The influence of macrocyclic polyether constitution upon ammonium ion/crown ether recognition processes

Article abstract of DOI:10.1002/1521-3765(20000616)6:12<2274::AID-CHEM2274>3.0.CO;2-2

Secondary dialkylammonium (R₂NH₂⁺) ions are hound readily by dibenzo[24]crown-8 (DB24C8) to form threaded complexes, namely [2]pseudorotaxanes. The effect of replacing one or both of the catechol rings in DB24C8 with resor

Full text of DOI:10.1002/1521-3765(20000616)6:12<2274::AID-CHEM2274>3.0.CO;2-2

FULL PAPER
Molecular Meccano, Part 60=
The Influence of Macrocyclic Polyether Constitution upon Ammonium
Ion/Crown Ether Recognition Processes
Stuart J. Cantrill,[a] David A. Fulton,[a] Aaron M. Heiss,[a] Anthony R. Pease,[a]
J. Fraser Stoddart,*[a] Andrew J. P. White,[b] and David J. Williams*[b]
1
Abstract: Secondary dialkylammonium
(R2NH2 ) ions are bound readily by
aneÐthe reaction of an amine with an
isocyanate to form a ureaÐwas tested
initially on a system incorporating
their protons by H NMR spectroscopy.
Unambiguous assignments can be made
as a result of the fact that the protons on
each face of the macrocyclic polyether
experience a unique set of through-
space interactions, as evidenced by
T-ROESY experiments. Additionally,
the two-dimensional NMR analyses are
in agreement with the X-ray crystallo-
graphic studies performed on these
[2]rotaxanes, indicating that the crown
ethers are located intimately around the
‡
dibenzo[24]crown-8 (DB24C8) to form
threaded complexes, namely [2]pseudo-
rotaxanes. The effect of replacing one or
both of the catechol rings in DB24C8
with resorcinol rings upon the crown
DB24C8 and was shown to work effi-
ciently. Both [2]rotaxanes have been
1
fully characterized by H and 13C NMR
spectroscopies, FAB mass spectrometry
and X-ray crystallography. Interestingly,
the unsymmetrical nature of the dumb-
bell-shaped component in each of the
two [2]rotaxanes renders each face of
the encircling macrocyclic polyether di-
astereotopic, a feature that is apparent
‡
etherꢀs ability to bind R2NH2 ions has
now been investigated. When only one
aromatic ring is changed from catechol
to resorcinol, a crown ether with a
[25]crown-8 constitution is createdÐ
namely benzometaphenylene[25]crown-
8 (BMP25C8). A [2]pseudorotaxane is
formed in the solid state when
BMP25C8 is co-crystallized with diben-
zylammonium hexafluorophosphate, as
evidenced by its X-ray crystal structure.
Furthermore, this crown ether has been
‡
NH2 centers as expected. Replacement
1
upon inspection of their H NMR spec-
of both catechol rings in the DB24C8
constitution with resorcinol rings results
in a crown ether with a [26]crown-8
constitutionÐnamely bismetaphenyl-
ene[26]crown-8 (BMP26C8). All the
evidence to date points to the fact that
this further change in constitution re-
sults in a crown ether that does not bind
tra. The resonances associated with the
diastereotopic protons on each face of
the macrorings are well enough resolved
to enable the faces of the crown ethers
to be readily identified with respect to
‡
shown to bind R2NH2 ions in solution,
an observation which has been exploited
in the synthesis of the first BMP25C8-
containing [2]rotaxane. The methodol-
ogy employed to generate this [2]rotax-
Keywords: crown ethers ´ hydrogen
‡
R2NH2 ions in either the solution or
bonds
´
rotaxanes
´
secondary
solid states.
ammonium ions ´ self-assembly
Introduction
[a]Prof. J. F. Stoddart, S. J. Cantrill, D. A. Fulton, Dr. A. M. Heiss,
A. R. Pease
For many years, it has been recognized[1, 2] that crown ethers
are exceptionally versatile hosts for a wide variety of cationic
guests. Amongst the many guests that have been investigated
are substituted ammonium ions, and although initially[3] the
Department of Chemistry and Biochemistry
University of California, Los Angeles
405 Hilgard Avenue, Los Angeles, CA 90095-1569 (USA)
Fax : (‡1)310-206-1843
E-mail: stoddart@chem.ucla.edu
‡
focus fell upon primary alkylammonium ions (RNH3 ), more
[b]Prof. D. J. Williams, Dr. A. J. P. White
Chemical Crystallography Laboratory
Department of Chemistry, Imperial College
South Kensington, GB-London, SW7 2AY (UK)
Fax : (‡44)20-7594-5835
‡
recently, secondary dialkylammonium ions (R2NH2 ) have
received[4] considerable attention. The discovery[5] that
‡
R2NH2
ions are complexed by dibenzo[24]crown-8
(DB24C8) (1) in a threaded rather than a face-to-face[6]
manner, heralded the arrival of a new paradigm for the
construction of 1) discrete interlocked molecules[7] and 2)
extended interwoven supramolecular arrays.[8] Indeed, the
[=]Part 59, see: B. Cabezon, J. Cao, F. M. Raymo, J. F. Stoddart, A. J. P.
White, D. J. Williams, Chem. Eur. J. 2000, 6, 2262 ± 2273.
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
2274
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2274 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 12

