Recognition-directed assembly of salt-binding [2]rotaxanes

Article abstract of DOI:10.1016/S0040-4020(01)01108-5

This paper describes the synthesis of four isomeric [2]rotaxanes and includes the first example of a salt binding [2]rotaxane with the following key observations. In polar solvents the [2]rotaxane undergoes a relatively slow co-conformational exchanges th

Full text of DOI:10.1016/S0040-4020(01)01108-5

TETRAHEDRON
Pergamon
Tetrahedron 58 (2002) 799±805
Recognition-directed assembly of salt-binding [2]rotaxanes
Martin J. Deetz, Rameshwer Shukla and Bradley D. Smithp
Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556-5670, USA
Received 29 May 2001; revised 9 September 2001; accepted 10 September 2001
AbstractÐThis paper describes the synthesis of four isomeric [2]rotaxanes and includes the ®rst example of a salt binding [2]rotaxane with
the following key observations. In polar solvents the [2]rotaxane undergoes a relatively slow co-conformational exchange that produces
1
broad signals in the H NMR spectrum. Upon complexation with K1, the NMR signals sharpen apparently because the rotaxane adopts a
single co-conformation. In contrast, the binding of Cl2 has no effect on the rotaxane's dynamic behavior. Although the binding of K1 induces
a preorganization of the [2]rotaxane it does not change the rotaxane/Cl2 association constant. The paper ®nishes with a discussion of the
future directions of this work. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
a macrocycle (this complex is sometimes referred to as a
psuedorotaxane) followed by an axle-capping reaction that
traps the components together,9 and (c) Slipping involves
mechanical slippage of a macrocycle over the dumbell-
shaped axle.10
A [2]rotaxane is an interlocked molecule in which a macro-
cyclic wheel is threaded onto an acyclic dumbell-shaped
axle. The ends of the axle have stopper groups that are
large enough to prevent dissociation of the components,
i.e. the components are held by a mechanical bond. The
past decade has seen a renaissance in rotaxane chemistry
after the initial investigations more than 30 years ago.1 In
recent years, a range of exotic rotaxane structures has been
prepared, and a number of new synthetic strategies have
been developed.2±7 The modern strategies rely on recogni-
tion-directed methods to template the rotaxane forming
reaction. Currently, three main methods are employed for
rotaxane assembly (Fig. 1). (a) Clipping is when a macro-
cycle is assembled around the dumbell-shaped axle,8 (b)
Threading involves binding of an unstoppered axle within
È
Recently, Vogtle and coworkers reported a new threading
method that employs an anion trapping effect.11±13 The
example shown in Fig. 2 uses a macrocyclic tetralactam to
bind a bulky phenolate anion via hydrogen bonding to its
isophthalamide NH protons.11,12 This complex then func-
tions as a wheeled supramolecular nucleophile, and reacts
with an electrophile through the cavity of the macrocycle to
form a semiaxle. Dissociation of the components, followed
by reaction of the semiaxle with a second wheeled phenolate
produces the [2]rotaxane. Yields of up to 95% have been
obtained using this trapping method,11 although more often
the yields are below 50%.12 It should be noted that 18-
crown-6 is a crucial additive to this reaction, as it increases
the solubility of the K2CO3 base and most likely prevents the
formation of a tight ion pair between the potassium and the
phenolate.12 A tight ion pair reduces phenolate reactivity
Figure 1. Three recognition-directed methods for assembling [2]rotaxanes.
Keywords: interlocked molecule; supramolecular chemistry; molecular
machine; anion binding.
p
Corresponding author. Tel.: 11-219-631-8632; fax: 11-219-631-6652;
e-mail: smith.115@nd.edu
11,12
È
Figure 2. Vogtle's wheeled phenolate, TrˆTrityl.
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(01)01108-5

800
M. J. Deetz et al. / Tetrahedron 58 (2002) 799±805
syntheses of four isomeric [2]rotaxanes that use bicycles 1
and 2 as wheel components and phthalate diesters 3 and 4 as
axle components (Fig. 5).
2. Results and discussion
The syntheses of isomeric salt-binding macrobicycles 1 and
2 are outlined in Scheme 1. The synthetic scheme has
already been reported for the cis isomer, 1,15 and so it is
not discussed here, except to state that compounds 1 and 2
can be prepared on a gram scale in six steps from dibenzo-
18-crown-6, and that mass spectra of the crude products of
the ®nal macrocyclization reactions show no evidence for
larger polycyclic structures. The X-ray crystal structure of
trans isomer, 2 (Fig. 6) shows that it has pseudo-C2 sym-
metry and a slightly smaller cavity than the cis isomer 1.
