Electrochemically switchable hydrogen-bonded molecular shuttles

Article abstract of DOI:10.1021/ja0352552

A series of [2]rotaxanes containing succinamide and naphthalimide hydrogen-bonding stations for a benzylic amide macrocycle is described. Electrochemical reduction and oxidation of the naphthalimide group alters its ability to form hydrogen bonds to the m

Full text of DOI:10.1021/ja0352552

Published on Web 06/18/2003
Electrochemically Switchable Hydrogen-Bonded Molecular
Shuttles
Alessio Altieri,‡ Francesco G. Gatti,† Euan R. Kay,† David A. Leigh,*,† David Martel,‡
Francesco Paolucci,*,‡ Alexandra M. Z. Slawin,§ and Jenny K. Y. Wong†
Contribution from the School of Chemistry, UniVersity of Edinburgh, The King’s Buildings,
West Mains Road, Edinburgh EH9 3JJ, United Kingdom, Dipartimento di Chimica “G.
Ciamician”, UniVersita` degli Studi di Bologna, V. F. Selmi 2, 40126, Bologna, Italy, and
Department of Chemistry, UniVersity of St. Andrews, St. Andrews KY16 9ST, United Kingdom
Received March 20, 2003; E-mail: david.leigh@ed.ac.uk; paolucci@ciam.unibo.it
Abstract: A series of [2]rotaxanes containing succinamide and naphthalimide hydrogen-bonding stations
for a benzylic amide macrocycle is described. Electrochemical reduction and oxidation of the naphthalimide
group alters its ability to form hydrogen bonds to the macrocycle to such a degree that redox processes
can be used to switch the relative macrocycle-binding affinities of the two stations in a rotaxane by over 8
orders of magnitude. The structure of the neutral [2]rotaxane in solution is established by 1H NMR
spectroscopy and shows that the macrocycle exhibits remarkable positional integrity for the succinamide
station in a variety of solvents. Cyclic voltammetry experiments allow the simultaneous stimulation and
observation of a redox-induced dynamic process in the rotaxane which is both reversible and cyclable.
Model compounds in which various conformational and co-conformational changes are prohibited
demonstrate unequivocally that the redox response is the result of shuttling of the macrocycle between the
two stations. At room temperature in tetrahydrofuran the electrochemically induced movement of the
macrocycle between the two stations takes ∼50 µs.
Introduction
level, moleculessand their partssare constantly moving above
0 K and it is the directional control of this motion which Nature
Natural1 and artificial2 devices that function through me-
chanical motion at the molecular level require a nanoscale
structure that restricts the freedom of movement of the various
components coupled with a processsusually chemicalsto power
and control their motions. For the design of prototypical
synthetic systems, chemists have taken inspiration from the
structural elements of machinery from the macroscopic world.3
Taking this analogy too far in terms of process must be
cautioned, however, since the modes of action for useful
movement at the molecular and macroscopic levels are very
different. In the macroscopic world, objects do not move until
provided with specific energy to do so. In a mechanical machine,
this is often through a directional stimulus (i.e., when work is
done to move components in a particular way). At the molecular
uses to perform useful mechanical tasks.1
‘Molecular shuttles’ provide a promising basis for artificial
molecular machines. Various types of rotaxane4 structures permit
large amplitude, largely independent, motions of the mechani-
cally interlocked components, and the noncovalent bond-directed
routes to their synthesis offer a process for controlling the
relative positioning of the components through the interactions
that ‘live on’ in the final products.5 In a rotaxane with two
different binding sites (‘stations’) in the thread, the macrocycle
distributes itself between the stations according to the difference
in the macrocycle-binding energies and the temperature. If a
suitably large difference in macrocycle affinity between the two
stations exists, the macrocycle resides overwhelmingly in one
(3) See, for example, analogic molecular versions of gears: (a) Iwamura, H.;
Mislow, K. Acc. Chem. Res. 1988, 21, 175-182. Turnstiles: (b) Bedard,
T. C.; Moore, J. S. J. Am. Chem. Soc. 1995, 117, 10662-10671. Brakes:
(c) Kelly, T. R.; Bowyer, M. C.; Bhaskar, K. V.; Bebbington, D.; Garcia,
A.; Lang, F.; Kim, M. H.; Jette, M. P. J. Am. Chem. Soc. 1994, 116, 3657-
3658. Ratchets: (d) Kelly, T. R.; Tellitu, I.; Sestelo, J. P. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1866-1868. (e) Kelly, T. R.; Sestelo, J. P.; Tellitu,
I. J. Org. Chem. 1998, 63, 3655-3665. Rotors: (f) Schoevaars, A. M.;
Kruizinga, W.; Zijlstra, R. W. J.; Veldman, N.; Spek, A. L.; Feringa, B. L.
J. Org. Chem. 1997, 62, 4943-4948. Unidirectional spinning motors: (g)
Kelly, T. R.; De Silva, H.; Silva, R. A. Nature 1999, 401, 150-152. (h)
Koumura, N.; Zijlstra, R. W. J.; van Delden, R. A.; Harada, N.; Feringa,
B. L. Nature 1999, 401, 152-155. (i) Kelly, T. R.; Silva, R. A.; De Silva,
H.; Jasmin, S.; Zhao, Y. J. Am. Chem. Soc. 2000, 122, 6935-6949. (j)
Koumura, N. L.; Geertsema, E. M.; van Gelder, M. B.; Meetsma, A.;
Feringa, B. L. J. Am. Chem. Soc. 2002, 124, 5037-5051.
† University of Edinburgh.
‡ Universita` degli Studi di Bologna.
§ University of St. Andrews.
(1) Molecular Motors; Schliwa, M., Ed.; Wiley-VCH: Weinheim, 2003.
(2) With the notable exception of liquid crystals, the first artificial devices
that rely on mechanical motion at the molecular level for a practical function
are the catenanes and rotaxanes used to configure the remarkable solid-
state switches and logic gates developed by Stoddart and Heath. (a) Collier,
C. P.; Wong, E. W.; Beˇlohradsky´, M.; Raymo, F. M.; Stoddart, J. F.;
Kuekes, P. J.; Williams, R. S.; Heath, J. R. Science 1999, 285, 391-394.
(b) Wong, E. W.; Collier, C. P.; Beˇlohradsky´, M.; Raymo, F. M.; Stoddart,
J. F.; Heath, J. R. J. Am. Chem. Soc. 2000, 122, 5831-5840. (c) Collier,
C. P.; Jeppesen, J. O.; Luo, Y.; Perkins, J.; Wong, E. W.; Heath, J. R.;
Stoddart, J. F. J. Am. Chem. Soc. 2001, 123, 12632-12641. (d) Luo, Y.;
Collier, C. P.; Jeppesen, J. O.; Nielsen, K. A.; Delonno, E.; Ho, G.; Perkins,
J.; Tseng, H.-R.; Yamamoto, T.; Stoddart, J. F.; Heath, J. R. Chem. Phys.
Chem. 2002, 3, 519-525.
(4) Molecular Catenanes, Rotaxanes and Knots; Sauvage, J.-P., Dietrich-
Buchecker, C., Eds.; Wiley-VCH: Weinheim, 1999.