2274±2287
‡
R2NH2 ion/DB24C8 supramolecular synthon has proven to
be extremely effective in the self-assembly[9] of many different
kinds of architectures.[10] However, it is not without its
problems. Our investigations[11] into the fabrication of higher
order multicomponent aggregates[12]Ðrich in interlocked and/
or intertwined motifsÐnecessitates the functionalization of
either the crown ether or the dialkylammonium ion compo-
nents in order to facilitate the formation of more elaborate
threaded host ± guest complexes. Formally, monosubstitution
of one of the catechol rings of DB24C8 reduces the local
symmetry of the macrocyclic polyether such that subsequent
incorporation into interlocked molecules may result in the
formation of isomeric compounds. This problem can be
alleviated by the replacement (Figure 1) of either one or
both[13] of the catechol with resorcinol rings. The fact of the
matter is that substitution at the C-5 position on a resorcinol
ring does not desymmetrize a macrocyclic polyether into
which it has been incorporated. Herein we report our findings
concerning the relative propensities with which DB24C8 (1),
benzometaphenylene[25]crown-8 (BMP25C8) (2), and bis-
O
O
O
O
O
O
O
O
X
Y
X
Y
(DB24C8)
1
2
(BMP25C8)
*
*
(BMP26C8)
3
‡
metaphenylene[26]crown-8 (BMP26C8) (3) bind R2NH2 ions
*
to generate 1) [2]pseudorotaxanes (Figure 2) and subsequent-
ly 2) [2]rotaxanes, formed by the kinetic trapping of these
supramolecular complexes.
Figure 1. A generic crown ether frameworkÐcontaining eight oxygen
atomsÐwhich incorporates two diametrically opposed aromatic rings
which are derived from catechol or resorcinol. Therefore, the three possible
permutations are 1) catechol/catechol that is, dibenzo[24]crown-8
(DB24C8) 1, 2) catechol/resorcinol that is, benzometaphenylene[25]-
crown-8 (BMP25C8) 2, and 3) resorcinol/resorcinol that is, bismetaphenyl-
Results and Discussion
ene[26]crown-8 (BMP26C8) 3. The constitutions of
2 and 3 allow
substitution at the C-5 position (*) of either one or both of the aromatic
units, respectively, without changing the local symmetry of the crown ether.
Dibenzo[24]crown-8 (DB24C8)
Background: Substantial evi-
dence has been gathered[4, 5]
O
O
O
O
O
O
X
X
which
demonstrates
that
O
O
+
+
+
O
O
‡
DB24C8 binds R2NH2 ions in
the solution, solid, and gas
phases by threading to generate
[2]pseudorotaxanes. Unequivo-
cal proof of such a binding
motif has been established[4b]
by a ꢂstopperingꢀ approach in
which the ends of a reactive
threadlike componentÐposi-
tioned through the center of
the crown etherꢀs macroring±
are allowed to react with a large
capping group, affording (Fig-
ure 3) a [2]rotaxane. Recently,
we communicated[7g] our pre-
liminary findings of an investi-
gation into the utility of bulky
isocyanates as the reactive
ꢂstopperingꢀ reagents and now
present our results, regarding
this new protocol, in detail.
Y
Y
O
O
O
O
O
O
+
N
H2
R2
R1
+
4-H·PF6
5-H·PF6
R1 = R2 = H
R1 = R2 = CO2Me
Figure 2. A schematic representation depicting how a generic crown ether can act as a host for substituted
secondary dibenzylammonium ions. The cavity of the crown ether is pierced by the threadlike cation, resulting in
the formation of a [2]pseudorotaxane which is stabilized predominantly by N H ´´´ O and C H ´´´ O hydrogen-
bonding interactions.
‡
À
À
X
X
X
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
x 2
x 1
O
O
O
O
O
O
O
O
O
a)
b)
Y
Y
Y
[2]Pseudorotaxane
[2]Rotaxane
[2]Semirotaxane
Figure 3. A schematic representation depicting two modifications of the threading-followed-by-stoppering
approach for the synthesis of [2]rotaxanes. Either a) a [2]pseudorotaxane is formed, which is subsequently capped
at both ends with large ꢂstopperꢀ groups, or b) a semirotaxaneÐwhich contains a threadlike molecule with one
bulky group already attachedÐis assembled, which is ꢂstopperedꢀ at the terminus over which the crown ether had
to pass to form the 1:1 complex.
Synthesis: The template-direct-
ed synthesis of the [2]rotaxane
10-H ´ O2CCF3 is illustrated in
Scheme 1. 4-Aminobenzyl-3,5-
Chem. Eur. J. 2000, 6, No. 12
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2275 $ 17.50+.50/0
2275