Figure 3. Left: salt-binding macrobicycle. Right: side-view of the X-ray
crystal structure of Na1´CHCl3´Cl2 complex.
and also inhibits binding of the phenolate to the macrocyclic
wheel.
Our group has become interested in this anion trapping reac-
tion for the following reasons. Recently, we prepared a
number of macrobicyclic receptors and demonstrated that
they can simultaneously bind a cation and an anion.14,15 One
example is the macrobicyclic receptor shown in Fig. 3
which binds sodium chloride as a solvent separated ion-
pair.15 The anion forms hydrogen bonds to the exo-face of
the isophthalamide bridge, and the cation is encapsulated
within the ring of the crown ether. This supramolecular
structure prompted us to consider whether the anion
template reaction could be used to convert our salt-binding
macrobicycle into a novel [2]rotaxanes with the generic
structure shown in Fig. 4. The interesting feature with our
system is that the crown ether required for the reaction to
proceed is now part of the rotaxane wheel. Moreover,
because the resulting [2]rotaxane has a crown ether incor-
porated into its wheel component, the rotaxane is likely to
have interesting supramolecular properties. Recently, we
communicated our ®rst discovery in this fascinating ®eld
of interlocked molecules,16 and in this paper we describe
our work in more detail. Speci®cally, we describe the
In solution, macrobicycles 1 and 2 bind salts with positive
cooperativity, i.e. the presence of a metal cation enhances
anion af®nity. Association constants were determined using
standard NMR titration methods. Solutions of 1 and 2 in
DMSO-d6/CD3CN (3:1) were titrated with tetrabutyl-
ammonium chloride in the presence and absence of
1 molar equiv. of potassium tetraphenylborate and the
resulting titration isotherms ®tted to a 1:1 binding model
using established iterative curve ®tting methods.17 As
shown in Table 1, cis isomer 1 has 2.5 times higher Cl2
af®nity than trans isomer 2. In each case the Cl2 association
Scheme 1. Reagents and conditions: (a) H2 (30 psi), Pd±C, DMF, cisÐ
98%, (transÐ99%), (b) 4-nitrobenzoyl chloride, Et3N, CH2Cl2, cisÐ99%
(transÐ99%), (c) NaH, CH3I, DMF, cisÐ99% (transÐ96%), (d) Fe
(powder), HCl, cisÐ81% (transÐ77%), (e) corresponding isophthaloyl
dichloride, Et3N, CH2Cl2, cisÐ32±75% (transÐ42±45%).
Figure 4. [2]Rotaxane with bicyclic wheel.
Figure 5. Isomeric wheel components 1 and 2 and axle components 3 and 4,
TrˆTrityl.
Figure 6. Front and side views of the X-ray crystal structure of macro-
bicycle 2.

M. J. Deetz et al. / Tetrahedron 58 (2002) 799±805
Table 1. Receptor/Cl2 association constants (KCl2) and change in NH
801
remembered that two covalent bonds are formed and four
units are assembled in a single reaction. The [2]rotaxanes
from cis macrobicycle, 1, are produced in higher yields than
trans macrobicycle, 2. The difference likely stems from a
more productive preorganization of the axle components
within the larger cis macrobicyclic cavity. The reaction
fails to produce [1´3] when cesium or tetrabutylammonium
4-tritylphenolate is used. The dibenzocrown unit in 1 is
known to have a low af®nity for large and diffuse cations,18
so we consider this observation as evidence in favor of the
templating model shown in Fig. 7. The size of the stopper
groups appears to be important because no [2]rotaxane is
isolated when the less spatially demanding 3,5-di-t-butyl
phenol, is used.10
chemical shift (Ddmax
)
1
11K1a
2
21K1a
[1´3]
[1´3]1K1a
b
KCl2 (M21
)
50
0.60
350
0.92
20
0.50
130
0.84
300
0.96
300
1.19
Ddmax (ppm)c
a
Presence of 1 molar equiv. of KBPh4.
Uncertainty ^20%, Tˆ295 K, [Receptor]ˆ10 mM in DMSO-d6/CD3CN
b
c
(3:1).
Ddmax represents the change in NH chemical shift after addition of
4 molar equiv. of N(Bu4)Cl.
