Induction of motion in a synthetic molecular machine: Effect of tuning the driving force

Article abstract of DOI:10.1002/chem.201204016

Rotaxane molecular shuttles were studied in which a tetralactam macrocyclic ring moves between a succinamide station and a second station in which the structure is varied. Station2 in all cases is an aromatic imide, which is a poor hydrogen-bond acceptor in the neutral form, but a strong one when reduced with one or two electrons. When the charge density on the hydrogen-bond-accepting carbonyl groups in station2 is reduced by changing a naphthalimide into a naphthalene diimide radical anion, the shuttling rate changes only slightly. When station2 is a pyromellitimide radical anion, however, the shuttling rate is significantly reduced. This implies that the shuttling rate is not only determined by the initial unbinding of the ring from the first station, as previously supposed. An alternative reaction mechanism is proposed in which the ring binds to both stations in the transition state. Shuttling via harpooning: Comparison of the shuttling kinetics in a series of imide-based rotaxanes suggest that the mechanism involves a transition state in which the moving ring is hydrogen-bonded to both stations (see figure).

Full text of DOI:10.1002/chem.201204016

DOI: 10.1002/chem.201204016
Induction of Motion in a Synthetic Molecular Machine:
Effect of Tuning the Driving Force
Jacob Baggerman,[a] Natalia Haraszkiewicz,[a] Piet G. Wiering,[a] Giulia Fioravanti,[b]
Massimo Marcaccio,[c] Francesco Paolucci,*[c] Euan R. Kay,[d] David A. Leigh,*[d] and
Albert M. Brouwer*[a]
5566
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5566 – 5577

FULL PAPER
Abstract: Rotaxane molecular shuttles
were studied in which a tetralactam
macrocyclic ring moves between a suc-
cinamide station and a second station
in which the structure is varied. Sta-
tion 2 in all cases is an aromatic imide,
which is a poor hydrogen-bond accept-
or in the neutral form, but a strong one
when reduced with one or two elec-
trons. When the charge density on the
hydrogen-bond-accepting
carbonyl
a pyromellitimide radical anion, how-
ever, the shuttling rate is significantly
reduced. This implies that the shuttling
rate is not only determined by the ini-
tial unbinding of the ring from the first
station, as previously supposed. An al-
ternative reaction mechanism is pro-
posed in which the ring binds to both
stations in the transition state.
groups in station 2 is reduced by chang-
ing a naphthalimide into a naphthalene
diimide radical anion, the shuttling rate
changes only slightly. When station 2 is
Keywords: electrochemistry · mo-
lecular shuttles · photochemistry ·
reaction mechanisms · rotaxanes
Introduction
One of the fascinating aspects of motor proteins is their ca-
pability to generate directional motion and force,[1–7] and to
create molecular systems that behave like such biomolecular
motors is a challenge that has been taken up by an increas-
ing number of researchers.[8–27] Rotaxane-based molecular
shuttles are a prototypical class of synthetic molecular ma-
chines.[19,28–44] They consist of (at least) two components me-
chanically linked together: a macrocyclic ring, which sur-
rounds a linear unit (thread) with bulky groups (stoppers) at
the ends that prevent the macrocycle from slipping off. The
two parts can move relatively easily with respect to each
other because there is no covalent bonding between them. If
the rotaxane has two binding sites (“stations”), then it is
possible to move the macrocycle from one station to the
other by changing the relative binding affinity of one of the
stations. A few years ago, we introduced rotaxane 1, which
showed reversible translational movement on a microsecond
timescale. Rotaxane 1 consists of a tetralactam macrocycle
around a thread with a succinamide station and a naphthali-
mide station connected by an alkyl chain. The co-conforma-
tion in which the macrocycle is bound to the succinamide
group (succ-1) is by far the most stable one according to
[a] Dr. J. Baggerman, Dr. N. Haraszkiewicz, P. G. Wiering,
Prof. Dr. A. M. Brouwer
Van ꢁt Hoff Institute for Molecular Science
University of Amsterdam, P.O. Box 94157
1090 GD Amsterdam (The Netherlands)
E-mail: A.M.Brouwer@uva.nl
[b] Dr. G. Fioravanti
Department of Physical and Chemical Sciences
University of LꢂAquila, Via Vetoio 1, 67100 LꢂAquila (Italy)
[c] Dr. M. Marcaccio, Prof. Dr. F. Paolucci
Dipartimento di Chimica “G.Ciamician”
Universitꢃ di Bologna, Via Selmi 2, 40126 Bologna (Italy)
E-mail: francesco.paolucci@unibo.it
1H NMR[45] and IR spectroscopic studies.[46–48] Reduction of
the naphthalimide group leads to a change in the equilibri-
um position of the macrocycle. It has a higher affinity
[d] Dr. E. R. Kay, Prof. Dr. D. A. Leigh
School of Chemistry, University of Edinburgh
The Kingꢂs Buildings, West Mains Road
Edinburgh, EH9 3JJ (UK)
Cꢀ
toward the naphthalimide radical anion (ni-1 ) than for the
E-mail: david.leigh@ed.ac.uk
succinamide station, and moves over on a microsecond time-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201204016.
scale.[49]
Chem. Eur. J. 2013, 19, 5566 – 5577
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5567

