Controlling ring translation of rotaxanes

Article abstract of DOI:10.1039/c0ob00270d

Novel rotaxanes containing two 9-aryl-9-methoxy-10-methyl-9,10- dihydroacridine moieties (acridanes) at both ends of the molecular axle as recognition stations for the tetracationic ring CBQT⁴⁺ were synthesized together with their acridinium counterparts. A new concept of controlling the ring movement within rotaxanes has been realized with these rotaxanes. Owing to Brownian molecular movement, the ring shuttles from one end of the axle to the other one in acridane rotaxanes. The shuttle process is stopped by converting two or one of the acridane stations into the corresponding acridinium unit. If both acridanes are transformed by addition of an acid, the ring resides on evasive stations present in the center of the axle. Photons convert only the unoccupied acridane station, thus the ring remains on the unchanged acridane station. The shuttle process can be switched on by addition of a base and by the thermal reaction of the methoxide with the formed acridinium ion, respectively.

Full text of DOI:10.1039/c0ob00270d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Controlling Ring Translation of Rotaxanes†
Antje Vetter and Werner Abraham*
Received 16th June 2010, Accepted 19th July 2010
DOI: 10.1039/c0ob00270d
Novel rotaxanes containing two 9-aryl-9-methoxy-10-methyl-9,10-dihydroacridine moieties (acridanes)
at both ends of the molecular axle as recognition stations for the tetracationic ring CBQT4+ were
synthesized together with their acridinium counterparts. A new concept of controlling the ring
movement within rotaxanes has been realized with these rotaxanes. Owing to Brownian molecular
movement, the ring shuttles from one end of the axle to the other one in acridane rotaxanes. The shuttle
process is stopped by converting two or one of the acridane stations into the corresponding
acridinium unit. If both acridanes are transformed by addition of an acid, the ring resides on evasive
stations present in the center of the axle. Photons convert only the unoccupied acridane station, thus
the ring remains on the unchanged acridane station. The shuttle process can be switched on by addition
of a base and by the thermal reaction of the methoxide with the formed acridinium ion, respectively.
A reduction in the size of the ring component due to photochem-
Introduction
ical electrocyclization resulted in a reduced shuttling motion.9 The
coordinative binding of metal ions stopped the shuttling process
completely.6a
Rotaxanes are mechanically coupled molecules consisting of two
parts, the wheel and the extended molecular axle, which are freely
movable relative to each other. They are promising candidates
for the development of molecular machines.1 The position of
the wheel is governed by the interaction between the so called
recognition stations present both in the wheel and the axle. If
at least two stations are present within the axle, the dynamics
of the ring translation along the axle depend on the presence of
identical or different stations. Due to Brownian motion, the wheel
shuttles between two identical stations in degenerate rotaxanes.
The activation energy of this shuttle movement depends both on
the chemical structure of the station and the distance between the
two stations.1d,1f If the energy of the interaction between the two
different stations and the wheel is sufficiently large, the wheel will
predominantly reside on that station which supplies the highest
energy gain for the wheel-axle-interaction. In switchable rotax-
anes, the energy of interaction for the recognition station and the
wheel is drastically weakened by an external stimulus. Accordingly,
the interaction with the second, unchanged station is favored and
the ring moves to this station and the so-called co-conformatio2 is
completely rearranged by Brownian motion. Solvents,3 chemicals,4
electrons,5 and photons1d,e are able to act as stimuli. Using light is
advantageous because no side products of the switching process
are evolved and because light action is addressable.
A photoreaction that switches the shuttling process off and on
in a two station degenerate rotaxane may be an alternative means
to control the movement of the wheel upon the axle. If we assume
that only one of the two identical stations can be transformed so
that the interaction energy between the wheel and this station is
drastically reduced, the wheel will reside mainly on the unchanged
station. Then, the reset would restart the shuttling process
between the two now identical stations. How, then, can different
photoreactivities of two identical moieties be achieved? Rotaxanes
with charge transfer interaction between wheel and axle may be
suitable. A real drawback of each photoresponsive system based
on charge transfer interaction is the strongly diminished quantum
yield of the photoprocess due to the increased thermal deactivation
of the photoexcited state.10 With two identical stations, only one
station can be occupied at a time, and the free station must
retain the natural, higher photoreactivity, completely unaffected
by the charge transfer interaction. In this way, the necessary
differentiation would be established.
Recently we have presented bistable rotaxanes which have
the photoactive station 9-aryl-9-alkoxy-acridane which inter-
acts with the tetracationic ring cyclobis(paraquat-p-phenylene)11
(CBQT4+).12 The acridane is transformed into the acridinium
alkoxide by photolysis. Therefore, the ring moves to a second
evasive station present within the axle. As expected, the effi-
ciency of the photolysis of the acridane occupied by the ring
is decreased by more than a factor of five compared with an
acridane compound, which is in the range of 50% if not in
complex with CBQT4+.12,13 This differentiation appears to be
sufficient to switch the shuttling process. Therefore, it is expected
that the shuttling of the ring can be controlled by dependence
on alternative stations present in the molecular thread of the
rotaxanes.
Until now, efforts were directed mainly toward safeguarding
the integrity of the co-conformers in both states. In contrast, only
a few means for controlling the dynamics of the shuttle process
by external stimuli are known. The addition of a base, an acid,
or Zn2+,6 a change of solvent,7 or the structure of the axle8 have
influenced the rate of the ring motion.
Institute of Chemistry, Humboldt-Universita¨t zu Berlin, Brook-Taylor-Str.
2, D-12489, Berlin, Germany. E-mail: abraham@chemie.hu-berlin.de
† Electronic supplementary information (ESI) available: Additional ex-
perimental information, UV-vis and NMR spectra of selected com-
pounds and transients; assignment of protons of rotaxanes. See DOI:
10.1039/c0ob00270d
Furthermore, acridanes are pseudo-bases. Accordingly, protons
should be able to control the ring movement.12b
4666 | Org. Biomol. Chem., 2010, 8, 4666–4681
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
the 9-aryl-acridinium units are not affected by the presence of the
ring. Accordingly, the ring component shuttles between stations
B and C as depicted in Scheme 1. NOE-Cross peaks between ring
proton signals and the stations B and C reveal the proximity of
CBQT4+ and B and C. Due to the shuttling of the ring between
stations B and C, the proton signals of B and C and of the chain
between these stations are very broad but visible.
Results and discussion
Synthesis of rotaxanes
A template directed threading-followed by connecting two half-
axles approach was employed to generate different types of rotax-
anes (Scheme 1). All of them include two acridinium/acridane-
stations at the ends of the axle and with CBQT4+ as the wheel.
They differ from each other by kind and number of evasive stations
in the middle of the axle. Protons should be able to transform the
two acridane stations into two acridinium units. Addition of a base
would result in restoring the starting acridane rotaxane. Photons
are expected to transform only one of the acridane stations into
the acridinium unit. The thermal back reaction between the
released methoxide ion and the formed acridinium ion reverses
the reaction. The lifetime of the acridinium methoxide is between
seconds and hours, depending on the alcohol content of the solvent
mixture.12
[2]Rotaxanes, molecules which consist of an axle and one
macrocycle, were prepared by reaction of the western half of the
molecular axle containing the acridane recognition station with
the eastern half containing the acridinium unit in the presence
of the tetracationic ring. Both an esterification and a “click”
reaction were used to connect both parts of the molecular axle
(see Schemes 2–4). Esterification of the alcohol 1 with the acid 2
results in the four station rotaxane 3. Surprisingly, despite various
ratios of CBQT(PF6)4 and 1 and 2, in one case both [2]rotaxane 3
and [3]rotaxane 4, containing two macrocycles, were formed.
The three-station-rotaxane 10 is available by “click”-chemistry,
whereby the question arises as to whether the triazole moiety is
able to act as recognition station of CBQT4+ (Scheme 4). “Click”-
chemistry also is the basis of the synthesis of 14 and 19 (see
Schemes 5 and 6). The reaction of the acridinium rotaxanes with
methanol in acetonitrile solution in the presence of NaHCO3
affords the related acridane rotaxanes 6, 7, 12, 16 and 21
(Schemes 3–6)
Rotaxane 4. In [3]rotaxane 4, the second ring CBQT4+ resides
on the second evasive station. Like rotaxane 3, both stations B and
C are shielded. In contrast to the spectrum of 3, the proton signals
of stations B and C and the chain between them are well resolved.
Apparently ring shuttling is restricted in the [3]rotaxane 4. Proton
resonances of stations A are not influenced by the presence of the
two rings. (Scheme S6, ESI†).
Rotaxane 10. In acridinium rotaxanes without an evasive
station, the aryl substituent in the 9-position of the acridinium unit
is able to accomodate CBQT4+.12b However, the triazole moiety
(T) present in rotaxane 10 competes with the acridinium stations.
Indeed, the wheel resides mainly on the triazole station. Therefore,
in the spectrum of 10, the CH- resonance of the triazole unit
is upfield shifted by 3.4 ppm. The methylene units adjacent to
the triazole unit are considerably upfield shifted also (Scheme
S9, ESI†). The small CIS-values observed for the 9-aryl protons
indicate that the ring occupies the acridinium units only to a small
extent.