The structures of the [2]rotaxanes were proven by mass
spectrometry and NMR spectroscopy. Positive ion FAB
mass spectra of the isomeric [2]rotaxanes show a weak or
non-existent molecular ion signal at m/z 1647 [M1H]1, and
intense signals at 1669 [M1Na]1 and 1685 [M1K]1 due to
complexation with adventitious cations in the sample
matrix. Rotaxane formation is readily indicated by the
1
changes in the H NMR signals for the axle phathalate
Figure 7. Reaction of wheeled phenolate to give semiaxle 7 and subse-
quently [2]rotaxane [1´3], TrˆTrityl.
protons. The signal for the two isophthalate protons, Hb,
in axle 3 is a doublet at 8.43 ppm when free in CDCl3 solu-
tion, but when incorporated into [2]rotaxane [1´3], the two
protons are no longer chemically equivalent and two
doublets at 8.64 and 8.55 ppm are observed. With [2]rotax-
ane [2´3], the two protons are chemically equivalent due to
the C2 symmetry of the wheel and only one doublet is
observed at 8.43 ppm. The signal for the four terephthalate
protons in axle 4 is a singlet at 8.04 ppm, when free in
CDCl3 solution, but when incorporated into [2]rotaxane
[1´4], two doublets are observed at 8.53 and 8.41 ppm. In
the case of [2]rotaxane [2´4], a single peak is observed at
8.40 ppm.
constants are increased by about a factor of seven when the
titration is repeated in the presence of 1 molar equiv. of
potassium tetraphenylborate.
A diagram of the proposed trapping effect is shown in Fig. 7.
We expected that 1 would be able to bind potassium 4-trityl-
phenolate as its solvent separated pair. Reaction of the
wheeled potassium phenolate with isophthaloyl dichloride,
for example, ®rst forms semiaxle 7 in situ and subsequently
the [2]rotaxane, [1´3]. While the anion templated esteri®ca-
È
tion method employed by Vogtle and coworkers requires
1
The room temperature H NMR spectra of all [2]rotaxanes
non-polar solvents to ensure high amounts of wheeled
nucleophile, our procedure produces [2]rotaxanes in a
range of organic solvents such as chloroform, THF, and
even THF±DMF. With the former two solvents, the reaction
is hampered by precipitation of a macrobicycle/KCl
complex. Changing to the more polar mixture of THF/
DMF produces homogeneous solutions, faster reaction
times, and slightly better yields. The most ef®cient pro-
cedure mixes the macrobicycle, and potassium trityl pheno-
late (2 molar equiv.) at 08C for 10±15 min in a 5/1 mixture
of THF and DMF, followed by single addition of all of the
appropriate solid phthaloyl dichloride. The reaction mixture
is slowly warmed to room temperature and then stirred
under an inert atmosphere for 2±3 days. The [2]rotaxanes
are produced in 3±20% yield after chromatography
(Table 2). Although the yields are low, it should be
in CDCl3 are well resolved and exhibit no apparent dynamic
behavior upon cooling to 2608C. In the case of [1´3], the
spectrum was assigned by COSY and ROESY NMR
methods. The ROESY spectrum is consistent with the
co-conformation (the relative orientation of the component
parts of an interlocked molecule has been de®ned as its
co-conformation19) shown in Fig. 8. For example, the axle
proton Ha (8.92 ppm) interacts through-space with the
wheel NH residues (8.18 ppm) and proton HA (7.74 ppm),
there are also cross peaks between axle protons Hb (8.54,
8.64 ppm), Hc (8.12 ppm) and the wheel's crown ether
protons (3.4±4.2 ppm).
In the more polar solvents, DMSO-d6 and acetone-d6, the
Table 2.
[2]Rotaxane
Wheel
Isolated
yield (%)
Axle
1
1
2
2
3
4
3
4
20
17
6
3
Figure 8. Likely co-conformation of [1´3] in CDCl3.

802
M. J. Deetz et al. / Tetrahedron 58 (2002) 799±805
Figure 10. Partial 1H NMR spectra of [2]rotaxane [1´3] in DMSO-d6:
(a) before and (b) after the addition of 5 molar equiv. of KPF6.
Figure 9. 1H NMR spectra of [2]rotaxane [2´4] in acetone-d6 at: (a) 1208C,
(b) 2408C, and (c) 2908C.
NMR spectra are broad at room temperature. Cooling the
acetone-d6 samples to 2408C results in increased spectrum
resolution, but only one set of peaks is observed. Further
cooling to 2908C leads to signal broadening. The three
spectra of [2]rotaxane [2´4] shown in Fig. 9, are typical
examples. The lack of peak splitting makes it hard to
attribute the peak broadness at room temperature to a
speci®c dynamic motion. The two most likely motions are
pirouetting of the wheel around the axle and shuttling of the
wheel along the axle.20
Figure 11. Likely supramolecular structure of the [1´3]/K1 complex. The
curved arrow indicates rapid bond rotation in the absence of Cl2.
changes the spatial positions of Ha and Hb so they pass
0
close to the crown hydrogens, whereas Hb and Hc always
Ê
remain more than 4 A from the crown. Further evidence for
this supramolecular structure is provided in the next
paragraph.
