Rotaxanes capable of recognising chloride in aqueous media

Article abstract of DOI:10.1002/chem.201002076

A new, versatile chloride-anion-templating synthetic pathway is exploited for the preparation of a series of eight new [2]rotaxane host molecules, including the first sulfonamide interlocked system. ¹H NMR spectroscopic titration investigations

Full text of DOI:10.1002/chem.201002076

DOI: 10.1002/chem.201002076
Rotaxanes Capable of Recognising Chloride in Aqueous Media
Laura M. Hancock,[a] Lydia C. Gilday,[a] Sꢀlvia Carvalho,[b] Paulo J. Costa,[b] Vꢀtor Fꢁlix,[b]
Christopher J. Serpell,[a] Nathan L. Kilah,[a] and Paul D. Beer*[a]
13082
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13082 – 13094

FULL PAPER
Abstract: A new, versatile chloride-
anion-templating synthetic pathway is
media. The interlocked hostꢁs halide
binding affinity can be further en-
hanced and tuned through the attach-
ment of electron-withdrawing substitu-
ents and by increasing its positive
charge. A dicationic rotaxane selective-
ly binds chloride in 35% water, where-
in no evidence of oxoanion binding is
observed. NMR spectroscopy, X-ray
structural analysis and computational
molecular dynamics simulations are
used to account for rotaxane formation
yields, anion binding strengths and se-
lectivity trends.
exploited for the preparation of
a
series of eight new [2]rotaxane host
molecules, including the first sulfona-
mide interlocked system. 1H NMR
spectroscopic titration investigations
demonstrate the rotaxanesꢁ capability
to selectively recognise the chloride
anion in competitive aqueous solvent
Keywords: anions · molecular rec-
ognition · rotaxanes · supramolec-
ular chemistry · template synthesis
Introduction
charge/radius ratio) and weakly pronounced coordination
preferences (no ligand field effects). Hydrogen bonds
formed with anions are weaker and more difficult to control
as compared with metal–cation coordinative bonds. Rare ex-
amples in which specific inorganic anions have been demon-
strated to play a templating role in the synthesis of organic-
based systems have been largely restricted to expanded por-
phyrins[6] and polypyrrolic,[7] hexaurea[8] and peptidic[9] mac-
rocycles.
In spite of the huge interest currently being shown in the
construction of mechanically bonded molecules for nano-
technological applications as molecular switches and ma-
chines,[10] the potential of their unique topological cavities in
host–guest chemistry has been largely neglected.[11] This is
especially the case for the recognition of anions.[12] We have
previously exploited the chloride anion in the templated
synthesis of rotaxane and catenane systems through
Grubbsꢁ-catalysed ring-closing metathesis (RCM) reac-
tions.[13]
Motivated by the fundamental roles negatively charged spe-
cies play in a range of chemical, biological, medical and en-
vironmental processes, the field of anion supramolecular
chemistry has expanded enormously during the past few de-
cades.[1] In spite of the numerous synthetic acyclic and mac-
rocyclic receptors reported to date, the degree of selectivity
exhibited by biological anion-binding proteins far exceeds
any synthetic receptor. Natureꢁs strong and highly selective
binding of sulfate[2] and phosphate[3] anions in water, for ex-
ample, is achieved by an intricate network of hydrogen
bonds buried deep inside specific proteins and not on the
surface exposed to the solvent. This suggests that, to signifi-
cantly “raise the bar” in the level of anion binding strength
and selectivity exhibited by current synthetic receptors, the
design and synthesis of much more elaborate three dimen-
sional receptor systems is required, in which the binding site
is in a protected hydrophobic microenvironment that ena-
bles the total encapsulation of the anionic guest species.
Taking this into account, we surmised that the construction
of such sophisticated target receptors would require unpre-
cedented, innovative, anion-template synthetic methodolo-
gies.
We have discovered, however, that the RCM catalyst is
unsuitable when the axle or macrocycle precursor compo-
nents contain competing ligating functional groups, such as
a pyridyl group. To overcome this problem, we recently re-
ported an alternative synthetic route for the preparation of
Although various serendipitous discoveries in which
anions have been shown to control the assembly of poly-
metallic cluster and cage complexes abound in the litera-
ture,[4] the strategic use of anions as potential templating re-
agents for the assembly of designed molecular architectures
remains very much in its infancy.[5] Importantly, this may be
a consequence of their perceived diffuse nature (small
a new [2]rotaxane, containing a pyridyl macrocyclic
wheel.[14] The synthetic versatility of this new route is illus-
trated herein by the preparation of a series of eight new
[2]rotaxanes, including the first sulfonamide interlocked
system, the three-dimensional interlocked-binding-domain
cavities of which can be tuned to exhibit high degrees of
chloride anion selectivity in competitive aqueous solvent
mixtures.[15] In addition, NMR spectroscopic and X-ray
structural investigations, in combination with computational
molecular dynamics simulations, are used to rationalise the
observed rotaxane formation yields, anion binding affinities
and selectivity trends.
[a] L. M. Hancock, L. C. Gilday, C. J. Serpell, Dr. N. L. Kilah,
Prof. P. D. Beer
Chemistry Research Laboratory, Department of Chemistry
University of Oxford, Mansfield Road, Oxford, OX1 3TA (UK)
Fax : (+44)186-527-2690
E-mail: paul.beer@chem.ox.ac.uk
[b] Dr. S. Carvalho, Dr. P. J. Costa, Prof. V. Fꢂlix
Departamento de Quꢃmica
Results and Discussion
CICECO and Secżo Autꢄnoma de CiÞncias da Safflde
Universidade de Aveiro, 3810-193 Aveiro (Portugal)
In a preliminary report, we described the preparation of a
rotaxane through a new synthetic pathway.[14] By varying the
nature of the condensation reactants and reaction condi-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201002076.
Chem. Eur. J. 2010, 16, 13082 – 13094
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13083

P. D. Beer et al.
tions, the synthetic versatility of this new rotaxanation
method was investigated, with the production of eight new
rotaxane host systems, the orthogonal interlocked-cavity di-
mensions, electronic substituents, degree of positive charge
and macrocycle-ring cavity size of which significantly influ-
ence the degree of selective chloride anion binding affinity
in competitive aqueous solvent media.
through the addition of an appropriate bis(acid chloride) to
a mixture of axle 1·Cl and either the four or five oxygen
polyether-containing bisACTHNURGTNE(NUG amine) in dichloromethane in the
presence of triethylamine (Scheme 1). After preparative
TLC purification the rotaxanes were isolated in 36–60%
yields (Table 1). The lower yields of the rotaxanes contain-
Table 1. Yields of [2]rotaxanes by using the modified synthetic route.
Synthesis of rotaxanes: It was discovered that the yield of
the original rotaxane could be significantly improved upon,
in fact nearly doubled, by limiting the amount of triethyl-
4·Cl
5·Cl
6·Cl
7·Cl
8·Cl
9·Cl
Yield [%]
60
56
55
42
36
41
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
drogen chloride salt)[16] with 3,5-bis(chlorocarbonyl)pyridine
in the presence of a pyridinium axle component 1·Cl[13a] in
dry CH2Cl2 and triethylamine (2.5 equiv) produced rotaxane
5·Cl in an improved isolated yield of 56% (Scheme 1).
Since the pyridinium motif is susceptible to reacting with
nucleophiles,[17] we postulated that excess triethylamine in
the reaction mixture may lead to degradation of the pyridi-
nium axle component. Exploiting this modified synthetic
methodology, a range of new [2]rotaxanes were prepared
ing the shorter polyether chain are attributed to unfavoura-
ble steric interactions between the methyl group of the pyri-
dinium unit and the macrocyclic component, which is corro-
borated by computational molecular dynamics simulations
(vide infra).
1
The H NMR spectrum of rotaxane 4·Cl, along with those
of its component macrocycle[18] and axle 1·Cl for compari-
son, is shown in Figure 1. [2]Rotaxane 4·Cl displays consid-
Figure 1. 1H NMR spectra in CDCl3 at 293 K of axle 1·Cl (top), [2]ro-
taxane 4·Cl (middle) and component macrocycle (bottom).
erable upfield shifts of the amide (d) and para-pyridinium
(g) protons, corresponding to polarisation of the chloride
anion away from the binding cleft as a result of halide anion
recognition by the macrocycleꢁs isophthalamide cavity. This
is further signified by downfield shifts of the amide (d) and
cavity isophthalamide (c) protons as they become deshield-
ed by the halideꢁs negative charge. The proton signals asso-
ciated with the hydroquinone environments (g and h) of the
macrocycle have moved upfield and split appreciably, which
is characteristic of p–p donor–acceptor interactions between
the electron-rich hydroquinones and the positively charged,
electron-deficient pyridinium motif. Finally, the signal corre-
sponding to the methyl protons (a) of the pyridinium axle
has moved significantly upfield, which is indicative of hydro-
gen bonding between these protons and the oxygen atoms
of the macrocycleꢁs polyether chain.
Scheme 1. Synthesis of [2]rotaxanes (4–9)·Cl.
13084
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13082 – 13094