Rotaxane 14. Besides the triazole moiety T, rotaxane 14 also
contains station B, which is expected to be the superior recognition
station. Indeed, station B is occupied predominantly by CBQT4+
(Scheme S12, ESI†).
Rotaxane 19. A different situation is observed with rotax-
ane 19. Despite the positively charged acridinium subunit, the
aminophenyl group in the 9-position is a better recognition station
than the triazole moiety (Fig. S5 and Scheme S14, ESI†) and
the aryl protons are most strongly upfield shifted. However, to a
smaller extent, the triazole unit is also occupied by the ring and we
can therefore conclude that CBQT4+ shuttles between these two
stations.
For comparison, the acridinium axles 11, 15 and 20 were also
converted into the corresponding acridane axles 22–24 (Scheme 7).
Co-conformation
It is worth noting that the second aminophenyl group, A¢ of
19 (Scheme 6), which differs from A only by the presence of a
methyl substituent, does not interact with the ring. The preferential
location of the ring at the aminophenyl acridinium station instead
of at the triazole station or the N-methylaminophenyl acridinium
station may be attributed to the increased electron donor capability
of the aminophenyl group compared to the alkoxypheny group of
10 on the one hand and steric interference due to the methyl group
at the amino group of the other hand.
Acridinium rotaxanes. The mutual conversion of the acridane
rotaxanes and their acridinium counterparts can be controlled by
the alternate addition of acid and base. It is important, therefore,
to determine the co-conformation of the two counterparts. In
order to probe the solution-state structures of rotaxanes NMR-
spectroscopy was employed. Complexation induced shifts (CIS)
of proton signals of the rotaxanes compared to those of the wheel-
free axles are equivalent to the interaction between the recognition
stations and the wheel CBQT4+. Although none of the molecular
axles is symmetrical the two acridinium or acridane stoppers have
almost identical NMR-signals (see for examples Schemes S10, S13,
S14, ESI†). The assignment of proton signals of the axle is possible
with the aid of two- dimensional techniques such as H,H-COSY,
C,H-COSY and ROESY.
Acridane rotaxanes. Next we studied the shuttling process of
the ring CBQT4+ between the two acridane stations at the ends
of the molecular axle. In general, the NMR-spectra of acridinium
rotaxanes exhibit sharp, easy-to-assign proton signals. In contrast,
ring rotations12a at room temperature lead to broad or even merged
baseline proton signals of CBQT4+ and the 9-aryl group of the
acridane rotaxanes (compare Fig. S1 and S2, ESI†). Even proton
signals of unoccupied evasive stations often are not visible at room
temperature, possibly due to rotations of the aryl ring within the
axle.12
Rotaxane 3. CIS-values reveal that stations B and C are
complexed by the ring CBQT4+ (Scheme S5, ESI†). The proton
resonances of both station B and station C are upfield shifted by
2–3 ppm compared with the axle 5. In contrast, proton signals of
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4666–4681 | 4667
©

View Article Online
Scheme 1 Types of novel rotaxanes possessing two acridane/acridinium stations.
4668 | Org. Biomol. Chem., 2010, 8, 4666–4681
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 2 The synthesis of the four station rotaxanes: generation of [2]rotaxane 3 and [3] rotaxane 4. Axle 5 is also illustrated.
In a first approximation, the proton resonances of those
methylene units directly bound to evasive stations B, C or T can be
used as probe signals.12a If the probe signals are not upfield shifted,
the ring CBQT4+ does not interact with stations B, C or T.
By lowering the temperature, both the proton signals of the ring
and of the stations A and B, C appear or become sharper and
hence assignable (Fig. S3, ESI†).
A priori, one may assume that the ring in degenerate acridane
rotaxanes will reside alternately on both stations. Previous studies
have revealed that evasive stations B or C are not able to compete
with the acridane moiety for the interaction with CBQT4+.12
the movement of CBQT4+ between the two acridane stations. The
rotational movements of the aromatic rings of CBQT4+ are frozen
at low temperature and the proton resonances of CBQT4+ are split
into at least three sets, as usually found for acridane rotaxanes.12
Rotaxane 7. Proton resonances of both CBQT4+ are very
broad both in CD3CN and in acetone-d6 solution. NMR spectra
in MeOD solution exhibit two broad singlets assignable in each
case to the 8 a-protons of the pyridinium rings of the two CBQT4+
wheels. Also, aryl protons of stations A–C are not visible. However,
using the methylene groups of the chain between B and C as a
probe, it is clear that B and C are not occupied by the rings. These
proton resonances are in accord with those found in the axle 5
(Scheme S8, ESI†). Accordingly, the CBQT4+-rings must reside
on both the acridane stations. Shuttling does not occur because
the two rings would interfere with each other.
Rotaxane 6. Neither proton signals of CBQT4+ nor proton
signals of aryl groups are visible in the NMR spectra at room
temperature. Only the probe protons of the methylene groups
between stations B and C appear at the same position as in the axle
molecule 5 indicating that the ring does not reside on the stations
B and C. However, at 233 K in CDOD/CD3CN (2 : 1) solution,
two sets of aryl protons related to A appear and can be assigned
by H–H- and C–H-COSY cross peaks. One set is strongly upfield
shifted; the other is only slightly upfield shifted (Scheme S6, ESI†).
Protons of the stations B and C appear in the normal aromatic
region. Two sets of aryl protons associated with A also appear
at 233 K in acetone-d6 solution and can be assigned by COSY
and ROESY cross peaks. This way, in addition to NOE cross
peaks, exchange cross peaks that correspond to the two different
A stations can be observed in acetone-d6, thus directly revealing
Rotaxane 12. Only a weak evasive recognition station (triazole
T) exists besides the powerful acridane recognition stations.
A priori, one can assume that the ring will reside alternately on
both stations A. However, only the proton signals of the occupied
aryl group at the 9-position of the acridane ring appear in the
1H-NMR spectrum of 12. Due to the shielding effect exerted
by CBQT4+, the signals are upfield shifted by 4.4 and 2.3 ppm
(Scheme S10, ESI†).
These resonances are not averaged values, but are consistent
with an occupied acridane station of rotaxanes with one acridane
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4666–4681 | 4669
©

View Article Online
Scheme 3 The formation of acridane rotaxanes from acridinium rotaxanes by the addition of base and methanol.
station only.12 Therefore, room temperature shuttling is slow on
the NMR-time scale. Again at 233 K two sets of aryl protons are
visible. One set is strongly upfield shifted further on; the second set
is in the region normally indicative of an uncomplexed acridane
station (Scheme S11, ESI†). The assignment of the proton signals
is possible with the help of two-dimensional NMR-methods (HH-,
CH-COSY and ROESY). The ROESY spectrum recorded at
233 K exhibits not only NOE cross peaks between the upfield
shifted aryl protons of the axle and the xylylene spacer of CBQT4+
but also exchange peaks arising from complementary protons of
the two aryl groups at the 9-position of the acridane moieties
appear (Fig. S4, ESI†). These peaks verify the shuttling of the ring
between the two acridane stations.
acridane unit (as an ionogenic compound) within the rotaxane
21 favors the acridane moiety in solvents containing methanol,
e.g., residual acridinium rotaxane 19 is transformed into the
corresponding acridane rotaxane 21. This finding is typical for an
acridane unit occupied by CBQT4+.12 In contrast, the molecular
axle 24 reacts with methanol to give 50% acridinium compound
(Scheme S16, ESI†).
The equilibrium between 19 and 21 is determined easily by UV-
Vis-spectroscopy, because only the acridinium compound absorbs
visible light (compare spectra given in Fig. S7, ESI†). These
differences in the behaviors of the rotaxane and the molecular
axle are only explicable if the two acridane units are alternately
protected by the ring from conversion into the acridinium
form.
Rotaxane 16. Only one set of aryl protons for the occupied
upfield shifted 9-aryl acridane moiety (Scheme S13, ESI†) appears
at 233 K in CD3CN/CD3OD solution (1 : 1). Therefore, not even
at low temperature can shuttling of CBQT4+ between the two
acridane stations be detected.
Switching
Switching by protons. The addition of an acid such as HClO4
or HPF6 converts the acridane rotaxanes in methanol solution
to their related acridinium counterparts. Because CBQT4+ now
resides on the evasive stations (e.g. B and C in rotaxane 3; T in
rotaxane 10; B in rotaxane 14 and A and T in rotaxane 19), the
large amplitude movement of the ring is stopped. The addition of a
base such as ethyl-di(isopropyl)amine re-establishes the acridane
rotaxanes. The switching process can easily be followed by UV-
Vis-spectroscopy (see Fig. 1 for an example). Acridane rotaxanes
absorb light below 350 nm; acridinium rotaxanes exhibit the
Rotaxane 21. The two acridane stations of the rotaxane 21
are not exactly identical (see Scheme 6). At room temperature,
two sets of aryl protons for A and A¢ are present. While the aryl
protons of station A appear at 1.9 and 4.5 ppm, those for A¢ are
very broad and appear at 6.5 and 7.0 ppm. Again, the assignment
was made with the help of CH-COSY und ROESY spectra (Fig.
S6 and Scheme S15, ESI†). Ring movement is recognizable by the
fact that the equilibrium between the acridinium moiety and the
4670 | Org. Biomol. Chem., 2010, 8, 4666–4681
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 4 Click-chemistry used to connect the two halves of the axle.
