Kinetic studies exploring the role of anion templation in the slippage formation of rotaxane-like structures

Article abstract of DOI:10.1002/chem.201002528

The first examples of the slippage formation of rotaxane-like structures in the presence of an anion template are reported between a macrocycle, synthesised by exploiting Eglinton coupling, and stoppered pyridinium axle components. The role of the anion template in the slippage process has been explored by kinetic studies. ¹HNMR spectroscopic investigations reveal the slippage species formed are not rotaxanes but pseudorotaxanes with some rotaxane character. The anion template significantly influences the amount of rotaxane character and the rate of slippage. Importantly, the fastest slippage rates, k_on, are achieved with the non-coordinating hexafluorophosphate anion, whereas the slowest slippage off rates, k _off, are observed in the presence of coordinating anions, such as chloride. Since the k_off rates are significantly smaller than the k_on rates in the presence of coordinating anions, these anions act as templates favouring formation of the slippage species thermodynamically. Consequently, the resulting pseudorotaxanes with coordinating anions have greater rotaxane character. Two strategies for converting the slippage pseudorotaxanes into rotaxanes using hydrogenation or complexation with cobalt carbonyl are investigated.

Full text of DOI:10.1002/chem.201002528

DOI: 10.1002/chem.201002528
Kinetic Studies Exploring the Role of Anion Templation in the Slippage
Formation of Rotaxane-Like Structures
Anna J. McConnell and Paul D. Beer*[a]
Abstract: The first examples of the
slippage formation of rotaxane-like
structures in the presence of an anion
template are reported between a mac-
rocycle, synthesised by exploiting
Eglinton coupling, and stoppered pyri-
dinium axle components. The role of
the anion template in the slippage pro-
cess has been explored by kinetic stud-
ies. 1H NMR spectroscopic investiga-
tions reveal the slippage species
formed are not rotaxanes but pseudor-
otaxanes with some rotaxane character.
The anion template significantly influ-
ences the amount of rotaxane character
and the rate of slippage. Importantly,
the fastest slippage rates, kon, are ach-
ieved with the non-coordinating hexa-
fluorophosphate anion, whereas the
slowest slippage off rates, koff, are ob-
served in the presence of coordinating
anions, such as chloride. Since the koff
rates are significantly smaller than the
kon rates in the presence of coordinat-
ing anions, these anions act as tem-
plates favouring formation of the slip-
page species thermodynamically. Con-
sequently, the resulting pseudorotax-
anes with coordinating anions have
greater rotaxane character. Two strat-
egies for converting the slippage pseu-
dorotaxanes into rotaxanes using hy-
drogenation or complexation with
cobalt carbonyl are investigated.
Keywords: anions
pseudorotaxanes
slippage
·
kinetics
rotaxanes
·
·
G
·
Introduction
The design and synthesis of mechanically interlocked struc-
tures, such as rotaxanes and catenanes, is an area of intense
current chemical research, not only due to their synthetic
challenge, but also for their potential applications in nano-
technology as molecular switches, machines and sensors.[1]
A
number of strategies have been developed for synthesising
rotaxanes such as clipping,[2–4] stoppering,[5,6] snapping[7] and
swelling.[8] Slippage, where the macrocycle slips over one of
the axle stoppers, is a much less common approach. This
usually involves heating the macrocycle and axle compo-
nents at elevated temperatures to overcome the energy bar-
rier to slippage, DG°on, and the resulting rotaxane structure
can be kinetically trapped by cooling the equilibrated mix-
ture to room temperature, provided the barrier to slippage
off (DG°off) cannot be overcome (Figure 1).The exploitation
of suitable non-covalent interactions between the macrocy-
cle and axle components increases this energy barrier stabil-
ising the rotaxane as compared to the free components.
Stoddart and co-workers reported the first rotaxane syn-
thesis by slippage of a bis-paraphenylene[34]crown-10 mac-
rocycle over a series of 4,4’-bipyridinium stoppered axles.[9]
Subsequent kinetic studies revealed that the size of both the
Figure 1. Energy level diagram for the formation of rotaxanes by slip-
page.
macrocycle and stoppers of the axle affect the rates of slip-
page.[9–11] A number of groups have since exploited slippage
to synthesise rotaxanes.[12] One of the inherent problems of
rotaxane synthesis by using the slippage strategy, however,
is the question of whether the resulting slippage species is a
true rotaxane or a pseudorotaxane with some rotaxane char-
acter; rotaxanes are kinetically stable, whereas pseudorotax-
anes can dissociate into the macrocycle and axle compo-
nents, for example when dissolved in polar solvents.[13] It is
[a] Dr. A. J. McConnell, Prof. P. D. Beer
Inorganic Chemistry Laboratory, Department of Chemistry
University of Oxford, South Parks Road, Oxford, OX1 3QR (UK)
Fax : (+44)1865-272-690
E-mail: paul.beer@chem.ox.ac.uk
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201002528.
2724
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2724 – 2733

FULL PAPER
for this reason that Stoddart and co-workers proposed that
pseudorotaxanes lie in the ꢁfuzzyꢂ domain between rotax-
anes and the individual components.[13]
corporates electron-rich hydroquinone rings for p–p stack-
ing interactions with the electron-deficient pyridinium unit
of the axle and an isophthalamide anion binding cleft. In the
course of investigating the anion binding properties of this
macrocycle, the hexyl-appended threads 2·X[15,16] were pre-
pared for pseudorotaxane studies. The terphenyl-stoppered
axle components 3·X[4,17] and 4·X were prepared to investi-
gate slippage.
We have exploited clipping and stoppering strategies, in
combination with anion templation and ion-pair recognition,
to prepare rotaxanes.[2–5,14] While investigating the use of
Eglinton coupling in the synthesis of rotaxanes, we have dis-
covered a macrocycle that undergoes slippage with stop-
pered pyridinium axles. Herein we report the first examples
of using slippage and anion templation to assemble rotax-
ane-like structures (Figure 2) and probe the role of the
anion template in the slippage process by kinetic investiga-
tions.
The synthesis of macrocycle 1 is outlined in Scheme 2.