The broad NMR spectrum of [1´3] in DMSO-d6 sharpens
considerably upon addition of 5 molar equiv. of KPF6 at
room temperature (Fig. 10). There are two possible
explanations for this observation: either the added K1
increases the rate of exchange between different co-con-
formations of [1´3], or the K1 freezes out a single co-con-
formation.20 Evidence for the latter explanation is gained
from the ROESY spectrum of [1´3] in DMSO-d6 with
added KPF6 which is consistent with a predominant
co-conformation. For example, the wheel's crown ether
protons cross-relaxes the axle protons Ha and Hb but not
The anion and salt binding properties of [2]rotaxane [1´3]
were evaluated in the following way. First, a solution of
[1´3] in DMSO-d6/CD3CN (3:1) was titrated with
N(Bu4)Cl. The NMR signals for [1´3] remained broad
throughout the titration, although the NH peak could always
be identi®ed. The change in NH chemical shift was
monitored and used to generate a titration isotherm which
was ®tted to 1:1 binding model using an iterative computer
method.17 The titration experiment was then repeated but
this time the host solution also included 1 molar equiv. of
KBPh4 which sharpens the 1H NMR spectrum of [1´3]. The
changes in rotaxane chemical shifts (Dd) for [1´3]/KBPh4
induced by the addition of 4 molar equiv. of tetrabutyl-
ammonium chloride are listed in Table 3. The large Dd
values observed for HA, HC, NH, Hb and Hc are strong
evidence that the Cl2 ion binds in a pocket formed by
these hydrogens. The small Dd value observed for Ha
0
protons Hb and Hc. It appears that the potassium cation
binds to the benzocrown ring and freezes out a single
rotaxane co-conformation by coordinating to one of the
ester carbonyls in the axle (Fig. 11). A CPK model of
[1´3] indicates that this co-conformation allows rapid
rotation about the carbonyl±phenyl bond which inter-

M. J. Deetz et al. / Tetrahedron 58 (2002) 799±805
Table 3. Change in 1H NMR chemical shifts (Dd max) for [1´3]/KBPh4 upon
addition of N(Bu4)Cl
803
Signalsa
NH Ha
Hb
Hb
Hc
HA
HB
HC
0
Dd max (ppm) 1.19 0.04 0.33 20.03 20.22 0.65 20.05 0.29
Final molar ratio of [1´3]/KBPh4/N(Bu4)Cl was 1:1:4 in DMSO-d6/CD3CN
(3:1).
See Fig. 5 for hydrogen labeling scheme.
a
indicates that carbonyl±phenyl rotation is restricted upon
binding of KCl, and that Ha points away from the Cl2 bind-
ing pocket. Overall, the data agrees nicely with the supra-
molecular structure shown in Fig. 11.
Figure 12. In polar aprotic solvent, the [2]rotaxane [1´3] populates multiple
axle/wheel orientations. Binding Cl2 does not measurably alter the
rotaxane's dynamic properties whereas binding K1 freezes out a single
co-conformation, and the combined effect of KCl produces the most
structural rigidity.
Table 1 compares the [1´3]/Cl2 and 1/Cl2 association
constants in the presence and absence of 1 molar equiv. of
KBPh4. Inspection of the association constants reveals some
surprising effects. First, in the absence of K1 ions, macro-
bicycle 1 has a six-fold lower Cl2 af®nity than [2]rotaxane
[1´3] in DMSO-d6/CD3CN (3:1). The large changes in NH
chemical shift clearly indicate that in both cases the NH
residues form hydrogen bonds to the Cl2. It is known
from our previous work that a DMSO solvent molecule
binds deeply within the cavity of macrobicycle 1, and thus
it competes strongly with the Cl2.15 In the case of [2]rotax-
ane [1´3], a DMSO molecule cannot bind to the macro-
bicyclic wheel in the same ditopic fashion, and does not
appear to strongly inhibit Cl2 binding to the wheel NH
protons. In the presence of 1 molar equiv. of KBPh4 the
1/Cl2 af®nity was enhanced seven-fold, whereas the
presence of KBPh4 had no effect on [1´3]/Cl2 af®nity. The
fact that no binding enhancement is observed in the rotaxane
case indicates that the axle is insulating any through-space
electrostatic interaction between the bound K1 and Cl2.