F. Paolucci, D. A. Leigh, A. M. Brouwer et al.
The observed activation energy and strong solvent polari-
ty effect led us to propose a model in which the displace-
ment of the macrocycle occurs in three stages.[48] First, in the
rate-determining step, the hydrogen bonds between the
macrocycle and the succinamide station are broken. This
occurs more slowly in less-polar solvents. Next, the macrocy-
cle has to move along the thread from the succinamide to
the naphthalimide. Molecular dynamics simulations[50] sug-
gest that this can occur on a subnanosecond timescale. In
the final stage, the macrocycle binds with the naphthalimide
radical anion. The process is reversible, and the position of
the ring on the thread is controlled by thermodynamic fea-
tures. Although the model is attractively simple and in
agreement with the available data, its validity needs to be
tested. The movement in the second stage can be a random
walk along the alkyl chain, which proceeds until the macro-
cycle is close enough to form a hydrogen bond to one of the
stations, whereupon it is “pulled in” irreversibly on the
nanosecond timescale. Recently, we showed that the de-
pendence of the shuttling rate on the spacer length could be
described very well by a biased random walk model.[48] The
origin of the bias, however, was not identified. Therefore, al-
ternative mechanisms need to be considered. If the binding
of the macrocycle to the accepting station requires a sub-
stantial structural reorganization, this might lead to an
energy barrier, which reduces the efficiency of trapping. In
experimental observations this will show up as a reduced
shuttling rate. Yet another possibility is that the macrocycle
stays bound to the initial stations, perhaps with only one hy-
drogen bond, and is “handed over” from one station to the
other. In this case, the whole shuttling process is more like
an elementary one-step chemical reaction. The rate can be
expected to depend on the geometry and charge distribution
of the accepting station, because that binds to the macrocy-
cle in the transition state.
lower. This implies that the previous description of the shut-
tling mechanism needs to be modified.
Results and Discussion
Cyclic voltammetry: As previously demonstrated for rotax-
ane 1, cyclic voltammetry (CV) experiments can give useful
information about the kinetics and thermodynamics of the
shuttling process.[45] Solutions of the naphthalene diimide
thread 3 and rotaxane 2 (0.5 mm in THF) displayed the CV
curves shown in Figure 1. Both NDI-centered reduction
processes are shifted in the case of rotaxane 2 towards less
negative potentials with respect to thread 3, as in the case of
NI-rotaxane 1 and the corresponding NI thread. Interesting-
ly, the shift is much larger in the case of the second reduc-
tion. For rotaxane 2 and thread 3, the first reduction poten-
tials are located at Er1e=d2 =ꢀ0.48 and ꢀ0.52 V (versus NHE),
respectively, which corresponds to the conversion of the
To gain more insight into these questions in the present
work, the structure of the original rotaxane 1 was modified.
The naphthalimide (NI) was replaced by a naphthalene dii-
mide group (NDI rotaxane 2), or by a pyromellitimide unit
(PMI rotaxane 4). In the radical anion of the NDI unit, the
charge will be more delocalized than in the naphthalimide
radical anion. This leads to a weaker binding affinity of the
macrocycle to the reduced NDI. In the radical anion of PMI
the aromatic unit is smaller, so a larger density of the
charge on the carbonyl oxygen atoms can be expected rela-
tive to NDI. The relative orientation of the C=O groups of
PMI is, however, different, which might have an effect on
the binding energy and binding geometry.
If the movement is like a random walk, and if trapping
occurs efficiently as soon as a contact between the macrocy-
cle and the accepting station is established, we expect that
the shuttling rate will be independent of the nature of the
accepting station. We studied the dynamic behavior of these
new molecular shuttles with electrochemical and photo-
chemical techniques. It turns out that the rates of shuttling
for the naphthalene imide and diimide systems 1 and 2 are
similar, but for the pyromellitimide rotaxane 4 it is distinctly
Figure 1. Cyclic voltammograms of rotaxane 2 and thread 3 (0.5 mm in
THF, supporting electrolyte: 0.05m tetrabutylammonium tetrafluorobo-
rate). Working electrode: platinum disc (125 mm diameter). a) v=1 Vsꢀ1
and T=258C; b) v=1 Vsꢀ1, T=ꢀ558C; c) rotaxane 2, v=100 Vsꢀ1, T=
258C. Reduction potential relative to NHE.
5568
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5566 – 5577

Synthetic Molecular Machine
FULL PAPER
naphthalene diimide to its radical anion (Figure 1a; scan
rate (n)=1 Vsꢀ1). The second reduction potentials are locat-
ed at Er1e=d2 =ꢀ0.83 and ꢀ1.05 V (versus NHE), for rotaxane
2 and thread 3, respectively. For thread 3 as well as rotaxane
2, both reduction processes are essentially reversible (DEp =
80 mV).
binding site most of the time. This behavior can be descri-
bed by the electrochemical reduction scheme for a redox-
switched binding process (Scheme 1). In this scheme, the re-
versible translational isomerism equilibria for each oxidation
state are connected through the electron-transfer steps.[51,52]
A computer simulation of all CV results was performed to
extract quantitatively the underlying kinetics of the phenom-
ena described above. From these simulations the data com-
piled in Scheme 1 were obtained as a “best fit”.
The difference in Erped (I) of the NDI thread and NDI ro-
taxane is not very large (40 mV), which is similar to that be-
tween the mono-imide-thread and rotaxane 1 studied previ-
ously, whereas for the second reduction the difference be-
tween thread 3 and rotaxane 2 is much bigger (230 mV).
The observed shifts are a consequence of the shuttling of
the macrocycle from the succinamide station (succ-2 co-con-
former), at which it is preferentially located in the neutral
state, onto the (reduced) NDI station (ndi-2 co-conformer)
and reflect the stabilization of the mono- and dianion of the
NDI station as a consequence of their increasing ability to
form hydrogen bonds with the macrocycle.
In all the systems we have investigated so far,[45,53] and in
which electron-transfer-induced shuttling was unequivocally
observed, the occurrence of shuttling brought about the pos-
itive shift of the reduction peak(s) as the result of stabiliza-
tion of imide-based radical anions by interaction with the
macrocycle.
The pyromellitic thread 5 shows two narrowly spaced re-
versible cathodic peaks at ꢀ0.73 and ꢀ1.38 V (versus NHE)
that correspond to the consecutive one-electron reductions
of the pyromellitimide moiety as previously described.[54]
For rotaxane 4 at low scan rates and 258C, the first reduc-
tion potential is observed at ꢀ0.72 V (versus NHE), and the
second reduction potential is located at ꢀ1.26 V (versus
NHE). For thread 5 as well as rotaxane 4, both reduction
processes are essentially reversible (DEp =80 mV). The re-
duction potentials of the rotaxanes and threads studied in
the present work are listed in Table 1.
As the scan rate is increased (Figure 1c; n=100 Vsꢀ1), a
small peak is observed (indicated by the arrow in Figure 1c)
following the second reduction peak at a potential close to
that of the second reduction of 3. At low temperature, T=
ꢀ558C, n=1 Vsꢀ1, a broadening of the second peak is ob-
served (Figure 1b; n=1 Vsꢀ1). The small peak observed in
some curves at high scan rate and the broadening at low
temperature most likely corresponds to the reduction of
Cꢀ
NDI to NDI2ꢀ, whereas the macrocycle still resides on the
succ station.
Table 1. Reduction potentials [V] of threads 3 and 5 and rotaxanes 2 and
4 relative to NHE.
As shown before, molecular shuttle 1 in the neutral state
adopts the succ-1 co-conformation. 1H NMR spectroscopic
evidence shows that the same is true for 2 and 4. In all
cases, the chemical shift of the succinamide methylene pro-
tons is dꢁ1.3 ppm upfield in the rotaxane relative to the
corresponding thread. The presence of two succinamide sta-
tions in 4 results in signals at both d=2.3 and 1.0 ppm,
which correspond to the “free” and “occupied” succinamide
stations, respectively. Rapid movement of the ring between
the succinamide units on either side of the pyromellitimide
apparently does not occur. After reduction, the macrocycle
shuttles from the succinamide station to the imide unit due
to the increased hydrogen-bonding ability of the imides in
Thread 3
Rotaxane 2
Thread 5
Rotaxane 4
red
1=2
red
E
E
(I)
(II)
ꢀ0.52
ꢀ1.05
ꢀ0.48
ꢀ0.83
ꢀ0.73
ꢀ1.38
ꢀ0.72
ꢀ1.26
1=2
In summary, the pyromellitic system displays the “typical”
electron-transfer-induced shuttling behavior that was ob-
served in all the previous systems. Notice that the apparent
E1/2 value of CV peaks in the case of square mechanisms
such as that shown in Figure 2 depends both on the E1/2
values of redox processes that involve the “pure” succ and
ndi co-conformers, respectively, and on the relative values
of equilibrium constants that involve the two co-conformers
in the oxidized and reduced states, respectively. The negligi-
ble shift of the first reduction peak in the rotaxane with re-
spect to the thread does not necessarily imply that shuttling
does not occur at the level of first reduction, rather it is indi-
cative that, also after reduction, the equilibrium is in that
case only slightly shifted toward the ndi co-conformer. In
line with this explanation, the much larger positive shift
(120 mV) observed at the level of the second reduction
peak of rotaxane 4 with respect to thread 5 is associated
with the much larger displacement of equilibrium between
the two co-conformers upon the introduction of a second
electron in the pyromellitic unit.
Cꢀ
the reduced state. In the case of 1 , the ni co-conformer is
strongly favored; but in the case of 2, a small proportion of
Cꢀ
the succ-2 co-conformation seems to be present next to the
Cꢀ
dominant ndi-2 co-conformer. A simulation of the CV
Cꢀ
Cꢀ
data (see below) indicates that the ratio succ-2 /ndi-2 is
approximately 1:3. In the doubly reduced state the co-con-
formational equilibrium strongly favors the ndi-22ꢀ form.
Hydrogen bonding to the macrocycle amide protons sta-
bilizes the increased electron density on the naphthalene dii-
mide, so that more positive potentials must be reached
before the oxidation of the dianion occurs, which results in
the larger shift observed in the second reoxidation peak.
After reoxidation of the dianion, the 1:5 succ-2 /ndi-2 co-
conformational equilibrium is re-established. Further reoxi-
dation of the radical anion regenerates the starting condi-
tions, and the macrocycle moves again to occupy the succ
Cꢀ
Cꢀ
Spectroelectrochemistry: Further evidence for shuttling in
rotaxanes 2 and 4 was found with UV-visible spectroelectro-
Chem. Eur. J. 2013, 19, 5566 – 5577
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5569