Two equivalents of 6, prepared via precursor 5[18] in two
steps (see Scheme S1 in the Supporting Information), was
Figure 2. Anion-templated slippage of a macrocycle over a stoppered pyr-
idinium axle.
Results and Discussion
Synthesis of components: To investigate the synthesis of ro-
taxanes by slippage, it was necessary to prepare the target
components shown in Scheme 1. The novel macrocycle 1 in-
Scheme 2. Synthesis and single-crystal X-ray structure of macrocycle 1.
heated with 7[15] and Cs2CO3 in dry DMF to afford the mac-
rocycle precursor 8 in 59% yield following purification by
silica gel chromatography. Macrocyclisation of 8 by Eglinton
coupling[19] of the terminal alkyne groups was carried out
using copper(II) acetate under high dilution conditions in
acetonitrile (Scheme 2). The desired macrocycle 1 was ob-
tained in 63% yield following purification by silica gel chro-
matography.
The stoppered axle 4·X was prepared from 9[17] according
to Scheme S2 in the Supporting Information. Compound 10
was obtained in 79% yield from the reaction of 9 with 3,5-
bis(chlorocarbonyl)pyridine. This was methylated with iodo-
methane to give 4·I in 97% yield and subsequent anion ex-
change gave 4·Cl and 4·Br in yields of 88 and 89%, respec-
tively. The hexafluorophosphate salt 4·PF6 was prepared
from anion exchange of 4·Cl with NH4PF6.
Scheme 1. Target macrocycle, hexyl appended thread and stoppered axle
components.
Chem. Eur. J. 2011, 17, 2724 – 2733
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2725

A. J. McConnell, P. D. Beer
X-ray crystal structure of 1:[20] Single crystals of 1 suitable
for X-ray crystallographic structural analysis were grown
from slow evaporation of a 5:1 CDCl3/CD3OD solvent mix-
ture. The structure clearly shows the planarity of the diyne
unit (Scheme 2). The macrocycle is in a chair-like conforma-
tion with a disordered methanol solvent molecule (omitted
for clarity) bound by weak hydrogen bonds to the isophthal-
amide amide and ortho CH proton.
sor 8 were also studied for comparative purposes. Upon ad-
dition of anions as their tetrabutylammonium (TBA) salts,
there were relatively modest downfield shifts of the iso-
phthalamide proton c and amide proton d indicating anion
binding within the isophthalamide anion binding cavity (Fig-
ure 3a). WinEQNMR2[21] analysis of the isophthalamide
proton
c titration data gave 1:1 association constants
(Table 1), which demonstrate that the non-coordinating hex-
afluorophosphate anion does not bind to either 8 or 1. Un-
surprisingly, the acyclic compound 8 binds the oxobasic di-
hydrogen phosphate anion preferentially to the other
anions, like many acyclic receptors. Macrocycle 1 binds the
halide anions more strongly than 8 due to the macrocyclic
effect, however, accurate association constants could not be
determined from the titration data for acetate and dihydro-
gen phosphate.[22] The magnitudes of all of the association
constants were lower than expected given that chloroform is
Anion binding and pseudorotaxane studies: Anion binding
and pseudorotaxane studies with macrocycle 1 were under-
taken initially to quantify the strength of anion binding
before investigating the role of the anion in the slippage
process. Owing to the limited solubility of 1 in [D6]acetone
and [D3]acetonitrile, the 1H NMR titration experiments
were carried out in CDCl3 monitoring chemical shift
changes. The anion-binding properties of macrocycle precur-
Figure 3. 1H NMR spectra of a) macrocycle 1 and one equivalent of TBACl b) macrocycle 1 c) pseudorotaxane of 1 and 2·Cl and d) 2·Cl in CDCl3 at
298 K.
2726
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2724 – 2733

Anion Templation in the Slippage Formation of Rotaxane-Like Structures
FULL PAPER
Table 1. Anion-binding properties of macrocycle precursor 8 and macro-
cycle 1 and pseudorotaxane studies with macrocycle 1 in CDCl3 at 298 K.
slipped over these stoppers. However, it was proposed that
the slight increase in size of macrocycle 1 compared to these
macrocycles and the enforced planarity of the diyne unit
may allow slippage to occur. Therefore, a number of
1H NMR experiments were carried out to investigate wheth-
er 1 was able to slip over the stopper groups of axle 3·Cl
forming the species shown in Scheme 3.
Macrocycle
precursor 8
K
Macrocycle 1
Pseudorotaxane
1·2·X
[c]
Xꢀ
Dd
Dd
[ppm][a]
K
K [mꢀ1
]
[ppm][a]
[mꢀ1
]
[mꢀ1
]
[b]
[b]
ꢀ
[d]
[d]
PF6
Iꢀ
ꢀ0.002
0.026
0.069
0.13
0.001
0.036
0.082
0.139
0.074
0.084
400
290
820
35
55
80
40
350
60
60
Brꢀ
Clꢀ
115
1350
OAcꢀ
H2PO4 0.245
0.047
[e]
[f]
ꢀ
[e]
[f]
[a] Chemical shift changes of the isophthalamide proton c upon addition
of one equivalent of anion. [b, c] Association constants derived from
winEQNMR2 analysis of the [b] isophthalamide proton titration data for
anion binding studies with macrocycle precursor 8 and macrocycle 1 and
[c] macrocycle hydroquinone proton g for pseudorotaxane studies with
2·X in CDCl3 at 298 K. Errors <10%. [d] No evidence of binding.
[e] Evidence of binding but accurate association constants could not be
determined from titration data.[22] [f] Not determined.
Scheme 3. Species formed from slippage of macrocycle 1 over the stop-
pers of axle 3·Cl.
a relatively non-polar solvent. It is likely that ion-pairing be-
tween the tetrabutylammonium cation and the anion is sig-
nificant in chloroform and this competes with the anion
binding process. Small downfield shifts of the tetrabutylam-
monium signals during the titration are consistent with ion-
pairing.