Fig. 1 UV/Vis spectra (MeCN–MeOH 4 : 1, 1 ¥ 10-5 M) of 12 (——) after addition of 1 equiv of HClO4 (— — —); after addition of 10 ¥ 5 equiv of
ethyl-di-isopropylamine (– – –) after addition of 10 ¥ 5 equiv of HClO4 (– - - –).
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4666–4681 | 4671
©

View Article Online
Scheme 5 Formation of the three-station-rotaxane 14 by click-chemistry and the conversion from acridinium 14 to acridane 16 by addition of base.
characteristic absorption spectra of the acridinium salts (around
440 nm for 3, 10 and 14 and around 550 nm for 19).
Switching by photons. Contrary to proton switching, the
strategy of photon control is based on the photolysis of only one,
namely the unoccupied, acridane station. Therefore, by photolysis
a rotaxane with one acridinium and one acridane station will be
formed. At least 5% methanol must be present in the solution in
order to perform a quantitative photoheterolysis.12 Also the reverse
reaction between the acridinium ion and the methoxide requires an
alcohol (methanol or ethanol) because the latter reacts with the
acridinium ion. The methoxide ion accepts the released proton.
The thermal back reaction which occurs after photoexcitation is
uniform as revealed by the appearance of two isosbestic points in
consecutive UV-Vis-spectra (Fig. S9 and S10, ESI†). The rate of
the thermal reaction depends on the nature of the alcohol. For
example, the lifetime of the acridinium methoxide formed from
the rotaxane 12 is 70 s in methanol and 600 s in ethanol solution.
The use of ethanol results in the exchange of the leaving group
without any detriment to the switching cycle.
The shuttling between both acridane stations revealed for
rotaxanes 6, 12 and 16 is stopped if both acridane stations are
transformed into the corresponding acridinium stations. Now
shuttling of CBQT4+ between B and C takes place as found for
rotaxane 3 (see above); no shuttling occurs for 10 and 14, because
only the evasive stations T and B, respectively, are involved in
the interaction between CBQT4+ and the molecular axles (see
Scheme 8 for two examples).
In the case of the rotaxane 21, only about 30% of the
acridane units can be converted to the acridinium units to prevent
protonation of the amino groups of the acridane units (see Fig.
S8, ESI†). Addition of ethyl-di(isopropyl)amine to the methanol
solution starts movement of the ring forward and back along the
whole length of the axle; protonation stops the shuttling process
and the wheel shuttles between A and T.
4672 | Org. Biomol. Chem., 2010, 8, 4666–4681
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 6 Synthesis of rotaxanes with 9-aminophenyl groups at the acridinium and acridane moieties.
The lifetime of the rotaxane containing one acridinium and
one acridane unit is too short to analyze the co-conformation
with NMR-spectroscopy. However, all studies have revealed that
the acridinium moiety is not occupied by CBQT4+, even if only
a weakly interacting alternate station such as the alkoxyphenyl
moiety is present.12 Therefore, the acridinium station which is
formed intermediately is not able to compete with the remaining
acridane station.
The increase of the transient absorbance under comparable
excitation conditions for the photolysis of 6, 12, 16 and 21 is at least
ten times higher than the absorbance increase by the photolysis
of the reference compounds 25 and 26 (see Fig. S11 and S12,
ESI†). Upon photolysis of 12, 50% of the increase of the transient
absorbance of compound 22 is observed (Fig. S13, ESI†). We
therefore conclude that the photolysis of the unoccupied acridane
station occurred.
As expected, the efficiency of the photoreaction of rotax-
anes 6, 12, 16 and 21 is significantly higher than that of
rotaxanes containing only one acridane station occupied by
CBQT4+.12
This conclusion is supported by the further finding that the
photolysis of an occupied acridane station results in a side reaction
that forms an acridinium ion by electron transfer. The charac-
14
teristic absorption spectrum of the radical trication CBQT∑3+ is
[3]Rotaxane 7 and [2]rotaxane 6 offer the opportunity to
compare a rotaxane with two occupied acridane stations with a
rotaxane with only one occupied acridane station. Under identical
conditions of equal light intensity and absorbance at 330 nm,
a turnover of 35% is obtained for 6 at 5 s irradiation time. In
contrast, only 10% is obtained for 7. As expected, the efficiency
is significantly higher if an unoccupied acridane station can be
excited. Also the transient absorption spectra given in Fig. 2 are
slightly different as will be discussed below.
superimposed on the transient UV-Vis-spectrum of the acridinium
ion. The radical cation decomposes and forms the acridinium
ion and the methoxy radical.15 CBQT4+ is thereby regenerated by
electron back-transfer from CBQT∑3+ to the methoxy radical. This
behavior, typical for occupied acridane stations, is not observed
in the cases of rotaxanes 6, 12, 16 and 21 (Fig. S14–S16, ESI†).
Accordingly, only the transient absorption spectrum monitored
with [3]rotaxane 7 (see Fig. 2) exhibits increased absorptions at
410 nm and >500 nm.
The photolytic behaviors of rotaxanes 12 and 21 are compared
with those for the one station rotaxanes 25 and 26 having only one
occupied acridane station and which were described in an earlier
publication (Schemes 9 and 10).12a
The nucleophilic reaction of the methoxide with the acridinium
rotaxane reverts to rotaxanes with two acridane stations thus
switching on the shuttle movement of the ring (Scheme 9). The
switching cycle can be repeated at least ten times without relevant
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4666–4681 | 4673
©

View Article Online
Scheme 7 Syntheses of axles with two acridane stations.
changes of the UV-Vis-spectrum of the acridane rotaxanes.
Notably, the acridane rotaxanes can be prepared in situ from
the acridinium rotaxanes by addition of ethyl-di-isopropylamine.
The switching cycle is not disturbed by the presence of the
amine.
In contrast, acid–base switching involves both acridane stations
and the ring resides on the evasive stations in the middle of the
axle.
Both variations of switching can be repeated at least ten times
without fading.
Taken together, our results uncover novel principles for control-
ling the motion of rotaxanes, findings that may be exploited for
the development of molecular machines.
Conclusions
We have synthesised novel bistable rotaxanes with a 9-aryl-9-
methoxy-acridane station at both ends of the axle. The ring
CBQT4+ shuttles from one acridane station to the other one along
the entire axle molecule. This large amplitude movement between
the acridane stations (about 2.5 nm for rotaxane 12 and 3.5 nm for
rotaxane 6 assuming an extended linear structure of the axle) is
stopped either by addition of protons or, more favorably, by light.
The principle of photo-switching is based on the much greater
photoheterolysis efficiency of acridanes not occupied by the
electron deficient ring compared to those of the occupied acridane
station. The high photoreactivity is an important advantage over
rotaxanes reported earlier by us.
Experimental
General methods
Commercially available chemicals and solvents (UVASOL,
Merck) were used as received unless otherwise noted; solvents were
dried according to standard procedures. Column chromatography
(CC) was carried out on 200 mesh silica gel (Merck). Melting
points (m.p.) were determined with a Boetius heating microscope.
ESI mass spectroscopy was carried out on LTQ FT, Finnigan
MAT, Bremen, Germany) equipped with an electrospray ion
source (Thermo Electron).
After excitation of the non-occupied 9-methoxy-9-aryl-acridane
station with light (>300 nm), the methoxide ion is released and
an acridinium ion is formed. The remaining second acridane
station has a considerably greater affinity for the ring than the
generated acridinium and evasive stations present within the
axle. Accordingly, the ring does not shuttle but remains on the
unchanged acridane station.
NMR spectra were recorded on a Bruker DPX 300 (300 MHz)
and a Bruker Advance 400 (400 MHz). The proton signals
were attributed to the different subunits with the aid of two-
dimensional NMR spectroscopy, such as C–H-COSY, H–H-
COSY, and ROESY.
UV/Vis measurements were performed with a Shimadzu UV
2101 PC spectrometer.
4674 | Org. Biomol. Chem., 2010, 8, 4666–4681
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 8 Examples for controlling the wheel movement by protons.
Irradiation of the rotaxane solutions were carried out with a
Rotaxanes 3 and 4
12a (0.299 g, 0.50 mmol) and CBQT(PF6)411 (0.617 g, 0.56 mmol)
conventional mercury arc (HBO 500 or HBO 200) combined with
a cut-off filter of 300 nm. Transient absorption spectra between
240 and 500 nm were recorded with the UV 2101 PC spectrometer
using the fastest scan mode. The transient UV-Vis-absorption
spectrum of photoexcited CBQT4+ in methanol solution was
recorded with the help of a flash photolysis apparatus.16 We thank
A. Jacobi and Prof. U.-W. Grummt, University Jena, for these
measurements.
1
in DMF (0.7 mL) were stirred under an argon atmosphere for
0.5 h. A mixture of 2 (ESI†) (0.391 g, 0.55 mmol), dicyclo-
hexylcarbodiimide (0.113 g, 0.55 mmol), and DMAP (0.007 g,
0.06 mmol) was added within 0.5 h. The suspension was stirred
for 48 h. The resulting solution was diluted with dichloromethane
(50 mL). The precipitate was washed with dichloromethane until
the solution is colorless. The solid was treated with a mixture
(2 mL) of acetonitrile (40 mL), ethylacetate (20 mL), cyclohexane
(10 mL), and NH4PF6 (0.7 g). CBQT(PF6)4 was filtered off. The
Turnover of acridinium ions generated by photolysis was
determined by comparing the absorbance at 360 nm with the
absorbance obtained upon addition of HClO4.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4666–4681 | 4675
©

View Article Online
Scheme 9 Switching of rotaxane 12 by photons and rotaxane 25 for comparison.