This is in contrast to a recently proposed induced-dipole
model that predicts electrostatic stabilizing effects can be
transmitted through aromatic groups.21 It seems that in this
present case the orientation of the central phthalate ring in
the axle does not allow such a p-cloud polarization effect to
occur.
the anion trapping reaction can be used to prepare salt bind-
ing rotaxanes and that these molecules have the potential to
exhibit chemical-dependent dynamic behavior. To date,
most attempts to control the motion of rotaxane and
catenanes have used photochemical or electrochemical
strategies.22,23 Such designs are attractive because light
and redox potential are readily amenable to spatiotemporal
control. Thus, it is not surprising that these systems are
being considered as components in light harvesting24 and
logic devices.25 The relatively few reports of chemical
switching of interlocked co-conformation include the use
of protons,22,23,26 metal cations,19,27±29 anions30,31 and
amines32 as chemical effectors. In these cases, the potential
applications include sensing, separations, catalysis and drug
delivery, although few working prototypes have appeared so
far. A lingering technological hurdle is the need for ef®cient
syntheses of interlocked compounds.2 Although the last
decade has seen remarkable advances in recognition-
directed synthesis, most of the presently accessible inter-
locked molecules, such as the [2]rotaxane [1´3] described
here, are still relatively simple structures when compared to
the molecular machines found in nature.22,23 In particular,
the development of high yielding, iterative synthetic
methods that can rapidly build sophisticated polyrotaxanes
and polycatenanes would undoubtedly bring major
advances to the ®eld.33,34 Of course a related technical
issue is the development of analytical methods to character-
ize these larger and more complicated molecular struc-
tures.20 The ongoing improvements in molecular imaging
3. Conclusions
This paper describes the ®rst example of a salt binding
[2]rotaxane and includes the following key observations.
In polar solvents the [2]rotaxane [1´3] undergoes a relatively
slow co-conformational exchange that produces broad
35
will de®nitely helpin this regard.
1
signals in the H NMR spectrum. Upon complexation with
K1, the NMR signals sharpen apparently because the
rotaxane adopts a single co-conformation. In contrast, the
binding of Cl2 has no apparent effect on the dynamic
behavior of [2]rotaxane [1´3], however, the combined effect
of KCl produces the most structural rigidity. Although the
binding of K1 induces a preorganization of [2]rotaxane
[1´3] it does not change the [1´3]/Cl2 association constant.
A schematic model of the proposed conformational and
binding equilibria is provided in Fig. 12.
From a broader perspective, the work described in this paper
can be viewed as a steptowards the develomp ent of
arti®cial molecular machines although some may consider
such a claim as hyperbole.36 As a compromise we state that
our near-term research goals are to invent molecular
versions of machine elements which have been de®ned as
ªOne of the various simple mechanisms, as the lever, pulley,
cam, gear-wheel, etc. which, alone or in combination,
facilitate conversion of energy to useful workº.37 The phrase
`conversion of energy to useful work' is a critical point and
needs to be emphasized. Although certain applications such
as catalysis and drug delivery can be achieved in homo-
geneous solution, many others will require the molecular
È
It appears that Vogtle's anion trapping method will be a
versatile recognition-directed approach to assembling inter-
locked molecules.11±13 The results of this study indicate that

804
M. J. Deetz et al. / Tetrahedron 58 (2002) 799±805
machine elements to be assembled and employed in a
coherent manner. Thus, the challenge for researchers is to
align them in ordered arrays on surface or interfaces, so that
their individual molecular outputs can be combined and
ampli®ed to produce a measurable macroscopic output.
134.8, 132.5, 132.4, 132.2, 132.0, 131.5, 131.4, 131.3,
131.2, 131.1, 130.9, 130.1, 129.7, 129.6, 129.3, 127.9,
127.5, 126.2, 126.1, 126.0, 120.9, 120.5, 120.3, 118.8,
114.4, 112.9, 112.3, 70.1, 69.2, 69.0, 68.1, 64.7, 64.7,
38.9, 35.4, 31.2, MS (FAB1) mass calcd for
C106H92N4NaO14 [M1Na]1 1669, found 1669.
4.1.4. Rotaxane [1´4]. Yield 17%. 1H NMR (CDCl3,
300 MHz): d (ppm) 8.53 (d, 2H, Jˆ8.4 Hz), 8.41 (d, 2H,
Jˆ8.4 Hz), 8.30 (d, 2H, Jˆ1.5 Hz), 8.13 (s, 2H), 7.68 (s
(br), 1H), 7.29 (m, 40H), 6.57 (d, 2H, Jˆ2.1 Hz), 6.44 (d,
2H, Jˆ8.4 Hz), 6.01 (d, 2H, Jˆ9.0 Hz) 4.09 (m, 4H), 3.93
(m, 4H), 3.65 (m, 4H), 3.45 (m, 2H), 3.37 (s, 6H), 3.30 (m,
2H), 1.30 (s, 9H). 13C NMR (CDCl3, 75 MHz): d (ppm)
170.9, 168.6, 164.6, 153.9, 148.8, 147.9, 146.6, 146.5,
146.1, 145.5, 139.2, 138.3, 135.0, 135.0, 134.7, 132.8,
132.6, 132.5, 132.2, 131.3, 131.2, 130.8, 129.8, 129.3,
127.8, 127.6, 126.3, 126.2, 121.6, 120.5, 120.2, 119.9,
118.7, 113.5, 112.6, 70.4, 69.5, 69.0, 68.3, 64.8, 38.5,
35.4, 31.3 MS (FAB1) mass calcd for C106H92N4NaO14
[M1Na]1 1669, found 1669.