Rotaxanes
FULL PAPER
Further characterisation of the structure of [2]rotaxane
NOEs 5 (e!g,h), 7 (e!c) and 8 (e!d) indicate that the
axle stoppers are spatially close to both the macrocycle
cavity and the hydroquinone protons, confirming that the
axle resides in the macrocycle in the depicted orientation.
Finally, interactions 1 (a!f,i–l) and 2 (a!g,h) reveal that
the pyridinium methyl protons are in close proximity to
both the polyether chain of the macrocycle (through hydro-
gen-bonding interactions previously described) and also to
the hydroquinone protons.
4·Cl was obtained by 2D 1H NMR ROESY spectroscopy
(Figure 2). The 1H NMR ROESY spectrum displays
a
To investigate the effect of increased charge on the anion-
recognition properties of this class of rotaxane host systems,
a novel dicationic [2]rotaxane 10·(I)
Rotaxane 5·Cl was converted to the hexafluorophosphate
salt before reacting with MeI to give 10·(I)(PF6) (Scheme 2).
ACHTUNGTREN(NUGN PF6) was synthesised.
AHCTUNGTRENNUNG
Scheme 2. Synthesis of dicationic [2]rotaxane 10·(I)ACTHNUTRGNE(UNG PF6).
It was also possible to synthesise the first sulfonamide
containing rotaxane 11·Cl by using this methodology
(Scheme 3). The yield of the formation of this rotaxane was
lower than those depicted in Scheme 1, which suggests chlo-
ride anion templation is not as efficient with the sulfon-
AHCTUNGTRENNUNG
tions with phthaloyl and terephthaloyl acid chlorides failed
to produce any evidence of rotaxane formation. These ob-
Figure 2. 1H–1H ROESY spectrum of [2]rotaxane 4·Cl in CD2Cl2 at
293 K. Cross-couplings are shown on the schematic diagram.
number of through-space correlations between the two sepa-
rate components of the [2]rotaxane. Important interactions
include 4 (b!g,h) and 3 (d!g,h), which are characteristic
of p–p stacking between the protons of the electron-rich hy-
droquinone unit and the electron-deficient pyridinium unit,
suggesting that the axle component is threaded through the
centre of the macrocycle. Interaction 6 (d!d) confirms the
cavity amide protons are in close proximity to each other as
they converge into the binding domain to complex chloride.
Scheme 3. Synthesis of sulfonamide-containing [2]rotaxane 11·Cl.
Chem. Eur. J. 2010, 16, 13082 – 13094
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13085

P. D. Beer et al.
servations highlight the importance of the isophthalamide
chloride anion recognition motif in mechanical bond forma-
tion.
Investigations into the role of the templating anion: We car-
ried out a series of experiments in order to gain insight into
the mechanism of this rotaxanation reaction and establish
the importance of the anion in templating the formation of
the interlocked species.
1
From comparison of the H NMR spectrum of an equimo-
lar mixture of axle component 1·Cl and bisACTHNUGRTENUNG(amine) precur-
sor 2 with the spectra of the separate compounds (all in
CDCl3), no evidence of p–p stacking interactions between
the axle pyridinium motif and the hydroquinone units of 2
was observed. This suggests that the rotaxanation reaction
proceeds via an intermediate assembly of components
(Scheme 4), in which an amine has reacted with one acid
Scheme 5. Synthesis of [2]rotaxane 4 by using DCC coupling.
taxanes in dichloromethane and benzene, respectively. Data
collection was undertaken at beamline I19 of Diamond
Light Source.
The structures (Figure 3) confirm the interlocked nature
of the respective systems and the location of the chloride
anion in the expected binding site, encapsulated by the or-
thogonal hydrogen-bonding array. The distance between the
chloride anion and the centre of the binding pocket (Nmass
)
ꢀ
defined by the four nitrogen atoms of the N H binding sites
(two amide nitrogen atoms of the pyridinium axle and two
Scheme 4. Proposed rotaxane intermediate.
chloride group and the chloride anion, through hydrogen
bonding, templates the intramolecular amide ring-closing
condensation reaction around the pyridinium axle compo-
nent to afford the interlocked product. This postulation
places great importance on the templating role of the halide
anion and so it was important to repeat the reaction in the
absence of chloride.
The condensation reaction was carried out by using an ac-
tivated ester, which was synthesised by reacting isophthalic
acid with N-hydroxysuccinamide in the presence of the cou-
pling
reagent
N,N’-dicyclohexylcarbodiimide
(DCC;
Scheme 5). The reaction was carried out both with the chlo-
ride salt and with the hexafluorophosphate salt of the pyridi-
nium axle component. In the presence of the chloride anion
the yield of the rotaxanation reaction was 42%, whereas the
rotaxane was obtained in only 16% yield when 1·PF6 was
used. Thus, the presence of chloride as a templating reagent
significantly enhances the yield of the rotaxane produc-
tion.[19]
X-ray crystal structures of rotaxanes: Crystals suitable for
X-ray diffraction were prepared for rotaxanes 4·Cl and 9·Cl
by gaseous diffusion of diisopropyl into solutions of the ro-
Figure 3. X-ray crystal structures of 4·Cl (top) and 9·Cl (bottom). Disor-
der and non-protic hydrogen atoms were omitted for clarity.
13086
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13082 – 13094

Rotaxanes
FULL PAPER
amide nitrogen atoms of the macrocycle) was calculated to
be 0.567ꢁ0.001 Š for 4·Cl and 0.649ꢁ0.003 Š for 9·Cl, sug-
gesting that the binding pocket of 4 is more favourably dis-
posed to chloride inclusion.
Table 2. The monocationic rotaxanes were observed to bind
chloride strongly even in this competitive aqueous solvent
mixture, whereas the oxoanions were complexed only
The effect of the longer polyether chain in 4·Cl, as op-
posed to 9·Cl, is also apparent. Whereas in 4·Cl the poly-
ether chain appears to have the capacity to swing back and
forth over the pyridinium methyl group with ease, it is less
clear that 9·Cl possesses the necessary degrees of freedom
for this process to occur readily. The interactions between
the polyether and the pyridinium methyl as expressed in the
polyether chain length are critical to synthesis (as demon-
strated above) and anion binding (vide infra). The greater
conformational flexibility in 4·Cl may allow an induced-fit
mechanism to operate more successfully.[20]
ꢀ
Table 2. Association constants, K [mꢀ1], of [2]rotaxanes with Clꢀ, H2PO4
and OAcꢀ in 45:45:10 CDCl3/CD3OD/D2O at 293 K.[a]
ꢀ
Clꢀ
H2PO4
OAcꢀ
4·PF6
5·PF6
6·PF6
7·PF6
8·PF6
9·PF6
1190
1500
1800
860
120
60
170
110
120
150
180
110
180
120
250
200
1450[b]
1175
[a] Obtained by monitoring proton b. Error <10%. [b] Error <15%.
Rotaxane anion binding properties: To study the anion-
binding properties of the rotaxanes, it was necessary to ex-
change the chloride templating anion for the non-coordinat-
ing hexafluorophosphate anion. This was achieved in quanti-
tative yield by washing a solution of the rotaxane in CHCl3
with an NH4PF6(aq) solution. Anion exchange of these [2]ro-
taxanes was observed, by 1H NMR spectroscopy, to have
only a small effect on the position of the cationic pyridinium
axle component within the macrocycle. Indeed, 2D ROESY
spectroscopic evidence (see the Supporting Information) in-
dicates that the structure remains interlocked with the same
coupling interactions between proximal protons observed as
were present in the 2D ROESY spectrum of the chloride
salt of the same rotaxane. The only significant differences
between the chloride and hexafluorophosphate salts of the
[2]rotaxanes are that the signals corresponding to the cavity
protons in which the chloride resided are observed to move
upfield when the templating anion is removed.
weakly, despite the greater basicity of these anions. These
results indicate that the respective interlocked binding pock-
ets are of complementary size to chloride and the pseudo-
tetrahedral binding environment created by the convergent
amide groups is able to satisfy the coordination sphere of
the spherical chloride anion.[22] Furthermore, whilst addition
of chloride to the [2]rotaxanes produces downfield shifts in
cavity pyridinium proton g, addition of either dihydrogen
phosphate or acetate anions causes an upfield perturbation
of this proton signal, presumably as a consequence of a con-
formational change in the [2]rotaxane, since these anions
are too large to penetrate the interlocked binding cavity
and, consequently, associate externally.
Of the monocationic rotaxanes, those with the larger mac-
rocyclic component were observed to complex chloride sig-
nificantly more strongly, which contrasts with what might be
predicted on the grounds of preorganisation. This may be at-
tributed to the binding of chloride causing unfavourable
steric interactions between the pyridinium methyl group of
the axle and the polyether chain of the smaller macrocycle,
which is suggested in the solid-state structure of 4·Cl
(Figure 3) and further rationalised in molecular dynamics
simulations (vide infra).
The anion-binding properties of rotaxanes (4–9)·PF6 were
1
investigated initially by H NMR titration experiments with
ꢀ
TBA salts of Clꢀ, H2PO4 and OAcꢀ in 1:1 CDCl3/CD3OD.
It was possible to monitor the chemical shift perturbations
of cavity protons c and g, and ortho-pyridinium proton b as
a function of the concentration of the anion. However,
winEQNMR2[21] was unable to determine the association
constant values for chloride binding in this solvent system as
the halide is bound extremely strongly (K>104 mꢀ1). In stark
contrast to this, the axle 1·PF6 only weakly binds chloride in
this solvent system (K=125mꢀ1), and exhibits a pronounced
selectivity preference for the basic dihydrogen phosphate
and acetate anions.[13a] Interestingly, the novel sulfonamide-
containing [2]rotaxane 11·PF6 proved to bind chloride rela-
tively weakly, exhibiting an association constant of 1500mꢀ1,
which correlates with the rotaxaneꢁs lower yield and indi-
cates that the halide anion is a less efficient templating re-
agent for this type of compound. Moreover, computational
molecular dynamics simulations support this conjecture
(vide infra).
As expected, the rotaxanes containing the electron-with-
drawing pyridine functionalities, 5·PF6 and nitro-isophthala-
mide 6·PF6, display an enhanced affinity for chloride. This is
a consequence of the increased acidity of the amide protons
and, thus, their hydrogen-bonding ability to the anion in
question.
As an extension of this concept, the novel dicationic [2]ro-
taxane 10·ACHTNUGTRNEUNG(PF6)2 was shown to bind chloride with an associa-
tion constant of 3000mꢀ1 in the 10% water solvent system.
This high association constant is attributed to a combination
of increased charge electrostatics and, thus, increased acidity
of the cavity protons of the rotaxane. Consequently, anion-
binding studies were carried out in a solvent system contain-
ing a higher percentage of water. The limiting factor of
these studies was the solubility of the rotaxane, which con-
tains large lipophilic stopper groups. Therefore, the anion
Analogous anion-binding titrations were undertaken in
the presence of 10% water, and winEQNMR2 was used to
calculate the association constants, which are reported in
properties of rotaxane 10·ACHTNUGTRENUNG(PF6)2 were investigated in 65:35
[D6]acetone/D2O, and the results are given in Table 3.
Chem. Eur. J. 2010, 16, 13082 – 13094
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13087