1
solution was subjected to column chromatography (CC) (silica gel,
acetonitrile (40 mL), ethylacetate (20 mL), cyclohexane (10 mL)
and NH4PF6 (0.7 g). The yellow fractions were collected and
evaporated to give a mixture of 3 and 4. The two rotaxanes were
separated by preparative TLC (silica gel, acetonitrile (40 mL),
ethylacetate (20 mL), cyclohexane (10 mL) and NH4PF6 (0.4 g)
affording 3 (Rf 0.63, 0.115 g, 9%), m.p. 168 ◦C and 4 (Rf 0.46–
0.34, 0.059 g, 3%), m.p. 214 ◦C as orange solids after washing with
water.
Rotaxane 4: H NMR (400 MHz, CD3CN, TMS) (13C, 100
MHz): d = 1.8 (29.8) (m, 2 H; CH2), 2.0 (28.6) (m, 2 H; CH2),
2.45 (33.0) (t, J = 8 Hz, 2 H; CH2), 2.73 (28.2) (t, J = 8 Hz, 2 H;
CH2), 2.9 (66.8) (s,br, 2H; ethyleneoxy) 2.95 (111.6) (d, J = 8 Hz,
2 H; Ar), 3.05 (111.8) (d, J = 8 Hz, 2 H; Ar), 3.1 (66.8) (s br, 2 H;
ethyleneoxy), 3.75 (69.2) (m br, 4 H; ethyleneoxy), 3.89–4.05 (68,
70) (m br, 12 H; ethyleneoxy), 4.3 (70.6) (s br; 4 H; ethyleneoxy),
4.34 (64.5) (t, J = 7 Hz, 2 H; CH2OC O), 4.61 (128.1) (d, J = 9 Hz,
2 H; Ar), 4.8 (38.6, 128.1) (m, 8 H; NMe, Ar), 5.76 (64.9) (m, 16
H; CBQT4+), 7.0 (114.8) (m, 4 H; aryl), 7.37 (132.0) (m, 4 H; aryl),
7.86, 7.87 (130.9, 131.0) (ds, 16 H; CBQT4+), 7.8 (130.4) (m, 4 H;
acridinium, H-2,7), 7.95–7.99 (126.6) (m, 20 H; acridinium, H-1,8,
CBQT4+), 8.35 (138.1) ((m, 4 H; acridinium, H-3,6); 8.55 (118.7) (d,
J = 9 Hz, 4 H; acridinium, H-4,5), 8.95 (145.0) (m, 16 H; CBQT4+);
1
Rotaxane 3, H NMR (400 MHz, CD3CN, TMS) (13C, 100
MHz): d = 1.83 (30) (m, 2 H; CH2), 2.0 (30) (m, 2 H; CH2), 2.3
(30) (m, 2 H; CH2), 2.5 (29) (m, 2 H; CH2), 2.9 (112) (s br, 2 H;
Ar), 3.0 (112) (s br, 2 H; Ar), 3.1 (67) (s br, 2 H; ethyleneoxy),
3.5 (68) (s br, 2 H; ethyleneoxy), 3.8–3.9 (67.4–71.06) (m br, 10
H; ethyleneoxy), 3.96 (69.7) (s br, 4 H; ethyleneoxy), 4.02 (69.8) (s
br, 2 H; ethyleneoxy), 4.16 (64.2) (t, J = 9 Hz, 2 H; CH2OC O),
4.3 (65.0) (s br, 4 H; ethyleneoxy), 4.5 (127), 4.6 (128) (s br, 2
H; Ar), 4.79 (s, 6 H; N+Me), 5.77 (65) (m, 8 H; CBQT4+), 6.95
(115) (d, J = 8 Hz, 1 H; aryl), 7.1 (115) (d br, 2 H; aryl), 7.39
(132.2) (d, J = 8 Hz, 2 H; aryl); 7.4 (132) (d, J = 8 Hz, 1 H; aryl),
7.25 (115) (d, J = 8 Hz, 1 H; aryl), 7.3 (132) (d br, 1 H; aryl),
7.82–7.85 (128, 131) (m, 11 H; acridinium, H-2,7, CBQT4+); 7.94
(126.8) (d, J = 7 Hz, 8 H; CBQT4+), 8.07 (130.4) (d, J = 9 Hz, 2
H; acridinium, H-1,8); 8.12 (130) (d, J = 9 Hz, 1 H; acridinium,
H-1) 8.35 (138.7) ((m, 4 H; acridinium, H-3,6); 8.57 (118.7) (d,
J = 8 Hz, 4 H; acridinium, H-4,5); 8.93 (145.0) (d, J = 7 Hz,
8 H; CBQT4+); HRMS (ESI): m/z: found: 685.2240; calcd for
-
HRMS (ESI): m/z: found: 1051.9301; calcd for [M – 3PF6 ]
[C142H136F42N10O10P7]3+: 1051.9306; found: 752.7045; calcd for [M –
4PF6 ] [C142H136F36N10O10P6]4+: 752.7036; found: 573.1709; calcd
-
for [M – 5PF6 ] [C142H136F30N10O10P5]5+:573.1699; found: 453.4821;
-
calcd for [M – 6PF6 ] [C142H136F24N10O10P4]6+: 453.4829; found
-
367.9898; calcd for [M – 7PF6 ] [C142H136F18N10O10P3]7+: 367.9904;
-
found 303.8706; calcd for [M – 8PF6 ] [C142H136F12N10O10P2]8+:
-
303.8710.
10-methyl-9-(4-(2-(2-(2-(4-(3-(3-(4-(2-(2-(2-(4-(10-methylacri-
dinium-9-yl)phenoxy)ethoxy)ethoxy)ethoxy)
phenyl)propanoyloxy)propyl)phenoxy)ethoxy)ethoxy)ethoxy)-
phenyl)acridinium hexafluorophosphate (5). The collected
dichloromethane solutions obtained during the procedure to
isolate rotaxanes 3 and 4 contained the molecular axle 5. The
solvent was removed and the residue was purified by CC (silica
gel, acetone (400 mL)/cyclohexane (40 mL)/NH4PF6 (0.5 g).
The yellow fractions were collected and the solvent was removed
-
[M – 3PF6 ] [C106H104F18N6O10P3]3+: 685.2241; found: 477.6771;
calcd for [M – 4PF6 ] [C106H104F12N6O10P2]4+: 477.6769; found:
-
353.1484; calcd for [M – 5PF6 ] [C106H104F6N6O10P]5+: 353.1486;
-
found: 270.1293; calcd for [M – 6PF6 ] [C106H104N6O10]6+; 270.1297.
-
4676 | Org. Biomol. Chem., 2010, 8, 4666–4681
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 10 Switching of rotaxane 21 by photons and rotaxane 26 for comparison.
in vacuo. The residue was treated with water (30 mL) and
Rotaxane 6
dichloromethane (30 mL). The organic phase was separated and
the aqueous phase was extracted with dichloromethane (2 ¥
30 mL). The combined organic phases were dried (MgSO4). The
solvent was evaporated to give 5 as an orange solid (0.232 g, 34%).
1H NMR (400 MHz, CD3CN, TMS) (13C, 100 MHz): d = 1.8
(31.9) (m, 2 H; CH2), 2.5–2.55 (29.9, 35.9) (m, 4 H; CH2), 2.77
(29.8) (t, J = 7 Hz, 2 H; CH2), 3.68–3.70 (70.4, 70.5) (m, 8 H;
ethyleneoxy), 3.78 (69.4) (t, J = 5 Hz, 4 H; ethyleneoxy); 3.89 (69.2)
(t, J = 5 Hz, 4 H; ethyleneoxy), 4.02 (60.0) (m, 2 H; CH2OC O),
4.06 (&(: = 9 (T, J = 5 Hz, 4 H; ethyleneoxy), 4.28 (67.4) (t, J =
5 Hz, 4 H; ethyleneoxy), 4.7 (38.7) (s, 6 H; NMe), 6.8 (114.3) (d,
J = 9 Hz, 4 H; aryl), 7.07 (132.0) (d, J = 9 Hz, 4 H; aryl), 7.26
(114.9) (d, J = 9 Hz, 4 H; Ar), 7.44 (130.4) (d J = 9 Hz, 4 H; Ar),
7.82 (130.6) (m, 4 H; acridinium, H-2,7); 8.11 (127.6) (d, J = 8 Hz,
4 H; acridinium, H-1,8), 8.37 (138.5) (m, 4 H; acridinium, H-3,6),
NaHCO3 (0.2 g) was added to the acridinium rotaxane 3 (0.05 g,
0.020 mmol) dissolved in acetonitrile (5 mL) followed by addition
of methanol (0.1 mL). The suspension was stirred for 36 h.