4. Experimental
4.1. Materials
All salts and NMR solvents were purchased from Aldrich
and, after checking for purity, were used as supplied.
4.1.1. Macrobicycles 1 and 2. The synthesis of cis isomer 1
from dibenzo18-crown-6 has been reported elsewhere in
detail.15 The same procedure was used to make the trans
1
isomer 2. H NMR (300 MHz, CDCl3) d (ppm) 8.31 (bs,
2H), 8.17 (s, 2H), 7.80 (s, 1H), 7.39 (d, 4H, Jˆ8.4 Hz), 7.22
(d, 4H, Jˆ8.6 Hz), 6.64 (d, 2H, Jˆ8.4 Hz), 6.52 (d, 2H,
Jˆ8.2 Hz), 6.47 (s, 1H), 4.09±3.79 (m, 16H), 3.34 (s,
6H), 1.39 (s, 9H) ppm. 13C NMR (75 MHz, DMSO-d6) d
(ppm) 170.0, 165.5, 153.5, 148.7, 147.1, 139.3, 138.2,
134.9, 131.1, 129.7, 129.1, 121.2, 119.5, 119.0, 111.7,
111.0, 69.6, 69.6, 67.8, 38.6, 35.2, 31.2. MS (FAB1) exact
mass calculated for C48H51N4O10 [M1H]1 843.3605, found
843.3594. X-Ray crystal data were collected and integrated
using a Bruker Apex system, with graphite monochromated
4.1.5. Rotaxane [2´3]. Yield 6%. 1H NMR (600 MHz,
CDCl3) d (ppm) 8.90 (s, 1H), 8.43 (dd, 2H, Jˆ1.7,
7.9 Hz), 8.27 (s, 2H), 8.12 (t, 1H, Jˆ7.8 Hz), 8.13 (s, 2H),
7.62 (s, 1H), 7.26±7.11 (m, 42H), 6.46 (d, 4H, Jˆ8.8 Hz),
6.43 (dd, 2H, Jˆ2.4, 8.4 Hz), 6.40 (d, 2H, Jˆ2.4 Hz), 6.30
(d, 2H, Jˆ8.4 Hz), 4.00±3.37 (m, 16H), 3.38 (s, 6H), 1.33
(s, 9H) ppm. 13C NMR (125 MHz, CDCl3) (ppm) 169.6,
166.5, 164.7, 153.7, 148.7, 147.9, 146.9, 146.3, 145.7,
139.4, 138.4, 135.7, 134.6, 132.0, 131.6, 131.3, 130.9,
130.7, 129.8, 129.6, 129.5, 127.6, 126.0, 120.4, 119.9,
119.5. 118.5, 112.7, 111.8, 69.6, 69.4, 68.7, 68.1, 64.6,
38.6, 35.2, 31.1. MS (FAB1) mass calculated for
C106H92N4NaO14 [M1Na]1, 1669, found 1669.
Ê
Mo Ka (lˆ0.71073 A) radiation at 170 K. The structure
was solved by direct methods. Hydrogen atoms were placed
at idealized positions and a riding model with ®xed thermal
parameters was used for subsequent re®nements. The asym-
metric unit contains one half of the macrocycle, with one
orientation of the t-butyl methyl groups at 50% occupancy.
The second orientation of the t-butyl methyl groups is
generated through symmetry. The X-ray data has been
deposited with the Cambridge Crystallographic Data Centre
as supplementary publication number CCDC 168432.
Copies of the data can be obtained free of charge on appli-
cation to CCDC at deposit@ccdc.cam.ac.uk.