P. D. Beer et al.
Table 3. Association constants, K [mꢀ1], of [2]rotaxane 10·(PF)2 with Clꢀ,
ꢀ
H2PO4 and OAcꢀ in 65:35 [D6]acetone/D2O at 293 K.[a]
ꢀ
Clꢀ
H2PO4
NB
OAcꢀ
NB
10·(PF)2
500
[a] Obtained by monitoring proton g. Error <10%. NB denotes no bind-
ing.
Remarkably, chloride is still bound with considerable
strength in 35% water, whereas no binding of oxoanions is
observed. This rotaxane displays total selectivity for chloride
over the more basic dihydrogen phosphate and acetate in
this solvent mixture, which is again attributable to the com-
plementary size and shape of chloride for the rotaxaneꢁs
binding cavity.
Molecular dynamics simulations: Further insights into the
formation and anion binding properties of representative ro-
taxanes 4, 7 and 11 were investigated by means of molecular
mechanics (MM) and molecular dynamics (MD) simula-
tions, by using Amber10 software[23] with the general Amber
force field (GAFF)[24] parameters (see Supporting Informa-
tion).
Since the X-ray crystal structures of 4·Cl and 9·Cl reveal
that chloride is bound in
a syn–syn co-conformation
(Figure 3), suitable syn–syn low energy co-conformations for
the rotaxane associations 4·Cl, 4·OAc, 4·H2PO4, 7·Cl, 7·OAc,
7·H2PO4 and 11·Cl were generated in the gas phase through
the quenched MD approach, as described in detail in the
Supporting Information. These co-conformations were then
immersed in an equilibrated cubic-box solvent mixture of
CHCl3/CH3OH/H2O (45:45:10) and subjected to classical
MD simulations for 10 ns.
We start by discussing chloride-templated rotaxane forma-
tion and their stability by analysing the MD simulations for
4·Cl, 7·Cl and 11·Cl. Snapshots of representative binding ar-
rangements taken from these MD simulations are shown in
Figure 4. In all three cases, the main structural features of
the respective binding arrangements found in the gas phase
are retained in solution over the entire time of the MD sim-
ulations and are consistent with the pseudo-tetrahedral
binding arrangement found in the single-crystal X-ray struc-
ture of 4·Cl, in which the chloride anion is bound to the
Figure 4. Snapshots of 4·Cl (top), 7·Cl (middle) and 11·Cl (bottom) taken
at 5 ns of MD simulation. Carbon atoms of the macrocycle and pyridini-
um axle are shown in grey and yellow, respectively. Hydrogen bonds are
represented as orange dashed lines. For clarity, only the N-methyl and
ꢀ
ꢀ
macrocycle and the pyridinium axle by four N H···Cl hy-
drogen bonds. Furthermore, the axleꢁs pyridinium ring is
almost parallel to the two hydroquinone entities, stabilised
by p–p interactions.
ꢀ
N H hydrogen atoms are shown.
For 4·Cl and 7·Cl associations, the Nmass···Clꢀ average dis-
tances are quite short (1.360 and 1.424 Š, respectively) with
small standard deviations, which indicates that, in both
cases, the chloride anion fits well inside the binding pocket,
rangement is interrupted after 8.5 ns of the MD simulation,
leading to the large standard deviation reported. These
structural trends are also evident from comparison of the
probability density function (PDF) of the Nmass···Xꢀ distan-
ces for the three rotaxanes, presented in the Supporting In-
formation. Indeed, for 4·Cl and 7·Cl, sharp distributions cen-
tred on the Nmass···Xꢀ average distances are obtained, while
for 11·Cl the distribution is significantly more diffuse. The
ꢀ
tightly hydrogen bonded to the four N H binding sites. In
contrast, for 11·Cl the average Nmass···Clꢀ distance increases
to 3.391 Š with a very large standard deviation of 2.08 Š,
which suggests that the orthogonal assembly between the
axle and the macrocycle with a sulfonamide motif is not
very stable. In fact, the chloride interlocked rotaxane ar-
N
mass···Xꢀ average distances and their standard deviations, as
13088
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13082 – 13094

Rotaxanes
FULL PAPER
well as the PDFs, correlate very well with the experimental-
ly observed rotaxane formation yields and chloride binding
affinities, which follow the order 4·Cl>7·Cl>11·Cl (see
above). The relatively more stable chloride structural ar-
rangement of 4·Cl, when compared to 7·Cl, correlates with
smaller average distances and standard deviations (with
sharper PDFs) and a larger observed yield. These results are
also consistent with the smaller yield of sulfonamide rotax-
ane 11·Cl that is indicative of less efficient chloride anion
templation, which can be caused by the different conforma-
tional restrictions imposed by the macrocycleꢁs R-SO2-R tet-
togram built with the spatial positions occupied by the N-
methyl carbon atom of the pyridinium over the entire simu-
lation time for both rotaxanes. It is clear that the density of
the methyl carbon atom in 4·Cl is located below the macro-
cycle polyether loop and evenly distributed on both sides of
the macrocyclic cavity, indicating that the methyl group of
the pyridinium can occupy the same plane as the macrocycle
polyether chain and is able to swing back and forth. As a
consequence of this spatial disposition, the average C···O
distance between the N-methyl group and the aliphatic
oxygen atoms of the polyether loop (1.599ꢁ0.41 Š) suggests
ꢀ
ꢀ
rahedral fragment on the N H bond directionality. This can
that the interlocked structure of 4·Cl is stabilised by H2C
be seen in Figure 4 (bottom), in which the two N H binding
H···O (d+–dꢀ) interactions. In contrast, for 7·Cl, the carbon
atom density is always out of the plane of the polyether
chain due to the shorter length of the polyether loop, show-
ing that the size of the loop (i.e., the macrocyclic cavity
size) determines the binding arrangement of the interlocked
structure owing to steric interactions, and ultimately has an
impact on the rotaxane formation (yield) and, subsequently,
itꢁs anion-binding ability.
ꢀ
groups from the macrocycle are clearly out of the plane of
the sulfonamide “head”, whilst in the isophthalamide coun-
terparts (top and middle), they are almost in the plane.
It was suggested earlier that the difference between the
yields of formation of 4·Cl and 7·Cl is attributable to the un-
favourable steric interactions between the methyl group of
the pyridinium axle and the polyether chains of the macro-
cycle. This observation is corroborated by the MD simula-
tions. Figure 5 depicts the grid contours representing the his-
We now focus on the anion-binding properties of the indi-
vidual rotaxane systems 4 and 7 towards Clꢀ, OAcꢀ and
H2PO4 ions. The Nmass···Xꢀ average distance (Table 4) for
ꢀ
Table 4. Average distances (with standard deviations) between the
anions (X=Clꢀ, OAcꢀ or H2PO4<M->) and the centre of the binding
pocket and between the hydroquinone macrocyclic rings (hyd) and pyri-
dinium (pyr) ring axle.
Rotaxane
Distances [Š]
hyd···pyr rings
Nmass···Xꢀ
4·Cl
7·Cl
1.360ꢁ0.23
1.424ꢁ0.33
3.391ꢁ2.08
1.800ꢁ0.18
1.840ꢁ0.23
2.015ꢁ0.21
2.140ꢁ0.25
4.061ꢁ0.37
4.763ꢁ0.45
4.652ꢁ0.51
4.037ꢁ0.36
4.104ꢁ0.54
4.815ꢁ0.74
4.812ꢁ0.50
3.834ꢁ0.34
3.918ꢁ0.32
4.026ꢁ0.41
3.684ꢁ0.25
4.322ꢁ0.58
3.977ꢁ0.37
4.027ꢁ0.41
11·Cl
4·OAc[a]
7·OAc[a]
[b]
4·H2PO4
7·H2PO4
[b]
[a] Distance measured to the centre of mass of the carbon and oxygen
atoms. [b] Distance measured to the P atom. The standard deviations
were calculated for N=50000.
4·Cl (1.360 Š) is shorter than for 4·OAc (1.800 Š) and
4·H2PO4 (2.015 Š), indicating that oxoanions are more
weakly bound than chloride, which is consistent with the ex-
perimental association constants (see Table 2). Furthermore,
this structural parameter also shows that, in contrast to chlo-
ride, the polyatomic anions are located outside the inter-
locked binding pocket, as can be clearly seen in the snap-
shots of 4·OAc and 4·H2PO4 represented in Figure 6.
In the case of 4·OAc, the anion is almost outside the bind-
ing pocket, but it is still possible for the pyridinium methyl
group to be found in the plane of the polyether loop, with
the oxygen atoms of the carboxylate group swapping be-
ꢀ
tween the N H binding sites and keeping the pseudo-tetra-
Figure 5. Grid representation (red) of the histograms for the location of
the methyl carbon atom in the pyridinium axle (yellow) over the 10 ns
simulation of 4·Cl (top) and 7·Cl (bottom). A density contour value of 10
was used.
hedral binding arrangement. In contrast, in 4·H2PO4 the
anion induces a significant distortion of the rotaxane struc-
ture, with increased exposure of the anion to the solvent
Chem. Eur. J. 2010, 16, 13082 – 13094
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13089