1
The H NMR spectrum of a 1:1 mixture of 1 and 3.Cl in
CDCl3 after 8 h at room temperature displayed a new set of
signals corresponding to the species formed from slippage,
in addition to signals for the free macrocycle and stoppered
axle (Figure 4b). The isophthalamide proton c and amide
proton d of the macrocycle have moved significantly down-
field, suggesting a chloride anion is bound within the anion
binding cavity. Additionally, the pyridinium cavity protons o
and n have shifted upfield due to the competitive binding of
chloride and the hydroquinone protons g and h have moved
upfield and split, which is characteristic of p–p stacking in-
teractions with the pyridinium unit of the axle component.
Given that slippage occurs at room temperature, it is un-
likely that the species formed by slippage is a rotaxane, but
rather a pseudorotaxane. This was confirmed by dissociation
of the slippage species into the macrocycle and stoppered
axle components upon addition of the competitive solvent
methanol, as evidenced by electrospray mass spectrometry
Pseudorotaxane studies with the unstoppered threads 2.X
were also undertaken in CDCl3. The downfield shifts of the
macrocycle protons c and d and upfield shifts and splitting
of the hydroquinone protons g and h upon addition of the
ion-pair thread were consistent with pseudorotaxane forma-
tion (Figure 3c). The titration data for hydroquinone proton
g was analysed by using winEQNMR2 to give 1:1 associa-
tion constants (Table 1). The pseudorotaxane association
constant magnitude is correlated with the strength of halide
binding by the macrocycleꢂs isophthalamide motif, in addi-
tion to the strength of ion-pairing. The largest association
constant value is observed with 2·Cl, the strongest ion-pair,
and reflects the complementary size match between the iso-
phthalamide macrocycle cavity and chloride anion guest.
The magnitude of the association constant decreases from
2·Cl to 2·I as the strength of the halide binding decreases
and halide size increases. Interestingly, 2·PF6 forms a pseu-
dorotaxane even in the absence of a coordinating anion and
with a larger association constant than for 2·I. This contrasts
to earlier studies in [D6]acetone with a related macrocycle
where pseudorotaxane formation occurs only in the pres-
ence of a coordinating anion.[16] Presumably p–p stacking in-
teractions are more prevalent in CDCl3 and as a result, the
halide anion is not required to facilitate interpenetration.
1
and H NMR spectroscopy. These preliminary studies sug-
gest that a pseudorotaxane rather than a rotaxane results
from slippage between 1 and 3·Cl. More detailed kinetic
studies of the slippage process were carried out using the
two axles, 3·X and sterically bulkier 4·X, at two different
temperatures varying the nature of the anion of the axle
component.
Equimolar mixtures of the macrocycle 1 and stoppered
axle (3·X or 4·X), each at a concentration of 5 mm, in CDCl3
1
Slippage studies: Having established the anion-binding
properties and pseudorotaxane formation with macrocycle
1, attention turned to investigating its slippage properties
and the role of the anion in the slippage process. It has been
reported that the triaryl stoppers of 3·Cl can stopper macro-
cycles with up to 42 carbon, nitrogen, oxygen or sulfur
atoms,[17] and previously studied macrocycles[3,4] have not
at 298 K or 323 K were monitored by H NMR spectroscopy
over a period of time. The signals corresponding to the slip-
page species increased with time as the signals for the free
macrocycle and stoppered axle decreased (see Figure S1 in
the Supporting Information). The concentration of the slip-
page species was determined by integrating the hydroqui-
none signals (g, h) of the slippage species relative to the
Chem. Eur. J. 2011, 17, 2724 – 2733
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2727

A. J. McConnell, P. D. Beer
Figure 4. 1H NMR spectra of a) axle 3·Cl b) a 1:1 mixture of 1 and 3·Cl after 8 h at room temperature and c) macrocycle 1 in CDCl3 at 298 K. The new
proton signals in b) correspond to the slippage species.
signal of the free macrocycle. Figure 5 shows that the slip-
page species concentration increases with time until equilib-
rium is reached.
Thermodynamic parameters, such as the association con-
stant Ka, and kinetic parameters, such as the rate of slippage
1
kon, were obtained from kinetic H NMR experiments. The
slippage process can be described by the equilibrium in
Equation (1)[23] and the association constant, Ka, can be de-
termined from the rotaxane, macrocycle and thread equilib-
rium concentrations according to Equation (2). In addition,
the rate constant for the slippage process, kon, can be deter-
mined by fitting the curve to the model described by Equa-
tion (3; Re =equilibrium rotaxane concentration; Mo =initial
macrocycle concentration; t=time in hours) (derived from
the integrated rate law[11,24]) using a non-linear curve fitting
program in Origin.[25]
kon
Macrocycle þ Axle
Rotaxane
H
ð1Þ
ð2Þ
G
koff
kon
koff
½Rotaxaneꢁe
Re
Me2
Ka ¼
¼
¼
½Macrocycleꢁe½Axleꢁe
h
i
kontðMo2 ꢀR2e Þ
Mo2Ree
ꢀ Mo2Re
ꢀ R2e
Re
Figure 5. Change in the concentration of the slippage species formed
from the stoppered axles 3·X in CDCl3 at 323 K. Symbols represent the
experimental data and the lines represent the calculated curves.
h
i
½Rotaxaneꢁt ¼
ð3Þ
kontðMo2 ꢀR2e Þ
Mo2e
Re
2728
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2724 – 2733

Anion Templation in the Slippage Formation of Rotaxane-Like Structures
FULL PAPER
The black line in Figure 5 shows the fitting of Equa-
tion (3) to the experimental data and the statistical parame-
be determined due to the very slow rate of slippage. Prelimi-
nary studies showed that equilibrium was reached after
more than 1 week.