4.1.6. Rotaxane [2´4]. Yield 3%. 1H NMR (600 MHz,
CDCl3) d (ppm) 8.40 (s, 4H), 8.30 (d, 2H, Jˆ1.4 Hz),
8.12 (s, 2H), 7.57 (s, 1H), 7.23±7.12 (m, 40H), 6.51 (d,
4H, Jˆ8.8 Hz), 6.42 (dd, 2H, Jˆ2.5, 8.4 Hz), 6.39 (d, 2H,
Jˆ2.5 Hz), 6.26 (d, 2H, Jˆ8.4 Hz), 3.93±3.89 (m, 4H), 3.83
(dt, 2H, Jˆ3.1, 9.5 Hz), 3.75 (td, 2H, Jˆ2.3, 9.3 Hz), 3.66
(dt, 2H, Jˆ2.9, 10.4 Hz), 3.58 (dt, 2H, Jˆ2.7, 9.8 Hz), 3.49
(dt, 2H, Jˆ2.9, 11.1 Hz), 3.35±3.32 (m, 2H), 3.38 (s, 6H),
1.34 (s, 9H) ppm. 13C NMR (125 MHz, CDCl3) d (ppm)
169.7, 166.9, 164.6, 153.8, 148.6, 148.0, 146.9, 146.4,
145.9, 139.4, 138.1, 134.5, 133.4, 132.0, 131.8, 130.9,
130.7, 129.8, 129.7, 127.7, 126.0, 120.5, 119.7, 119.3,
118.4, 112.1, 111.6, 69.4, 69.3, 68.6, 67.8, 64.7, 38.5,
35.2, 31.0. MS (FAB1) mass calculated for
C106H92N4NaO14 [M1Na]1, 1669, found 1669.
4.1.2. General rotaxane synthesis. Under an atmosphere of
argon, macrobicycle 1 or 2 (0.025 g, 0.029 mmol) and
potassium 4-tritylphenolate (0.022 g, 0.058 mmol) were
stirred in 5:1 THF/DMF (18 mL) for 1 h. The solution
was cooled to ice temperature then solid phthalolyl or
isophthalolyl dichloride (0.006 g, 0.029 mmol) was added
and the solution allowed to warm and stir for four days.
The solvent was evaporated and the residue puri®ed by
column chromatography using silica gel and CH2Cl2/
MeOH (95:5).
4.1.3. Rotaxane [1´3]. Yield 20%. 1H NMR (CDCl3,
600 MHz): d (ppm) 8.92 (s, 1H), 8.64 (d, 1H, Jˆ8.0 Hz),
8.55 (d, 1H, Jˆ8.0 Hz), 8.24 (s, 2H), 8.17 (s, 2H), 8.12 (t,
1H, Jˆ8.0 Hz), 7.72 (s, 1H), 7.23±7.06 (m, 40H), 6.96 (d,
2H, Jˆ9.0 Hz), 6.75 (d, 2H, Jˆ9.0 Hz) 6.48 (s, dd, 2H,
Jˆ8.5, 2.5 Hz), 6.35 (d, 2H, Jˆ8.5 Hz), 6.14 (d, 2H,
Jˆ8.5 Hz), 4.05±4.03 (m, 4H), 3.80 (m, 6H), 3.68 (m,
2H), 3.52 (m, 2H), 3.42 (m, 2H), 3.35 (s, 6H), 1.34 (s,
9H). 13C NMR (CDCl3, 150 MHz): d (ppm) 170.3, 167.6,
165.8, 164.8, 153.8, 148.5, 148.3, 148.0, 147.6, 147.2,
146.5, 146.5, 145.7, 145.5, 139.4, 138.3, 136.3, 136.2,
4.2. NMR titration experiments
Aliquots of tetrabutylammonium chloride stock solution
were added to a 10 mM solution (3:1 DMSO-d6/CD3CN,
0.75 mL) of receptor in the presence and absence of
potassium tetraphenylborate. The changes in receptor
chemical shift were monitored and used to produce titration
isotherms which were ®tted to a 1:1 binding model using an
iterative computer method that has been described
elsewhere.17

M. J. Deetz et al. / Tetrahedron 58 (2002) 799±805
805
19. Fyfe, M. C. T.; Glink, P. T.; Menzer, S.; Stoddart, J. F.; White,
A. J. P.; Williams, D. J. Angew. Chem., Int. Ed. 1997, 36,
2068±2070.
Acknowledgements
This work was supported by the National Science Founda-
tion and the University of Notre Dame (George M. Wolf and
Reilly fellowships for M. J. D.). The X-ray structure of
compound 2 was solved by Dr A. Beatty, University of
Notre Dame.
20. Pons, M.; Millet, O. Prog. Nucl. Magn. Reson. Spectrosc.
2001, 38, 267±324.
21 Ngola, S. M.; Kearney, P. C.; Mecozzi, S.; Russell, K.;
Dougherty, D. A. J. Am. Chem. Soc. 1999, 121, 1192±1201.
22. Sauvage, J.-P. Acc. Chem. Res. 1998, 31, 611±619.
23. Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F. Angew.
Chem., Int. Ed. 2000, 39, 3348±3391.
References
24. Brouwer, A. M.; Frochot, C.; Gatti, F. G.; Leigh, D. A.;
Mottier, L.; Paolucci, F.; Rof®a, S.; Wirpel, G. W. H. Science
2001, 291, 2124±2128.
1. Schill, G. Catenanes, Rotaxanes, Knots; Academic: New
York, 1971.