NaHCO3 was filtered off. The dark brown solution was evapo-
rated. The remaining oily solid was extracted with chloroform
(3 ¥ 5 mL). The remaining solid (0.040 g, 89%) was used without
further purification.1H-NMR (400 MHz, acetone-d6) (13C 100
MHz): d = 1.8 (30.6) (s br, 2 H; CH2), 2.4 (110.0) (s br, 2 H; Ar),
2.5 (30) (s br, 2 H; CH2), 2.6 (30) (s br, 2 H; CH2), 2.8 (30.2) (s br, 2
H; CH2), 3.0 (66.7) (s, br, 2 H; ethyleneoxy), 3.40 (52) (s, 6 H; CH3,
OMe), 3.5–3.7 (66.2, 69.3, 70.0, 70.3) (m br, 14 H; ethyleneoxy),
3.9 (32.8, 69.8) (m, 8 H; ethyleneoxy, NMe), 3.95 (69.8) (s br, 2
H; OCH2), 4.02 (70.4) (s br, 2 H; CH2OC O), 4.2 (69.8) (s br 2
H; OCH2), 4.66 (127.7) (s br, 2 H; Ar), 6.0 (64, 113.6) (m br, 10
H; CBQT4+, Ar), 6.8 (129.8) (s br, 2 H; Ar), 7.4–7.7 (m br, 18 H;
acridane, CBQT4+), 8.5 (126,9) (s br, 4 H; CBQT4+), 8.6 (126.4)
8.58 (d, J = 9 Hz, 4 H; acridinium, H-4,5); HRMS (ESI): m/z:
+
found: 550.2585; calcd for [M – 2PF6 ] [C70H72N2O10]2 : 550.2588.
-
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4666–4681 | 4677
©

View Article Online
4 H; CBQT4+); HRMS (ESI): m/z: found 986.3737 calcd. for (M –
-
2PF6 ), [C108H110F12N6O12P2]2+: 986.3727; found 609.2610; calcd.
for (M – 3PF6 ), [C108H110F6N6O12P]3+: 609.2602; found 420.7046
-
calcd. for (M – 4PF6 ), [C108H110N6O12]4+): 420.7040.
-
Rotaxane 7
NaHCO3 (0.2 g) was added to the acridinium rotaxane 4 (0.05 g,
0.014 mmol) dissolved in acetonitrile (5 mL) followed by addition
of methanol (0.1 mL). The suspension was stirred for 36 h.
NaHCO3 was filtered off. The dark brown solution was evapo-
rated. The remaining oily solid was extracted with chloroform
(3 ¥ 5 mL). The remaining solid (0.042 g, 89%) was used without
further purification.1H NMR (400 MHz, CD3CN, TMS) (13C, 100
MHz): d = 1.9 (29.5) (s br, 2 H; CH2), 2.5 (38) (s br, 2 H; CH2), 2.6
(35.7) (s br, 2 H; CH2), 2.8 (30.9) (s br, 2 H; CH2), 2.9 (67.2) (s, br,
2 H; ethyleneoxy), 3.30 (52.2) (s, 6 H; CH3, OMe), 3.5–3.8 (m br,
14 H; ethyleneoxy), 4.13 (64.0) (s br, 2 H; CH2OC O), 9.1 (144.3)
(s br, 16 H; CBQT4+); HRMS (ESI): found 975.9666 calcd. for
(M – 3PF6 ), [C144H142F30N10O12P5]3+: 975.9667; found 695.7339;
-
calcd. for (M – 4PF6 ), [C144H142F24N10O12P4]4+: 695.7337; found
-
-
527.5944 calcd. for (M – 5PF6 ), [C144H142N10O12P3]5+): 527.5940.
Rotaxane 10
A solution of 8 (ESI†) (0.33 g, 0.69 mmol), 9 (ESI†) (0.41 g,
11
0.69 mmol) and CBQT(PF6)4 (0.76 g, 0.69 mmol) in dry DMF
(2 mL) was stirred for 30 min under an argon atmosphere.
A solution of CuSO4¥5H2O (0.11 g, 0.41 mmol) and sodium
ascorbate (0.14 g, 0.72 mmol) in water (0.8 mL) was added within
15 min.. The resulting suspension was stirred at room temperature
for 2 days. The reaction mixture was evaporated in vacuo at
90 ◦C. The resulting residue was treated with MeCN (10 mL)
and NH4PF6 (5% in MeCN). The precipitate was filtered and
the filtrate was evaporated. The resulting oil was extracted with
dichloromethane (3 ¥ 10 mL) in order to dissolve compound
9. The insoluble residue was treated with 8 mL of a solution
of NH4PF6 (7 g) in MeCN (400 mL), ethyl acetate (200 mL)
and cyclohexane (100 mL). Free CBQT(PF6)4 was filtered off.
The solution was chromatographed (NH4PF6 (7 g) in MeCN
(400 mL)/ethylacetate (200 mL)/cyclohexane (100 mL). The
yellow fractions were collected and the solvents were removed.
The resulting yellow solid was washed with water and dried in
vacuum to give 10 (0.13 g, 8%), m.p. 219 ◦C. UV-Vis (lmax/nm
(e/M-1cm-1): 260 (138000), 359.5 (27600), 432 (13500). 1H NMR
(400 MHz, CD3CN, TMS) (13C, 100 MHz): d = 2.58 (62.7) (s, 2
H; OCH2), 3.28–3.25 (m, 2 H; OCH2), 3.55–3.52 (m, 2 H; OCH2),
3.8 (50.4) (m, 2 H; NCH2) 3.86–3.73 (m, 14 H; ethyleneoxy), 3.9
(67.9, 69.3) (m 4 H; ethyleneoxy), 4.06 (t, J = 5 Hz, 2 H, OCH2),
4.66 (121.9) (s, 1H;, triazole), 4.80 (s, 6 H; N+Me), 5.73 (s, 8 H;
CBQT4+), 6.83 (114.9) (d, J = 8 Hz, 2 H; aryl), 6.90 (114.8) (d, J =
8 Hz, 2 H; aryl), 7.35 (132.2) (d, J = 8 Hz, 4 H; aryl); 7.70 (s, 8
H; CBQT4+); 7.88–7.82 (127.9) (m, 4 H; acridinium, H-2,7), 8.00
(130.4) (m, 4 H; acridinium, H-1,8); 8.08 (127.3) (d, J = 7 Hz, 8 H;
CBQT4+); 8.36 (138.8) (m, 4 H; acridinium, H-3,6); 8.56 (118.7)
(d, J = 8 Hz, 4 H; acridinium, H-4,5); 8.92 (145.1) (d, J = 7 Hz, 8
H; CBQT4+); HRMS (ESI): m/z: found: 999.7720; calcd for [M –
Fig. 2 Transient absorbance monitored with the [3]rotaxane 7 (3 ¥ 10-5 M,
arrows denote the increased absorption by the radical ion) (a) and the
[2]rotaxane 6 (3 ¥ 10-5 M) in methanol solution (b).
(s br, 2 H; CBQT4+), 9.1 (143.7) (s br, 2 H; CBQT4+), 9.4 (144.3)
1
(s br, 2 H; CBQT4+), 9.6 (144.7) (s br, 4 H; CBQT4+. H-NMR
(400 MHz, CD3OD/CD3CN 2 : 1, 243 K) (13C 100 MHz): d = 1.8
(31.2) (s br, 2 H; CH2), 2.1 (111) (s, br, 2 H; Ar), 2.5 (31.4) (s br, 2 H;
CH2), 2.6 (37) (s br, 2 H; CH2), 2.8 (30.1) (s br, 2 H; CH2), 2.7(68)
(s br, 2 H; ethyleneoxy), 3.4 (34) (s, 3 H; NMe), 3.5–3.8 (34, 70) (m
br, 25 H; ethyleneoxy, NMe), 4.4 (129) (s br, 2 H; Ar), 5.6 (65) (m
br, 8 H; CBQT4+), 5.9 (114) (s br, 2 H; Ar), 6.7 (127) (s br, 2 H; Ar),
6.9 (121, 126.8) (s br, 6 H; CBQT4+, acridane, H-2,7), 7.0 (114) (s
br, 4 H; acridane, H-4,5), 7.2 (129) (s br, 4 H; acridane, H-3,6), 7.6
(130) (s br, 4H; acridane, H-1,8), 8.1 (128) (s br, 2 H; CBQT4+),
8.3 (127.0) (s br, 4 H; CBQT4+), 8.5 (144) (s br, 4 H; CBQT4+), 8.9
(145) s br, 4 H; CBQT4+). 1H-NMR (400 MHz, aceton-d6, 233 K)
(13C 100 MHz): d = 1.8 (30.6) (s br, 2 H; CH2, 2.4 (110) (s, br, 2 H;
Ar), 2.5 (30) (s br, 2 H; CH2), 2.6 (30) (s br, 2 H; CH2), 2.8 (30.2) (s
br, 2 H; CH2), 3.0 (66.7) (s br, 2 H; ethyleneoxy), 3.4 (52) (s, 6 H;
OMe), 3.5–3.8 (66) (m br, 16 H; ethyleneoxy), 3.9 (32.8, 69.8) (m,
8 H; ethyleneoxy, NMe), 3.95 (69.8) (m, 2 H; ethyleneoxy), 4.02
(70.4) (s br, 2 H; CH2OC O), 4.2 (69.8) (m, 2 H; ethyleneoxy), 4.7
(127.7) (s br, 2 H; Ar), 5.98 (65, 113.6) (m br, 10 H; CBQT4+,
Ar), 6.7 (114) (s br, 2 H; Ar), 6.8 (129.8) (s br, 4 H; Ar), 6.9 (121,
126.8) (s br, 6 H; CBQT4+, acridane, H-2,7), 7.0 (114) (s br, 4 H;
acridane, H-4,5), 7.1 (114) (s br, 2 H; Ar), 7.2 (129) (s br, 4 H;
acridane, H-3,6), 7.3 (131) (s br, 2 H; Ar), 7.6 (130) (s br, 4H;
acridane, H-1,8), 8.0–8.2 (130.4) (s br, 8 H; CBQT4+), 8.5 (126.9)
(s br, 2 H; CBQT4+), 8.6 (126.4) (s br, 4 H; CBQT4+), 9.1 (143.7) (s
br, 2 H; CBQT4+), 9.4 (144.3) s br, 2 H; CBQT4+), 9.6 (144.7) (s br,
2PF6 ] [C91H89F24N9O7P4]2+: 999.7721; found: 618.1945; calcd for
-
-
[M – 3PF6 ] [C91H89F18N9O7P3]3+: 618.1931; found: 427.4050 calcd
4678 | Org. Biomol. Chem., 2010, 8, 4666–4681
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
for [M – 4PF6 ] [C91H89F12N9O7P2]4+: 427.4037; found: 312.9308;
temperature for 2 days. The reaction mixture was evaporated in
vacuo at 90 ◦C. The resulting residue was treated with MeCN
(10 mL) and NH4PF6 (5% in MeCN). The precipitate was filtered
and the filtrate was evaporated. The resulting oil was extracted
with dichloromethane (3 ¥ 10 mL) in order to dissolve compound
15. The insoluble residue was treated with 8 mL of a solution
of NH4PF6 (7 g) in MeCN (400 mL), ethyl acetate (200 mL),
and cyclohexane (100 mL). Free CBQT(PF6)4 was filtered off.