F. Paolucci, D. A. Leigh, A. M. Brouwer et al.
Scheme 1. Electrochemical reduction cycle of rotaxane 2 and equilibrium constants obtained from simulation of the CV curves.
chemistry. In the case of naphthalimide rotaxane 1, several
of the radical anion absorption bands were found to occur
at shorter wavelengths than those of the corresponding
thread.[49] These blueshifts occur because the hydrogen
bonds between the macrocycle and the carbonyls of the
imide stabilize the ground state of the anion more than its
excited state. The spectra of naphthalene diimide thread 3
and rotaxane 2 and their radical anions and dianions, depict-
ed in Figure 3, show similar behavior. The spectra of the
neutral thread 3 and rotaxane 2 are identical, which is in
agreement with binding of the macrocycle to the succina-
mide in the neutral form. Upon reduction, the typical spec-
tra for the naphthalene diimide radical anion and dianion
are observed.[55] Although the spectra of the reduced forms
of 2 and 3 are similar, some clear peak shifts can be seen.
The absorption maxima and molar absorption coefficients
are listed in Table 2.
Figure 2. Cyclic voltammograms of rotaxane 4 and thread 5 (0.5 mm in
THF, supporting electrolyte: 0.05m tetrabutylammonium tetrafluorobo-
rate). Working electrode: platinum disc (125 mm diameter); v=1 Vsꢀ1
and T=258C. Reduction potentials relative to NHE.
The low-energy absorption maxima of the radical anion
and the dianion of rotaxane 2 are shifted to shorter wave-
5570
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5566 – 5577

Synthetic Molecular Machine
FULL PAPER
Figure 4. Normalized absorption spectra of a) the pyromellitic radical
anion and b) dianion of the thread 5 (dashed line) and rotaxane 4 (solid
line) in THF.
Figure 3. The absorption spectra of a) the radical anion and b) the dia-
nion of naphthalene diimide rotaxane 2 (solid line), and thread 3 (dashed
line) in THF. The spectra of the neutral molecules (identical for 2 and 3)
are shown in gray.
The molar absorption coefficient of the similar N,N’-bis(2,5-
di-tert-butylphenyl) pyromellitic imide radical anion is 4.17ꢄ
104 mꢀ1 cmꢀ1 according to Rak et al.[56] On the basis of this
value, the molar absorption coefficient of the dianion is esti-
mated to be approximately 1.1ꢄ105 mꢀ1 cmꢀ1. This is slightly
lower than the value (1.4ꢄ105 mꢀ1 cmꢀ1) measured by Gosz-
tola et al.[55]
Table 2. Absorption maxima (wavelength l [nm] and molar absorption
coefficients e [103 mꢀ1 cmꢀ1]) of the radical anions and dianions of thread
3 and rotaxane 2.
l1 (e1)
l2 (e2)
l3 (e3)
l4 (e4)
l5 (e5)
Cꢀ
Cꢀ
3
2
474 (30)
479 (27)
399 (16)
402 (19)
609 (7.4)
612 (8.0)
422 (24)
422 (26)
684 (2.5)
674 (2.5)
516 (4.8)
509 (2.5)
763 (5.0)
745 (4.2)
557 (6.5)
548 (6.8)
32ꢀ
22ꢀ
605 (11)
595 (12)
The spectrum of the radical anion of rotaxane 4 is only
slightly blueshifted at the peak, but broadened and more in-
tense in the short-wavelength shoulder relative to that of
thread 5. The change in shape indicates that there is an in-
teraction between the macrocyclic ring and the chromo-
phore, which means that in the radical anion there must be
lengths than those of thread 3. These shifts indicate the
translocation of the macrocycle from the succinamide to the
naphthalene diimide stations in the reduced form. The ab-
sorption bands at 674 and 745 nm of the rotaxane radical
anion are somewhat broader than those of the thread, prob-
ably because of the presence of a small fraction of the succ-
Cꢀ
Cꢀ
a significant population of the pmi-4 co-conformer. Molec-
ular computation (see below) indicates that the macrocyclic
Cꢀ
Cꢀ
ring binds more strongly to the PMI than to NDI . Fur-
ther independent evidence for essentially complete translo-
cation of the macrocycle from infrared spectroelectrochem-
istry will be published elsewhere.[58]
2
co-conformer, which is in agreement with the CV results.
The absorption spectra of the radical anions and dianions
of the pyromellitimide thread 5 and rotaxane 4 in tetrahy-
drofuran (THF) are depicted in Figure 4. The absorption
maxima are listed in Table 3. The spectra are similar to pre-
viously reported spectra of anions of pyromellitic dii-
mides.[55–57] The radical anion spectra of thread 5 and rotax-
ane 4 are very similar, with a strong absorption at 716 nm.
In the dianions, the absorption maximum of the rotaxane
is broadened and distinctly redshifted by 17 nm. The redshift
is in contrast with the blueshifts observed in the cases of the
naphthalene mono and diimide radical anions and NDI dia-
nion, but the spectral changes in combination with the result
for the radical anion leave no doubt that the predominant
co-conformer is pmi-42ꢀ.
Table 3. Absorption maxima of the pyromellitic radical anion and dia-
nion of thread 5 and rotaxane 4.
Note that although the pyromellitimide rotaxane 4 has
two equivalent succinamide stations, only one of them is in-
Radical anion [nm]
Dianion [nm]
1
volved in the shuttling process. The H NMR spectra (see
thread 5
rotaxane 4
717
716
553
570
the Supporting Information) show that the macrocyclic ring
is localized on one of the succinamide stations, and shuttling
Chem. Eur. J. 2013, 19, 5566 – 5577
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5571