ter c2 indicates the quality of the fitting (Table S1).[26]
A
value of zero indicates a perfect fit and therefore the smaller
the number, the better the fit. The rate constant for slippage
off, koff, can also be calculated according to Equation (2)
using the values calculated for Ka and kon. A comparison of
the curves for slippage with the hexafluorophosphate and
halide salts of the stoppered axles 3·X and 4·X at 298 K is
shown in Figure S2 and the curves for slippage at 323 K are
shown in Figure 5 and Figure 6. The kinetic and thermody-
namic parameters for the slippage process in Table 2 were
Importantly, the nature of the anion of the stoppered axle
does affect both the rate of slippage and the association con-
stant value (Figure 5, 6 and S2, Table 2). Slippage was ex-
pected to occur for the hexafluorophosphate stoppered
axles based on the pseudorotaxane studies but unexpectedly,
the macrocycle slips over these axles and reaches equilibri-
um relatively more quickly than for the corresponding axles
with coordinating anions. Ion-pairing is significant in CDCl3
and the rates of slippage are correlated to the strength of
the ion-pair with the slippage rate kon decreasing as the
strength of the ion-pair increases. This suggests that the
mechanism of slippage may involve dissociation of the ion-
pair prior to slippage. It is unlikely that anion binding in the
isophthalamide cavity occurs prior to slippage as the signals
1
of the free macrocycle and stoppered axle in the H NMR
spectra do not shift over time (Figure S1). The mechanism
for slippage is complicated by a number of competing equili-
bria, however, the results suggest that the dissociation of the
ion-pair is the rate-limiting step.
The association constant value is also dependent on the
ion-pair strength and follows the same trend as in the pseu-
dorotaxane studies with the highest association constants for
the chloride stoppered axles. The association constants de-
crease as the ion-pair strength decreases with the exception
of 4·PF6. The decrease in the association constant for the
iodide thread compared with the non-coordinating hexa-
fluorophosphate thread is most likely because the iodide
anion is too large to bind in the anion binding cavity with
the stoppered pyridinium axle. The koff rate constants were
calculated from the Ka and kon values and importantly, these
show that coordinating anions stabilise the rotaxane-like
structure slowing the slippage off rates. For example, the
slippage off rate for 3·X at 298 K is more than ten times
slower in the presence of the coordinating chloride anion
than for the non-coordinating hexafluorophosphate anion.
Hydrogen bonding interactions between the stoppered axle
with a coordinating anion and macrocycle components stabi-
lise the rotaxane-like structure increasing the barrier to slip-
page off.
Temperature has a predictable effect on the slippage rates
and association constants with both of the slippage rates, kon
and koff, increasing and the association constant decreasing
upon raising the temperature from 298 K to 323 K. The
effect of temperature was most significant for slippage with
4.Cl where the time to reach equilibrium was reduced from
over a week to approximately two days by heating at 323 K.
It was hoped that this could be exploited to synthesise a ro-
taxane, as cooling the equilibrated mixture to room temper-
ature would make the barrier to slippage off too high to be
overcome at room temperature.
Figure 6. Change in the concentration of the slippage species formed
from the stoppered axles 4·X in CDCl3 at 323 K. Symbols represent the
experimental data and the lines represent the calculated curves.
Table 2. Kinetic and thermodynamic parameters for the slippage process
obtained from 1H NMR kinetic experiments in CDCl3 at 298 K and
323 K.
298 K
323 K
Ka
[mꢀ1
kon
A
koff
[hꢀ1
Ka
[mꢀ1
kon
A
koff
[hꢀ1
G
]
[mꢀ1 hꢀ1
]
R
]
G
]
]
N
]
3·PF6
3·I
3·Br
3·Cl
4·PF6
4·I
135
146 (5)
1.08
61
176 (3)
2.90
[a]
[a]
[a]
[a]
[a]
[a]
[b]
[b]
[b]
[b]
[b]
[b]
205
97
51
15.55 (0.08)
4.76 (0.05)
2.21 (0.02)
0.0759
0.0488
0.0431
100
61
31
37
78
56 (2)
16.59 (0.07)
8.6 (0.1)
5.4 (0.1)
3.96 (0.04)
0.56
0.27
0.28
0.14
0.051
4·Br
4·Cl
100
1.72 (0.02)
0.017
[c]
[c]
[c]
ACHTUNGTRENNUNG[a,b,c] Parameters for these stoppered axles could not be obtained due to
[a] limited solubility [b] insolubility and [c] the slow rate of slippage.
Standard errors derived from Origin fitting of the experimental data are
shown in parentheses. Errors for Ka and koff are estimated to be less than
5%.
calculated from these data. Solubility problems prevented
the study of slippage with 3·I and 3·Br. In contrast, the stop-
pered axles 4·X were all soluble in CDCl3 although the rates
of slippage were much slower due to the increased level of
steric hindrance from the extra tert-butyl group on the stop-
pers. Accurate data for slippage with 4·Cl at 298 K could not
To investigate whether the species formed were rotaxanes
or pseudorotaxanes, 100 mL CD3OD was added to an equili-
brated mixture of the macrocycle, stoppered thread and slip-
page species in 500 mL CDCl3 at 298 K. Methanol would be
Chem. Eur. J. 2011, 17, 2724 – 2733
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2729

A. J. McConnell, P. D. Beer
expected to disrupt stabilising hydrogen bonding interac-
tions with the anion and p–p interactions between the stop-
pered axle and macrocycle lowering the barrier to slippage
off. A rotaxane species would be stable while a pseudorotax-
ane species would undergo slippage off increasing the
amount of free macrocycle and stoppered axle over time.
The amount of macrocycle was monitored over time by inte-
grating the hydroquinone signals of the macrocycle relative
to the hydroquinone signals of the slippage species. Fig-
ure S3 demonstrates that all of the slippage species formed
are pseudorotaxanes rather than rotaxanes, although the
rate of slippage off is dependent on the anion and the type
of stoppered axle. Unsurprisingly, the species formed with
3.X are completely dissociated within several hours but
those formed with the larger stoppered 4.X are more stable.
taxanes were explored. The enforced planarity of the diyne
unit may restrict the macrocycle into a conformation where
slippage over the stoppers is possible. Removing this planar-
ity after slippage could modify the conformation and size of
the macrocycle cavity stabilising the slippage species as a ro-
taxane where the barrier to slippage off is too high to be
overcome at room temperature. Scheme 5 shows the two
strategies investigated for converting the pseudorotaxane
species formed from equilibration of mixtures of 1 and 3·Cl
or 4·Cl, into rotaxanes. Hydrogenation of the diyne unit was
explored as one strategy using palladium on carbon under a
hydrogen atmosphere in the presence of methanol. Com-
plexation of cobalt carbonyl clusters to the diyne unit was
also investigated, as a number of groups have exploited the
decrease in the C-C-C bond angle[27] upon complexation
with cobalt carbonyl to reduce the size of macrocycle cavi-
ties.[28] For both syntheses, a mixture of 1 and excess stop-
pered axle in chloroform was stirred at room temperature
or 508C, for slippage with 3·Cl and 4·Cl, respectively, for
three days until equilibrium was reached.