The solution was chromatographed (NH4PF6 (7 g) in MeCN
(400 mL)/ethylacetate (200 mL)/cyclohexane (100 mL). The
yellow fractions were collected and the solvents were removed. The
resulting yellow solid was washed with water and dried in vacuum
to give 14 (0.035 g, 4%), m.p. 160 ◦C. 1H NMR (400 MHz, CD3CN,
TMS) (13C, 100 MHz): d = 1.4 (30.3) (m, 2 H; CH2), 1.7 (31) (m,
2 H; CH2),3.61 (70.8) (t, J = 6 Hz; 2 H; OCH2), 3.66 (111.6) (d,
J = 9 Hz, 2 H; Ar), 3.72–3.76 (m, 12 H; ethyleneoxy), 3.9 (m 4
H; ethyleneoxy), 3.99 (127.9) (d, J = 9 Hz, 2 H; Ar), 4.12 (t, J =
5 Hz, 2 H; OCH2), 4.20 (68.0) (t, J = 4 Hz, 2 H; ethyleneoxy), 4.27
(68.0) (t, J = 4 Hz, 2 H; ethyleneoxy), 4.59 (60) (s, 2 H; OCH2),
4.80 (50.6, 39.0) (m, 8 H; NCH2, N+Me), 5.73 (m, 8 H; CBQT4+),
7.19 (114.5) (d, J = 9 Hz, 2 H; aryl), 7.25 (114.5) (d, J = 9 Hz, 2 H;
aryl), 7.45 (132.2) (m, 4 H; aryl), 7.70(131) (s, 8 H; CBQT4+); 7.85
(127.8) (m, 4 H; acridinium, H-2,7), 7.95 (127.3) (d, J = 7 Hz, 8 H;
CBQT4+), 8.10 (130.2) (m, 4 H; acridinium, H-1,8), 8.21 (125.4) (s,
1 H; triazol), 8.36 (138.5) (m, 4 H; acridinium, H-3,6), 8.56 (118.4)
(d, J = 10 Hz, 4 H; acridinium, H-4,5), 8.92 (145.1) (d, J = 7 Hz, 8
H; CBQT4+); HRMS (ESI): m/z: found: 1066.8073; calcd for [M –
-
-
calcd for [M – 5PF6 ] [C91H89F6N9O7P]5+; 312.9300.
The dichloromethane solution containing the axle 11 was
evaporated. The remaining residue was purified by column
chromatography (NH4PF6 (7 g) in MeCN (400 mL)/ethylacetate
(200 mL)/cyclohexane (100 mL) to give pure 11 (0.16 g, 47%)
as an orange, resinous compound. 1H NMR (400 MHz, CD3CN,
TMS) (13C, 100 MHz): d = 3.66–3.63 (m, 8 H; ethyleneoxy), 3.73–
3.68 (m, 4 H; ethyleneoxy); 3.84 (69.1) (t, J = 5 Hz, 2 H; OCH2),
3.92–3.89 (m, 4 H; ethyleneoxy), 4.28–4.24 (m, 4 H; ethyleneoxy),
4.62 (51.8) (t, J = 5 Hz, 2 H; NCH2), 4.69 (62.6) (s, 2 H; OCH2);
4.79 (s, 6 H; N+Me), 7.23, 7.25 (114.4) (m 4 H; aryl), 7.44, 7.45
(131.3) (dd, J = 9 Hz, 4 H; aryl), 7.84 (127.4) (m, 4 H; acridinium,
H-2,7), 8.11 (130.0) (m, 4 H; acridinium, H-1,8), 8.35 (138.5) (m,
4 H; acridinium, H-3,6), 8.58 (d, J = 9 Hz, 4 H; acridinium, H-
-
4,5); HRMS (ESI): m/z: found: 1044.3886; calcd for [M – PF6 ]
[C55H57F6N5O7P]+: 1044.3894; found: 449.7122; calcd for [M –
2PF6 ] [C55H57N5O7]2+: 449.7124.
-
Rotaxane 12
NaHCO3 (0.2 g) was added to the acridinium rotaxane 10
(0.05 g, 0.022 mmol) dissolved in acetonitrile (5 mL) followed
by addition of methanol (0.1 mL). The suspension was stirred
for 36 h. NaHCO3 was filtered off. The dark brown solution
was evaporated. The remaining oily solid was extracted with
chloroform (3 ¥ 5 mL). The remaining solid (0.040 g, 89%)
was used without further purification. UV-Vis (MeCN), l/nm
(e/M-1cm-1): 262 (45000), 277 (sh); (38800), 331 (sh) 8100.1H-
NMR (400 MHz, MeOD/CD3CN 3 : 1) (13C 100 MHz): d = 2.3
(110.8) (d, J(H,H) = 8.3 Hz, 4 H; Ar), 2.7 (67.4) (s, br, 4 H;
ethyleneoxy), 3.40 (49.3) (s, 6 H; CH3, methoxy), 3.5–3.8(33.2,
69.3,69.6,70.0, 70.3,70.4) (m, 24 H; NMe, ethyleneoxy), 4.1 (63.5)
(s br, 2 H; OCH2), 4.3 (50.4) (t, J(H,H) = 6.6 Hz, 2 H; NCH2),
4.87 (128.0) (d, J(H,H) = 8.4 Hz, 4 H; Ar), 5.7 (64) (s br, 8 H;
CBQT4+), 7.3–7.9 (m br, 32 H; acridane, CBQT4+), 9.0 (145) (s br,
2PF6 ] [C100H99F24N9O8P4]2+: 1066.8086; found: 662.8846; calcd for
-
[M – 3PF6 ] [C100H99F18N9O8P3]3+: 662.8842; found: 460.9224 calcd
-
for [M – 4PF6 ] [C100H99F12N9O8P2]4+: 460.9220; found: 339.7449;
-
calcd for [M – 5PF6 ] [C100H999F6N9O8P]5+; 339.7446.
