A fast-moving [2]rotaxane whose stoppers are remote from the copper complex core

Article abstract of DOI:10.1021/ol052051c

(Figure Presented) A new copper-complexed rotaxane is described. It consists of a two-coordination site ring threaded by a sterically non-hindering 2,2′-bipyridine derivative. An electrochemical signal (oxidation or reduction of the copper center, Cu(I) o

Full text of DOI:10.1021/ol052051c

ORGANIC
LETTERS
2005
Vol. 7, No. 26
5753-5756
A Fast-Moving [2]Rotaxane Whose
Stoppers Are Remote from the Copper
Complex Core
Ulla Le´tinois-Halbes, David Hanss, John M. Beierle, Jean-Paul Collin, and
Jean-Pierre Sauvage*
Laboratoire de Chimie Organo-Mine´rale, UniVersite´ Louis Pasteur/U.M.R. du CNRS
7513, Institut Le Bel, 4, rue Blaise Pascal, 67070 Strasbourg-Cedex, France
sauVage@chimie.u-strasbg.fr
Received August 25, 2005
ABSTRACT
A new copper-complexed rotaxane is described. It consists of a two-coordination site ring threaded by a sterically non-hindering 2,2′-bipyridine
derivative. An electrochemical signal (oxidation or reduction of the copper center, Cu(I) or Cu(II)) induces rearrangement of the system. By
using long and flexible linkers between the stoppers and the central complex, ligand exchange is fast, which leads to short response times
(on the millisecond time scale and even below).
In the field of dynamic molecular systems, whose motions
are triggered and controlled from the outside and which are
often referred to as “molecular machines” or “molecular
motors”,1 catenanes and rotaxanes occupy a special position.2
Transition-metal-based molecular machines whose organic
backbone is a catenane or a rotaxane have been prepared in
the course of the past decade.3a These systems contain either
a redox-active species3b or a ruthenium(II) central complex
as photoactive unit.3c Since the first bistable copper-com-
plexed [2]catenane reported by our group,4 several related
species, including [2]rotaxanes, have been described with
shorter and shorter response times between the redox signal
(copper(II)/copper(I)) and the subsequent motion leading to
the most thermodynamically stable situation.3b From the
(1) (a) Balzani, V.; Venturi, M.; Credi, A. Molecular DeVices and
Machines; Wiley-VCH: Weinheim, 2003. (b) Sauvage, J.-P. Molecular
Machines and Motors, Structure & Bonding; Springer: Berlin, Heidelberg,
2001; Vol. 99 (c) Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F.
Angew. Chem., Int. Ed. 2000, 39, 3348. (d) Special issue on Molecular
Machines. Acc. Chem. Res. 2001, 34, 341. (e) Fabbrizzi, L.; Licchelli, M.;
Pallavicini, P. Acc. Chem. Res. 1999, 32, 846. (f) Kelly, T. R.; de Silva,
H.; Silva, R. A. Nature 1999, 401, 150. (g) Koumura, N.; Zijistra, R. W.
J.; van Delden, R. A.; Harada, N.; Feringa, B. L. Nature 1999, 401, 152.
(h) Leigh, D. A.; Wong, J. K. Y.; Dehez, F.; Zerbetto, F. Nature 2003,
424, 174. (i) Jimenez-Molero, M. C.; Dietrich-Buchecker, C. O.; Sauvage,
J.-P. Chem. Commun. 2003, 1613. (j) Shipway, A. N.; Willner, I. Acc. Chem.
Res. 2001, 34, 421. (k) Harada, A. Acc. Chem. Res. 2001, 34, 456.
(2) Sauvage, J.-P., Dietrich-Buchecker, C. O., Eds.; Molecular Catenanes,
Rotaxanes and Knots; Wiley-VCH: Weinheim, 1999. (b) Bissell, R. A.;
Co´rdova, E.; Kaifer, A. E.; Stoddart, J. F. Nature 1994, 369, 133. (c) Badjic,
J. D.; Balzani, V.; Credi, A.; Silvi S.; Stoddart, J. F. Science 2004, 303,
1845 and references therein.
(3) (a) Jime´nez, M. C.; Dietrich-Buchecker, C. O.; Sauvage, J.-P. Angew.
Chem., Int. Ed. 2000, 39, 3284. Jime´nez, M. C.; Dietrich-Buchecker, C.
O.; Sauvage, J.-P. Chem. Eur. J. 2002, 8, 1456. (b) Collin, J.-P.; Gavin˜a,
P.; Sauvage, J.-P. Chem. Commun. 1996, 2005. Raehm, L.; Kern, J.-M.;
Sauvage, J.-P. Chem. Eur. J. 1999, 5, 3310. Sauvage, J.-P. Acc. Chem.
Res. 1998, 31, 611. (c) Pomeranc, D.; Jouvenot, D.; Chambron, J.-C.; Colllin,
J.-P.; Heitz, V.; Sauvage, J.-P. Chem. Eur. J. 2003, 9, 4247. Mobian, P.;
Kern, J.-M.; Sauvage, J.-P. Angew. Chem., Int. Ed. 2004, 43, 2392.
(4) Livoreil, A.; Dietrich-Buchecker, C. O.; Sauvage, J.-P. J. Am. Chem.
Soc. 1994, 116, 9399.
10.1021/ol052051c CCC: $30.25
© 2005 American Chemical Society
Published on Web 12/02/2005