F. Paolucci, D. A. Leigh, A. M. Brouwer et al.
between the two equivalent sites is not detectable on the
NMR spectroscopic timescale.
Photoreduction: In the case of the naphthalimide rotaxane
1, photoinduced electron transfer was used to generate the
radical anion, and the shift in the position of the absorp-
tion was used to monitor the shuttling process in time with
transient-absorption spectroscopy on a microsecond time-
scale. Charge recombination restores the neutral ground-
state form of the system, so that no net chemical change
occurs and repeated excitation of the sample can be used.
Unfortunately, direct excitation of naphthalene diimides re-
sulted in rapid photodegradation.[61–63] Also, in the presence
Computation: To test our intuitive expectations of the rela-
tive binding strengths of the macrocyclic ring to the differ-
ent stations, some ab-initio calculations were performed.
Owing to the size and complexity of the systems we had to
restrict ourselves to the B3LYP/6-31G(d) level of calculation
on selected model systems. Calculations were performed for
isolated molecules and for molecules solvated in acetonitrile
with the polarizable continuum model (PCM).[59,60] The geo-
metries of the isolated molecules were used.
Two conformations were considered for the macrocyclic
ring: a chair (C2h) and a boat (C2v symmetry). For the free
macrocycle, the latter turned out to be lower in energy by
1 kcalmolꢀ1. The succinamide thread unit was modeled with
the full diphenylethyl stopper on one side and a methyl
group mimicking the alkane linker. For the isolated mole-
cules, the thread model was most stable in a conformation
with an intramolecular hydrogen bond, but when solvated,
the extended conformation was preferred. The imides were
in all cases modeled as the NH derivatives. To estimate the
relative interaction energies, a complex of the succinamide
model with the chair macrocycle was compared with the
complexes of the boat macrocycle with the imides (C2v sym-
metry). Model structures are shown in Figure S3 in the Sup-
porting Information. In Table 4 we list the differences in the
of 1,4-diazaACTHNUGTRNEUNG[2.2.2]bicyclooctane (DABCO) as an electron
donor, reversible formation of radical anions could not be
achieved. To avoid the reactive excited state of the naphtha-
lene diimide, excitation of 1,4-dimethoxybenzene (DMB),
an electron donor, in the presence of naphthalene diimide
was used as an alternative route to the radical anion. Benzo-
nitrile was used as the solvent. To investigate the mecha-
nism, we used naphthalene diimide model compound 6. Ex-
citation of DMB with a 310 nm laser pulse in the presence
of naphthalene diimide 6 in benzonitrile gave a transient-ab-
sorption spectrum (Figure 5) that resembles the naphthalene
diimide radical anion spectrum well (Figure 3).
Table 4. Computed binding energies (B3LYP/6-31G(d) [kcalmolꢀ1]) of
the tetralactam macrocycle to imides in neutral, radical anion, and dia-
nion states, relative to binding to a succinamide model. Structures are
shown in the Supporting Information.
NI
NI
(CH3CN)[a]
NDI
NDI
(CH3CN)[a]
PMI
PMI
(CH3CN)[a]
Figure 5. Transient-absorption spectra of a 100 mm solution of 6 in benzo-
nitrile after excitation of DMB (20 mm) with 310 nm (the highest DA was
at 2 ms after the laser pulse, subsequent spectra at later times, with an in-
crement of 2 ms).
G
G
ACHTUNGTRENNUNG
neutral
anion
dianion
4.4
1.4
7.0
ꢀ17.1
2.4
ꢀ3.8
8.0
3.7
ꢀ26.7 ꢀ9.2
ꢀ18.9 ꢀ6.3
ꢀ55.1 ꢀ12.0
ꢀ65.1
[a] Polarizable continuum model for acetonitrile, with gas-phase geome-
tries.
The driving force for electron transfer in polar solvents
can be calculated from Equation (1), in which DG is the
free energy change, eEox and eEred are the relative energies
of the oxidized electron donor and reduced electron accept-
or, and E00 is the electronic excitation energy.
binding energies of the macrocyclic ring with the imide
model and with the thread model. For the neutral molecules
these values are all positive, which means that the ring pre-
fers to bind to the succinamide; for the radical anions and
dianions they are negative. This means that overall the re-
sults are in agreement with our experimental findings. More
details are given in the Supporting Information.
DG ¼ eEoxꢀeEredꢀE00
ð1Þ
The excitation energy of DMB in the S1 state is 4.0 eV,
and its oxidation potential is 1.58 V (versus NHE).[64] The
driving force for electron transfer between DMB and naph-
thalene diimides (Ered =ꢀ0.24 V versus NHE)[55] can be esti-
mated as ꢀ2.2 eV. Electron transfer between DMB and ben-
zonitrile (Ered =ꢀ1.98 V versus NHE)[65] has a driving force
of ꢀ0.4 eV. The formation of NDI radical anions can take
place directly by reaction of the diimide with excited DMB,
but since the concentration of the diimide is only approxi-
mately 10ꢀ4 m, this is kinetically unlikely during the nanosec-
Of the three radical anions, the naphthalene diimide is
computed to have the least favorable interaction with the
macrocycle. This agrees qualitatively with our observation
Cꢀ
Cꢀ
that the equilibrium constant of the ndi-2 /succ-2 equili-
brium is only about 5. The pyromellitimide anion is predict-
ed to bind more strongly to the macrocycle. On the basis of
this result and those of the IR study,[58] it is reasonable to
Cꢀ
assume that in the case of 4 the co-conformational equili-
Cꢀ
brium favors the pmi-4 co-conformer.
5572
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5566 – 5577