-
The dichloromethane solution containing the axle 15 was evap-
orated. The remaining residue was purified by CC (NH4PF6 (0.5
g) in acetone (400 mL)/cyclohexane (40 mL). After evaporating
the solvents the residue wetre treated with water (30 mL) and
dichloromethane (30 mL). The aqueous phase was extracted with
dichloromethane (2 ¥ 30 mL) The combined organic phases were
dried (MgSO4). The solvent was evaporated in vacuo to give pure
1
8 H; CBQT4+). H-NMR (400 MHz, CD3OD/CD3CN 3 : 1, 233
K) (13C 100 MHz): d = 2.2 (s, br, 2 H; Ar), 2.8 (68) (s, br, 2 H;
ethyleneoxy), 3.0 (70) (s br, 2 H; ethyleneoxy), 3.3–3.8 (32, 49, 70
(m br, 30 H; NMe, OMe, ethyleneoxy), 4.2 (s br, 2 H; OCH2), 4.5
(50.0, 127.4) (s br, 4 H; NCH2, Ar), 5.7 (64) (d br, 8 H; CBQT4+),
6.7 (113.4) (s br, 2 H; Ar), 6.9 (120.4) (s br, 4 H; acridane, H-2,7),
7.1 (112.6, 113.4, 126) (s br, 6 H; Ar, acridane, H-4,5), 7.2 (128) (s
br, 4 H; CBQT4+) 7.3 (128) (s br, 4H; acridane, H-3,6), 7.5 (132) (s
br, 4H; acridane, H-1,8), 7.7 (128) (s br, 8 H; CBQT4+) 8.2 (127.7)
(s br, 4H; CBQT4+), 8.7 (145) (s br, 4 H; CBQT4+), 9.0 (145) s br, 4
H; CBQT4+); HRMS (ESI): m/z: found 885.8261 calcd. for (M –
1
15 (0.158 g, 41%) as an orange, resinous compound. H NMR
(400 MHz, CD3CN, TMS) (13C, 100 MHz): d = 1.64 (31.6) (m,
2 H; CH2), 2.44 (31.2) (t, J = 9 Hz, 2 H; CH2), 3.34 (69.5) (t,
J = 6 Hz, 2 H; OCH2), 3.55–3.58 (m, 4 H; ethyleneoxy), 3.60–
3.63 (m, 10 H; ethyleneoxy); 3.67–3.69 (m, 2 H; ethyleneoxy), 3.82
(70) (t, J = 5 Hz, 2 H; ethyleneoxy), 4.23 (68) (t, J = 5 Hz, 2 H;
ethyleneoxy), 4.26 (68) (t, J = 5 Hz, 2 H; ethyleneoxy), 4.51 (50.3)
(t, J = 5 Hz, 2 H; NCH2), 4.80 (38.9) (s, 6 H; N+Me), 4.95 (61.6)
(s, 2 H; OCH2), 6.77 (114.7) (d, J = 7 Hz, 2 H; Ar), 6.97 (127.9) (d,
J = 7 Hz, 2 H; Ar), 7.2 (115.1) (m, 4 H; Ar), 7.4 (132.2) (m, 4 H;
Ar), 7.79 (127.9) (m, 4 H; acridinium, H-2,7), 7.92 (125.4) (s, 1 H;
triazol), 8.05 (127.9) (m, 4 H; acridinium, H-1,8), 8.33 (138.8) (m,
4 H; acridinium, H-3,6), 8.57 (m, 4 H; acridinium, H-4,5); HRMS
-
2PF6 ), [C93H95F12N9O9P2]2+: 885.8268; found 542.2242; calcd. for
-
(M – 3PF6 ), [C93H95F6N9O9P]3+: 542.2230; found 370.4323 calcd.
for (M – 4PF6 ), [C93H95N9O9]4+): 370.4313.
-
(ESI): m/z: found: 516.7490; calcd for [M – 2PF6 ] [C64H67N5O8]2+:
-
Rotaxane 14
516.7489.
A solution of 9 (0.188 g, 0.32 mmol), 13 (ESI†) (0.199 g,
0.32 mmol) and CBQT(PF6)4 (0.352 g, 0.32 mmol) in dry DMF
(2 mL) was stirred for 30 min under an argon atmosphere.
A solution of CuSO4¥5H2O (0.047 g, 0.19 mmol) and sodium
ascorbate (0.065 g, 0.33 mmol) in water (0.8 mL) was added
within 15 min. The resulting suspension was stirred at room
Rotaxane 16
NaHCO3 (0.2 g) was added to the acridinium rotaxane 10
(0.05 g, 0.022 mmol) dissolved in acetonitrile (5 mL) followed
by addition of methanol (0.1 mL). The suspension was stirred
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4666–4681 | 4679
©

View Article Online
-
for 36 h. NaHCO3 was filtered off. The dark brown solution
was evaporated. The remaining oily solid was extracted with
chloroform (3 ¥ 5 mL). The remaining solid (0.039 g, 81%)
was used without further purification. 1H-NMR (400 MHz,
MeOD/CD3CN 1 : 1, 233 K) (13C 100 MHz): d = 1.7 (32.5) (s br, 2
H; CH2), 2.2 (109.7) (s br, 2 H; Ar), 2,5 (30.8) (s br, 2 H; CH2), 2.7
(67.4) (s br, 2 H; ethyleneoxy), 3.3 (51, 64) (s, 8 H; OCH2, OMe),
3.4 (32.5) (s, 6 H; NMe), 3.5–3.8 (m br, 16 H; ethyleneoxy), 3.7 (74)
(s br, 2 H; ethyleneoxy), 4.2 (67.4) (s br, 2 H; OCH2), 4.4 (127.8) (s
br, 2 H; Ar), 4.5 (48.7) (m, br, 2 H; NCH2), 5.6–5.7 (64.1) (m br, 8
H; CBQT4+), 6.7 (113.7) (d, J = 9 Hz, 2 H; aryl), 6.8 (120) (m, br, 4
H; acridane, H-2,7), 7.03 (d, J = 9 Hz, 2 H; aryl), 7.2–7.6 (m br, 20
H; acridane, CBQT4+), 8.1 (127) (s br, 4 H; CBQT4+), 8.3 (127) (s
br, 4 H; CBQT4+), 8.6 (144.3) (s br, 4 H; CBQT4+), 8.9 (145.1) (s br,
4 H; CBQT4+); HRMS (ESI): m/z: found 952.8625 calcd. for (M –
for [M – 3PF6 ] [C92H93F18N11O5P3]3+: 622.2090;); found: 430.4160
-
calcd for [M – 4PF6 ] [C92H93F12N11O5P2]4+: 430.4156.
The dichloromethane solution containing the free axle was
evaporated. The remaining residue was purified by column
chromatography (NH4PF6 (7 g) in MeCN (400 mL)/ethylacetate
(200 mL)/cyclohexane (100 mL) to give pure 20 (0.15 g, 19%)
as an orange, resinous compound. 1H NMR (400 MHz, CD3CN,
TMS) (13C 100 MHz): d = 3.1 (38.6) (s, 3 H; NMe), 3.36 (43.2) (t,
J(H,H) = 5.5 Hz, 2 H; NH–CH2), 3.5–3.7 (68.4, 69.1, 69.3, 69.6,
70.2, 70.3, 70.4) (m, 16 H; ethyleneoxy), 3.85 (t, J(H,H) = 5.3 Hz, 2
H; OCH2), 4.47 (50.1) (t, J(H,H) = 5.0 Hz, 2 H; NCH2), 4.58 (63.9)
(s, 2 H; OCH2), 4.71, 4.72 (38.8) (ds, 6 H; N+Me), 6.92 (112.2) (d,
J(H,H) = 8.5 Hz, 2 H; Ar), 7.02 (111.7) (d, J(H,H) = 8.8 Hz, 2
H; Ar), 7.29 (132.8) (d, J(H,H) = 8.7 Hz, 2 H; Ar), 7.36 (132.8)
(d, J(H,H) = 9.0 Hz, 2 H; Ar), 7.8 (127.2) (m, 4H; acridinium,
H-2,7), 8.21 (130.9) (d, J(H,H) = 8.7 Hz, 2 H; acridinium, H-1,8),
8.28 (m, 2 H; acridinium, H-3,6), 8.49 (d, J(H,H) = 9.3 Hz, 2 H;
acridinium, H-4,5); HRMS (ESI): m/z: found 1056.4355; calcd for
2PF6 ), [C102H105F12N9O10P2]2+: 952.8629; found 586.9198; calcd.
-
for (M – 3PF6 ), [C102H105F6N9O10P]3+: 586.9203; found 403.9497
-
calcd. for (M – 4PF6 ), [C93H95N9O9]4+): 403.9491.
-
-
[M – PF6 ] [C56H61F6O5N7P]+: 1056.4371;]; found 455.7365; calcd
for [M – 2PF6 ] [C56H61O5N7]2+: 455.7362.
-
Rotaxane 19
A solution of 17 (ESI†) (0.35 g, 0.72 mmol), 18 (ESI†) (0.44 g,
0.72 mmol) and CBQT(PF6)4 (0.79 g, 0.72 mmol) in dry DMF
(2 mL) was stirred for 30 min under an argon atmosphere.
A solution of CuSO4¥5H2O (0.10 g, 0.37 mmol) and sodium
ascorbate (0.14 g, 0.72 mmol) in water (0.8 mL) was added
within 15 min. The resulting suspension was stirred at room
temperature for 2 days. The reaction mixture was evaporated
Rotaxane 21
NaHCO3 (0.2 g) was added to the acridinium rotaxane 19 (0.150 g,
0.065 mmol) dissolved in acetonitrile (5 mL) under addition
of methanol (0.1 mL). The suspension was stirred for 36 h.
NaHCO3 was filtered off. The dark blue solution was evaporated.
The remaining oily solid was extracted with chloroform (3 ¥
5 mL). The remaining solid (0.130 g, 96%) was used without
further purification. UV-Vis (lmax/nm (e/M-1cm-1): 265 (68900);
1H-NMR (400 MHz, MeOD/CD3CN 2 : 1) (13C 100 MHz): d =
1.9 (108.2) (s br, 2 H; Ar), 2.0 (43.9) (s br, 2 H; NCH2), 2.7 (38.2)
(s, 3 H; NMe), 3.2 (48.8) (s, 3 H; OMe), 3.3 (52.3) (s, 3 H; OMe),
3.6–3.8 (32.5, 70.0) (m, 24 H; NMe, ethyleneoxy), 4.1 (63.0) (s,
2 H; OCH2), 4.3 (49.9) (t, J(H,H) = 5.1 Hz, 2 H; NCH2), 4.4
(49.9) (m, 2 H; NCH2), 4.45 (128.5) (s, br, 2 H; Ar), 5.7 (63.7) (s
br, 8 H; CBQT4+), 6.5 (111) (s br, 2 H, Ar), 6.8 (120.0) (s br, 4 H;
acridane, H-2,7), 7.1–7.4 (113, 128.7, 128.9) (m br, 12 H; acridane)
7.6 (127, 130) (s br 16 H; CBQT4+), 8.7 (s br, 8 H; CBQT4+).