P. D. Beer et al.
Figure 7. Superposition of the rotaxane structures 4·Cl (blue), 4·OAc
(green) and 4·H2PO4 (orange) showing the different axle orientations.
ꢀ
second solvent shells of Clꢀ, OAcꢀ and H2PO4 during the
course of the MD simulations.
For all rotaxanes, methanol preferentially solvates the
anions in comparison with water, leading to a higher aver-
age number of molecules of methanol in both the first and
second solvation shells. Chloroform, as expected, solvates
the anions very poorly and the number of molecules of
chloroform only increases significantly when one approaches
the bulk. For example, the second solvation shell of OAcꢀ
bound to 4 and 7 has a significant average number of
chloroform molecules, 2.1 and 2.5, respectively, which is ex-
plained by the fact that the acetate methyl group is far away
from the binding pocket and the 5 Š cut-off approaches the
solution bulk. Since chloroform does not interact significant-
ly with the anions, the remaining discussion will be focused
only on methanol and water molecules.
Figure 6. Snapshots of 4·OAc (top) and 4·H2PO4 (bottom) taken at 6 ns
of MD simulation. The polyatomic anions are drawn in ball and stick
style. The remaining details are given in Figure 4.
medium and with the pyridinium methyl group outside the
polyether chain. These distortions imply a different orienta-
tion of the axle inside the macrocycle, on binding these ox-
oanions, which might account for the notable upfield pertur-
bations of the pyridinium proton (g) observed in the acetate
and dihydrogen phosphate NMR titration experiments (dis-
cussed above). Figure 7 shows the different orientations ob-
served for the pyridinium axle in representative snapshots
of 4·Cl, 4·OAc, and 4·H2PO4.
The first water and methanol solvation shell for the chlo-
ride anion increases in number of molecules in the following
order 4·Cl<7·Cl<11·Cl, meaning that chloride is increasing-
ly more exposed to solvent molecules. The same conclusion
can be drawn from the second solvation shell. The smallest
chloride association constant is therefore expected to be
found for rotaxane 11, due to its labile behaviour. In con-
trast, the strongest association is with rotaxane 4 due to the
As one moves from Clꢀ to
OAcꢀ to H2PO4 , the orthogo-
ꢀ
Table 5. Average number (maximum and minimum in parenthesis) of methanol, water and chloroform mole-
ꢀ
[a]
cules closer than 3.4 Š (1st shell) and 5 Š (2nd shell) to the anion (Clꢀ, OAcꢀ and H2PO4
Methanol Water
2nd shell
)
for 4, 7 and 11.
nal coordination environment
becomes increasingly distorted.
The number of solvent mole-
cules of methanol, water and
chloroform around the anions
may also be useful for under-
standing the anion-binding pro-
cess in rotaxanes 4, 7 and 11.
Table 5 shows the average
number of solvent molecules
Chloroform
1st shell
2nd shell
1st shell
1st shell
2nd shell
4
Clꢀ
1.5 (0–4)
1.4 (0–6)
2.0 (0–5)
1.6 (0–4)
1.5 (0–6)
2.7 (0–6)
1.9 (0–7)
3.2 (0–7)
3.1 (0–10)
3.2 (0–7)
3.3 (0–8)
3.3 (0–9)
4.2 (1–9)
3.7 (0–14)
0.7 (0–4)
0.4 (0–4)
0.8 (0–4)
0.9 (0–7)
0.3 (0–3)
0.6 (0–4)
1.2 (0–9)
1.2 (0–6)
0.9 (0–6)
1.2 (0–8)
1.6 (0–11)
0.6 (0–5)
1.1 (0–6)
2.2 (0–19)
0.0 (0–2)
0.3 (0–3)
0.0 (0–2)
0.0 (0–1)
0.4 (0–3)
0.0 (0–2)
0.0 (0–2)
0.4 (0–4)
2.1 (0–6)
0.8 (0–4)
0.3 (0–4)
2.5 (0–7)
0.5 (0–4)
0.2 (0–3)
ꢀ
OAcꢀ
H2PO4
Clꢀ
ꢀ
ꢀ
7
OAcꢀ
H2PO4
Clꢀ
11
[a] For OAcꢀ, the distances are measured to the centre of mass formed by the O and C atoms. For H2PO4 the
found within the first and distances are measured to P.
13090
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13082 – 13094

Rotaxanes
FULL PAPER
more “buried” position of Clꢀ inside the binding pocket,
leading to a relatively small exposure to solvent molecules.
The effect of the size of the polyether chain on the bind-
ing affinities of 4 and 7 for the polyatomic anions is also evi-
dent. As for chloride, the first and second solvation shells of
the anions increases in number of molecules with decreasing
chain size, suggesting that the association constants with
ing through an MBraun MPSP-800 column and then used immediately.
Triethylamine was distilled from and stored over potassium hydroxide.
Water was deionised and microfiltered by using a Milli-Q Millipore ma-
chine.
1H, 13C{1H}, 19F{1H} and 31P{1H} NMR spectra were recorded on Varian
Mercury-VX 300 and Bruker AVII500 spectrometers. Mass spectrometry
was carried out on a Bruker micrOTOF spectrometer. All new rotaxanes
were characterised by H and 13C{1H} NMR spectroscopy and ESMS.
Axles 1·Cl and 1·PF6,[13a] bis(amine) 2,[16] and rotaxanes 5·Cl and 5·PF6
ACTHNUTRGNEUNG
1
[14]
OAcꢀ and H2PO4 should be of larger magnitude for 4 than
ꢀ
were prepared according to reported procedures. The synthesis of bis-
AHCTUNGTREGUN(NN amine) 3 is given in the Supporting Information.
for 7. This is indeed observed in the experimentally deter-
mined association constants.
General procedure for rotaxane synthesis
Method 1: Axle 1·Cl (100 mg, 0.093 mmol) and the appropriate bis-
AHCTUNGTRENNUNG
Conclusion
a
0.23 mmol) was added to the reaction mixture, followed immediately by
addition of the appropriate acid chloride (0.093 mmol) in dry CH2Cl2
(5 mL). The reaction was stirred under N2 for 1 h, before washing with
10% HCl(aq) (2”25 mL) and H2O (2”25 mL). The organic layer was
dried over anhydrous MgSO4, the solvent removed in vacuo and the re-
sulting crude yellow solid purified by preparative TLC.
With the objective of highlighting the potential advantages
of interlocked host systems for anion-recognition applica-
tions, a series of eight new [2]rotaxane host molecules have
been prepared, including the first sulfonamide interlocked
system, through a new, versatile chloride-anion-templating
pathway. 1H NMR anion-titration-binding investigations
demonstrate the ability of the rotaxanesꢁ three-dimensional
binding domains to exhibit high degrees of chloride anion
selectivity in competitive aqueous solvent media. The inter-
locked hostꢁs chloride-anion-binding affinity can be further
enhanced and tuned through the attachment of electron
withdrawing substituents to the macrocycleꢁs isophthalamide
motif and by increasing the positive charge with pyridinium
group modification. In particular, the orthogonal cavity dis-
Method 2: Isophthalic acid (8.3 mg, 0.05 mmol) was suspended in dry
CH3CN (5 mL). This was followed by addition of N-hydroxysuccinamide
(13.8 mg, 0.12 mmol) and then N,N’-dicyclohexylcarbodiimide (22.7 mg,
0.11 mmol). The reaction mixture was stirred at RT for 16 h, then fil-
tered. The filtrate was concentrated in vacuo, re-dissolved in dry CH2Cl2
(2 mL) and added to a solution of axle 1 (0.046 mmol), bisACHTUNGTRENNUNG(amine) 2
(21.5 mg, 0.046 mmol) and Et3N (16 mL, 0.115 mmol) in CH2Cl2 (10 mL).
The solution was stirred under a nitrogen atmosphere for 16 h, then
washed with 10% HCl(aq) (2”15 mL) and H2O (2”15 mL), dried over
anhydrous MgSO4 and the solvent removed in vacuo. The resulting crude
material was purified by preparative TLC (silica; 95:5 CHCl3/MeOH) to
yield the product as a yellow solid.
position of the two 3,5-bis
motifs of the dicationic rotaxane 10·AHCUTNGTRENNUNG
ACHTUNGTRENNUNG
General procedure for anion exchange: The rotaxane was dissolved in
CHCl3 (15 mL) and washed with NH4PF6(aq) solution (0.1m; 10”10 mL)
and H2O (2”10 mL). The organic layer was dried over MgSO4 and the
solvent removed in vacuo to give a yellow solid in quantitative yield.
bles this interlocked host to recognise chloride in 35%
water, wherein no evidence of the binding of oxoanions, ace-
tate or dihydrogen phosphate, is detected. The larger, basic
oxoanions cannot penetrate the rotaxaneꢁs interlocked,
buried binding pocket. The combination of NMR spectros-
Rotaxane 4·Cl (method 1): BisACHTNUTRGNE(NUG amine) 2 (60 mg, 0.093 mmol) and iso-
phthaloyl dichloride (19 mg, 0.093 mmol) were used. Preparative TLC
(silica; 95:5 CHCl3/MeOH) gave a yellow solid (91 mg, 60%).
Rotaxane 4·Cl (method 2): Axle 1·Cl (50 mg, 0.046 mmol) was used.
Preparative TLC (silica; 95:5 CHCl3/MeOH) gave a yellow solid (32 mg,
41%). 1H NMR (300 MHz, CDCl3): d=10.40 (s, 2H; axle NH), 10.06 (s,
1H; pyridinium H4), 9.02 (s, 2H; pyridinium H2, H6), 8.91 (s, 1H; iso-
ACHTUNGTRENNUNGcopy, X-ray structural analysis and molecular-dynamics sim-
ulations were used to explain the rotaxane formation yields,
anion-binding strengths and selectivity trends. Importantly,
it is the near total encapsulation of the chloride anion guest
by the respective rotaxaneꢁs isophthalamide macrocycle and
pyridinium axle binding groups that results in the efficacy of
halide recognition and templation. Molecular dynamics sim-
ulation studies show that as chloride is increasingly exposed
to solvent molecules, as is the case with the unfavourable
conformational restrictions imposed by the sulfonamide in-
terlocked rotaxane system, the halide anion is complexed
more weakly and is a less efficient templating reagent. The
design and manipulation of the unique topological cavities
of mechanically bonded molecules for anion recognition and
sensing applications is continuing in our laboratories.
3
phthalamide H2), 8.54 (brs, 2H; isophthalamide NH), 7.99 (d, J=7.5 Hz,
3
2H; isophthalamide H4, H6), 7.83 (d, J=8.4 Hz, 4H; ArHNH), 7.27–7.05
(m, 31H; ArH, isophthalamide H5), 6.44 (d, 3J=8.6 Hz, 4H; hydroqui-
none ArH), 6.16 (d, 3J=8.6 Hz, 4H; hydroquinone ArH), 4.66 (s, 3H;
3
N+CH3), 4.11 (t, J=4.5 Hz, 4H; CH2 ), 3.74–3.71 (m, 20H; CH2 ),
1.33 ppm (s, 36H; tBu); 13C{1H} NMR (75.5 MHz, CDCl3): d=167.0,
158.2, 153.7, 151.7, 148.3, 146.9, 145.3, 144.5, 143.5, 134.8, 133.6, 133.4,
131.7, 131.5, 131.0, 130.6, 128.5, 127.3, 125.7, 124.9, 124.2, 120.2, 114.7,
114.6, 70.7, 70.5, 70.0, 68.2, 65.6, 63.8, 49.6, 40.7, 34.3, 31.3 ppm; HRMS
(ESI): m/z: calcd for C106H116N5O11: 1635.8699; found: 1635.8654
[MꢀCl]+.
ꢀ
ꢀ
ꢀ
ꢀ
Rotaxane 4·PF6 (general procedure for anion exchange): By using rotax-
ane 4·Cl (75 mg, 0.045 mmol) to give 4·PF6 as a yellow solid (80 mg,
100%).
Rotaxane 4·PF6 ACHTNUTRGNEUNG(method 2): Axle 1·PF6 (55 mg, 0.046 mmol) was used.
Preparative TLC (SiO2; 95:5 CHCl3/MeOH) gave a yellow solid (15 mg,
20%). 1H NMR (300 MHz, CDCl3): d=9.26 (s, 2H; axle NH), 8.78 (s,
2H; pyridinium H2, H6), 8.66 (s, 1H; pyridinium H4), 8.22 (s, 1H; iso-
phthalamide H2), 8.09 (d, 3J=7.6 Hz, 2H; isophthalamide H4, H6), 7.62
(d, 3J=8.4 Hz, 4H; ArHNH), 7.44 (t, 3J=7.3 Hz, 1H; isophthalamide
H5), 7.26–7.07 (m, 32H; ArH, isophthalamide NH), 6.57 (d, 3J=8.9 Hz,
Experimental Section
General considerations: Commercially available solvents and chemicals
were used without further purification unless otherwise stated. Where
dry solvents were used, they were degassed with nitrogen, dried by pass-
4H; hydroquinone ArH), 6.34 (d, 3J=8.9 Hz, 4H; hydroquinone ArH),
+
ꢀ
ꢀ
ꢀ
ꢀ
4.26 (s, 3H; N CH3), 4.11 (m, 4H; CH2 ), 4.00–3.95 (m, 20H; CH2 ),
Chem. Eur. J. 2010, 16, 13082 – 13094
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13091