J. F. Stoddart, D. J. Williams et al.
FULL PAPER
O
H
a)
+
N
H2
a, b, c
O
O
O
+
NH3
H2
H3
H5
O
O
H4
H6
–
+
67 %
2Cl
a
O
b
N
O
H
2 O
H1
8-H2·2Cl
6
N
N
O
H
H
b
a
O
O
O
–
CF3CO2
H8
d, e, f
64 %
O
O
O
b)
+
H7
N
O
H
2 O
α2
β2
α1
O
N
N
O
H
H
β1
γ2
O
γ1
10-H·O2CCF3
Scheme 1. The synthesis of the [2]rotaxane 10-H ´ O2CCF3: a) 4-azidoben-
zylamine, C6H6, reflux; b) NaBH4/MeOH, room temperature; c) HCl,
CH2Cl2/Et2O, overall 67%; d) NaOH, H2O/CH2Cl2; e) DB24C8 (1),
CF3CO2H, CH2Cl2; f) 2,6-diisopropylphenylisocyanate (9), room temper-
ature, overall 64%.
Figure 4. The structure of the [2]rotaxane 10-H ´ O2CCF3, showing the
labeling scheme for both a) the dumbbell and b) the crown ether
components, used in describing its NMR spectroscopic properties. The
schematic representation b) highlights the unsymmetrical nature of the
dumbbell. The protons (a1, b1, g1) on one face of the DB24C8 macrocycle
are oriented toward the 3,5-di-tert-butylphenyl stopper, whereas those
protons (a2, b2, g2) on the opposite face of the macrocycle are directed
toward the 2,6-diisopropylphenyl stopper, that is, the protons located on
opposite faces of the crown ether are diastereotopic.
di-tert-butylbenzylamine (8) was prepared by NaBH4 reduc-
tion of the imine generated from the condensation of 3,5-di-
tert-butylbenzaldehyde (6)[14] and 4-azidobenzylamine (7).[15]
It was isolated as its dihydrochloride salt 8-H2 ´ 2Cl in an
overall yield of 67%. Deprotonation of this salt was followed
by the addition of one equivalent of CF3CO2H which
presumably brought about the monoprotonation of the
diamine 8 at the more basic secondary dibenzylamine site.
the macrocycle. The facial desymmetrization of the crown
ether is evident and has a profound effect upon the 1H NMR
spectrum (Figure 5) of this compound. Although the reso-
nances for the a-OCH2 protons overlap, those associated with
the b- and g-OCH2 protons are separated such that distinct
multiplets are observed for b1- and b2-OCH2 and g1- and g2-
OCH2 protons. The unambiguous assignment of the super-
scripts ꢂ1ꢀ and ꢂ2ꢀ (denoting the face of the macrocycle on
which a particular proton resides) was made upon inspection
of the T-ROESY spectrum. Two important probe protons on
the dumbbellꢀs backbone are H-2 and H-3. Should, as
‡
As a consequence, an R2NH2 ion is generated which is
capable of recognizing the DB24C8 molecules also present in
the solution. Addition of 2,6-diisopropylphenylisocyanate,
followed by work-up and subsequent chromatographic puri-
fication, afforded the [2]rotaxane 10-H ´ O2CCF3 as the major
product in a 64% yield. This result supports the hypothesis
that a reactive [2]pseudorotaxane must be formed under the
reaction conditions since the interlocked molecule 10-H ´
O2CCF3 can only be obtained via the intermediacy of a
threaded 1:1 complex.
expected,[16] the DB24C8 macroring encircle the NH2 center,
‡
in preference to the urea moiety, H-2 and H-3 will flank the
macrocycle. Indeed, this hypothesis is confirmed upon in-
spection of the partial T-ROESY spectrum shown in Figure 6.
Through-space interactions are observed[17] between H-2, on
the dumbbell, and all of the OCH2 protons (a1, b1 and g1) on
only one side of the macrocycle. The corresponding pattern is
observed for H-3, which only correlates to resonances arising
from the a2-, b2- and g2-OCH2 protons residing on the
opposite face of the crown ether.
NMR spectroscopy: The full assignment of the 1H NMR
spectrum and in-depth analysis of the 13C NMR spectrum (see
Experimental Section) was made possible by two-dimensional
1
1H-1H correlation (COSY) and H-13C correlation (HMQC)
experiments. The unsymmetrical nature of the dumbbell-
shaped component of the [2]rotaxane 10-H ´ O2CCF3 renders
the protons on each face of the encircling macrocycle
heterotopic (Figure 4). When a symmetrical thread/dumbbell
is bound/encircled by a DB24C8 macroring, the resonances
associated with the a- and b-OCH2 protons of the crown ether
appear as narrow multiplets, and the resonance of the g-OCH2
protons gives rise to a singlet. In the case of 10-H ´ O2CCF3,
there are now two distinct environments for each of these
pairs of protons, respectively, that is, the protons (a1, b1, and
g1) on the side of the macrocycle facing toward the 3,5-di-tert-
butylphenyl stopper are different from those (a2, b2, and g2)
facing the 2,6-diisopropylphenyl stopper on the other side of
X-ray crystallography: Single crystals, suitable for X-ray
crystallographic analysis, were obtained upon layering a
CH2Cl2 solution of the [2]rotaxane 10-H ´ O2CCF3 with
hexanes. The X-ray crystal analysis of the [2]rotaxane shows
the crystals to contain two independent molecules in the
‡
asymmetric unit. In both molecules, it is the NH2 center that
is encircled by the DB24C8 macrocycle with the urea
component hydrogen bonded to the trifluoroacetate counter-
ion. Stabilization in both independent molecules is by a
‡
À
À
combination of N H ´´´ O and C H ´´´ O hydrogen bonding
interactions (Figure 7). The conformation of the cationic
2276
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2276 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 12