Synthetic Molecular Machine
FULL PAPER
ond lifetime of the DMB excited singlet state. Alternatively,
the benzonitrile radical anion could be formed first, fol-
lowed by electron transfer to NDI. Electron transfer be-
tween DMB in the S1 state and acceptors with reduction po-
tentials in the same range as benzonitrile has been observed
before.[66] To distinguish between the two pathways, the fluo-
rescence and transient absorption of DMB in acetonitrile
and benzonitrile were studied.
Time-resolved fluor escence of DMB (lex =310 nm) in
acetonitrile and benzonitrile was measured by using a streak
camera. The fluorescence spectra (Figure 6) (lmax =328 nm)
and lifetime of 2.7 ns observed in degassed acetonitrile are
Figure 7. Transient-absorption spectra of DMB in a) acetonitrile and
b) benzonitrile; lex =315 nm. Time increments between successive spectra
1 ms.
Figure 6. Fluorescence spectra of DMB in acetonitrile (solid line) and
benzonitrile (dashed line).
in good agreement with literature data for DMB.[67] In ben-
zonitrile the fluorescence has a maximum around 450 nm
and a biexponential decay with lifetimes of 3.5 and 9.6 ns.
Between 300 and 400 nm, there is weak shoulder with a life-
time of 2.3 ns. The changes in the fluorescence spectrum and
the longer lifetimes indicate the formation of an exciplex be-
tween DMB and benzonitrile.
The transient-absorption spectra of DMB in degassed ace-
tonitrile and benzonitrile are depicted in Figure 7. The spec-
tra in acetonitrile have a maximum around 425 nm, decay in
10 ms, and correspond with the known triplet–triplet absorp-
tion spectra of dimethoxybenzene.[68] The assignment to the
triplet is supported by the fact that in the presence of air the
decay becomes much faster (100 ns) due to quenching by
oxygen. The spectra observed in benzonitrile resemble more
the summed spectra of the radical cation of DMB[68–70] and
the radical anion of benzonitrile.[71,72] The decay of the tran-
sient in benzonitrile is not sensitive to oxygen, thus indicat-
ing that the observed band is mainly due to the DMB radi-
cal cation. Moreover, the decay follows second-order kinet-
ics, as expected for separated radical ion pairs in solution.
The observed exciplex fluorescence in benzonitrile and the
differences in transient-absorption dynamics strongly sug-
gest that electron transfer between DMB and the solvent
benzonitrile occurs, followed by reduction of the diimide by
the benzonitrile radical anion.
Figure 8. Contour plot of the transient-absorption spectra in benzonitrile.
The dotted lines show the shift of the maxima.
spectra of a solution of rotaxane 2 and dimethoxybenzene in
benzonitrile after excitation with a nanosecond laser pulse
at 315 nm. At short times, the radical anion absorption
Cꢀ
bands are those of the succ-2 co-conformer, very similar to
Cꢀ
those of thread 3 observed in the spectroelectrochemical
experiments. On a timescale of tens of microseconds, the
Cꢀ
spectrum evolves into the blueshifted one of the ndi-2 co-
conformer.
The same experiment with thread 3 instead of rotaxane 2
did not show such a shift. This behavior corresponds to that
of the previously described naphthalimide shuttle 1.
To obtain the shuttling rate constant, the spectral evolu-
tion was quantitatively modeled. The two absorption bands
in the spectra of the radical anions of the thread and rotax-
ane between 650 and 800 nm were fitted as a linear combi-
nation of two Gaussian functions (see Figure S2 and
Table S1 in the Supporting Information). The similar band
Shuttling: When the radical anion of rotaxane 2 is created
by laser-induced electron transfer, the spectral evolution can
be monitored in time with transient-absorption spectrosco-
py. Figure 8 shows an image of the transient-absorption
Chem. Eur. J. 2013, 19, 5566 – 5577
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5573

F. Paolucci, D. A. Leigh, A. M. Brouwer et al.
shapes found for the initial and final states suggest that the
population of the succ co-conformer in the radical anion in
benzonitrile is smaller than in THF, in which the electro-
chemical experiments were carried out (see Figure 3). Then,
the obtained transient-absorption spectra at different times
were fitted as a superposition of those two functions. From
the obtained amplitude for each spectrum the relative con-
tributions were calculated (Figure 9). The change of the rel-
ative contribution of each spectrum corresponds to the
change in population of each species.
Figure 10. Transient-absorption spectra of 122 mm PMI thread 5 in benzo-
nitrile after excitation of DMB (6 mm) with a 315 nm laser pulse. The
first spectrum is 6 ms after the pulse. The consecutive spectra (black to
gray) were measured with a time increment of 2 ms.
The radical anion transient-absorption bands of the PMI
rotaxane 4 were analyzed in the same way as for rotaxane 2.
The analysis gave a time constant of 44 ms (Figure 11). Al-
though the accuracy is not deemed very high because of the
relatively small spectral changes, the shuttling process in 4ꢀ
is clearly slower than that in 2ꢀ and 1ꢀ.
Cꢀ
Figure 9. Change of the fractions of the succ-2 co-conformer (circles)
Cꢀ
and ndi-2 co-conformer (squares) with time. The curves are fits with a
biexponential model.
Cꢀ
The change in population of the two co-conformers of 2
shows an exponential time profile, with a time constant of
13 ms. In the case of establishment of an equilibrium follow-
ing a perturbation, the observed rate constant is the sum of
the forward and backward rate constants, but if we neglect
the equilibrium population of the succ co-conformer in the
radical anion state, the obtained 8ꢄ105 sꢀ1 is just the for-
ward shuttling rate constant. At shorter times there is a rela-
tively slow rise that was modeled with an additional expo-
nential. This is probably caused by the fact that it takes a
few microseconds to form the NDI radical anion. So in the
first five microseconds there is the formation of the radical
anion and also shuttling of the already formed radical
anions.
The shuttling time constant of rotaxane 1 in benzonitrile
was measured in the same way as before,[49] and found to be
10 ms.
Photoreduction of the pyromellitic diimide thread 5 was
attempted by using the same procedure used for reduction
of the naphthalene diimide thread and rotaxane. Excitation
at 315 nm of a 20 mm dimethoxybenzene solution with
10ꢀ4 m PMI thread 5 gave the transient-absorption spectra
shown in Figure 10.
Figure 11. The relative amplitude of the initial (circles) and final
(squares) spectrum of the radical anions of the PMI [2]rotaxane 4 in
time.
In our initial paper on the shuttling in naphthalimide ro-
taxane 1, we observed that the rate of the process increased
strongly as solvent polarity increased, and we determined a
free energy of activation of 10 kcalmolꢀ1, which we consid-
ered to indicate that in the transition state of the reaction,
hydrogen bonds to the initial binding station were broken.[49]
The shuttling would then continue with a diffusive random
walk of the ring along the thread, thus leading to its trap-
ping by the accepting naphthalimide anion station. Some-
what puzzling was the negative entropy of activation, but
this can be attributed to the probability factor (P<1) in-
volved in the second stage of the reaction. More recently,
the dependence of the rate of the length of the spacer be-
tween the binding sites was found to be in agreement with a
biased random walk.[48] In this model, the accepting station
has a passive role, and the strength of its interactions with
the macrocycle has no influence on the reaction rate. The
present findings with pyromellitimide rotaxane 4, however,
indicate that the rate does depend on the nature of the ac-
Two main peaks are present: one at 723 nm with a should-
er at 659 nm that corresponds to the radical anion spectrum
shown in Figure 4. The other band has a maximum at
437 nm. This is due to the dimethoxybenzene radical cation
spectrum (compare with Figure 7). This shows that it is pos-
sible to create the pyromellitic radical anion photochemical-
ly by excitation of the dimethoxybenzene electron donor.
5574
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5566 – 5577

