecules.
1
2 2
Fig. 5 H NMR spectra of 1, 4 and C2 in CD Cl .
We are grateful to the Deutsche Forschungsgemeinschaft,
the Humboldt Foundation (PM) and the Fonds der
Chemischen Industrie for financial support.
Notes and references
1
(a) G. Rapenne, Org. Biomol. Chem., 2005, 3, 1165; (b) A. Carella,
C. Coudret, G. Guirado, G. Rapenne, G. Vives and J.-P. Launay,
Dalton Trans., 2007, 177.
Fig. 6 UV/Vis absorption spectra of 4, C2 and C3 (at 2.5 ꢂ 10ꢁ M)
6
2
(a) W. R. Browne and B. L. Feringa, Nat. Nanotechnol., 2006, 1,
25; (b) S. Bonnet, J.-P. Collin, M. Koizumi, P. Mobian and J.-P.
Sauvage, Adv. Mater., 2006, 18, 1239.
in dichloromethane at 25 1C.
3
4
J. R. Nitschke, Acc. Chem. Res., 2007, 40, 103.
K.-H. Gross, F. Rieder, S. M. Moresco, A. Stojkovic and C. J.
Gourdon, Nat. Mater., 2005, 4, 892.
in phenanthrolines 1 and 2 are instructive concerning the
1
structure of the complexes. While H NMR shifts of p- and
o-methyls in free 1 show up at 2.33/2.05 ppm, those of 1 in
complex C2 are shifted to 1.85/1.54 ppm and similarly in
complex C1 to d ¼ 1.97/1.58 ppm. For C3 (1.96/1.94 and
5 G. Rapenne and G. Jimenez-Bueno, Tetrahedron, 2007, 63, 7018.
6
7
A. Carella, G. Rapenne and J.-P. Launay, New J. Chem., 2005, 29, 288.
D. Mossinger, J. Hornung, S. Lei, S. D. Feyter and S. Hoger,
Angew. Chem., Int. Ed., 2007, 46, 6802.
¨
¨
1
.45/1.43 ppm) the effect is even more pronounced.
8
9
E. R. Kay, D. A. Leigh and F. Zerbetto, Angew. Chem., Int. Ed.,
2007, 46, 72 and references therein.
S. Anderson, H. L. Anderson and J. K. M. Sanders, J. Chem. Soc.,
Perkin Trans. 1, 1995, 2255.
UV/Vis absorption spectra can also be utilised for the
characterisation of the assemblies. The absorption spectra of
hexakis-terpyridine 4 and complexes C2, C3 are shown in
Fig. 6. The strong absorption bands at 250–400 nm corre-
1
0 H.-e. Song, C. Kirmaier, J. K. Schwartz, E. Hindin, L. Yu, D. F.
Bocian, J. S. Lindsey and D. Holten, J. Phys. Chem. B, 2006, 110,
19131 and references therein.
2
2
spond to typical ligand p–p* transitions, whereas the band at
B500 nm, definitely responsible for the red colour, was
assigned to a metal-to-ligand-charge-transfer (MLCT) transi-
tion. For both assemblies, the MLCT bands were nearly
identical, clearly indicative of extremely similar molecular
environments at the metal centres. Moreover, these MLCT
bands proved to be quite similar to those of analogous
1
1
1
1
1 P. Wang, C. N. Moorefield and G. R. Newkome, Angew. Chem.,
Int. Ed., 2005, 44, 1679.
2 M. Schmittel, V. Kalsani, R. S. K. Kishore, H. Co
Bats, J. Am. Chem. Soc., 2005, 127, 11545.
3 M. Schmittel, V. Kalsani, P. Mal and J. W. Bats, Inorg. Chem.,
006, 45, 6370 and references therein.
4 (a) M. Schmittel, C. Michel, A. Wiegrefe and V. Kalsani, Synthesis,
001, 1561; (b) V. Kalsani, H. Ammon, F. Jackel, J. P. Rabe and
M. Schmittel, Chem.–Eur. J., 2004, 10, 5481.
¨
lfen and J. W.
2
2
¨
1
23
heteroleptic Cu grid assemblies reported by our group.
The MLCT bands in the visible region at lmax ¼ 479 nm (emax
ꢁ1
1
1
5 M. Schmittel and A. Ganz, Chem. Commun., 1997, 999.
6 R. Diercks, J. C. Armstrong, R. Boese and K. P. C. Vollhardt,
Angew. Chem., Int. Ed. Engl., 1986, 25, 268.
ꢁ
1
ꢁ1
¼
63 200 M cm ) for C2 and at 483 nm (emax ¼ 62 700 M
ꢁ1 1
cm ) for C3 represent six Cu phenanthroline complex units.
17 (a) J. Nierle, D. Barth and D. Kuck, Eur. J. Org. Chem., 2004, 867;
b) C. Lepetit, C. Zou and R. Chauvin, J. Org. Chem., 2006, 71,
6317.
0
(
For comparison, MLCT bands in bis(2,2 -bipyridine) cop-
ꢁ
1
ꢁ1
per(I) complexes have typical emax ¼ 6800–8500 M cm
,
1
8 U. S. Schubert, H. Hofmeier and G. R. Newkome, Modern
Terpyridine Chemistry, Wiley-VCH, Weinheim, 2006.
while those for bisphenanthroline copper(I) complexes are in
ꢁ
1
1
the range 3200–13 200 M cm
ꢁ1 23
.
19 G. R. Newkome, T. J. Cho, C. N. Moorefield, P. P. Mohapatra
and L. A. Godinez, Chem.–Eur. J., 2004, 10, 1493.
20 S. M. Brombosz, A. J. Zucchero, R. L. Phillips, D. Vazquez, A.
Wilson and U. H. F. Bunz, Org. Lett., 2007, 9, 4519.
21 K. C. Russell, E. Leize, A. VanDorsselaer and J.-M. Lehn, Angew.
Chem., Int. Ed. Engl., 1995, 34, 209.
To gain some structural insight, we have performed MM
s
computations (Hyperchem , Hypercube Inc.). Minimised
space filling models of C2 (other rotamers are also possible)
and C3 in Fig. 3 (vide supra) reveal a diameter and perimeter of
2
2 X. Y. Li, J. Illigen, M. Nieger, S. Michel and C. A. Schalley,
Chem.–Eur. J., 2003, 6, 1332.
3
.9 nm and 12.3 nm, respectively.
In conclusion, we present herein the first synthesis of the D6h
2
3 M. Schmittel, V. Kalsani, C. Michel, P. Mal, H. Ammon, J. P.
¨
Rabe and F. Jackel, Chem.–Eur. J., 2007, 13, 6223.
symmetric hexakis-terpyridine 4, constituting our axle and
9
62 | Chem. Commun., 2008, 960–962
This journal is ꢀc The Royal Society of Chemistry 2008