◦
in vacuo at 90 C. The resulting residue was treated with MeCN
(10 mL) and NH4PF6 (5% in MeCN). The precipitate was
filtered and the filtrate was evaporated. The resulting oil was
extracted with dichloromethane in order to dissolve compound
20. The insoluble residue was treated with 8 mL of a solution
of NH4PF6 (7 g) in MeCN (400 mL), ethyl acetate (200 mL),
and cyclohexane (100 mL). Free CBQT(PF6)4 was filtered off.
The solution was chromatographed (NH4PF6 (7 g) in MeCN
(400 mL)/ethylacetate (200 mL)/cyclohexane (100 mL). The
violet fractions were collected and the solvents were removed.
The resulting dark violet solid was washed with water and dried in
vacuum to give 19 (0.295 g, 18%), m.p. 185 ◦C. UV-Vis (lmax/nm
(e/M-1cm-1): 257.5 (180700), 343.5 (14400), 358.5 (27000), 412
HRMS (ESI): m/z: found 948.8233; calcd. for (M - MeO-
-
PF6 ), [C93H96F18N11O6P3]2+: 948.8235; found 891.8501; calcd for
-
[M–2 PF6 ], [C94H99F12N11O7P2]2: 891.8501; found 584.2276; calcd
-
1
(6000), 530 (16400); H NMR (400 MHz, CD3CN, TMS) (13C,
for [M - MeO- - 2PF6 ], [C93H96F12N11O6P2]3+:584.2276; found
-
100 MHz): d = 1.24 (43.3) (s, br, 2 H; NHCH2), 3.03 (38.9) (s,
br, 3 H; NMe), 3.5 (67.8) (m, 2 H; OCH2), 3.5 (69.5) (m, 2 H;
ethyleneoxy), 3.6 (51.8) (m, 2 H; NMeCH2), 3.6–3.8 (70.3, 70.5)
(m, 16 H; ethyleneoxy), 4.08 (109.3) (d, J = 9 Hz, 2 H; aryl), 4.21
(49.7) (t, J = 5 Hz, 2 H; NCH2), 4.73 (38.6) (s, 3 H; N+Me), 4.82
(38.6) (s, 3 H; N+Me), 5.73 (64.9) (m, 8 H; CBQT4+), 6.57(124.1)
(s, 1 H; triazole), 6.67 (133.2) (d, J = 9 Hz, 2 H; aryl), 6.97 (111.4)
(d, J = 8 Hz, 2 H; aryl), 7.37 (133.2) (d, J = 8 Hz, 4 H; aryl), 7.73
(131.1) (s, 8 H; CBQT4+), 7.80 (127.9) (m, 2 H, acridinium, H-2,7),
7.85 (130.4) (d, J = 9 Hz, 2 H; acridinium, H-1,8), 8.07 (127.3) (d,
J = 7 Hz, 8 H; CBQT4+); 8.1 (127.9) (m, 2 H; acridinium, H-2,7),
8.20 (130.4) (d, J = 9 Hz, 2 H; acridinium, H-1,8), 8.31 (138.8) (m,
2 H; acridinium, H-3,6), 8.44 (138.8) (m, 2 H; acridinium, H-3,6),
8.51 (118.7) (d, J = 9 Hz, 2 H; acridinium, H-4,5), 8.62 (118.9) (d,
J = 9 Hz, 2 H; acridinium, H-4,5), 8.94 (145.5) (d, J = 7 Hz, 8 H;
CBQT4+); HRMS (ESI): m/z: found: 1005.7961; calcd for [M –
546.2464; calcd for [M – 3PF6 ], [C94H99F6N11O7P]3+: 546.2451;
-
found 401.9294 calcd for [M - MeO- - 3PF6 ] [C93H96F6N11O6P]4+:
-
401.9297.
Notes and references
1 (a) J. F. Stoddart, Acc. Chem. Res., 2001, 34, 410–411; (b) C. A. Schalley,
K. Beizai and F. Vo¨gtle, Acc. Chem. Res., 2001, 34, 465–476; (c) V.
Balzani, M. Venturi and A. Credi, Molecular Devices and Machines,
Wiley-VCH, Weinheim 2003; (d) E. R. Kay, D. A. Leigh and F. Zerbetto,
Angew. Chem., 2007, 119, 72–196;Angew. Chem., Int. Ed., 2007, 46, 72–
191; (e) S. Saha and J. F. Stoddart, Chem. Soc. Rev., 2007, 36, 77–92.
2 M. C. T. Fyfe, P. T. Glink, S. Menzer, J. F. Stoddart, A. J. P. White and
D. J. Williams, Angew. Chem., 1997, 109, 2158–2160; Angew. Chem.,
Int. Ed. Engl., 1997, 36, 2068–2070.
3 W. Clegg, C. Gimenez-Saiz, D. A. Leigh, A. Murphy, A. M. Z. Slawin
and S. J. Teat, J. Am. Chem. Soc., 1999, 121, 4124–4129.
4 R. A. Bissel, E. Cordova, A. E. Kaifer and J. F. Stoddart, Nature, 1994,
369, 133–137.
-
2PF6 ] [C92H93F24N11O5P4]2+: 1005.7959; found: 622.2094; calcd
4680 | Org. Biomol. Chem., 2010, 8, 4666–4681
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
5 (a) H.-R. Tseng, S. A. Vignon and J. F. Stoddart, Angew. Chem., 2003,
115, 1529–1539; Angew. Chem., Int. Ed., 2003, 42, 1491–1495; (b) G.
Periyasamy, J.-P. Collin, J.-P. Sauvage, R. D. Levin and F. Remacle,
Chem. Eur. J., 2009, 15, 1310–1313.
6 (a) L. Jiang, J. Okano, A. Orita and J. Otera, Angew. Chem., 2004,
116, 2173–2176; Angew. Chem., Int. Ed., 2004, 43, 2121–2124; (b) P.
Ghosh, G. Federwisch, M. Kogej, C. A. Schalley, D. Haase, W. Saak,
A. Lu¨tzen and R. M. Gschwind, Org. Biomol. Chem., 2005, 3, 2691–
2700; (c) D. A. Leigh, P. J. Lusby, A. M. Z. Slawin and D. B. Walker,
Angew. Chem., 2005, 117, 4633–4640; Angew. Chem., Int. Ed., 2005, 44,
4557–4564; (d) N.-C. Chen, C.-C. Lai, Y.-H. Liu, S.-M. Peng and S.-H.
Chiu, Chem.–Eur. J., 2008, 14, 2904–2908.
7 (a) D. B. Amabilino, P. R. Ashton, V. Balzani, C. L. Brown, A. Credi,
J. M. J. Frechet, J. W. Leon, F. M. Raymo, N. Spencer, J. F. Stoddart
and M. Venturi, J. Am. Chem. Soc., 1996, 118, 12012–12020; (b) A. S.
Lane, D. A. Leigh and A. Murphy, J. Am. Chem. Soc., 1997, 119,
11092–11093.
10 V. Balzani, M. Clemente-Leo´n, A. Credi, B. Ferrer, M. Venturi, A. H.
Flood and J. F. Stoddart, Proc. Natl. Acad. Sci. U. S. A., 2006, 103,
1178–1183.
11 P. L. Anelli, P. R. Ashton, R. Ballardini, V. Balzani, M. Delgado, M. T.
Gandolfi, T. T. Goodnow, A. F. Kaifer, D. Philp, M. Pietraszkiecz, L.
Prodi, M. V. Reddington, M. Z. Slawin, N. Spencer, J. F. Stoddart,
C. Vicent and D. J. Williams, J. Am. Chem. Soc., 1992, 114, 193–
218.
12 (a) W. Abraham, K. Buck, M. Orda-Zgadzaj, S. Schmidt-Scha¨ffer and
U.-W. Grummt, Chem. Commun., 2007, 3094–3096; (b) W. Abraham,
A. Wlosnewski, K. Buck and S. Jacob, Org. Biomol. Chem., 2009, 7,
142.
13 T. M. Grigor’eva, V. L. Ivanov and M. G. Kuzmin, Zh. Org. Khim.,
1981, 17, 423–428.
14 R. Ballardini, V. Balzani, M. T. Gandolfi, L. Prodi, M. Venturi, D.
Philp, H. G. Ricketts and J. F. Stoddart, Angew. Chem., 1993, 105,
1362–1364; Angew. Chem., Int. Ed. Engl., 1993, 32, 1301–1303.
15 L. Grubert and W. Abraham, Tetrahedron, 2007, 63, 10778–
10787.
8 T. Oshikiri, H. Yamaguchi, Y. Takashima and A. Harada, Chem.
Commun., 2009, 5515–5517.
9 K. Hirose, Y. Shiba, K. Ishibashi, Y. Doi and Y. Tobe, Chem.–Eur. J.,
2008, 14, 3427–3433.
16 W. Abraham, L. Grubert, U.-W. Grummt and K. Buck, Chem.–Eur. J.,
2004, 10, 3562.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4666–4681 | 4681
©