Synthetic Molecular Machine
FULL PAPER
cepting station. Because the hydrocarbon linkers between
the stations are flexible, an alternative mechanism might be
considered in which the shuttling proceeds in one kinetic
step through a transition state in which the macrocyclic ring
is bound to both stations, while the flexible spacer is folded.
A sketch of one of many possible transition structures is
shown in Figure 12.
the ring slides along the thread in a diffusive process.[75–77]
With very few exceptions, the threads are highly flexible,[78]
and therefore we suggest that in many cases harpooning-
type mechanisms should be considered as alternatives. In
one case, it was reported that electrostatically induced fold-
ing actually prevented shuttling.[79] In highly viscous or oth-
erwise constrained environments, such as polymer gels,
large-amplitude motions are restricted.[80] In such a situa-
tion, the different steric requirements of diffusive and har-
pooning mechanisms might be important. For example, if
the stoppers are much larger than the moving ring, the con-
traction of the thread that is required for harpooning might
be hindered more than the diffusive motion of the ring.
Conclusion
New rotaxanes were synthesized that incorporated different
diimides as redox-active stations and succinamide as a
common template for ring formation and the preferred
1
binding site in the neutral molecules. H NMR spectroscopy
of naphthalene diimide rotaxane 2 and pyromellitic diimide
rotaxane 4 demonstrated that the tetralactam macrocycle
binds exclusively to the succinamide station in the neutral
molecules. Thermodynamic data derived from cyclic voltam-
metry of 2 showed that the macrocycle binds predominantly
to the diimide in the radical anion state, and exclusively to
the diimide in the dianion. Spectral changes upon shuttling
were observed by UV-visible spectroelectrochemistry and in
time-resolved experiments in which reduction was achieved
by laser-induced photochemistry. The rate of the shuttling
process at room temperature in benzonitrile was found to
be 8ꢄ105 sꢀ1, which is only slightly slower than that in rotax-
ane 1, for which it is 1ꢄ106 sꢀ1.
The smaller strength of binding in the naphthalene dii-
mide does not have a significant effect on the rate, which is
consistent with our original hypothesis that the rate-limiting
step in the shuttling process is the release of the macrocycle
from its initial succinamide binding station. The results with
the pyromellitimide rotaxane 4, however, shed doubt on this
mechanism, because in this case the rate is approximately
three times smaller. As an alternative to the two-step mech-
anism, a single-step harpooning process should be consid-
ered.
Figure 12. Examples of possible “harpooning” transition structures for
shuttling in 4 and 1 .
Cꢀ
Cꢀ
Such a “harpooning” mechanism is qualitatively compati-
ble with the observed distance dependence, because forma-
tion of the transition-state structure is statistically less likely
as the length of the spacer is increased. Conformations in
which the tetralactam ring is hydrogen-bonded to two sta-
tions have been observed in crystal structures, and have
even been proposed as preferred energy minima in particu-
lar in cases in which the binding to the stations is relatively
weak.[73] Infrared absorption[46] and NMR spectroscopic
studies, and X-ray crystallography[73] have also provided evi-
dence for the existence of “bridged” conformations in hy-
drogen-bond-based rotaxanes.[74] Although in this harpoon-
ing mechanism the ring is always hydrogen-bonded to a sta-
tion, the number of hydrogen bonds in the transition state is
smaller than in the initial state, so our initial explanation of
the solvent effect still holds. Finally, the negative entropies
of activation are readily explained by this mechanism be-
cause in the transition state many degrees of freedom are
more or less frozen.
Whereas the available experimental data show that a pas-
sive mechanism in which the acceptor station has no role in
determining the shuttling rate is not correct for the com-
pounds studied in the present work, the generality of this
observation remains to be investigated. Rotaxanes of the
type discussed in the present paper can be designed in such
a way that harpooning can be excluded by incorporating a
sufficiently long rigid thread unit. Unfortunately, such sys-
tems are difficult to obtain experimentally because of solu-
bility problems.[74]
Experimental Section
Experimental details are given in the Supporting Information.
Acknowledgements
This research was financially supported by the Netherlands Organization
for the Advancement of Research (NWO). Berta Garcia-Landa made a
crucial contribution to the frontispiece illustration.
For the other classes of rotaxane molecular shuttles de-
scribed in the literature, it seems to be tacitly assumed that
Chem. Eur. J. 2013, 19, 5566 – 5577
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5575