9
8644
J. AM. CHEM. SOC. 2003, 125, 8644-8654
10.1021/ja0352552 CCC: $25.00 © 2003 American Chemical Society

Hydrogen-Bonded Molecular Shuttles
A R T I C L E S
positional isomer or co-conformation.6 In stimuli-responsive
molecular shuttles,7-11 an external trigger is used to chemically
modify the system and alter the noncovalent intercomponent
interactions such that the second macrocycle-binding site
becomes energetically more favored causing translocation of
the macrocycle along the thread to the second station (Figure
1). The system can be returned to its original state by using a
second chemical transformation to restore the initial order of
station binding affinities. Performed consecutively, these two
steps allow the ‘machine’ to carry out a complete cycle of
shuttling motion. The movement of the macrocycle from station
to station is thus caused by using chemical reactions to put the
molecule into nonequilibrium co-conformations and then al-
lowing the background thermal energy to drive the macrocycle
to the new global minimum (i.e., through biased Brownian
motion).
Figure 1. Translational submolecular motion in a stimuli-responsive
molecular shuttle: (i) the macrocycle initially resides on the preferred (green)
station; (ii) a reaction occurs (red f blue) changing the relative binding
potentials of the two stations such that (iii) the macrocycle ‘shuttles’ to the
now-preferred (blue) station. If the reverse reaction (blue f red) now occurs,
(iv) the components return to their original positions. The energy available
to do mechanical work through shuttling in such a cycle is the sum of the
difference in macrocycle binding energies of the two stations in each of
the two states (i.e., ∆∆Ggreen-red + ∆∆Gblue-green).
(5) For reviews on the development of interlocked molecules as molecular
machines, see: (a) Balzani, V.; Go´mez-Lo´pez, M.; Stoddart, J. F. Acc.
Chem. Res. 1998, 31, 405-414. (b) Sauvage, J.-P. Acc. Chem. Res. 1998,
31, 611-619. (c) Chambron, J.-C.; Sauvage, J.-P. Chem. Eur. J. 1998, 4,
1362-1366. (d) Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F.
Angew. Chem., Int. Ed. 2000, 39, 3348-3391. (e) Asfari, Z.; Vicens, J. J.
Inclusion Phenom. Macro. Chem. 2000, 36, 103-118. (f) Pease, A. R.;
Jeppesen, J. O.; Stoddart, J. F.; Luo, Y.; Collier, C. P.; Heath, J. R. Acc.
Chem. Res. 2001, 34, 433-444. (g) Ballardini, R.; Balzani, V.; Credi, A.;
Gandolfi, M. T.; Venturi, M. Acc. Chem. Res. 2001, 34, 445-455. (h)
Collin, J.-P.; Dietrich-Buchecker, C.; Gavin˜a, P.; Jimenez-Molero, M. C.;
Sauvage, J.-P. Acc. Chem. Res. 2001, 34, 477-487. (i) Molecular Machines
and Motors: Structure and Bonding; Sauvage, J.-P., Ed.; Springer: Berlin,
2001;Vol. 99. (j) Venturi, M.; Credi, A.; Balzani, V. Molecular DeVices
and MachinessA Journey into the Nanoworld, Wiley-VCH: Weinheim,
2003.
(6) “Co-conformation” refers to the relative positions of the mechanically
interlocked components with respect to each other, see: Fyfe, M. C. T.;
Glink, P. T.; Menzer, S.; Stoddart, J. F.; White, A. J. P.; Williams, D. J.
Angew. Chem., Int. Ed. Engl. 1997, 36, 2068-2070. Energy differences
between two translational isomers or co-conformations of greater than 1.18
kcal mol-1 imply >90% station occupancy for the more favored site at
298 K.
(7) For examples of chemically responsive molecular shuttles, see: (a) Lane,
A. S.; Leigh, D. A.; Murphy, A. J. Am. Chem. Soc. 1997, 119, 11092-
11093. (b) Mart´ınez-D´ıaz, M.-V.; Spencer, N.; Stoddart, J. F. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1904-1907. (c) Gong, C.; Gibson, H. W. Angew.
Chem., Int. Ed. Engl. 1997, 36, 2331-2333. (d) Ashton, P. R.; Ballardini,
R.; Balzani, V.; Baxter, I.; Credi, A.; Fyfe, M. C. T.; Gandolfi, M. T.;
Go´mez-Lo´pez, M.; Mart´ınez-D´ıaz, M.-V.; Piersanti, A.; Spencer, N.;
Stoddart, J. F.; Venturi, M.; White, A. J. P.; Williams, D. J. J. Am. Chem.
Soc. 1998, 120, 11932-11942. (e) Jime´nez, M. C.; Dietrich-Buchecker,
C.; Sauvage, J.-P Angew. Chem., Int. Ed. 2000, 39, 3284-3287. (f) Lee, J.
W.; Kim, K.; Kim, K. Chem. Commun. 2001, 1042-1043. (g) Elizarov,
A. M.; Chiu, S.-H.; Stoddart, J. F. J. Org. Chem. 2002, 67, 9175-9181.
(h) Jimenez-Molero, M. C.; Dietrich-Buchecker, C.; Sauvage, J.-P. Chem.
Eur. J. 2002, 8, 1456-1466. (i) Da Ross, T.; Guldi, D. M.; Farran Morales,
A.; Leigh, D. A.; Prato, M.; Turco, R. Org. Lett. 2003, 5, 689-691. (j)
Tseng, H.-R.; Vignon, S. A.; Stoddart, J. F. Angew. Chem., Int. Ed. 2003,
42, 1491-1495.
(8) For examples of photochemically responsive molecular shuttles, see: (a)
Benniston, A. C. Chem. Soc. ReV. 1996, 25, 427-435. (b) Murakami, H.;
Kawabuchi, A.; Kotoo, K.; Kunitake, M.; Nakashima, N. J. Am. Chem.
Soc. 1997, 119, 7605-7606. (c) Blanco, M.-J.; Jime´nez, M. C.; Chambron,
J.-C.; Heitz, V.; Linke, M.; Sauvage, J.-P Chem. Soc. ReV. 1999, 28, 293-
305. (d) Ashton, P. R.; Ballardini, R.; Balzani, V.; Credi, A.; Dress, K. R.;
Ishow, E.; Kleverlaan, C. J.; Kocian, O.; Preece, J. A.; Spencer, N.; Stoddart,
J. F.; Venturi, M.; Wenger, S. Chem. Eur. J. 2000, 6, 3558-3574. (e)
Wurpel, G. W. H.; Brouwer, A. M.; van Stokkum, I. H. M.; Farran, A.;
Leigh, D. A. J. Am. Chem. Soc. 2001, 123, 11327-11328. (f) Brouwer, A.
M.; Frochot, C.; Gatti, F. G.; Leigh, D. A.; Mottier, L.; Paolucci, F.; Roffia,
S.; Wurpel, G. W. H. Science 2001, 291, 2124-2128. (g) Stanier, C. A.;
Alderman, S. J.; Claridge, T. D. W.; Anderson, H. L. Angew. Chem., Int.
Ed. 2002, 41, 1769-1772. (h) Altieri, A.; Bottari, G.; Dehez, F.; Leigh,
D. A.; Wong, J. K. Y.; Zerbetto, F. Angew. Chem., Int. Ed. 2003, 42, in
press.