P. D. Beer et al.
1.32 ppm (s, 36H; tBu); 13C{1H} NMR (75.5 MHz, CDCl3): d=166.9,
158.6, 152.9, 151.7, 148.6, 146.8, 145.0, 144.4, 143.3, 134.5, 133.9, 131.9,
131.6, 131.0, 130.6, 129.0, 127.4, 125.8, 124.3, 123.5, 119.8, 115.9, 114.7,
70.6, 70.4, 70.0, 67.5, 66.9, 63.8, 49.3, 39.7, 34.3, 31.3 ppm; 19F{1H} NMR
(282.4 MHz, CDCl3): d=ꢀ70.0 ppm (d, 1J=710 Hz, PF6ꢀ); 31P{1H} NMR
(121.6 MHz, CDCl3): d=ꢀ144.0 ppm (sept, 1J=710 Hz, PF6ꢀ); HRMS
(ESI): m/z: calcd for C106H116N5O11: 1635.8699; found: 1635.8691
[MꢀPF6]+.
19F{1H} NMR (282.4 MHz 1:1 CDCl3/CD3OD): d=ꢀ74.5 ppm (d, 1J=
710 Hz, PF6ꢀ); 31P{1H} NMR (121.6 MHz, 1:1 CDCl3/CD3OD): d=
ꢀ143.7 ppm (sept, 1J=710 Hz, PF6ꢀ); HRMS (ESI): m/z: calcd for
C104H112N5O10: 1591.8443; found: 1591.8391 [MꢀPF6]+.
Rotaxane 8·Cl (method 1): BisACHTNUTRGNE(NUG amine) 3 (46 mg, 0.093 mmol) and 3,5-
pyridine dicarbonyl dichloride (16 mg, 0.093 mmol) were used. Prepara-
tive TLC (silica; 9:1 CHCl3/MeOH) gave a yellow solid (55 mg, 36%).
1H NMR (300 MHz, 1:1 CDCl3/CD3OD): d=9.58 (s, 1H; pyridinium H4),
9.51 (s, 2H; pyridinium H2 & H6), 9.25 (t, 4J=2.1 Hz, 1H; pyridine H4),
9.09 (d, 4J=2.1 Hz, 2H; pyridine H2, H6), 7.68 (d, 3J=8.8 Hz, 4H;
HNArH), 7.26–7.03 (m, 30H; ArH), 6.42 (d, 3J=9.1 Hz, 4H; hydroqui-
Rotaxane 6·Cl (method 1): BisACTHUNRTGNEUNG(amine) 2 (60 mg, 0.093 mmol) and 5-nitro-
isophthaloyl dichloride (23 mg, 0.093 mmol) were used. Preparative TLC
(silica; 93:7 CHCl3/MeOH) gave a yellow solid (87 mg, 55%). 1H NMR
(300 MHz, CDCl3): d=10.27 (s, 2H; axle NH), 10.02 (s, 1H; pyridinium
H4), 9.37 (s, 1H; isophthalamide H4), 8.99 (s, 2H; isophthalamide H2,
none ArH), 6.13 (d, 3J=9.1 Hz, 4H; hydroquinone ArH), 4.56 (s, 3H;
3
N+CH3), 4.06 (t, J=4.6 Hz, 4H; CH2 ), 3.90–3.86 (m, 12H; CH2 ),
ꢀ
ꢀ
ꢀ
ꢀ
3.66 (t, J=4.6 Hz, 4H; CH2 ), 1.29 ppm (s, 36H; tBu); 13C{1H} NMR
(75.5 MHz, 1:1 CDCl3/CD3OD): d=166.5, 159.8, 153.7, 152.8, 152.6,
149.5, 147.8, 145.9, 144.5, 139.0, 135.5, 134.8, 133.6, 132.6, 131.8, 131.5,
129.4, 128.3, 126.7, 125.2, 121.6, 115.5, 114.8, 71.7, 71.0, 68.1, 66.4, 64.8,
3
4
3
H6), 8.91 (d, J=1.2 Hz, 2H; pyridinium H2, H6), 8.68 (t, J=3.8 Hz, 2H;
isophthalamide NH), 7.86 (d, 3J=8.8 Hz, 4H; ArHNH), 7.26–6.99 (m,
30H; ArH), 6.47 (d, 3J=9.1 Hz, 4H; hydroquinone ArH), 6.17 (d, 3J=
9.1 Hz, 4H; hydroquinone ArH), 4.71 (s, 3H; N+CH3), 4.13 (t, 3J=
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
41.6, 35.0, 31.9, 30.5 ppm; HRMS (ESI): m/z: calcd for C103H111N6O10
:
4.7 Hz, 4H; CH2 ), 3.77–3.72 (m, 20H; CH2 ), 1.32 ppm (s, 36H;
tBu); 13C{1H} NMR (75.5 MHz, CDCl3): d=158.0, 153.3, 151.8, 148.5,
146.7, 145.3, 144.8, 143.3, 137.1, 135.2, 134.7, 133.6, 131.8, 130.9, 130.5,
130.0, 127.3, 126.4, 125.8, 124.2, 120.3, 114.7, 70.7, 70.5, 70.0, 68.4, 65.4,
63.7, 49.8, 41.0, 34.3, 31.3 ppm; HRMS (ESI): m/z: calcd for
C106H116N6O13: 1680.8556; found: 1680.8631 [MꢀCl]+.
1592.8395; found: 1592.7850 [MꢀCl]+.
Rotaxane 8·PF6 (general procedure for anion exchange): By using rotax-
ane 8·Cl (50 mg, 0.031 mmol) to give 8·PF6 as a yellow solid (53 mg,
100%). 1H NMR (300 MHz, 1:1 CDCl3/CD3OD): d=9.36 (s, 2H; pyridi-
4
nium H2, H6), 9.10 (d, J=2.1 Hz, 2H; pyridine H2, H6), 9.06 (s, 1H; pyri-
dinium H4), 9.00 (t, 4J=2.1 Hz, 1H; pyridine H4), 7.55 (d, 3J=8.8 Hz,
Rotaxane 6·PF6 (general procedure for anion exchange): By using rotax-
ane 6·Cl (75 mg, 0.044 mmol) to give 6·PF6 as a yellow solid (80 mg,
3
4H; HNArH), 7.26–7.06 (m, 30H; ArH), 6.43 (d, J=9.1 Hz, 4H; hydro-
3
1
quinone ArH), 6.22 (d, J=9.1 Hz, 4H; hydroquinone ArH), 4.58 (s, 3H;
100%). H NMR (300 MHz, CDCl3): d=9.93 (brs, 2H; axle NH), 8.97 (s,
+
3H; pyridinium H2, H4, H6), 8.65 (s, 2H; isophthalamide H2, H6), 8.51 (s,
ꢀ
ꢀ
ꢀ
ꢀ
N CH3), 3.94 (brs, 4H; CH2 ), 3.89–3.87 (m, 12H; CH2 ), 3.55 (brs,
4H; CH2 ), 1.29 ppm (s, 36H; tBu); 13C{1H} NMR (75.5 MHz, 1:1
CDCl3/CD3OD): d=166.4, 159.5, 153.43, 152.4, 151.8, 149.4, 147.7, 147.4,
146.1, 144.3, 141.1, 135.2, 133.7, 132.6, 131.7, 131.4, 130.1, 128.2, 126.7,
125.1, 121.4, 115.6, 114.8, 71.4, 70.9, 68.0, 66.7, 64.7, 40.4, 35.0, 31.8 ppm;
19F{1H} NMR (282.4 MHz, 1:1 CDCl3/CD3OD): d=ꢀ73.0 ppm (d, 1J=
712 Hz, PF6ꢀ); 31P{1H} NMR (121.6 MHz, 1:1 CDCl3/CD3OD): d=
ꢀ144.2 ppm (sept, 1J=712 Hz, PF6ꢀ); HRMS (ESI): m/z: calcd for
1H isophthalamide H4), 7.65 (d, J=8.5 Hz, 4H; ArHNH), 7.28–7.07 (m,
2
ꢀ
ꢀ
32H; ArH, isophthalamide NH), 6.64 (d, 2J=8.1 Hz, 4H; hydroquinone
ArH), 6.35 (d, 2J=8.1 Hz, 4H; hydroquinone ArH), 4.06–4.03 (m, 7H;
+
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
N CH3, CH2 ), 3.75–3.70 (m, 16H; CH2 ), 3.57 (brs, 4H; CH2 ),
1.32 ppm (s, 36H; tBu); 13C{1H} NMR (75.5 MHz, CDCl3): d=164.5,
158.8, 153.0, 151.6, 148.7, 146.7, 145.2, 144.6, 143.3, 136.1, 134.5, 134.0,
132.0, 131.0, 130.5, 127.4, 126.3, 125.9, 124.3, 119.8, 115.9, 114.7, 70.6,