Figure 1. Synthesis of the free rotaxane 1.
minutes or even hours required originally to set a given part
of the compound in motion,4 the time scale for the most
recent compounds could be shifted toward hundreds of
milliseconds (CuII f CuII , the subscript indicates the
coordination number of the copper center) and even mil-
liseconds (CuI5 f CuI4).5
Synthesis of the Copper(I)-Complexed Rotaxane. The
synthetic route leading to the free rotaxane 1 and the various
starting compounds and intermediates are represented in
Figure 1. We chose triethyleneglycol chains for the elonga-
tion of the axle. The axle of the rotaxane was synthesized
starting from 5,5′-bis(bromomethyl)-2,2′-bipyridine 3.6 The
Williamson ether synthesis with 2 equiv of 8-(2H-tetrahy-
4
5
The rearrangement process involves several coordination/
decoordination steps, an important factor being the acces-
sibility of the copper center and its ability to undergo fast
ligand exchange. To make the copper center as easy as
possible to reach for entering ligands, we thought that the
bulky stoppers should be located far away from the central
complex. We thus prepared and studied a new bistable
rotaxane whose general structure is similar to that of a
previously published compound,5 but whose stoppers are very
remote from the copper center (see Figure 2). This new
dynamic system can indeed be set in motion more rapidly
than the previously described systems. In the present report,
we would like to describe the synthetic procedure leading
to this new rotaxane; the dynamic behavior of the copper-
complexed rotaxane will also be discussed.
3
dropyran-2-yloxy)-3n6 -dioxaoctan-1-ol7 in the presence of
sodium hydride led to compound 4. The THP protecting
groups of 4 were removed upon treatment with HCl to give
5. Diol 5 was mesylated on both hydroxyl groups with
methanesulfonyl chloride to afford 6, which was subse-
quently reacted with LiBr to give 7. The stoppers were added
to the molecule in two steps. Usually, the Williamson ether
synthesis does not lead to very high yields, so it seemed
more reasonable to have only one stoppering reaction to carry
out once the macrocycle is threaded on the axle. The first
stoppering reaction for the synthesis of 9, which is a statistical
reaction, was carried out with only 0.3 equiv of the stopper
8 in respect to the axle in order to minimize as much as
(6) Schubert, U. S.; Eschbaumer, C.; Hochwimmer, G. Synthesis 1999,
5, 779.
(5) Poleschak, I.; Kern, J.-M.; Sauvage, J.-P. Chem. Commun. 2004, 474.
(7) Loiseau, F. A.; Hi, K. K.; Hill, A. M. J. Org. Chem. 2004, 69, 639.
5754
Org. Lett., Vol. 7, No. 26, 2005