We recently reported the real-time observation of macrocycle
translocation in such a stimuli-responsive molecular shuttle
following the photochemically induced reduction of the naph-
thalimide unit in rotaxane 1.8f In fact, we originally designed 1
as an electrochemically switchable system.12 Electrochemistry
is potentially an attractive method through which to modulate
the behavior of molecular devices because it can be easily and
rapidly turned on and off while also offering a reagent and
waste-free procedure. Obviously, an important attribute of any
‘molecular machine’ is that for the submolecular motion to be
useful it must be detectable through some change in the system’s
(10) For examples of the use of electrical and electrochemical stimuli to control
ring rotation in catenanes, catenates, and rotaxanes, see: (a) Amabilino,
D. B.; Dietrich-Buchecker, C. O.; Livoreil, A.; Pe´rez-Garc´ıa, L.; Sauvage,
J.-P.; Stoddart, J. F. J. Am. Chem. Soc. 1996, 118, 3905-3913. (b) Ca´rdenas,
D. J.; Livoreil, A.; Sauvage, J.-P. J. Am. Chem. Soc. 1996, 118, 11980-
11981. (c) Livoreil, A.; Sauvage, J.-P.; Armaroli, N.; Balzani, V.; Flamigni,
L.; Ventura, B. J. Am. Chem. Soc. 1997, 119, 12114-12124. (d) Asakawa,
M.; Ashton, P. R.; Balzani, V.; Credi, A.; Hamers, C.; Mattersteig, G.;
Montalti, M.; Shipway, A. N.; Spencer, N.; Stoddart, J. F.; Tolley, M. S.;
Venturi, M.; White, A. J. P.; Williams, D. J. Angew. Chem., Int. Ed. Engl.
1998, 37, 333-337. (e) Raehm, L.; Kern, J.-M.; Sauvage, J.-P. Chem. Eur.
J. 1999, 5, 3310-3317. (f) Collier, C. P.; Mattersteig, G.; Wong, E. W.;
Luo, Y.; Beverly, K.; Sampaio, J.; Raymo, F. M.; Stoddart, J. F.; Heath, J.
R. Science 2000, 289, 1172-1175. (g) Asakawa, M.; Higuchi, M.;
Mattersteig, G.; Nakamura, T.; Pease, R.; Raymo, F. M.; Shimizu, T.;
Stoddart, J. F. AdV. Mater. 2000, 12, 1099-1102. (h) Bermudez, V.;
Capron, N.; Gase, T.; Gatti, F. G.; Kajzar, F.; Leigh, D. A.; Zerbetto, F.;
Zhang, S. Nature 2000, 406, 608-611. (i) Ashton, P. R.; Baldoni, V.;
Balzani, V.; Credi, A.; Hoffmann, H. D. A.; Mart´ınez-D´ıaz, M.-V.; Raymo,
F. M.; Stoddart, J. F.; Venturi, M. Chem. Eur. J. 2001, 7, 3482-3493.
(11) For examples of the use of electrochemical stimuli to control threading
and dethreading in pseudorotaxanes and other host-guest systems, see:
(a) Bernardo, A. R.; Stoddart, J. F.; Kaifer, A. E. J. Am. Chem. Soc. 1992,
114, 10624-10631. (b) Collin, J.-P.; Gavin˜a, P.; Sauvage, J.-P. Chem.
Commun. 1996, 2005-2006. (c) Asakawa, M.; Ashton, P. R.; Balzani, V.;
Credi, A.; Mattersteig, G.; Matthews, O. A.; Montalti, M.; Spencer, N.;
Stoddart, J. F.; Venturi, M. Chem. Eur. J. 1997, 3, 1992-1996. (d)
Devonport, W.; Blower, M. A.; Bryce, M. R.; Goldenberg, L. M. J. Org.
Chem. 1997, 62, 885-887. (e) Credi, A.; Montalti, M.; Balzani, V.;
Langford, S. J.; Raymo, F. M.; Stoddart, J. F. New. J. Chem. 1998, 22,
1061-1065. (f) Asakawa, M.; Ashton, P. R.; Balzani, V.; Boyd, S. E.;
Credi, A.; Mattersteig, G.; Menzer, S.; Montalti, M.; Raymo, F. M.; Ruffilli,
C.; Stoddart, J. F.; Venturi, M.; Williams, D. J. Eur. J. Org. Chem. 1999,
985-994. (g) Ashton, P. R.; Balzani, V.; Becher, J.; Credi, A.; Fyfe, M.
C. T.; Mattersteig, G.; Menzer, S.; Nielsen, M. B.; Raymo, F. M.; Stoddart,
J. F.; Venturi, M.; Williams, D. J. J. Am. Chem. Soc. 1999, 121, 3951-
3957. (h) Cooke, G.; Duclairoir, F. M. A.; Rotello, V. M.; Stoddart, J. F.
Tetrahedron Lett. 2000, 41, 8163-8166. (i) Balzani, V.; Credi, A.;
Mattersteig, G.; Matthews, O. A.; Raymo, F. M.; Stoddart, J. F.; Venturi,
M.; White, A. J. P.; Williams, D. J. J. Org. Chem. 2000, 65, 1924-1936.
(j) Balzani, V.; Becher, J.; Credi, A.; Nielsen, M. B.; Raymo, F. M.;
Stoddart, J. F.; Talarico, A. M.; Venturi, M. J. Org. Chem. 2000, 65, 1947-
1956. (k) Lahav, M.; Shipway, A. N.; Willner, I.; Nielsen, M. B.; Stoddart,
J. F. J. Electroanal. Chem. 2000, 482, 217-221.
(9) For examples of electrochemically responsive molecular shuttles, see ref
8a and the following: (a) Bissell, R. A.; Co´rdova, E.; Kaifer, A. E.; Stoddart,
J. F. Nature 1994, 369, 133-136. (b) Anelli, P.-L.; Asakawa, M.; Ashton,
P. R.; Bissell, R. A.; Clavier, G.; Go´rski, R.; Kaifer, A. E.; Langford, S.
J.; Mattersteig, G.; Menzer, S.; Philp, D.; Slawin, A. M. Z.; Spencer, N.;
Stoddart, J. F.; Tolley, M. S.; Williams, D. J. Chem. Eur. J. 1997, 3, 1113-
1135. (c) Collin, J.-P.; Gavin˜a, P.; Sauvage, J.-P. New J. Chem. 1997, 21,
525-528. (d) Armaroli, N.; Balzani, V.; Collin, J.-P.; Gavin˜a, P.; Sauvage,
J.-P.; Ventura, B. J. Am. Chem. Soc. 1999, 121, 4397-4408. (e) Ballardini,
R.; Balzani, V.; Dehaen, W.; Dell’Erba, A. E.; Raymo, F. M.; Stoddart, J.
F.; Venturi, M. Eur. J. Org. Chem. 2000, 591-602.
(12) Earlier attempts to make redox-active hydrogen-bonded molecular shuttles
based on anthraquinone subunits were unsuccessful (Dunnett, J. A. Ph.D.
Thesis, University of Manchester Institute of Science and Technology, U.K.,
2000).
9
J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003 8645

A R T I C L E S
Altieri et al.
Chart 1. Molecular Shuttle 1, Shown as the succ-1 Positional
Isomer, and Thread 2a
Figure 2. X-ray crystal structure of [2]rotaxane 3 (for clarity carbon atoms
of the macrocycle are shown in blue and the carbon atoms of the thread
are shown in yellow; oxygen atoms are depicted in red, nitrogen atoms in
dark blue, and amide protons in white). Intramolecular hydrogen-bond
distances (Å): O1-HN3/O2-HN5 1.88. Intermolecular hydrogen-bond
distances (Å): O5-HN4′/ O3′-HN6 2.0, O3′-HN1/O5-HN2′ 2.21.