70.5, 70.1, 67.3, 66.7, 63.8, 49.4, 39.9, 34.3, 31.3 ppm; 19F{1H} NMR
(282.4 MHz, CDCl3): d=ꢀ70.0 ppm (d, 1J=713 Hz, PF6ꢀ); 31P{1H} NMR
(121.6 MHz, CDCl3): d=ꢀ144.0 ppm (sept, 1J=713 Hz, PF6ꢀ); HRMS
C
103H111N6O10: 1592.8395; found: 1592.8319 [MꢀPF6]+.
Rotaxane 9·Cl (method 1): BisACTHUNRTGNEUNG(amine) 3 (46 mg, 0.093 mmol) and 5-nitro-
isophthaloyl dichloride (23 mg, 0.093 mmol) were used. Preparative TLC
(silica; 93:7 CHCl3/MeOH) gave a yellow solid (64 mg, 41%). 1H NMR
(300 MHz, 1:1 CDCl3/CD3OD): d=9.62 (s, 1H; pyridinium H4), 9.52 (s,
2H; pyridinium H2, H6), 9.30 (brt, 4J=1.4 Hz, 1H; isophthalamide H2),
(ESI): m/z: calcd for C106H116N6O13
[MꢀPF6]+.
: 1680.8556; found: 1680.8534
Rotaxane 7·Cl (method 1): BisACHTNUTRGNE(NUG amine) 3 (46 mg, 0.093 mmol) and iso-
4
3
8.84 (d, J=1.4 Hz, 2H; isophthalamide H4, H6), 7.72 (d, J=8.7 Hz, 4H;
phthaloyl dichloride (19 mg, 0.093 mmol) were used. Preparative TLC
(silica; 9:1 CHCl3/MeOH) gave a yellow solid (62 mg, 42%). 1H NMR
(300 MHz, 1:1 CDCl3/CD3OD): d=9.58 (s, 1H; pyridinium H4), 9.46 (s,
2H; pyridinium H2, H6), 8.80 (t, 4J=1.5 Hz, 1H; isophthalamide H2),
7.95 (dd, 3J=7.9, 4J=1.5 Hz, 2H; isophthalamide H4, H6) 7.68 (d, 3J=
8.7 Hz, 4H; ArHNH), 7.25–7.05 (m, 31H; ArH, isophalamide H5), 6.38
HNArH), 7.25–6.99 (m, 30H; ArH), 6.42 (d, 3J=9.1 Hz, 4H; hydroqui-
none ArH), 6.13 (d J=9.1 Hz, 4H; hydroquinone ArH), 4.58 (s, 3H; N+
3
3
ꢀ
ꢀ
ꢀ
ꢀ
CH3), 4.10 (t, J=4.4 Hz, 4H; CH2 ), 3.89–3.84 (m, 12H; CH2 ), 3.69
(t, 3J=4.4 Hz, 4H; CH2
)
1.30 ppm (s, 36H; tBu); 13C{1H} NMR
ꢀ
ꢀ
(75.5 MHz, 1:1 CDCl3/CD3OD): d=165.7, 159.4, 153.6, 152.6, 149.3,
149.2, 147.5, 144.1, 135.8, 135.2, 133.4, 132.4, 131.6, 131.2, 128.0, 126.9,
126.5, 124.9, 121.2, 115.3, 114.6, 71.5, 70.9, 67.9, 66.1, 64.5, 41.8, 34.9, 31.8,
30.3 ppm; HRMS (ESI): m/z: calcd for C106H114N7O12: 1636.8394; found:
1636.7740 [MꢀCl]+.
(d, 3J=9.0 Hz, 4H; hydroquinone ArH), 6.11 (d, 3J=9.0 Hz, 4H; hydro-
quinone ArH), 4.44 (s, 3H; N+CH3), 4.08 (t, J=4.7 Hz, 4H; CH2 ),
3
ꢀ
ꢀ
3
ꢀ
ꢀ
ꢀ
ꢀ
3.88–3.85 (m, 12H; CH2 ), 3.67 (t, J=4.7 Hz, 4H; CH2 ), 1.29 ppm
(s, 36H; tBu); 13C{1H} NMR (75.5 MHz, 1:1 CDCl3/CD3OD): d=168.9,
159.5, 153.7, 152.5, 149.3, 147.7, 145.5, 144.4, 139.6, 135.5, 134.2, 133.5,
132.4, 132.1, 131.7, 131.4, 129.6, 128.1, 126.5, 125.0, 121.1, 115.3, 114.7,
71.6, 70.9, 67.9, 66.3, 64.6, 41.3, 34.9, 31.8, 30.3 ppm; HRMS (ESI): m/z:
calcd for C104H112N5O10: 1591.8443; found: 1591.7578. [MꢀCl]+.
Rotaxane 9·PF6 (general procedure for anion exchange): By using rotax-
ane 9·Cl (50 mg, 0.030 mmol) to give 9·PF6 as a yellow solid (53 mg,
100%). 1H NMR (300 MHz, 1:1 CDCl3/CD3OD): d=9.41 (s, 2H; pyridi-
nium H2, H6), 9.15 (s, 1H; pyridinium H4), 9.06 (s, 1H; isophthalamide
H2), 8.84 (s, 2H; isophthalamide H4, H6), 7.57 (d, 3J=8.8 Hz, 4H;
HNArH), 7.25–7.03 (m, 30H; ArH), 6.43 (d, 3J=8.4 Hz, 4H; hydroqui-
Rotaxane 7·PF6 (general procedure for anion exchange): By using rotax-
ane 7·Cl (50 mg, 0.031 mmol) to give 7·PF6 as a yellow solid (53 mg,
100%). 1H NMR (300 MHz, 1:1 CDCl3/CD3OD): d=9.91 (s, 2H; pyridi-
nium NH), 9.27 (s, 2H; pyridinium H2, H6), 9.04 (s, 1H; pyridinium H4),
8.50 (s, 1H; isophthalamide H2), 8.10 (brs, 2H; isophthalamide NH), 7.94
none ArH), 3.97–3.58 (d, 3J=8.4 Hz, 4H; hydroquinone ArH), 4.59 (s,
+
ꢀ
ꢀ
ꢀ
ꢀ
3H; N CH3), 3.97 (brs, 4H; CH2 ), 3.89–3.87 (m, 12H; CH2 ), 3.58
(brs, 4H; CH2 ), 1.30 ppm (s, 36H; tBu); 13C{1H} NMR (125 MHz, 1:1
CDCl3/CD3OD): d=165.5, 159.2, 153.5, 152.4, 149.1, 147.4, 145.4, 144.0,
135.7, 135.1, 133.3, 132.2, 131.5, 131.1, 129.0, 127.9, 126.8, 126.3, 124.8,
121.0, 115.2, 114.4, 71.4, 70.7, 69.7, 66.0, 64.3, 53.5, 41.7, 34.7, 31.7 ppm;
19F{1H} NMR (282.4 MHz, 1:1 CDCl3/CD3OD): d=ꢀ73.0 ppm (d, 1J=
712 Hz, PF6ꢀ); 31P{1H} NMR (121.6 MHz, 1:1 CDCl3/CD3OD): d=
ꢀ144.0 ppm (sept, 1J=712 Hz, PF6ꢀ); HRMS (ESI): m/z: calcd for
C106H114N7O12: 1636.8394; found: 1636.8242 [MꢀPF6]+.
ꢀ
ꢀ
4
3
(dd, 3J=7.9, J=1.6 Hz, 2H; isophthalamide H4, H6), 7.51 (d, J=8.8 Hz,
4H; ArHNH), 7.39 (t, J=7.9 Hz, 1H; isophthalamide H5), 7.26–7.06 (m,
3
30H; ArH), 6.43 (d, 3J=9.1 Hz, 4H; hydroquinone ArH), 6.22 (d, 3J=
9.1 Hz, 4H; hydroquinone ArH), 4.51 (s, 3H; N+CH3), 3.94 (t, 3J=
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
4.6 Hz, 4H; CH2 ), 3.88–3.87 (m, 12H; CH2 ), 3.59–3.54 (m, 4H;
CH2 ), 1.29 ppm (s, 36H; tBu); 13C{1H} NMR (75.5 MHz, 1:1 CDCl3/
CD3OD): d=168.4, 159.2, 153.4, 152.2, 149.1, 147.5, 145.4, 144.1, 135.1,
134.1, 133.5, 132.3, 131.6, 131.2, 129.4, 127.9, 126.4, 125.3, 124.9, 121.5,
120.8, 115.2, 114.6, 71.3, 70.7, 67.7, 66.3, 64.4, 40.8, 34.8, 31.7 ppm;
ꢀ
Rotaxane 10·
ACHTNUGTNERN(UGN PF6)2: Rotaxane 8·PF6 (50 mg, 0.030 mmol) was dissolved
in MeI (5 mL) and the solution was heated at reflux under a nitrogen at-
13092
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13082 – 13094