F. Paolucci, D. A. Leigh, A. M. Brouwer et al.
[37] P. Lussis, T. Svaldo-Lanero, A. Bertocco, C. A. Fustin, D. A. Leigh,
A. S. Duwez, Nat. Nanotechnol. 2011, 6, 553–557.
[38] D. D. Gꢈnbas¸, L. Zalewski, A. M. Brouwer, Chem. Commun. 2011,
47, 4977–4979.
[1] Molecular Motors (Ed.: M. Schliwa), Wiley VCH, Weinheim, 2003.
[2] V. Bormuth, V. Varga, J. Howard, E. Schaffer, Science 2009, 325,
870–873.
[3] A. Yildiz, M. Tomishige, R. D. Vale, P. R. Selvin, Science 2004, 303,
676–678.
[39] T. Ogoshi, D. Yamafuji, T. Aoki, T. Yamagishi, J. Org. Chem. 2011,
76, 9497–9503.
[4] R. D. Vale, R. A. Milligan, Science 2000, 288, 88–95.
[5] Z. Zhang, D. Thirumalai, Structure 2012, 20, 628–640.
[6] W. B. Redwine, R. Hernꢅndez-Lꢆpez, S. Zou, J. Huang, S. L. Reck-
Peterson, A. E. Leschziner, Science 2012, 337, 1532–1536.
[7] N. D. Derr, B. S. Goodman, R. Jungmann, A. E. Leschziner, W. M.
Shih, S. L. Reck-Peterson, Science 2012, 338, 662–665.
[8] E. R. Kay, D. A. Leigh, F. Zerbetto, Angew. Chem. 2007, 119, 72–
196; Angew. Chem. Int. Ed. 2007, 46, 72–191.
[9] T. Kudernac, N. Ruangsupapichat, M. Parschau, B. Macia, N. Katso-
nis, S. R. Harutyunyan, K.-H. Ernst, B. L. Feringa, Nature 2011, 479,
208–211.
[10] G. T. Carroll, M. M. Pollard, R. van Delden, B. L. Feringa, Chem.
Sci. 2010, 1, 97–101.
[40] T. Ogoshi, D. Yamafuji, T. Aoki, K. Kitajima, T.-A. Yamagishi, Y.
Hayashi, S. Kawauchi, Chem. Eur. J. 2012, 18, 7493–7500.
[41] T. Ogoshi, D. Yamafuji, T. Aoki, T.-A. Yamagishi, Chem. Commun.
2012, 48, 6842–6844.
[42] S. W. Hansen, P. C. Stein, A. Sørensen, A. I. Share, E. H. Witlicki, J.
Kongsted, A. H. Flood, J. O. Jeppesen, J. Am. Chem. Soc. 2012, 134,
3857–3863.
[43] A. Joosten, Y. Trolez, J.-P. Collin, V. Heitz, J.-P. Sauvage, J. Am.
Chem. Soc. 2012, 134, 1802–1809.
[44] H. Li, A. C. Fahrenbach, A. Coskun, Z. Zhu, G. Barin, Y.-L. Zhao,
Y. Y. Botros, J.-P. Sauvage, J. F. Stoddart, Angew. Chem. 2011, 123,
6914–6920; Angew. Chem. Int. Ed. 2011, 50, 6782–6788.
[45] A. Altieri, F. G. Gatti, E. R. Kay, D. A. Leigh, D. Martel, F. Paoluc-
ci, A. M. Z. Slawin, J. K. Y. Wong, J. Am. Chem. Soc. 2003, 125,
8644–8654.
[11] B. L. Feringa, J. Org. Chem. 2007, 72, 6635–6652.
[12] W. R. Browne, B. L. Feringa, Nat. Nanotechnol. 2006, 1, 25–35.
[13] Y.-C. You, M.-C. Tzeng, C.-C. Lai, S.-H. Chiu, Org. Lett. 2012, 14,
1046–1049.
[46] D. C. Jagesar, F. Hartl, W. J. Buma, A. M. Brouwer, Chem. Eur. J.
2008, 14, 1935–1946.
[14] C. J. Chuang, W. S. Li, C. C. Lai, Y. H. Liu, S. M. Peng, I. Chao,
S. H. Chiu, Org. Lett. 2009, 11, 385–388.
[15] K. Hirose, Y. Shiba, K. Ishibashi, Y. Doi, Y. Tobe, Chem. Eur. J.
2008, 14, 3427–3433.
[16] T. Umehara, H. Kawai, K. Fujiwara, T. Suzuki, J. Am. Chem. Soc.
2008, 130, 13981–13988.
[47] M. R. Panman, P. Bodis, D. J. Shaw, B. H. Bakker, A. C. Newton,
E. R. Kay, D. A. Leigh, W. J. Buma, A. M. Brouwer, S. Woutersen,
Phys. Chem. Chem. Phys. 2012, 14, 1865–1875.
[48] M. R. Panman, P. Bodis, D. J. Shaw, B. H. Bakker, A. C. Newton,
E. R. Kay, A. M. Brouwer, W. J. Buma, D. A. Leigh, S. Woutersen,
Science 2010, 328, 1255–1258.
[17] S. Silvi, M. Venturi, A. Credi, Chem. Commun. 2011, 47, 2483–2489.
[18] T. Avellini, H. Li, A. Coskun, G. Barin, A. Trabolsi, A. N. Basuray,
S. K. Dey, A. Credi, S. Silvi, J. F. Stoddart, Angew. Chem. 2012, 124,
1643–1647; Angew. Chem. Int. Ed. 2012, 51, 1611–1615.
[19] S. Saha, J. F. Stoddart, Chem. Soc. Rev. 2007, 36, 77–92.
[20] R. E. Dawson, S. F. Lincoln, C. J. Easton, Chem. Commun. 2008,
3980–3982.
[49] A. M. Brouwer, C. Frochot, F. G. Gatti, D. A. Leigh, L. Mottier, F.
Paolucci, S. Roffia, G. W. H. Wurpel, Science 2001, 291, 2124–2128.
[50] M. Socol, P. D. Zoon, A. M. Brouwer, unpublished results.
[51] M. von Delius, E. M. Geertsema, D. A. Leigh, Nat. Chem. 2010, 2,
96–101.
[52] S. A. Lerke, D. H. Evans, S. W. Feldberg, J. Electroanal. Chem. Inter-
facial Electrochem. 1990, 296, 299–315.
[21] P. Raiteri, G. Bussi, C. S. Cucinotta, A. Credi, J. F. Stoddart, M. Par-
rinello, Angew. Chem. 2008, 120, 3592–3595; Angew. Chem. Int. Ed.
2008, 47, 3536–3539.
[53] G. Fioravanti, N. Haraszkiewicz, E. R. Kay, S. M. Mendoza, C.
Bruno, M. Marcaccio, P. G. Wiering, F. Paolucci, P. Rudolf, A. M.
Brouwer, J. Am. Chem. Soc. 2008, 130, 2593–2601.
[22] J. Wang, B. L. Feringa, Science 2011, 331, 1429–1432.
[23] A. Coskun, M. Banaszak, R. D. Astumian, J. F. Stoddart, B. A. Grzy-
bowski, Chem. Soc. Rev. 2012, 41, 19–30.
[24] M. von Delius, E. M. Geertsema, D. A. Leigh, D. T. D. Tang, J. Am.
Chem. Soc. 2010, 132, 16134–16145.
[25] M. J. Barrell, A. G. CampaÇa, M. von Delius, E. M. Geertsema,
D. A. Leigh, Angew. Chem. 2011, 123, 299–304; Angew. Chem. Int.
Ed. 2011, 50, 285–290.
[54] A. Viehbeck, M. J. Goldberg, C. A. Kovac, J. Electrochem. Soc.
1990, 137, 1460–1466.
[55] D. Gosztola, M. P. Niemczyk, W. Svec, A. S. Lukas, M. R. Wasielew-
ski, J. Phys. Chem. A 2000, 104, 6545–6551.
[56] S. F. Rak, T. H. Jozefiak, L. L. Miller, J. Org. Chem. 1990, 55, 4794–
4801.
[57] S. L. Buchwalter, R. Iyengar, A. Viehbeck, T. R. OꢂToole, J. Am.
Chem. Soc. 1991, 113, 376–377.
[26] A. G. CampaÇa, A. Carlone, K. Chen, D. T. F. Dryden, D. A. Leigh,
U. Lewandowska, K. M. Mullen, Angew. Chem. 2012, 124, 5576–
5579; Angew. Chem. Int. Ed. 2012, 51, 5480–5483.
[58] D. C. Jagesar, J. Baggerman, P. G. Wiering, E. Kay, D. A. Leigh,
A. M. Brouwer, unpublished results.
[59] Gaussian03 (Revision B.03), M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-
y, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Pittsburgh,
PA, 2003.
ˇ ˇ
[27] P. Kovarꢇcek, J.-M. Lehn, J. Am. Chem. Soc. 2012, 134, 9446–9455.
[28] S. A. Vignon, T. Jarrosson, T. Iijima, H.-R. Tseng, J. K. M. Sanders,
J. F. Stoddart, J. Am. Chem. Soc. 2004, 126, 9884–9885.
[29] T. Iijima, S. A. Vignon, H. R. Tseng, T. Jarrosson, J. K. M. Sanders,
F. Marchioni, M. Venturi, E. Apostoli, V. Balzani, J. F. Stoddart,
Chem. Eur. J. 2004, 10, 6375–6392.
[30] K. M. Mullen, K. D. Johnstone, M. Webb, N. Bampos, J. K. M. Sand-
ers, M. J. Gunter, Org. Biomol. Chem. 2008, 6, 278–286.
[31] Y. Makita, N. Kihara, T. Takata, J. Org. Chem. 2008, 73, 9245–9250.
[32] V. Balzani, A. Credi, M. Venturi, Chem. Soc. Rev. 2009, 38, 1542–
1550.
[33] A. Moretto, I. Menegazzo, M. Crisma, E. J. Shotton, H. Nowell, S.
Mammi, C. Toniolo, Angew. Chem. 2009, 121, 9148–9151; Angew.
Chem. Int. Ed. 2009, 48, 8986–8989.
[34] F. Durola, J. Lux, J.-P. Sauvage, Chem. Eur. J. 2009, 15, 4124–4134.
[35] I. Murgu, J. M. Baumes, J. Eberhard, J. J. Gassensmith, E. Arunku-
mar, B. D. Smith, J. Org. Chem. 2011, 76, 688–691.
[36] J. M. Baumes, I. Murgu, A. Oliver, B. D. Smith, Org. Lett. 2010, 12,
4980–4983.
[60] M. Cossi, G. Scalmani, N. Rega, V. Barone, J. Chem. Phys. 2002,
117, 43–54.
5576
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5566 – 5577