1
a The letters indicate selected nonequivalent H environments.
properties. Electrochemistry is also advantageous in this regard
since the same stimulus can simultaneously act as both effector
and detector of the motion.
Here we report the cyclic voltammetry of 1 and a series of
related rotaxanes and threads. The rates and energies obtained
from the electrochemical redox experiments are consistent with
those obtained photochemically. The spectroscopic and elec-
trochemical behavior of key model compounds is used to show
unambiguously that the dynamics observed is a consequence
of the position of the macrocycle on the thread changing upon
reduction of the naphthalimide unit.
Design
Hydrogen bonds are largely electrostatic in nature,13 so
changes in the electron charge density of a hydrogen-bonding
group can have a dramatic effect on its binding abilitysa
phenomenon exploited in redox-switched molecular recogni-
tion14 and the modulation of protein and peptide structure.15
Molecular shuttle 1 consists of a benzylic amide macrocycles
a strong hydrogen-bond donor through the amide NH groupss
mechanically linked onto a thread molecule, 2, containing two
potential hydrogen-bond acceptor moietiessa succinamide
(succ) station and a redox-active 3,6-di-tert-butyl-1,8-naphthal-
imide16 (ni) stationsseparated by a C12 aliphatic spacer (Chart
Figure 3. X-ray crystal structure of [2]rotaxane 4. Intramolecular hydrogen-
bond distances (Å): O2-HN3/O1-HN5 1.90, O2-HN4/O1-HN6 2.01.
1). The ability of the succ station to template formation of the
macrocycle by a five-component ‘clipping’ reaction between
isophthaloyl dichloride and p-xylylene diamine in apolar
solvents is well established.8f X-ray crystal structures of model
succinamide rotaxanes 3 and 4 (Figures 2 and 3) demonstrate
an excellent fit between the thread and macrocycle, both in terms
of steric interactions and the complementary positioning of the
hydrogen-bonding amide groups on the two components. While
the structure of 3 shows both intra- and intermolecular hydrogen
bonding in the solid state, 4 has a solely intramolecularly
hydrogen-bonded structure and is presumably more representa-
tive of the hydrogen-bonding motifs adopted by ‘isolated’
molecules in solution. The macrocycle adopts a near-perfect
chair conformation, driven by formation of two sets of bifurcated
hydrogen bonds between the amide protons of each isophthala-
mide moiety and the succinamide carbonyl oxygens.
(13) Morokuma, K. Acc. Chem. Res. 1977, 10, 294-300.
(14) See, for example: (a) Brooksby, P. A.; Hunter, C. A.; McQuillan, A. J.;
Purvis, D. H.; Rowan, A. E.; Shannon, R. J.; Walsh, R. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 2489-2491. (b) Ge, Y.; Lilienthal, R.; Smith, D.
K. J. Am. Chem. Soc. 1996, 118, 3976-3977. (c) Deans, R.; Niemz, A.;
Breinlinger, E. C.; Rotello, V. M. J. Am. Chem. Soc. 1997, 119, 10863-
10864. (d) Boulas, P. L.; Go´mez-Kaifer, M.; Echegoyen, L. Angew. Chem.,
Int. Ed. Engl. 1998, 37, 216-247. (e) Niemz, A.; Rotello, V. M. Acc. Chem.
Res. 1999, 32, 44-52. (f) Boal, A. K.; Rotello, V. M. J. Am. Chem. Soc.
1999, 121, 4914-4915. (g) Ge, Y.; Miller, L.; Ouimet, T.; Smith, D. K J.
Org. Chem. 2000, 65, 8831-8838. (h) Ge, Y.; Smith, D. K. Anal. Chem.
2000, 72, 1860-1865. (i) Carr, J. D.; Coles, S. J.; Hursthouse, M. B.; Light,
M. E.; Tucker, J. H. R.; Westwood, J. Angew. Chem., Int. Ed. 2000, 39,
3296-3299. (j) Tucker, J. H. R.; Collinson, S. R. Chem. Soc. ReV. 2002,
31, 147-156. (k) Cooke, G.; Rotello, V. M. Chem. Soc. ReV. 2002, 31,
275-286.
The choice of a substituted naphthalimide unit as the redox-
active station was inspired by the work of Smith14b and
Rotello.14c,e,f Imides are comparatively poor hydrogen-bond
acceptors;17 indeed, threads containing just the naphthalimide
(15) See, for example: (a) Schenck, H. L.; Dado, G. P.; Gellman, S. H. J. Am.
Chem. Soc. 1996, 118, 12487-12494. (b) Pascher, T.; Chesick, J. P.;
Winkler, J. R.; Gray, H. B. Science 1996, 271, 1558-1560.
(16) The tert-butyl substituents provide the steric bulk necessary to stopper the
rotaxane while only slightly altering the redox potential of the naphthalimide
group (E1/2 (2) ) -1.41 V cf. E1/2 (N-methyl naphthalimide) ) -1.31 V):
Viehbeck, A.; Goldberg, M. J.; Kovac, C. A. J. Electrochem. Soc. 1990,
137, 1460-1466.
(17) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford University
Press: New York, 1997.
9
8646 J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003

Hydrogen-Bonded Molecular Shuttles
A R T I C L E S
Chart 2. Model Rotaxanes 3-8 Shown as the succ- Positional Isomersa
a The dashed lines show hydrogen-bond interactions possible in 1 but disallowed in the model compound. Solid lines show changes of co-conformation
that require shuttling which are disallowed in the model compound.
unit do not template formation of the benzylic amide macrocycle
to give rotaxanes. To minimize its free energy, the macrocycle
in 1 must therefore sit over the succinamide station in non-
hydrogen-bonding solvents, so that co-conformation succ-1
predominates (Figure 4). One-electron reduction of naphthal-
imides to the corresponding radical anion, however, results in
a substantial increase in electron charge density on the imide
carbonyls and a concomitant increase in hydrogen-bond ac-
cepting ability.14b,c,e In 1, this change in oxidation state should
reverse the relative hydrogen-bonding abilities of the two thread
stations so that co-conformation ni-1-• is preferred in the
reduced state. This dynamic process was studied in 1 through
cyclic voltammetry experiments and 1H NMR spectroscopy. To
probe the nature of the stimulated motion, we carried out similar
studies on model rotaxanes 5-8 (Chart 2). Each of these
analogues of 1 is designed to restrict a different type of
conformational or co-conformational change so that the true
origin of the effects seen for 1 could be unambiguously
determined.
Synthesis
Rotaxanes 1, 5, and 6 were prepared according to Scheme 1.
Treatment of threads 2, 9, or 10 with 10 equiv of each of
p-xylylene diamine and isophthaloyl dichloride (CHCl3, Et3N,
9
J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003 8647

A R T I C L E S
Altieri et al.
Scheme 1. Synthesis of Rotaxanes 1, 5, and 6a
a (i) 1,12-Diaminododecane, Et3N, EtOH, ∆, 14%; (ii) 1,2-ethylenediamine, EtOH, ∆, 78%; (iii) pyridine, AcOH, 90%; (iv) 1-[3-(dimethylamino)propyl]-
3-ethylcarbodiimide hydrochloride (EDCI), 4-(dimethylamino)pyridine (DMAP), CH2Cl2, 0 °C to RT, 98% (2), 93% (10); (v) MeI, NaH, THF, 0 °C to RT,
94%; (vi) isophthaloyl dichloride, p-xylylene diamine, Et3N, CHCl3, 59% (1), 33% (5), 62% (6).