Towards the synthesis of kinetically stable rotaxanes: The
kinetic slippage studies have proven that the species formed
by slippage are not rotaxanes, however, the pseudorotaxane
formed with 4·Br has more rotaxane character than the
other species since it has the longest half-life of 6.5 h (see
Table S2 in the Supporting Information). It was hoped that
the species formed from 4·Cl would be more kinetically
stable and therefore, the synthesis and isolation of 11·Cl was
attempted (Scheme 4).
For the hydrogenation reaction, palladium on carbon and
a small amount of methanol as a protic source were added
and the resulting mixture was stirred under a hydrogen at-
mosphere for 18 h. The successful formation of 12·Cl and
13·Cl was indicated by mass spectrometry and by character-
istic upfield shifts and splitting of the hydroquinone protons
1
g and h in the crude H NMR spectra (Figure S5). It did not
prove possible, however, to isolate these species due to the
small quantities formed. While the amount of methanol
used in the hydrogenation step was limited to prevent disso-
ciation of the slippage species, dissociation may have
become significant over the time required for complete hy-
drogenation of the alkyne groups. The hydrogenated macro-
cycle 15 (Scheme 6) also formed as a by-product and the
synthesis of 15 was attempted by hydrogenation of macrocy-
cle 1 using similar reactions conditions. Preliminary slippage
experiments with 15 and 3·Cl at 323 K suggested that this
macrocycle may be suitable for preparing stable rotaxanes
since an almost indetectable quantity of the slippage species
formed after 6 h. However, synthetic difficulties, despite
using a number of synthetic strategies other than hydrogena-
tion, prevented the synthesis of the hydrogenated macrocy-
cle in sufficient purity and quantity for further experimenta-
tion.
An excess of Co2(CO)8 was added to the equilibrated
mixture of 1 and 3·Cl in chloroform and stirred for one hour
under a nitrogen atmosphere. A small amount of 14·Cl con-
taminated with the cobalt carbonyl protected macrocycle 16
(Scheme 6) was isolated following purification by prepara-
tive thin-layer chromatography. Decomposition in solution,
however, prevented isolation and complete characterisation
of pure 14.Cl. Although complete assignment of the
1H NMR spectrum was made difficult by the broadness of
peaks and contamination with 16 (Figure S6b), upfield shifts
and splitting of the macrocycle hydroquinone protons g and
h and the upfield shift of the axle amide protons o are con-
sistent with an interlocked structure. Additionally, the peak
Scheme 4. Synthesis of 11·Cl by slippage.
A mixture of 1 and excess 4·Cl were heated at 508C until
equilibrium was reached, before cooling to room tempera-
ture. The resulting mixture of 1, 4·Cl and 11·Cl was purified
by silica gel chromatography, however, a polar solvent
system including methanol was required to achieve chroma-
tographic separation of the components. A small amount of
11·Cl was isolated contaminated with 1 and 4·Cl, as shown
by the 1H NMR spectrum (Figure S4) and peak at m/z
1751.94 corresponding to [11·ClꢀCl]+ in the electrospray
mass spectrum. While there was evidence of the successful
synthesis of 11·Cl, it is unlikely that it is kinetically stable
enough at room temperature for isolation as evidenced by
the contamination with 1 and 4·Cl after purification. There-
fore, further purification was not attempted as this could po-
tentially lead to further dissociation of 11·Cl.
On account of not being able to isolate any of the slip-
page product species, strategies for their conversion to ro-
2730
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2724 – 2733

Anion Templation in the Slippage Formation of Rotaxane-Like Structures
FULL PAPER
Scheme 5. Strategies for converting the pseudorotaxane formed by slippage into a rotaxane.
Conclusion
A new macrocycle containing a diyne unit was synthesised
exploiting Eglinton coupling and the formation of rotaxane-
like structures from slippage reactions between the macrocy-
cle and stoppered pyridinium axles in the presence of an
anion template in chloroform are reported. Slippage was in-
vestigated via kinetic 1H NMR experiments revealing that
the rates of slippage, kon and koff, are dependent on the
nature of the anion. Interestingly, the highest kon rates are
achieved with the non-coordinating hexafluorophosphate
anion, whereas coordinating anions thermodynamically sta-
bilise the rotaxane-like slippage species resulting in low koff
rates. Addition of methanol, however, results in dissociation
of the slippage species into the macrocycle and axle compo-
nents, which suggests the koff rates are not low enough to ki-
netically trap the slippage species as a rotaxane. Therefore,
the species formed are pseudorotaxanes with the amount of
rotaxane character dependent on the nature of the anion
template. Pseudorotaxanes with more rotaxane character
result from slippage reactions with sterically bulkier stop-
Scheme 6. Hydrogenated macrocycle 15 and cobalt carbonyl protected
macrocycle 16.
at m/z 2210.04 in the electrospray mass spectrum corre-
sponds to [14·ClꢀCl]+ (Figure S6e).
Chem. Eur. J. 2011, 17, 2724 – 2733
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2731

A. J. McConnell, P. D. Beer
N+CH3), 1.30 ppm (s, 54H, tBu); 13C{1H} NMR (125 MHz, CDCl3): d =
158.01, 148.32, 147.09, 144.96, 143.87, 140.95, 135.27, 134.31, 131.63,
130.60, 125.28, 119.52, 63.49, 49.20, 34.29, 31.38 ppm; ESMS: m/z calcd
for C82H94N3O2Cl: 1152.7341; found: 1152.7341 [MꢀCl]+.
pered axles and coordinating anions since the koff rate is sig-
nificantly smaller than the kon rate. Slippage reactions fol-
lowed by hydrogenation or reaction with cobalt carbonyl
were investigated as two strategies for converting the rotax-
ane-like slippage species into rotaxanes. Pleasingly, there
was evidence for the formation of the desired species, but
their isolation proved a challenge due to the small quantities
of rotaxane formed.