Synthetic Molecular Machine
FULL PAPER
[61] P. Ganesan, J. Baggerman, H. Zhang, E. J. R. Sudholter, H. Zuilhof,
J. Phys. Chem. A 2007, 111, 6151–6156.
[62] J. Baggerman, Photoactive Hydrogen Bonded Rotaxanes, Ph. D.
Thesis, Universiteit van Amsterdam, 2006.
[63] B. M. Aveline, S. Matsugo, R. W. Redmond, J. Am. Chem. Soc. 1997,
119, 11785–11795.
[64] A. Zweig, W. G. Hodgson, W. H. Jura, J. Am. Chem. Soc. 1964, 86,
4124–4129.
[65] J. Ouyang, A. J. Bard, J. Phys. Chem. 1987, 91, 4058–4062.
[66] F. A. Carroll, M. T. McCall, G. S. Hammond, J. Am. Chem. Soc.
1973, 95, 315–318.
[67] I. B. Berlman, Handbook of Fluorescence Spectra of Aromatic Mole-
cules, Academic Press, New York, 1971.
[68] G. Grabner, S. Monti, G. Marconi, B. Mayer, C. Klein, G. Koehler,
J. Phys. Chem. 1996, 100, 20068–20075.
[69] M. Choy de Martinez, O. P. Marquez, J. Marquez, F. Hahn, B.
Beden, P. Crouigneau, A. Rakotondrainibe, C. Lamy, Synth. Met.
1997, 88, 187–196.
[73] G. Bottari, F. Dehez, D. A. Leigh, P. J. Nash, E. M. Pꢉrez, J. K. Y.
Wong, F. Zerbetto, Angew. Chem. 2003, 115, 6066–6069; Angew.
Chem. Int. Ed. 2003, 42, 5886–5889.
[74] D. D. Gꢈnbas¸, A. M. Brouwer, J. Org. Chem. 2012, 77, 5724–5735.
[75] A. B. C. Deutman, C. Monnereau, J. A. A. W. Elemans, G. Ercolani,
R. J. M. Nolte, A. E. Rowan, Science 2008, 322, 1668–1671.
[76] P. H. Ramos, R. G. E. Coumans, A. B. C. Deutman, J. M. M. Smits,
R. de Gelder, J. A. A. W. Elemans, R. J. M. Nolte, A. E. Rowan, J.
Am. Chem. Soc. 2007, 129, 5699–5702.
[77] R. G. E. Coumans, J. A. A. W. Elemans, R. J. M. Nolte, A. E.
Rowan, Proc. Natl. Acad. Sci. USA 2006, 103, 19647–19651.
[78] I. L. Yoon, D. Benitez, Y.-L. Zhao, O. S. Miljanic, S.-Y. Kim, E.
Tkatchouk, K. C. F. Leung, S. I. Khan, W. A. Goddard, J. F. Stod-
dart, Chem. Eur. J. 2009, 15, 1115–1122.
[79] K. Nikitin, E. Lestini, J. Stolarczyk, H. Mꢈller-Bunz, D. Fitzmaurice,
Chem. Eur. J. 2008, 14, 1117–1128.
[80] D. C. Jagesar, S. M. Fazio, J. Taybi, E. Eiser, F. G. Gatti, D. A. Leigh,
A. M. Brouwer, Adv. Funct. Mater. 2009, 19, 3440–3449.
[70] P. OꢂNeill, S. Steenken, D. Schulte-Frohlinde, J. Phys. Chem. 1975,
79, 2773–2779.
[71] B. Chutny, A. J. Swallow, Trans. Faraday Soc. 1970, 66, 2847–2854.
[72] J. Holcman, K. Sehested, J. Chem. Soc. Faraday Trans. 1 1975, 71,
1211–1213.
Received: November 9, 2012
Published online: April 5, 2013
Chem. Eur. J. 2013, 19, 5566 – 5577
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5577