Rotaxanes
FULL PAPER
mosphere for 16 h. The solvent was removed in vacuo, and the yellow
residue re-dissolved in CHCl3 and washed with NH4PF6(aq) (0.1m; 10”
10 mL) and H2O (2”10 mL). The organic layer was dried over MgSO4
and the solvent removed in vacuo to give a yellow solid (51 mg, 95%).
1H NMR (500 MHz, 1:1 CDCl3/CD3OD): d=9.42 (s, 1H; axle pyridinium
H4), 9.39 (s, 2H; axle pyridinium H2, H4), 9.19 (s, 1H; macrocycle pyridi-
ler, P. A. Gale, W.-S. Cho, Anion Receptor Chemistry, RSC, Cam-
bridge, 2006; c) F. P. Schmidtchen, M. Berger, Chem. Rev. 1997, 97,
1609–1646; d) P. D. Beer, P. A. Gale, Angew. Chem. 2001, 113, 502–
532; Angew. Chem. Int. Ed. 2001, 40, 486–516; e) M. D. Best, S. L.
Tobey, E. V. Anslyn, Coord. Chem. Rev. 2003, 240, 3–15; f) J. M. Lli-
nares, D. Powell, K. Bowman-James, Coord. Chem. Rev. 2003, 240,
57–75; g) K. Choi, . A. D. Hamilton, Coord. Chem. Rev. 2003, 240,
101–110; h) T. J. Wedge, M. F. Hawthorne, Coord. Chem. Rev. 2003,
240, 111–128; i) A. P. Davis, J.-B. Joos, Coord. Chem. Rev. 2003, 240,
143–156; j) P. D. Beer, E. J. Hayes, Coord. Chem. Rev. 2003, 240,
167–189; k) R. Martꢃnez-MµÇez, F. Sancenꢄn, Chem. Rev. 2003, 103,
4419–4476; l) M. H. Filby, J. W. Steed, Coord. Chem. Rev. 2006, 250,
3200–3218; m) A. P. Davis, Coord. Chem. Rev. 2006, 250, 2939–
2951; n) E. Garcꢃa-EspaÇa, P. Dꢃaz, J. M. Llinares, A. Bianchi,
Coord. Chem. Rev. 2006, 250, 2952–2986; o) E. A. Katayev, Y. A.
Ustynyuk, J. L. Sessler, Coord. Chem. Rev. 2006, 250, 3004–3037;
p) E. J. OꢁNeil, B. D. Smith, Coord. Chem. Rev. 2006, 250, 3068–
3080; q) R. Martꢃnez-MµÇez, F. Sancenꢄn, Coord. Chem. Rev. 2006,
250, 3081–3093; r) T. Gunnlaugsson, M. Glynn, G. M. Tocci, P. E.
Kruger, F. M. Pfeffer, Coord. Chem. Rev. 2006, 250, 3094–3117;
s) C. R. Rice, Coord. Chem. Rev. 2006, 250, 3190–3199; t) P. A.
Gale, R. Quesada, Coord. Chem. Rev. 2006, 250, 3219–3244;
u) P. A. Gale, S. E. Garcꢃa-Garrido, J. Garric, Chem. Soc. Rev. 2008,
37, 151–190; v) J. Pꢂrez, L. Riera, Chem. Soc. Rev. 2008, 37, 2658–
2667; w) J. W. Steed, Chem. Soc. Rev. 2009, 38, 506–519; x) C. Calta-
girone, P. A. Gale, Chem. Soc. Rev. 2009, 38, 520–563; y) S. Kubik,
Chem. Soc. Rev. 2009, 38, 585–605; z) Y. Hua, A. H. Flood, Chem.
Soc. Rev. 2010, 39, 1262–1271.
3
nium H4), 8.68 (d, J=8.8 Hz, 4H; NHArH), 7.32–7.11 (m, 34H; stopper
ArH), 6.65 (d, 3J=8.8 Hz, 4H; hydroquinone ArH), 6.33 (d, 3J=8.8 Hz,
4H; hydroquinone ArH), 4.48 (s, 3H; axle N+CH3), 4.04 (t, 3J=4.9 Hz,
+
4H; CH2 ), 3.93 (s, 3H; macrocycle N CH3), 3.76 (t, 3J=4.5 Hz, 4H;
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
CH2 ), 3.74–3.71 (m, 8H; CH2 ), 3.68–3.67 (m, 4H; CH2 ),
1.28 ppm (s, 36H; tBu); 13C{1H} NMR (75.5 MHz, CDCl3): d=160.6,
158.5, 153.0, 151.8, 148.5, 147.0, 145.3, 144.2, 143.6, 135.4, 134.6, 133.8,
131.7, 131.1, 130.7, 127.3, 125.8, 124.2, 120.0, 115.4, 114.6, 70.6, 69.8, 67.5,
66.5, 63.8, 54.8, 49.7, 49.5, 40.3, 34.3, 31.4 ppm; 19F{1H} NMR (282.4 MHz,
CDCl3): d=ꢀ70.9 ppm (d, 1J=710 Hz, PF6ꢀ); 31P{1H} NMR (121.6 MHz,
CDCl3): d=ꢀ144.1 ppm (sept, 1J=710 Hz, PF6ꢀ); HRMS (ESI): m/z:
calcd for C106H117N6O11: 1650.8814; found: 1650.9064 [MꢀPF6ꢀH]+.
Rotaxane 11·Cl (method 1): BisACHTNUTRGNE(NUG amine) 3 (46 mg, 0.093 mmol) and 1,3-
benzenedisulfonyl chloride (27 mg, 0.093 mmol) were used. Preparative
TLC (silica; 93:7 CHCl3/MeOH) gave a yellow solid (36 mg, 23%).
1H NMR (300 MHz, 1:1 CDCl3:CD3OD): d=9.62 (s, 2H; pyridinium H2,
H6), 8.89 (s, 1H; pyridinium H4), 8.55 (s, 1H; sulfonyl H2), 7.90 (d, 3J=
7.8 Hz, 2H; sulfonyl H4, H6), 7.64, (d, 3J=8.0 Hz, 4H; HNArH), 7.29–
7.12 (m, 31H;ArH, sulfonyl H5), 6.41 (d, 3J=8.4 Hz, 4H; hydroquinone
ArH), 6.20 (d, 3J=8.4 Hz, 4H; hydroquinone ArH), 4.57 (s, 3H; N+
CH3), 3.87 (brs, 4H; CH2 ), 3.82–3.76 (m, 12H; CH2 ), 3.09 (t, 3J=
ꢀ
ꢀ
ꢀ
ꢀ
4.7 Hz, 4H; CH2 ), 1.30 ppm (s, 36H; tBu); 13C{1H} NMR (75.5 MHz,
1:1 CDCl3/CD3OD): d=159.4, 153.2, 152.4, 149.4, 147.7, 145.5, 144.4,
142.8, 135.4, 133.3, 132.5, 131.7, 131.4, 131.0, 128.1, 126.5, 126.2, 125.0,
120.6, 116.1, 114.9, 71.4, 70.8, 67.9, 67.5, 64.6, 42.9, 34.9, 31.8, 30.3 ppm;
HRMS (ESI): m/z: calcd for C102H112N5O12S2: 1663.7782; found:
1663.7636 [MꢀCl]+.
ꢀ
ꢀ
[2] J. W. Pflugrath, F. A. Quiocho, Nature 1985, 314, 257–260.
[3] H. Luecke, F. A. Quiocho, Nature 1990, 347, 402–406.
[4] a) X. Yang, C. B. Knobler, M. F. Hawthorne, Angew. Chem. 1991,
103, 1519–1521; Angew. Chem. Int. Ed. Engl. 1991, 30, 1507–1508;
b) Z. Zheng, C. B. Knobler, M. F. Hawthorne, J. Am. Chem. Soc.
1995, 117, 5105–5113; c) J. S. Fleming, K. L. V. Mann, C.-A. Carraz,
E. Psillakis, J. C. Jeffery, J. A. McCleverty, M. D. Ward, Angew.
Chem. 1998, 110, 1315–1318; Angew. Chem. Int. Ed. 1998, 37, 1279–
1281; d) R. L. Paul, Z. R. Bell, J. C. Jeffery, J. A. McCleverty, M. D.
Ward, Proc. Natl. Acad. Sci. USA 2002, 99, 4883–4888; e) C. S.
Campos-Fernµndez, B. L. Schottel, H. T. Chifotides, J. K. Bera, J.
Bacsa, J. M. Koomen, D. H. Russell, K. R. Dunbar, J. Am. Chem.
Soc. 2005, 127, 12909–12923; f) S. P. Argent, T. Riis-Johannessen,
J. C. Jeffery, L. P. Harding, M. D. Ward, Chem. Commun. 2005,
4647–4649.
Rotaxane 11·PF6 (general procedure for anion exchange): By using rotax-
ane 11·Cl (30 mg, 0.018 mmol) to give 11·PF6 as a yellow solid (32 mg,
100%). 1H NMR (300 MHz, 1:1 CDCl3/CD3OD): d=9.29 (s, 2H; pyridi-
4
nium H2, H6), 8.66 (s, 1H; pyridinium H4), 8.49 (brt, J=1.7 Hz, 1H; sul-
3
4
4
fonyl H2), 8.01 (dd, J=8.0, J=1.7 Hz, 2H; sulfonyl H4, H6), 7.58 (d, J=
8.9 Hz, 4H; HNArH), 7.29–7.12 (m, 31H; ArH, sulfonyl H5), 6.46 (d,
3J=9.2 Hz, 4H; hydroquinone ArH), 6.22 (d, 3J=9.2 Hz, 4H; hydroqui-
+
ꢀ
ꢀ
none ArH), 4.53 (s, 3H; N CH3), 3.97–3.88 (m, 16H; CH2 ), 3.16 (t,
3J=5.3 Hz, 4H;
CH2 ), 1.29 ppm (s, 36H; tBu); 13C{1H} NMR
ꢀ
ꢀ
(75.5 MHz, 1:1 CDCl3/CD3OD): d=159.3, 153.2, 152.4, 149.4, 147.7,
145.8, 144.6, 143.0, 135.2, 133.8, 132.7, 131.8, 131.4, 131.1, 128.1, 126.6,
125.6, 125.1, 120.6, 116.2, 114.8, 71.2, 70.9, 67.7, 67.6, 64.6, 42.9, 34.9, 31.8,
30.3, 23.3 ppm; 19F{1H} NMR (282.4 MHz, 1:1 CDCl3/CD3OD): d=
ꢀ72.7 ppm (d, 1J=712 Hz, PF6ꢀ); 31P{1H} NMR (121.6 MHz, 1:1 CDCl3/
CD3OD): d=ꢀ144.2 ppm (sept, 1J=712 Hz, PF6ꢀ); HRMS (ESI): m/z:
calcd for C102H112N5O12S2: 1663.7782; found: 1663.7624 [MꢀPF6]+.
[5] a) R. Vilar, Angew. Chem. 2003, 115, 1498–1516; Angew. Chem. Int.
Ed. 2003, 42, 1460–1477; b) R. Vilar, Eur. J. Inorg. Chem. 2008,
357–367; c) R. Custelcean, J. Bosano, P. V. Bonnesen, V. Kertesz,
B. P. Hay, Angew. Chem. 2009, 121, 4085–4089; Angew. Chem. Int.
Ed. 2009, 48, 4025–4029.
[6] M Shionoya, H. Furuta, V. Lynch, A. Harriman, J. L. Sessler, J. Am.
Chem. Soc. 1992, 114, 5714–5722.
[7] a) E. A. Katayev, G. Dan Pantos, M. D. Reshetova, V. N. Khrustalev,
V. Lynch, Y. A. Ustynyuk, J. L. Sessler, Angew. Chem. 2005, 117,
7552–7556; Angew. Chem. Int. Ed. 2005, 44, 7386–7390; b) E. A.
Katayev, N. V. Boev, V. N. Khrustalev, Y. A. Ustynyuk, I. G. Tana-
naev, J. L. Sessler, J. Org. Chem. 2007, 72, 2886–2896.
Acknowledgements
We thank the EPSRC for a studentship (L.M.H.) and Johnson Matthey
for a CASE studentship (C.J.S.). N.L.K. acknowledges the Royal Com-
mission for the Exhibition of 1851 for a research fellowship. P.J.C. and
S:C thank FCT for the post-doctoral grants SFRH/BPD/27082/2006 and
SFRH/BPD/42357/2007, respectively. V.F. acknowledges the FCT, with
co-participation of the European Community fund FEDER, for financial
support under project PTDC/QUI/68582/2006. We also thank Diamond
Light Source for an award of beam time on I19 (MT1880), the beamline
instrument scientists for their assistance, and Dr. N. H. Rees for NMR
advice.
[8] D. Meshcheryakov, V. Bçhmer, M. Bolte, V. Hubscher-Bruder, F.
Arnaud-Neu, H. Herschbach, A. Van Dorsselaer, I. Thondorf, W.
Mçgelin, Angew. Chem. 2006, 118, 1679–1682; Angew. Chem. Int.
Ed. 2006, 45, 1648–1652.
[9] a) M. Bru, I. Alfonso, M. I. Burguete, S. V. Luis, Angew. Chem.
2006, 118, 6301–6305; Angew. Chem. Int. Ed. 2006, 45, 6155–6159;
b) I Alfonso, M. Bolte, M. Bru, M. I. Burguete, S. V. Luis, J. Rubio,
J. Am. Chem. Soc. 2008, 130, 6137–6144.
[10] a) Molecular Catenanes, Rotaxanes and Knots: A Journey Through
the World of Molecular Topology (Eds.: C. Dietrich-Buchecker, J.-P.
Sauvage), Wiley-VCH, Weinheim, 1999; b) V. Balzani, M. Venturi,
A. Credi, Molecular Devices and Machines—A Journey into the
Nanoworld, Wiley-VCH, Weinheim, 2003; c) E. R. Kay, D. A. Leigh,
F. Zerbetto, Angew. Chem. 2007, 119, 72–196; Angew. Chem. Int.
[1] a) A. Bianchi, K. Bowman-James, E. Garcꢃa-EspaÇa, Supramolec-
ular Chemistry of Anions, Wiley-VCH, New York 1997; b) J. L. Sess-
Chem. Eur. J. 2010, 16, 13082 – 13094
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13093