Scheme 2. Synthesis of Rotaxanes 7 and 8a
a (i) (PhCH2)2NH, EDCI, DMAP, CH2Cl2, 0 °C to RT, 94%; (ii) 3-amino-propane-1,2-diol, ∆, 84%; (iii) tert-butylchlorodiphenylsilane (TBDPSCl),
imidazole, DMAP, CH2Cl2, 62%; (iv) pyridine, ∆, 60%; (v) EDCI, DMAP, CH2Cl2, 0 °C to RT, 90%; (vi) isophthaloyl dichloride, p-xylylene diamine,
Et3N, CHCl3, 23%; (vii) CH2Cl2, tetrabutylammonium fluoride (TBAF), 70%.
4 h, high dilution) provided [2]rotaxanes 1, 5, and 6 in 59%,
33%, and 62% yields, respectively. Scheme 2 shows the
synthesis of ‘gate-closed’ thread 11 which, under similar
rotaxane-forming conditions, furnished [2]rotaxane 7 in 23%
yield. The ‘gate-opened’ rotaxane 8 was obtained by treatment
of 7 with a stoichiometric amount of tetra-n-butylammonium
fluoride in CH2Cl2.18
(18) ‘Gate-opened’ rotaxane 8 was obtained together with the 1°-O-acyl, 2°-
hydroxyl regioisomer from which it was not separated.
9
8648 J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003

Hydrogen-Bonded Molecular Shuttles
A R T I C L E S
Figure 4. An electrochemically switchable, hydrogen-bonded molecular shuttle 1. In the neutral state, the translational co-conformation succ-1 is predominant
as the ni station is a poor hydrogen-bond acceptor (Kn ) (1.2 ( 1) × 10-6). Upon reduction, the equilibrium between succ-1-• and ni-1-• is altered (Kred
) (5 ( 1) × 102) because ni-• is a powerful hydrogen-bond acceptor. Upon reoxidation, the macrocycle shuttles back to its original position. Repeated
reduction and oxidation causes the macrocycle to shuttle forward and backward between the two stations. All the values shown refer to experiments in
anhydrous THF at 298 K with tetrabutylammonium hexafluorophosphate as the supporting electrolyte.
Table 1. Solvent Effects on Chemical Shifts (δH) and Differential
Chemical Shifts (∆δH) for Succinamide Protons in 1 and 2
NH and NH
H and H
c
f
d
e
solvent
δH(1)/ppm
δH(2)/ppm
∆δH/ppm
CDCl3
5.95
5.73; 5.61
6.42; 4.22
7.20; 7.12
7.78; 7.60
-1.45
-1.2
-1.2
-0.7
d3-MeCNa
d8-THF
6.47; 6.42
7.84; 7.63
7.92; 7.40
d6-DMSO
a Spectra collected at 329 K due to low solubility of 1 in d3-MeCN.
station (Ha, Hb, and Hg) are also shielded but to a lesser extent;
all other thread protons, including those of the naphthalimide,
are essentially unaffected by the presence of the macrocycle.
The infrared spectrum of 1 in CHCl3 also shows no changes in
the CO stretching band of the naphthalimide moiety compared
to the spectrum of 2 (νCO
) 1780 cm-1 in each case),
stretch
suggesting no significant hydrogen bonding occurs to the
naphthalimide subunit in the rotaxane.8f Clearly the succ-1
translational isomer is the predominant structure in chloroform.
In fact, unlike peptide-based molecular shuttles,7a the posi-
tional integrity of the macrocycle in 1 is remarkably independent
of the nature of the solvent. The difference in chemical shifts
for methylene protons Hd and He in 1 and 2 as well as the shifts
for succinamide amide protons Hc and Hf in solvents of varying
hydrogen-bonding ability are shown in Table 1. The increasing
δH values for the amide protons illustrate the ability of these
solvents to disrupt hydrogen bonding (CDCl3 < d3-MeCN <
d8-THF < d6-DMSO). It can be seen, however, that in CDCl3,
d3-MeCN, and d8-THF, shielding of the succ methylene protons
is virtually unchanged, indicating that positional integrity of the
Figure 5. 1H NMR spectra (400 MHz) of (a) 2 and (b) 1 in CDCl3 at 298
K. The letters correspond to the assignments shown in Chart 1.
Co-conformation in the Neutral State
1H NMR spectroscopy is a useful tool for studying transla-
tional isomerism in rotaxanes. For benzylic amide macrocycle-
containing rotaxanes, aromatic ring currents in the p-xylylene
rings result in significant upfield shifting (typically ∆δH > 1
ppm) for protons on the portion of the thread covered by the
macrocycle.19 Comparison of the 1H NMR spectra of rotaxane
1 and the corresponding thread 2 in CDCl3 (Figure 5) shows
such an upfield shift for succinic methylene protons Hd and He
(∆δH ) -1.45 ppm).20 Protons in close proximity to the succ
(20) Dilution experiments demonstrated that intermolecular hydrogen-bonding
interactions were absent at concentrations of 10 mmol L-1 and below for
the current rotaxanes and threads. All NMR spectra discussed were therefore
collected at this standard concentration.
(19) Johnston, A. G.; Leigh, D. A.; Murphy, A.; Smart, J. P.; Deegan, M. D. J.
Am. Chem. Soc. 1996, 118, 10662-10663.
9
J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003 8649

A R T I C L E S
Altieri et al.
Figure 6. 1H NMR spectrum of 4 (400 MHz, CDCl3) at 243 K showing
slow pirouetting of the macrocycle about the thread (HE1 and HE2 resolved).
SPT-SIR between resolved signals gives the energy barrier for macrocycle
pirouetting. Higher temperature spectra (expansion, 243-296 K (CDCl3)
and 329-369 K (C2D2Cl4)) illustrate the wide temperature ranges that can
exist between full resolution and coalescence of signals affected by the
dynamic processes.
Figure 7. Cyclic voltammograms at 298 K of (a) 2 and (b) 1 showing
both experimental (black line) and simulated (red line) results. Experiments
carried out on 1 mmol L-1 solutions of substrate in anhydrous THF with
tetrabutylammonium hexafluorophosphate (5 × 10-2 mol L-1) as electrolyte
and ferrocene as internal standard. Scan rate: 2 V s-1. Working electrode:
Pt.
macrocycle pirouetting at 298 K of 12.9 ( 0.1 kcal mol-1
.
shuttle is maintained in all three solvents. Even in d6-DMSO
the upfield shift for Hd and He is only reduced by ≈50%,
indicating that even in this strongly hydrogen-bond-disrupting
solvent the macrocycle still spends a significant amount of time
over the succ station.
Similarly, the barrier for 3 was determined as 11.2 ( 0.1 kcal
mol-1 at 298 K. Even when compared to the fumaramide
template which holds the hydrogen-bond acceptor groups of the
thread in a close-to-ideal arrangement for the donor groups in
q
the macrocycle (∆G298 for the fumaramide analogue of 3 )
The energy barrier for pirouetting of the macrocycle around
the thread provides a definitive measure of the strength of the
intercomponent interactions.21 Due to the complexity of the 1H
NMR spectra of 1 and its analogues, this process was studied
in the simpler, C2-symmetric, rotaxanes 3 and 4 as models for
13.4 ( 0.1 kcal mol-1 21b
it can be seen that the succinic
)
templates in 3 and 4 result in strong intercomponent hydrogen
bonding. Clearly, dissociation from the succinamide station in
a molecular shuttle will involve a significant energy barrier.