4·Br: A solution of 4·I (0.087 g, 0.068 mmol) in CHCl3 (30 mL) was
washed with 1m NH4Br (8ꢃ30 mL). The organic layer was dried over
MgSO4, filtered and the solvent was removed in vacuo to give 4·Br as a
yellow solid in 88% yield (0.073 g, 0.059 mmol). 1H NMR (300 MHz,
CDCl3): d = 10.96 (br s, 2H, NH), 10.51 (s, 1H, pyH4), 8.91 (s, 2H, pyH2
3
and H6), 7.85 (d, J=8.2 Hz, 4H, ArH), 7.27 (s, 16H, ArH), 7.16 (d, 3J=
7.1 Hz, 12H, ArH), 4.07 (s, 3H, N+CH3), 1.30 ppm (s, 54H, tBu); 13C{1H}
NMR (125 MHz, [D6]DMSO): d=159.88, 147.87, 147.50, 143.67, 143.44,
141.60, 135.60, 133.50, 130.90, 129.96, 124.49, 119.57, 62.97, 48.58, 34.08,
31.13 ppm; ESMS: m/z calcd for C82H94N3O2Br: 1152.7341; found:
1152.7342 [MꢀBr]+.
Experimental Section
All commercial-grade chemicals and solvents were used without further
purification unless otherwise stated. Where dry solvents were used, they
were degassed with nitrogen, dried by passing through an MBraun
MPSP-800 column and then used immediately. Triethylamine was dis-
tilled over and stored over potassium hydroxide. Thionyl chloride was
distilled over triphenyl phosphite. TBA salts were stored prior to use
under vacuum in a desiccator containing phosphorus pentoxide and self-
indicating silica gel. Deionised water was used in all cases.
1H, 13C{1H}, 19F and 31P NMR spectra were recorded on a Varian Mercury
300, a Varian Unity Plus 500 or a Bruker AVII500 with cryoprobe spec-
trometer. Mass spectra were obtained on a Bruker micrOTOF or a
MALDI Micro MX spectrometer. Melting points were recorded on a
Gallenkamp capillary melting point apparatus and are uncorrected.
4·PF6: A solution of 4·Cl (0.029 g, 0.024 mmol) in CHCl3 (10 mL) was
washed with 0.2m NH4PF6 (5ꢃ5 mL) and H2O (5 mL). The organic layer
was dried over MgSO4, filtered and the solvent was removed in vacuo to
give 4·PF6 as a yellow solid in 95% yield (0.031 g, 0.023 mmol). M.p.
2608C (decomp); 1H NMR (300 MHz, CDCl3): d = 9.42 (m, 3H, pyH4
and NH), 8.80 (s, 2H, pyH2 and H6), 7.59 (d, 3J=8.2 Hz, 4H, ArH), 7.25
(m, 16H, ArH), 7.12 (d, 3J=8.2 Hz, 12H, ArH), 4.09 (s, 3H, N+CH3),
1.29 ppm (s, 54H, tBu); 13C{1H} NMR (125 MHz, [D6]DMSO): d=
159.88, 147.88, 147.51, 143.67, 143.43, 141.60, 135.62, 133.51, 130.90,
129.96, 124.49, 119.57, 62.97, 48.59, 34.08, 31.13, 30.69 ppm; 19F NMR
(282.5 MHz, CDCl3): d = ꢀ70.54 ppm (d, 1J=715 Hz, PF6ꢀ); 31P NMR
(121.5 MHz, CDCl3):
d
=
ꢀ144.48 ppm (septet, 1J=715 Hz, PF6ꢀ).
ESMS: m/z calcd for C82H94N3O2PF6: 1152.7341; found: 1152.7345
[MꢀPF6]+.
Compounds 2·X,[15,16] 3·X (where X=PF6, I, Br, Cl),[4] 5,[18] 7[15] and 9[17]
were prepared according to literature procedures. The synthesis and char-
acterisation of 6 has been reported previously.[18] A new preparation of 6
is reported in the Supporting Information.
8: A suspension of 7 (0.97 g, 2.22 mmol), 6 (1.42 g, 5.58 mmol) and
Cs2CO3 (1.61 g, 4.93 mmol) in dry degassed DMF (45 mL) was heated at
1008C under a N2 atmosphere for 3 days. The reaction mixture was
cooled to room temperature, filtered and the solvent was removed in
vacuo. Purification by silica gel chromatography (98:2 CH2Cl2/MeOH)
gave 8 as a white solid in 59% yield (0.79 g, 1.31 mmol). M.p. 148–
1508C; 1H NMR (300 MHz, CDCl3): d = 8.21 (s, 1H, isoH), 7.93 (dd,
Macrocycle 1: A solution of CuACHTNUGTRNEUNG(OAc)2 (0.13 g, 0.65 mmol) in CH3CN
(50 mL) was added dropwise to a solution of 8 (0.14 g, 0.23 mmol) in
CH3CN (100 mL), and the mixture was heated under reflux for 16 h,
cooled to room temperature and the solvent was removed in vacuo. The
residue was redissolved in H2O (30 mL) and extracted with CH2Cl2 (5ꢃ
30 mL). The combined organic layers were dried over MgSO4, filtered
and the solvent was removed in vacuo. Following purification by silica
gel chromatography (98:2 CHCl3/MeOH), 1 was isolated as a white solid
4
3
3J=7.7 Hz, J=1.8 Hz, 2H, isoH), 7.53 (t, J=7.7 Hz, 1H, isoH), 6.86 (m,
8H, hydroq ArH), 6.72 (t, 3J=5.3 Hz, 2H, NH), 4.27 (d, 4J=2.4 Hz, 4H,
OCH2CCH), 4.11 (m, 8H, OCH2), 3.88 (m, 8H, NCH2, OCH2), 2.47 ppm
(t, 4J=2.4 Hz, 2H, OCH2CCH); 13C{1H} NMR (75.5 MHz, CDCl3): d =
166.69, 153.23, 152.74, 134.68, 130.05, 129.01, 125.49, 115.71, 115.40, 74.77,
68.29, 67.82, 67.20, 58.54, 39.75 ppm; ESMS: m/z calcd for C36H43N4O8:
659.3075; found: 659.3107 [M+CH3CN+NH4]+.