2. Molecular Catenanes, Rotaxanes and Knots, Sauvage, J. P.,
Dietrich-Bucheker, C. O., Eds.; VCH-Wiley: Weinheim,
1999.
25. Wong, E. W.; Collier, C. P.; Belohradsky, M.; Raymo, F. M.;
Stoddart, J. F.; Heath, J. R. J. Am. Chem. Soc. 2000, 122,
8531±8540.
3. Leigh, D. A.; Murphy, A. Chem. Ind. 1999, 178±184.
È
4. Reuter, C.; Schmieder, R.; Vogtle, F. Pure Appl. Chem. 2000,
72, 2233±2241.
26. Anelli, P.; Spencer, N.; Stoddart, J. F. J. Am. Chem. Soc. 1991,
113, 5131±5133.
27. Asakawa, M.; Ashton, P. R.; Iqbal, S.; Stoddart, J. F.; Tinker,
N. D.; White, A. J. P.; Williams, D. J. Chem. Commun. 1996,
483±486.
5. Raymo, F.; Stoddart, J. F. Chem. Rev. 1999, 99, 1643±1663.
6. Bryant, W. S.; Guzei, I. A.; Reingold, A. L.; Gibson, H. W.
Org. Lett. 1999, 1, 47±50.
28. Asakawa, M.; Iqbal, S.; Stoddart, J. F.; Tinker, N. D. Angew.
Chem., Int. Ed. 1996, 35, 976±978.
29. Amabilino, D. B.; Dietrich-Buchecker, C. O.; Livoreil, A.;
7. Jeong, K.-S.; Choi, J. S.; Chang, S. Y.; Chang, H.-Y. Angew.
Chem., Int. Ed. 2000, 39, 1692±1695.
8. Philp, J. F.; Stoddart, J. F. Synlett 1991, 445±458.
Â
Perez-Garcia, L.; Sauvage, J.-P.; Stoddart, J. F. J. Am.
Chem. Soc. 1997, 119, 12114±12124.
30. Andrievsky, A.; Ahuis, F.; Sessler, J. L.; Vogtle, F.; Gudat, D.;
È
È
9. Vogtle, F.; Handel, M.; Meier, S.; Ottens-Hildebrandt, S.; Ott,
F.; Schmidt, T. Liebigs Ann. 1995, 739±743.
È
È
È
10. Handel, M.; Plevoets, M.; Gestermann, S.; Vogtle, F. Angew.
Moini, M. J. Am. Chem. Soc. 1998, 120, 9712±9713.
31. Montalti, M.; Prodi, L. Chem. Commun. 1998, 1461±1462.
32. Balzani, V.; Credi, A.; Langford, S. J.; Raymo, F. M.;
Stoddart, J. F.; Venturi, M. J. Am. Chem. Soc. 2000, 122,
3542±3543.
Chem., Int. Ed. Engl. 1997, 36, 1199±1201.
È
È
11. HuÈbner, G. M.; Glaser, J.; Seel, C.; Vogtle, F. Angew. Chem.,
Int. Ed. Engl. 1999, 38, 383±386.
12. Reuter, C.; Wienand, W.; HuÈbner, G. M.; Seel, C.; Vogtle, F.
È
Chem. Eur. J. 1999, 5, 2692±2697.
È
33. Raymo, F. M.; Stoddart, J. F. Chem. Rev. 1999, 99, 1643±
1663.
13. Seel, C.; Vogtle, F. Chem. Eur. J. 2000, 6, 21±24.
14. Shukla, R.; Kida, T.; Smith, B. D. Org. Lett. 2000, 2, 3039±
3248. Mahoney, J. M.; Beatty, A. M.; Smith, B. D. J. Am.
Chem. Soc. 2001, 123, 5847±5848.
È
34. Parham, A. H.; Schmieder, R.; Vogtle, F. Synlett 1999, 12,
1887±1890.
35. Shigekawa, H.; Miyake, K.; Sumaoka, J.; Harada, A.;
Komiyama, M. J. Am. Chem. Soc. 2000, 122, 5411±5412.
36. Bard, A. J. Chem. Eng. News 1999, 77, 5.
37. Funk and Wagnalls New International Dictionary of the
English Language, World Publishing: Chicago, Comprehen-
sive Ed., 1995, p762.
15. Deetz, M. J.; Shang, M.; Smith, B. D. J. Am. Chem. Soc. 2000,
122, 6201±6207.
16. Shukla, R.; Deetz, M. J.; Smith, B. D. Chem. Commun. 2000,
2397±2398.
17. Hughes, M. P.; Smith, B. D. J. Org. Chem. 1997, 62, 4492±
4501.
18. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L.
Chem. Rev. 1991, 91, 1721±2085.