1
shuttles 1 and 5, respectively. Variable-temperature (VT) H
Redox-Switched Shuttling
NMR experiments in CDCl3 and C2D2Cl4 (because of the wide
temperature range involved) using both the coalescence method22
and spin polarization transfer by selective inversion recovery
(SPT-SIR)23 were employed to calculate the rates of macrocycle
Translocation of the macrocycle in 1 could be readily
triggered by electrochemical reduction of the naphthalimide
station to its radical anion. Since the changes in hydrogen-
bonding pattern affect the electrochemical properties of the
system, the shuttling process could be both effected and
observed in cyclic voltammetry (CV) experiments of 1 and its
analogues and components.25 The CV response of 2 in anhy-
drous THF (Figure 7a) exhibits a reversible, one-electron
reduction peak corresponding to reduction and reoxidation of
the naphthalimide group (2/2-•, E1/2 ) -1.41 V).16 Rotaxane
1, on the other hand, displays a similar cathodic signal, I (Ec )
1
pirouetting. The VT H NMR plot for 4 (Figure 6) shows the
signals for macrocycle benzylic methylene protons E1 and E2
are well separated at 243 K, while a single sharp peak is
observed at 369 K.24 SPT-SIR of 4 gives an energy barrier for
(21) For examples of macrocycle pirouetting in benzylic amide macrocycle-
based [2]rotaxanes, see ref 10h and the following: (a) Leigh, D. A.; Murphy,
A.; Smart, J. P.; Slawin, A. M. Z. Angew. Chem., Int. Ed. Engl. 1997, 36,
728-732. (b) Gatti, F. G.; Leigh, D. A.; Nepogodiev, S. A.; Slawin, A.
M. Z.; Teat, S. J.; Wong, J. K. Y. J. Am. Chem. Soc. 2001, 121, 5983-
5989. (c) Gatti, F. G.; Leo´n, S.; Wong, J. K. Y.; Bottari, G.; Altieri, A.;
Farran Morales, A. M.; Teat, S. J.; Frochot, C.; Leigh, D. A.; Brouwer, A.
M.; Zerbetto, F. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 10-14.
(22) Sandstro¨m, J. Dynamic NMR Spectroscopy; Academic Press: London, 1982.
(23) Dahlquist, F. W.; Longmur, K. J.; Du Vernet, R. B. J. Magn. Reson. 1975,
17, 406-413. For applications of SPT-SIR to dynamic processes in
rotaxanes, see, for example: Clegg, W.; Gimenez-Saiz, C.; Leigh, D. A.;
Murphy, A.; Slawin, A. M. Z.; Teat, S. J. J. Am. Chem. Soc. 1999, 121,
4124-4129. For an example of how the SPT-SIR experiments were
performed for the current systems, see the Supporting Information.
(24) Due to the wide temperature range, spectra were collected in CDCl3 from
243 to 296 K and in C2D2Cl4 from 329 to 369 K
(25) Cyclic voltammetry was carried out on 10-3 mol L-1 solutions of substrate
in anhydrous THF with tetrabutylammonium hexafluorophosphate (TBHF,
5 × 10-2 mol L-1) as electrolyte and ferrocene as internal standard. The
invariance of the chemical shift for amide protons of 1 and 2 in d8-THF in
the presence of TBHF at 5 × 10-2 mol L-1 suggests that the electrolyte
does not significantly affect amide-amide intramolecular hydrogen bonds
at these concentrations.
9
8650 J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003

Hydrogen-Bonded Molecular Shuttles
A R T I C L E S
-1.40 V, at 2 V s-1), yet lacks the corresponding oxidation
peak (Figure 7b). Rather, a new one-electron anodic peak, II,
appears toward more positive potentials (Ea ) -0.89 V, at 2 V
s-1). On subsequent scans, performed without renewal of the
diffusion layer, it is only these two peaks that appear, indicating
that the system returns to its original state after each cycle on
this time scale.
These results can be interpreted in terms of a standard square
scheme for a redox-switched binding process whereby the
reversible translational isomerism equilibria for each oxidation
state are connected via the electron-transfer steps (Figure 4).
As shown earlier, the shuttle in its neutral state overwhelmingly
adopts co-conformation succ-1. Reduction of the naphthalimide
group to give succ-1-• is therefore unaffected by the macrocycle.
Due to the increased hydrogen-bonding ability of naphthalimide
in its reduced state, the equilibrium between co-conformations
in this oxidation state lies far toward the ni-1-• isomer so that
shuttling of the macrocycle occurs. Resultant hydrogen bonding
to the macrocycle amide protons stabilizes the increased electron
charge on the naphthalimide so that more positive potentials
must be reached before the oxidation ni-1-• f ni-1 occurs (∆Ep
) 0.51 V). Once again in the neutral state, the macrocycle
shuttles back to the preferred succinamide station therefore
restoring the system to its original state as succ-1. The absence
of any succ-1-• f succ-1 oxidation peak or ni-1 f ni-1-•
reduction peak demonstrates the remarkable positional integrity
of the shuttle in both oxidation states while also indicating that
the shuttling process is rapid on the time scale of the experiment.
The CV curve of 1 was simulated (Figure 7b). Fitting with
the experimental results allowed calculation of the redox
Figure 8. Cyclic voltammograms at 213 K of 1 showing both experimental
(black line) and simulated (red line) results. Experiments carried out on 1
mmol L-1 solutions of substrate in anhydrous THF with tetrabutylammo-
nium hexafluorophosphate (5 × 10-2 mol L-1) as electrolyte and ferrocene
as internal standard. Scan rate: 2 V s-1. Working electrode: Pt.
Figure 9. Variation of forward shuttling rate (kf1-•) with relative permit-
tivity, ꢀ, of solvent. Data points in black are calculated from experiments
using an electrochemical stimulus; data points in red are calculated from
experiments using a photochemical stimulus.8f All experiments were carried
out at 298 K.
ni
potential for the electron-transfer process ni-1-• / ni-1 as E1/2
) -0.90 V. Also obtained were the rate constants in the reduced
state for the forward shuttling process (succ-1-• f ni-1-•; kf1-•
) (2 ( 1) × 104 s-1; T ) 298 K, THF) and the backward
observe the succ-1-• f succ-1 oxidation before shuttling to ni-
1-• occurs. Such high scan rates are not achievable with 1 due
to the high resistivity of THF solutions. An alternative strategy
is to run the experiments at low temperatures to slow the
shuttling motion. Indeed, at 213 K, the cathodic peak, I, for
reduction of the succ-1 species exhibits an anodic counterpart,
III, while the intensity of anodic peak II is correspondingly
reduced (Figure 8). The even greater resistivity at this temper-
ature prevents the use of higher scan rates to oxidize all succ-
1-• species before shuttling occurs.
process (ni-1-• f succ-1-•; kb ) (40 ( 5) × 102 s-1; T )
1-•
298 K, THF). These figures yield a co-conformational equilib-
rium constant for the reduced state of Kred ) (5 ( 1) × 102
confirming that the ni-1-• co-conformation is indeed strongly
preferred. The ni-1 f succ-1 shuttling process occurs faster than
the ni-1-• f ni-1 oxidative electron-transfer step, so that it is
not possible to calculate rate constants for this process.26 The
equilibrium constants in each oxidation state for any such redox-
switched binding process are related by eq 1.27 Accordingly,
the constant for the neutral state co-conformational equilibrium
Simulation of the low-temperature CV curve (Figure 8)
indicated, as expected, that the rate constant for forward shuttling
in the reduced state is much lower than that at room temperature
(kf1-• ≈ 1 s-1; T ) 213 K, THF). By repeating the simulation
at various temperatures and subjecting the results to an Eyring
plot analysis, the following activation parameters for the
shuttling process in the reduced state were obtained: ∆Hq )
was calculated as Kn ) (1.2 ( 1) × 10-6 28
, consistent with the
predominance of the succ-1 co-conformation shown by 1H NMR
and IR.