P. D. Beer et al.
Ed. 2007, 46, 72–191; d) S. Saha, J. F. Stoddart, Chem. Soc. Rev.
2007, 36, 77–92; e) S. J. Loeb, Chem. Soc. Rev. 2007, 36, 226–235;
f) B. Champin, P. Mobian, J.-P. Sauvage, Chem. Soc. Rev. 2007, 36,
358–366; g) J. F. Stoddart, H. M. Colquhoun, Tetrahedron 2008, 64,
8231–8263.
C.-H. Lee, P. C. Bennett, J. L. Sessler, Chem. Commun. 2009, 1109–
1111.
[16] K.-Y. Ng, V. Fꢂlix, S. M. Santos, N. H. Rees, Chem. Commun. 2008,
1281–1283.
[17] a) U. Eisner, J. Kuthan, Chem. Rev. 1972, 72, 1–42; b) D. M. Stout,
A. I. Meyers, Chem. Rev. 1982, 82, 223–243.
[11] M. J. Chmielewski, J. J. Davis, P. D. Beer, Org. Biomol. Chem. 2009,
7, 415–424.
[12] J. J. Gassensmith, S. Matthys, J.-J. Lee, A. Wojcik, P. V. Kamat, B. D.
Smith, Chem. Eur. J. 2010, 16, 2916–2921.
[18] J. A. Wisner, P. D. Beer, M. G. B. Drew, Angew. Chem. 2001, 113,
3718–3721; Angew. Chem. Int. Ed. 2001, 40, 3606–3609.
[19] To replicate the original reaction conditions, the acid chloride con-
densation was carried out in the presence of AgPF6 (2 equiv) to pre-
cipitate the chloride generated during the condensation reaction.
The yield of this reaction was observed to be approximately 15%,
only a quarter of that in which the chloride anion was present. This
is another positive indication that the yield of the reaction is highly
dependant on the presence of the chloride anion.
[20] Although co-crystallised solvent was present, it could not be re-
solved structurally in either case and was modelled as a diffuse
cloud of electrons. However, its presence and effect on the overall
conformation of both of these structures, along with packing consid-
erations, cannot be ignored, especially when compared with the mo-
lecular dynamics solution structures. Conclusions concerning the
electronic effect of the nitro group are difficult to make due to the
multiplicity of variables (polyether chain length and crystallisation
requirements).
[13] a) J. A. Wisner, P. D. Beer, M. G. B. Drew, M. R. Sambrook, J. Am.
Chem. Soc. 2002, 124, 12469–12476; b) M. R. Sambrook, P. D. Beer,
J. A. Wisner, R. L. Paul, A. R. Cowley, J. Am. Chem. Soc. 2004, 126,
15364–15365; c) K.-Y. Ng, A. R. Cowley, P. D. Beer, Chem.
Commun. 2006, 3676; d) M. R. Sambrook, P. D. Beer, M. D. Lank-
shear, R. F. Ludlow, J. A. Wisner, Org. Biomol. Chem. 2006, 4,
1529–1538; e) M. D. Lankshear, P. D. Beer, Acc. Chem. Res. 2007,
40, 657–668; f) M. D. Lankshear, N. H. Evans, S. R. Bayly, P. D.
Beer, Chem. Eur. J. 2007, 13, 3861; g) M. S. Vickers, P. D. Beer,
Chem. Soc. Rev. 2007, 36, 211–225; h) B. Huang, S. M. Santos, V.
Fꢂlix, P. D. Beer, Chem. Commun. 2008, 4610–4612; i) 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;
j) D. E. Phipps, P. D. Beer, Tetrahedron Lett. 2009, 50, 3454–3457.
[14] L. M. Hancock, P. D. Beer, Chem. Eur. J. 2009, 15, 42–44.
[15] For examples of chloride recognition in aqueous media, see: a) W.
Goodall, J. A. G. Williams, J. Chem. Soc. Dalton Trans. 2000, 2893–
2895; b) V. Amendola, E. Bastianello, L. Fabbrizzi, C. Mangano, P.
Pallavicini, A. Perotti, A. M. Lanfredi, F. Ugozzoli, Angew. Chem.
2000, 112, 3039–3042; Angew. Chem. Int. Ed. 2000, 39, 2917–2920;
c) D. K. Smith, Org. Biomol. Chem. 2003, 1, 3874–3877; d) R. Nish-
iyabu, M. A. Palacios, W. Dehaen, P. Anzenbacher, Jr., J. Am.
Chem. Soc. 2006, 128, 11496–11504; e) X. C. Jiang, A. B. Yu, Lang-
muir 2008, 24, 4300–4309; f) D.-W. Yoon, D. E. Groos, V. M. Lynch,
[21] M. J. Hynes, J. Chem. Soc. Dalton Trans. 1993, 311–312.
[22] Anion binding investigations with axle 1·PF6 in 45:45:10 CDCl3/
CD3OD/D2O revealed, as expected, very weak binding of Clꢀ.
[23] D. A. Case, AMBER 10, University of California, San Francisco,
2008.
[24] J. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman, D. A. Case, J.
Comput. Chem. 2004, 25, 1157–1174.
Received: July 21, 2010
Published online: October 28, 2010
13094
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13082 – 13094