1
in 63% yield (0.086 g, 0.14 mmol). M.p. 158–1608C; H NMR (300 MHz,
CDCl3): d = 8.05 (s, 1H, isoH), 7.96 (dd, 3J=7.8 Hz, 4J=1.6 Hz, 2H,
isoH), 7.51 (t, 3J=7.8 Hz, 1H, isoH), 6.82 (m, 8H, hydroq ArH), 6.74 (t,
3J=5.6 Hz, 2H, NH), 4.30 (s, 4H, OCH2CCH), 4.09 (m, 8H, OCH2),
3.85 ppm (m, 8H, NCH2, OCH2); 13C{1H} NMR (75.5 MHz, CDCl3): d=
166.85, 153.04, 152.78, 134.59, 130.37, 129.06, 124.76, 115.81, 115.40, 75.36,
70.62, 68.47, 67.84, 67.14, 59.05, 39.78 ppm; ESMS: m/z calcd for
C34H34N2O8Na: 621.2207; found: 621.2203 [M+Na]+.
Acknowledgements
4·I: A suspension of 10 (0.69 g, 0.61 mmol) and iodomethane (5 mL,
excess) in acetone (20 mL) was heated under reflux under a N2 atmos-
phere for 55 h. The reaction mixture was cooled to room temperature
and poured into diethyl ether (100 mL) to precipitate the product. The
solid was collected by filtration, washed with diethyl ether to give 4·I as a
yellow solid in 97% yield (0.76 g, 0.59 mmol). 1H NMR (300 MHz,
CDCl3): d = 10.56 (br s, 2H, NH), 10.12 (s, 1H, pyH4), 8.86 (s, 2H, pyH2
and H6), 7.76 (d, 3J=8.2 Hz, 4H, ArH), 7.24 (m, 16H, ArH), 7.13 (m,
12H, ArH), 4.06 (br s, 3H, N+CH3), 1.29 ppm (s, 54H, tBu), 13C{1H}
NMR (125 MHz, CDCl3): d=158.49, 148.37, 146.69, 145.27, 143.80,
141.16, 134.79, 134.56, 131.61, 130.56, 124.31, 119.63, 63.48, 49.22, 34.30,
31.38 ppm; ESMS: m/z calcd for C82H94N3O2I: 1152.7341; found:
1152.7340 [MꢀI]+.
4·Cl: A solution of 4·I (0.50 g, 0.39 mmol) in CHCl3 (70 mL) was washed
with 1m NH4Cl (8ꢃ70 mL). The organic layer was dried over MgSO4, fil-
tered and the solvent was removed in vacuo. Purification by silica gel
chromatography (95:5 CHCl3/MeOH) gave 4·Cl as a yellow solid in 89%
yield (0.41 g, 0.35 mmol). 1H NMR (300 MHz, CDCl3): d = 10.94 (br s,
2H, NH), 10.59 (s, 1H, pyH4), 8.78 (s, 2H, pyH2 and H6), 7.82 (d, 3J=
8.2 Hz, 4H, ArH), 7.26 (m, 16H, ArH), 7.15 (m, 12H, ArH), 3.85 (s, 3H,
We wish to thank the Woolf Fisher Trust and the Overseas Research Stu-
dent (ORS) Awards Scheme for a scholarship (A. J.M). We also thank
Dr. N. H. Rees for valuable NMR advice and Nicholas White for X-ray
crystallography.
[1] a) M. S. Vickers, P. D. Beer, Chem. Soc. Rev. 2007, 36, 211–225;
b) J. D. Crowley, S. M. Goldup, A.-L. Lee, D. A. Leigh, R. T. McBur-
ney, Chem. Soc. Rev. 2009, 38, 1530–1541; c) B. Champin, P.
Mobian, J.-P. Sauvage, Chem. Soc. Rev. 2007, 36, 358–366; d) S. J.
Loeb, Chem. Soc. Rev. 2007, 36, 226–235; e) S. Saha, J. F. Stoddart,
Chem. Soc. Rev. 2007, 36, 77–92; f) V. Balzani, A. Credi, S. Silvi, M.
Venturi, Chem. Soc. Rev. 2006, 35, 1135–1149; g) E. R. Kay, D. A.
Leigh, F. Zerbetto, Angew. Chem. 2007, 119, 72–196; Angew. Chem.
Int. Ed. 2007, 46, 72–191; h) W. R. Dichtel, O. S. Miljanic, W.
Zhang, J. M. Spruell, K. Patel, I. Aprahamian, J. R. Heath, J. F.
Stoddart, Acc. Chem. Res. 2008, 41, 1750–1761; i) J. F. Stoddart,
Chem. Soc. Rev. 2009, 38, 1802–1820.
[2] A. J. McConnell, C. J. Serpell, A. L. Thompson, D. R. Allan, P. D.
Beer, Chem. Eur. J. 2010, 16, 1256–1264.
2732
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2724 – 2733

Anion Templation in the Slippage Formation of Rotaxane-Like Structures
FULL PAPER
[3] L. M. Hancock, P. D. Beer, Chem. Eur. J. 2009, 15, 42–44.
[4] J. A. Wisner, P. D. Beer, M. G. B. Drew, M. R. Sambrook, J. Am.
Chem. Soc. 2002, 124, 12469–12476.
[5] A. Brown, K. M. Mullen, J. Ryu, M. J. Chmielewski, S. M. Santos, V.