Kred
RT
succ
∆Eo ) E1/2ni - E1/2
)
ln
(1)
(
)
nF Kn
14.0 ( 0.6 kcal mol-1, ∆Sq ) 8.7 ( 2 cal K-1 mol-1 and ∆G298
q
Given the fast rate constants for shuttling in the reduced state,
a scan rate on the order of 10 kV s-1 would be required to
) 11.4 ( 1.2 kcal mol-1
.
The electrochemical results are consistent with the photo-
chemically induced shuttling experiments carried out in alky-
lnitrile solutions.8f Unfortunately, when carrying out the elec-
trochemistry experiments in acetonitrile, adsorption phenomena
at the electrodes prevented any analysis of the curves and
therefore any direct comparison of the two stimuli in the same
solvent. The electrochemical measurements were instead re-
peated in DMF. Figure 9 compares the forward shuttling rates
(26) On increasing the scan rate from 20 mV s-1 to 250 V s-1, the anodic peak
II shifts to higher potentials by ≈60 mV/log unit of scan rate, indicating
that the oxidation process is kinetically slow. Due to the non-Nernstian
nature of the oxidation process, the rate constants relative to the ni-1/succ-1
shuttling process can therefore not be obtained by fitting of the CV curve.
The absence of a cathodic counterpart for peak II at scan rates as high as
250 V s-1, however, allows estimation of kb ≈ 1.6 × 103 s-1, so that kf1
1
≈ 2 × 10-3 s-1
VCH: Weinheim, 1999.
.
(27) Kaifer, A.; Go´mez-Kaifer, M. Supramolecular Electrochemistry; Wiley-
succ
(28) Assuming E1/2
) E1/2 (2) ) -1.41 V.
9
J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003 8651

A R T I C L E S
Altieri et al.
Figure 10. Possible folded co-conformations for reduced rotaxane 1-•. In fact, only folded-ni-1-• is a low-energy structure.
calculated from a range of experiments in different solvents
using the two stimuli. An approximately linear relationship exists
between the rate of shuttling (kf1-•) and polarity of the medium,
suggesting that the nature of the solvent is the main variable
affecting the shuttling rates in these experiments (the presence
of the supporting electrolytes in the electrochemistry experi-
ments could also play a role, but see ref 25). Pleasingly, the
shuttling process in this class of molecular shuttles is indepen-
dent of the meansslight or electronssused to trigger it.
Is the Redox-Induced Motion Really Shuttling?
It might be argued that hydrogen-bonding interactions
between the ni station and either the succ station or the
macrocycle in 1-• could arise from folded conformations (e.g.,
folded-succI-1-• and folded-succII-1-•, Figure 10), accounting
for the experimental observations without requiring an actual
change in the position of the macrocycle on the thread in the
reduced rotaxane. To demonstrate unequivocally that the reduc-
tion of 1 does, in fact, result in shuttling of the macrocycle
between the succ and ni stations, the electrochemical behavior
of model rotaxanes 5-8 (Chart 2) was investigated.
‘Gate-closed’ shuttle 7 incorporates a bulky silyl ether
between the two stations which is large enough to preclude any
shuttling yet should not prevent the formation of a variety of
folded conformations; removal of this group to give 8 restores
the possibility of redox-switched shuttling, thus generating a
system closely analogous to 1.
Compared to 1, rotaxane 6 has a reduced length of alkyl
spacer (C2 in place of C12), thus disallowing folded conforma-
tions involving multipoint interactions between the naphthal-
imide carbonyls and macrocycle amide protons.
In model compound 5, methylation of the succinamide
nitrogens prevents a folded conformation involving hydrogen
bonding to these units (i.e., the succ unit in 5 can only act as a
hydrogen-bond acceptor).
1H NMR spectra of each of 5-8 in CDCl3 show that the
macrocycle sits over the succ station in each case, exactly as
observed for 1. The behavior of these model systems, together
with other control experiments and observations, leads to a series
of arguments showing that the observed CV effects in 1 are a
result of the macrocycle shuttling along the thread.
Figure 11. Cyclic voltammograms at 223 K of (a) 7 and (b) 8. Experiments
carried out on 0.5 mmol L-1 solutions in DMF with tetraethylammonium
tetrafluoroborate (5 × 10-2 mol l-1) as internal standard. Scan rate: 0.5 V
s-1. Working electrode: Pt.
stabilization of the naphthalimide radical anion as a result of
folding in 7-•. However, the ‘gate-opened’ version 8 shows
virtually identical CV behavior (Figure 11b) to rotaxane 1, the
only difference being that the dynamics are over a slightly longer
time scale presumably as a result of the branching in the thread
providing a small steric barrier to macrocycle translation.
(ii) Model rotaxane 6, which is too short and rigid to allow
multiple hydrogen bonds between the macrocycle and the ni
station when the macrocycle is situated over the succinamide
station, also shows very similar CV behavior to 1, although over
a much shorter time scale, reflecting the smaller amplitude of
shuttling that is required.
(iii) Removing the possibility of hydrogen bonding to the
amide protons of the succ station in 5 does not affect the CV
response, suggesting that all the hydrogen bonds to the ni-•
station come from the macrocycle.
(iv) Molecular modeling studies show that folded conforma-
tions of 1-• can only be stabilized by one or two hydrogen bonds
from the macrocycle to the ni station due to geometrical
constraints. Molecular dynamics simulations on such structures
(i) ‘Gate-closed’ rotaxane 7 (where shuttling is prohibited)
shows a CV response (Figure 11a) identical to that of the thread
2 not rotaxane 1, indicating that there is no hydrogen-bond
9
8652 J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003

Hydrogen-Bonded Molecular Shuttles
A R T I C L E S
indicate that the macrocycle would still shuttle along the thread
in order to form multipoint hydrogen bonds to the naphthalimide
unit.29
(v) The large anodic shift for reoxidation of the naphthalimide
in 1-• indicates extensive stabilization of the radical anion (∆E°
) 0.51 V; corresponding to ∆G ) -11.8 kcal mol-1). This is
equivalent to the formation of three or four strong hydrogen
bonds which modeling shows can only arise in geometries where
shuttling has occurred.30
(vi) If only one or two intramolecular hydrogen bonds were
producing the shifts seen in the CV of 1, a similar process could
take place in the thread (2) using the succinamide amide protons
as hydrogen-bond donors. This is clearly not the case as 2
exhibits an identical CV response to that of its N-methylated
analogue (9).