possible the formation of the dumbbell like molecule 10.
Compound 9 was obtained in 30% yield, and 62% of the
nonconsumed starting material 7 could be recovered. In the
next step, macrocycle 2 was threaded onto the axle by using
Cu(I) as templating agent. The threading process leading to
1
11+ was shown to be quantitative by H NMR. The phenyl
ring protons in 2 show a significant shift to lower frequencies
when the ring is threaded on the axle. This shift results from
the cone of anisotropy of the bipyridine unit of the axle.
Another equivalent of 8 for the second stoppering reaction
was then added to the pre-rotaxane 11+ in the presence of
Cs2CO3 and DMF8 in order to obtain 12+. The copper
complex 12+ revealed to be unstable on silica. This is
understandable in view of the structure since the stability of
copper(I) complexes is particularly sensitive to the hindering
nature of the ligands used. Sterically hindering bidentate
chelates such as 6,6′-disubstituted 2,2′-bipyridine or 2,9-
disubstituted 1,10-phenanthroline lead to roughly tetrahe-
drally coordinated copper(I) complexes and, thus, highly
stable compounds. By contrast, if the encumbering nature
of the ligand is decreased, the complex stability of the ligand
suffers as a result of the possibility of forming close-to-square
planar complexes and to the increased accessibility of the
metal center to O2. To fabricate fast-moving molecular
machines, it could be detrimental to build extremely stable
and sterically protected copper complexes. This is clearly
illustrated by the dynamic properties of the first copper-based
molecular machines made in our group.4 In other words, the
ideal systems, combining fast rearrangement and reasonable
stability, will have to be a compromise between these two
features, as illustrated by the present rotaxane. Upon
purification of 12+ partial loss of Cu(I) was observed, leading
to mixtures of 12+ and the demetalated rotaxane 1. To obtain
pure 12+, we decided to demetalate it with KCN, purify the
free rotaxane 1 on column chromatography, and subsequently
remetalate it for the electrochemical studies. This strategy
proved to be successful. Column chromatography of the
demetalated rotaxane on silica using pentane, ethyl acetate,
methanol, and triethylamine as eluents (70:28:1:1) led to pure
1 in 32% yield. The partial loss of copper on silica from
12+ as well as the isolation of one fraction containing the
dumbbell 10 lacking the threaded macrocycle, where deth-
reading occurred during the second stoppering reaction,
indicate that ligand exchange of copper in this system is
rather facile. To show that the macrocycle is indeed able to
pirouette around the axle, as shown schematically in Figure
2, and to determine the speed of this rotational motion, we
carried out electrochemical studies on 12+.
Figure 2. Schematic representation of the redox-driven pirouetting
motion. In the drawing, the thread has been represented as fixed,
whereas the ring undergoes the motion. This is of course arbitrary
since the molecules are free to diffuse in solution and it is thus
impossible to identify a motionless part and a mobile part. (a)
Behavior of the previously published system5 at room temperature;
(b) behavior of the present system at -40 °C.
Compound 1 was metalated with Cu(I) using [Cu(MeCN)4]-
PF6 in acetonitrile leading to pure 124 . The solution became
ments were carried out in acetonitrile using a platinum
electrode as working electrode and Bu4NPF6 (0.1 M) as
supporting electrolyte. A silver wire served as reference
electrode.
+
n
deep red due to the metal-to-ligand charge transfer (MLCT)
band in the visible region. The central metal Cu(I) is
coordinated in a tetrahedral manner by the 2,9-substituted
phenanthroline and the 2,2′-bipyridine moieties. The cyclic
voltammetry (CV) study is presented in Figure 3. Experi-
In the first series of scans (Figure 3) the potential sweep
was applied between -400 and +800 mV at a scan rate of
200 mV s-1. At -400 mV no current was observed, as 124
+
is electrochemically inactive at this potential. When the
potential sweep was continued toward the anodic region, an
oxidation peak could be observed at a potential of +460 mV,
(8) (a) Dietrich-Buchecker, C. O.; Sauvage, J.-P.; Kintzinger, J.-P.
Tetrahedron Lett. 1983, 24, 5095. (b) Dietrich-Buchecker, C. O.; Sauvage,
J.-P. Chem. ReV. 1987, 87, 795.
Org. Lett., Vol. 7, No. 26, 2005
5755