Felix, A. L. Thompson, J. E. Warren, S. I. Pascu, P. D. Beer, J. Am.
Chem. Soc. 2009, 131, 4937–4952.
[15] J. A. Wisner, P. D. Beer, M. G. B. Drew, Angew. Chem. 2001, 113,
3718–3721; Angew. Chem. Int. Ed. 2001, 40, 3606–3609.
[16] M. R. Sambrook, P. D. Beer, J. A. Wisner, R. L. Paul, A. R. Cowley,
F. Szemes, M. G. B. Drew, J. Am. Chem. Soc. 2005, 127, 2292–2302.
[17] H. W. Gibson, S.-H. Lee, P. T. Engen, P. Lecavalier, J. Sze, Y. X.
Shen, M. Bheda, J. Org. Chem. 1993, 58, 3748–3756.
[6] a) Y.-L. Huang, W.-C. Hung, C.-C. Lai, Y.-H. Liu, S.-M. Peng, S.-H.
Chiu, Angew. Chem. 2007, 119, 6749–6753; Angew. Chem. Int. Ed.
2007, 46, 6629–6633; b) J.-P. Collin, J. Frey, V. Heitz, J.-P. Sauvage,
C. Tock, L. Allouche, J. Am. Chem. Soc. 2009, 131, 5609–5620.
[7] a) V. Aucagne, K. D. Hꢄnni, D. A. Leigh, P. J. Lusby, D. B. Walker, J.
Am. Chem. Soc. 2006, 128, 2186–2187; b) K. Hirose, K. Nishihara,
N. Harada, Y. Nakamura, D. Masuda, M. Araki, Y. Tobe, Org. Lett.
2007, 9, 2969–2972.
[18] T. Harada, K. Muramatsu, K. Mizunashi, C. Kitano, D. Imaoka, T.
Fujiwara, H. Kataoka, J. Org. Chem. 2008, 73, 249–258.
[19] a) G. Eglinton, A. R. Galbraith, J. Chem. Soc. 1959, 889–896; b) F.
Vçgtle, I. Michel, R. Berscheid, M. Nieger, K. Rissanen, S. Kotila,
K. Airola, N. Armaroli, M. Maestri, V. Balzani, Liebigs Ann. 1996,
1697–1704.
[20] CCDC-791035 contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[8] C.-W. Chiu, C.-C. Lai, S.-H. Chiu, J. Am. Chem. Soc. 2007, 129,
3500–3501.
[9] P. R. Ashton, M. Belohradsky, D. Philp, J. F. Stoddart, J. Chem. Soc.,
Chem. Commun. 1993, 1269–1274.
[21] M. J. Hynes, J. Chem. Soc. Dalton Trans. 1993, 311–312.
[22] The Job Plot for the titration with H2PO4 was consistent with a 1:2
ꢀ
[10] a) F. M. Raymo, K. N. Houk, J. F. Stoddart, J. Am. Chem. Soc. 1998,
120, 9318–9322; b) F. M. Raymo, J. F. Stoddart, Pure Appl. Chem.
1997, 69, 1987–1988.
[11] M. Asakawa, P. R. Ashton, R. Ballardini, V. Balzani, M. Belohrad-
sky, M. T. Gandolfi, O. Kocian, L. Prodi, F. M. Raymo, J. F. Stoddart,
M. Venturi, J. Am. Chem. Soc. 1997, 119, 302–310.
[12] a) M. A. Bolla, J. Tiburcio, S. J. Loeb, Tetrahedron 2008, 64, 8423–
8427; b) P. Linnartz, S. Bitter, C. A. Schalley, Eur. J. Org. Chem.
2003, 4819–4829; c) S.-Y. Hsueh, C.-C. Lai, Y.-H. Liu, Y. Wang, S.-
M. Peng, S.-H. Chiu, Org. Lett. 2007, 9, 4523–4526; d) H. W.
Gibson, N. Yamaguchi, Z. Niu, J. W. Jones, C. Slebodnick, A. L.
Rheingold, L. N. Zakharov, J. Polym. Sci. Part A: Polym. Chem.
2010, 48, 975–985.
[13] P. R. Ashton, I. Baxter, M. C. T. Fyfe, F. M. Raymo, N. Spencer, J. F.
Stoddart, A. J. P. White, D. J. Williams, J. Am. Chem. Soc. 1998, 120,
2297–2307.
receptor/anion stoichiometry but accurate association constants
could not be determined. The titration data for OAcꢀ could not be
satisfactorily modelled with either a 1:1 or 1:2 receptor/anion stoi-
chiometry.
[23] The slippage species are assumed to be rotaxanes for the purpose of
these calculations, although further kinetic studies determined that
the slippage species are pseudorotaxanes with rotaxane character.
[24] K. J. Laidler, Chemical Kinetics, 3rd ed., Harper Collins, New York,
1987.
[25] OriginPro8, OriginLab Corporation, 1991–2008.
[26] For the purpose of curve fitting, it is assumed that the stoppered
pyridinium axle exists as a tight ion-pair. The low c2 values demon-
strate there is good agreement between the experimental data and
calculated curve, and therefore this assumption is taken as valid.
[27] B. F. G. Johnson, J. Lewis, P. R. Raithby, D. A. Wilkinson, J. Organo-
met. Chem. 1991, 408, C9-C12.
[14] a) P. D. Beer, M. R. Sambrook, D. Curiel, Chem. Commun. 2006,
2105–2117; b) D. Curiel, P. D. Beer, Chem. Commun. 2005, 1909–
1911; c) K. M. Mullen, J. Mercurio, C. J. Serpell, P. D. Beer, Angew.
Chem. 2009, 121, 4875–4878; Angew. Chem. Int. Ed. 2009, 48, 4781–
4784.
[28] a) D. G. Hamilton and J. K. M. Sanders, Chem. Commun. 1998,
1749–1750; b) L. Zhao, B. H. Northrop, P. J. Stang, J. Am. Chem.
Soc. 2008, 130, 11886–11888.
Received: September 1, 2010
Published online: January 24, 2011
Chem. Eur. J. 2011, 17, 2724 – 2733
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2733