Influence of Crown Ether Constitution on Binding
2274±2287
1
Figure 5. The partial H NMR spectrum (400 MHz, CD2Cl2) of 10-H ´ O2CCF3.
Table 1, respectively). Howev-
er, the conformations of the two
independent DB24C8 macrocy-
cles differ, both crown ethers
having an extended geometry,
but with distinctly different ge-
ometries for one of the poly-
ether linkages. In both inde-
pendent molecules, there is an
apparent overlaying of one of
the catechol rings and the p-
toluidinyl ring of the cationic
dumbbell. The centroid ± cent-
roid distances (4.48 Š in i and
4.54 Š in ii) are, however, too
great for any significant intra-
molecular p ± p stacking inter-
actions. There is also a marked
absence of any inter-[2]rotax-
ane interactions, although there
is evidence for a weak intermo-
À
lecular C H ´´´ p interaction
À
(H ´´´ p 2.73 Š, C H ´´´ p angle
1488) between one of the meth-
ylene hydrogen atoms in one of
the polyether chains in mole-
cule ii and one of the catechol
Figure 6. The partial T-ROESY spectrum (400 MHz, CD2Cl2) of 10-H ´ O2CCF3. The most significant probe
protons are H-2 and H-3, which each show a correlation to only one set of protons on only one face of the crown
ether.
rings of
a symmetry-related
[2]rotaxane.
dumbbell in both molecules is essentially the same with the
Benzometaphenylene[25]crown-8 (BMP25C8)
plane of the urea fragment steeply inclined to the plane of the
terminal 2,6-diisopropylphenyl ring but lying close to the
plane of the central p-toluidinyl ring (w and x in Table 1,
Background: Replacement of one of the catechol rings of
DB24C8 with resorcinolÐaffording a [25]crown-8 deriva-
tiveÐis desirable in the context of preserving local symmetry
should subsequent functionalization of the crown ether be
desired. However, this strategy is only sustainable if such a
‡
respectively). The plane of the all-anti CCH2NH2 CH2C
backbone is steeply inclined to both the central p-toluidinyl
and the 3,5-di-tert-butylphenyl ring systems (y and z in
Chem. Eur. J. 2000, 6, No. 12
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2277 $ 17.50+.50/0
2277