conclusion from the cyclic voltammogramms. With these
results in hand we have shown that the macrocycle is indeed
pirouetting very fast around the axle. The system moves
faster than the scan rate, which implies that we can only
estimate an upper limit for the half-life time of the species
1242+ using the method of Nicholson and Shain.9 Using their
reported equations we find a reaction rate k ) 12 s-1 for the
2+
2+
rearrangement 124 f 125 and a half-life time t1/2) 60
ms at -40 °C. In agreement with previous studies on related
compounds it can be assumed that this step is about 2-3
orders of magnitude slower than the rearrangement 125+ f
+
124 . We can thus estimate that the half-life time of 125+ is
in the range of 100 µs. Compared to the previously reported
rotaxanes of ref 5, the motion of the macrocycle around the
axle in 12m+ is 2 orders of magnitude faster.
Figure 3. Cyclic voltammetric study. The electrochemical re-
sponses of the 4-coordinate and the 5-coordinate copper complexes
are found around +460 and -140 mV, respectively (potential vs a
silver quasi-reference electrode).
In conclusion, a new rotaxane has been synthesized, for
which the stoppers attached at the ends of the axle are remote
from the mobile central core and the chelate incorporated in
the thread is sterically little hindering. These two features
make the copper center very accessible and strongly facilitate
ligand exchange within its coordination sphere. The mutual
movements of the ring and the axle have been investigated
by cyclic voltammetry, and it has been shown that the ring
moves very fast around the axle (milliseconds) even at low
temperature. The present system constitutes a prototype of
a fast-moving molecular machine that will be used further
to attach related species on electrode surfaces.
corresponding to the expected oxidation of Cu(I) to Cu(II)
in agreement with related copper-complexed rotaxanes.5 On
the reverse scan, no corresponding reduction peak was
observed. However, going to the cathodic region, a reduction
peak corresponding to the reduction of the pentacoordinated
1252+ f 125+ could be observed at -140 mV. These results
clearly show that, upon oxidation or reduction of the central
metal Cu, the macrocycle is set in motion. Upon oxidation
+
of 124 , the resulting tetrahedrally coordinated Cu(II) is
unstable as Cu(II) forms stable square planar complexes or
higher coordination (five or six). Therefore, the macrocycle
pirouettes around the axle permitting the restoration of a
stable coordination, which is pentacoordination by the 2,2′,
6′2′′-terpyridine and 2′,6′-bipyridine moiety, providing a
stable coordination situation. The same phenomenon happens
for the reduction of 1252+: when the latter is reduced, the
Acknowledgment. We are grateful to the European
Communities (BIOMACH NMP-2002-3.4.1.1-3: contract
505487-1) for financial support and, in particular. for a
fellowship to U.L.-H. We also thank the CNRS for financial
support.
Supporting Information Available: Exprimental pro-
cedure and characterization data for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
+
pentacoordinated situation 125 is unstable, and the ring is
set in motion leading to the preferred tetrahedral coordination
of Cu(I). To determine the velocity of the rearrangement of
the macrocycle, we applied higher scan rates at various
temperatures to the system. The maximal chosen scan rate
was 2000 mV s-1 because at higher values the capacity
currents became too important to be able to draw a reasonable
OL052051C
(9) Nicholson, R. S.; Shain, I. Anal. Chem. 1964, 36, 706.
5756
Org. Lett., Vol. 7, No. 26, 2005