(vii) For rotaxanes 1, 5, and 8 (dynamics in 6 are too fast to
measure), more polar solvents accelerate the speed of the
process, suggesting that breaking of hydrogen bonds is the rate-
determining step, not formation of hydrogen bonds which one
might expect in a folding mechanism.
controlling and observing submolecular motion in hydrogen-
bonded molecular shuttles. In the present benzylic amide
macrocycle-based rotaxane system, the motion resulting from
reduction and reoxidation of the naphthalimide station is
unequivocally shown to be reversible shuttling of the macrocycle
along the thread. This constitutes a molecular shuttle which
exhibits remarkable positional integrity of the macrocycle on
the thread in both oxidation states and rapid dynamics of the
triggered shuttling between them. The simplicity of the structures
involved and the process used to control the motion of the
components augurs well for the development of molecular
devices based on such systems.
Experimental Section
Synthesis. All synthetic procedures are reported in the Supporting
Information.
Electrochemistry. Materials. All chemicals used were reagent
grade. Tetrabutylammonium hexafluorophosphate (TBHF, Fluka) was
used as supporting electrolyte as received. Tetrahydrofuran (LiChrosolv)
was treated according to a procedure described elsewhere.33 For the
electrochemical experiments, the solvent was distilled into the elec-
trochemical cell, prior to use, by a trap-to-trap procedure.
(viii) It is known that alkyl chains fold several orders of
magnitude faster than the dynamics observed in the current
system.31
Instrumentation and Measurements. The one-compartment elec-
trochemical cell was of airtight design, with high-vacuum glass
stopcocks fitted with either Teflon or Kalrez (DuPont) O-rings to
prevent contamination by grease. The connections to the high-vacuum
line and to the Schlenck containing the solvent were made by spherical
joints fitted with Kalrez O-rings. The pressure measured in the
electrochemical cell prior to performing the trap-to-trap distillation of
the solvent was typically 1.0 to 2.0 × 10-5 mbar. The working electrode
consisted either of a 0.6 mm diameter platinum wire (0.15 cm2
approximately) sealed in glass or platinum disk ultramicroelectrodes
(with radii from 5 to 62.5 µm) also sealed in glass. The counter electrode
consisted of a platinum spiral, and the quasi-reference electrode was a
silver spiral. The quasi-reference electrode drift was negligible for the
time required by a single experiment. Both the counter and reference
electrodes were separated from the working electrode by ∼0.5 cm.
Potentials were measured with the ferrocene or decamethylferrocene
standards and are always referred to saturated calomel electrode (SCE).
(ix) Folding a long alkyl chain is entropically unfavorable,
and cyclic hydrogen-bonded conformations forming large rings
of the type produced in the possible folded conformations of
1-• are not often free energy minima.32 In comparison, the
entropy of activation calculated for shuttling in 1-• (∆Sq ) 8.7
( 2 cal K-1 mol-1) is positive; the calculated enthalpy of
activation (∆Hq ) 14.0 ( 0.6 kcal mol-1) is also more consistent
with a shuttling mechanism, where the rate-determining step is
breaking of hydrogen bonds as opposed to forming bonds in a
folding mechanism.
(x) The energy barrier determined for shuttling in the reduced
state (∆G298q ) 11.4 ( 1.2 kcal mol-1) is consistent with known
energy barriers for shuttling between two degenerate stations
in similar molecular shuttles7a and also with the energy barrier
for pirouetting of the macrocycle in model rotaxane 3 (see
above).
Given that folding cannot account for the multipoint hydrogen
bonding that stabilizes the reduced naphthalimide in 1-• nor
can it account for the behavior of 8 mirroring that of 1 while
that of 7 does not, the only process that satisfies the experimental
behavior for these molecules is shuttling. We do not exclude
the likelihood that folding occurs in some of these systems to
some degree (e.g., folded-ni-1-•, Figure 10). Indeed, there is
evidence to support this in a few of the low-energy co-
conformations identified in modeling studies (see above and
ref 29). However, in all these low-energy co-conformationss
folded or extendedsthe macrocycle has shuttled from the succ
station to the ni-• unit.
E
1/2 values correspond to (Ec + Ea)/2 from CV. Ferrocene (decameth-
ylferrocene) was also used as an internal standard for checking the
electrochemical reversibility of a redox couple. The temperature
dependence of the relevant internal standard redox couple potential was
measured with respect to SCE by a nonisothermal arrangement.34
Voltammograms were recorded with an AMEL Model 552 potentiostat
or a custom-made fast potentiostat controlled by either an AMEL Model
568 function generator or an ELCHEMA Model FG-206F. Data
acquisition was performed by a Nicolet Model 3091 digital oscilloscope
interfaced to a PC. Temperature control was accomplished within 0.1
°C with a Lauda thermostat. The minimization of ohmic drop was
achieved through the positive feedback circuit implemented in the
potentiostat. Digital simulation of cyclic voltammetric experiments: The
CV simulations were carried out by the DigiSim 3.0 software by
Bioanalytical Systems Inc. All the fitting parameters (see Supporting
Information) were chosen so as to obtain a visual best fit over a 102-
103-fold range of scan rates.
-
Conclusions
The use of electrochemical stimuli to manipulate intercom-
ponent interactions provides a powerful method for both
Acknowledgment. We thank Andrea Altieri for the prepara-
tion of model rotaxane 4. This work was supported through the
EU TMR Network contract FMRX-CT96-0059, the EPSRC,
the University of Bologna (“Funds for Selected Research
(29) Indeed, recent molecular modelling studies by other groups reinforce this
view: Zheng, X.; Sohlberg, K. J. Phys. Chem. A 2003, 107, 1207-1215.
(30) In complexes with 2,6-diamidopyridines, three hydrogen bonds to a ni unit
cause a stabilization of ∼0.2 V: see refs 14b and 14c.
(31) Zachariasse, K. A.; Mac¸anita, A. L.; Ku¨nhle, W. J. Phys. Chem. B 1999
103, 9356-9365.
(33) Carano, M.; Ceroni, P.; Mottier, L.; Paolucci, F.; Roffia, S. J. Electrochem.
Soc. 1999, 146, 3357-3360.
(34) Yee, E. L.; Cave, R. J.; Guyer, K. L.; Tyma, P. D.; Weaver, M. J. J. Am.
Chem. Soc. 1979, 101, 1131-1137.
(32) Gellman, S. H.; Dado, G. P.; Liang, G.-B.; Adams, B. R. J. Am. Chem.
Soc. 1991, 113, 1164-1173.
9
J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003 8653

A R T I C L E S
Altieri et al.
Topics”), MIUR (PRIN 2002, prot. 2002035735), and CNR
(Agenzia 2000). F.G.G. and D.A.L. are Marie Curie (FMBI-
CT97-2834) and EPSRC Advanced (AF/982324) Research
Fellows.
grams illustrating the effect of scan rate and of temperature on
the electrochemical response of compounds 1, 2, 6, 7, and 8,
procedure for spin polarization transfer by selective inversion
recovery (SPT-SIR) experiments, and Eyring plot for the
variation of shuttling rates with temperature (PDF and CIF).
This material is available free of charge via the Internet at
http://pubs.acs.org.
Supporting Information Available: Full synthetic procedures,
crystallographic information, fitting parameters for the simula-
tion of cyclic voltammetry curves, additional cyclic voltammo-
JA0352552
9
8654 J. AM. CHEM. SOC. VOL. 125, NO. 28, 2003