J. F. Stoddart, D. J. Williams et al.
FULL PAPER
O
O
O
O
O
O
O
O
O
a
OH
OH
OTs
OTs
64 %
O
O
O
11
12
O
O
O
O
O
O
b
O
O
83 %
2
Scheme 2. The synthesis of BMP25C8 (2): a) TsCl, N,N-DMAP, Et3N,
CH2Cl2, 64%; b) catechol (13), Cs2CO3, MeCN, reflux, 83%.
12. Subsequent macrocyclization of 12 with catechol 13, in the
presence of base, gave the crown ether in good yield. Single
crystals suitable for X-ray crystallographic analysis were
grown from an ethanolic solution upon slow evaporation.
Figure 7. The molecular structures of the two crystallographically inde-
pendent [2]rotaxanes present in the crystals of 10-H ´ O2CCF3. Hydrogen
X-ray crystallography: The X-ray analysis of BMP25C8 shows
(Figure 8a) the macrocycle to have an extended geometry
À
bonding distances and angles [(X ´´´ O), (H ´´´ O) distances [Š], (X H ´´´
O) angles [8]]: for molecule i; a) 2.85, 2.01, 156; b) 2.89, 2.16, 137; c) 3.31,
2.36, 171; e) 2.89, 2.03, 159; f) 2.86, 1.98, 164; for molecule ii; a) 2.91, 2.11,
147; b) 3.01, 2.23, 145; c) 3.41, 2.51, 157; d) 3.25, 2.41, 147; e) 2.76, 1.92, 163;
f) 2.91, 2.02, 168. The centroid ± centroid distances g) for molecules i and ii
are 4.48 and 4.54 Š, respectively.
Table 1. The torsion angles (w, x, y, and z) observed for the DB24C8- and
BMP25C8-containing rotaxanes, 10-H ´ O2CCF3 and 14-H ´ O2CCF3, re-
spectively. In each structure, there are two crystallographically independent
molecules (i and ii) in the asymmetric unit.
Torsion angles [8]
Structure
w
x
y
z
{DB24C8}
10-H ´ O2CCF3 (i)
10-H ´ O2CCF3 (ii)
84
70
10
16
86
85
64
63
{BMP25C8}
14-H ´ O2CCF3 (i)
14-H ´ O2CCF3 (ii)
72
90
4
3
75
86
59
54
modification in the macrocyclic polyetherꢀs backbone does
not hinder significantly its ability to bind R2NH2 ions.
Figure 8. a) The crystal structure of BMP25C8 2. b) Adjacent molecules
‡
À
are stacked by virtue of pairs of C H ´´´ p interactions, a) H ´´´ p distance
À
À
2.85 Š, C H ´´´ p angle 1408; b) H ´´´ p distance 2.79 Š, C H ´´´ p angle
1378, supplemented by p ± p stacking interactions between the resorcinol
and catechol moieties, interactions c and d, respectively. c) End-on view of
the constricted nanotube formed by the stacking of BMP25C8 macrocycles.
Synthesis: The synthetic pathway leading to the formation of
BMP25C8 (2) is depicted in Scheme 2. Tosylation of the
diol[18] 11, under standard conditions, afforded the ditosylate
2278
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2278 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 12

Influence of Crown Ether Constitution on Binding
2274±2287
with a fairly large central pathway through the center of the
macroring. The molecules stack along the crystallographic b
50mÀ1. Under comparable conditions, the Ka value for the
equilibrium between DB24C8 and 5-H ´ PF6 was deter-
mined[22] to be 1100mÀ1. Therefore, BMP25C8 does bind[23]
R2NH2 salts, though with a substantially lower Ka value
(approximately one order of magnitude less) than that
observed with the [24]crown-8 analogue.
À
direction and are linked by pairs of C H ´´´ p interactions (a
‡
and b in Figure 8b) involving one of the phenoxymethylene
hydrogen atoms of both the catechol and resorcinol rings and
their adjacent counterparts within the stack. These interac-
tions are supplemented, to a lesser degree, by partial p ± p
overlap between the resorcinol rings and between the
catechol rings (c and d in Figure 8b) of adjacent molecules
(the centroid ± centroid and interplanar separations are 5.14,
3.44 Š and 5.14, 3.52 Š, respectively). The combination of
these intermolecular interactions produces (Figure 8c) con-
stricted nanotubes that extend through the crystal. Adjacent
nanotubes, in one direction in
X-ray crystallography: Single crystals of the [2]pseudorotax-
ane, suitable for X-ray crystallographic analysis, were ob-
tained from a CD2Cl2 solution of an equimolar mixture of 2
and 4-H ´ PF6 upon layering with hexanes. The 1:1 complex
formed between the dibenzylammonium cation and
BMP25C8 shows (Figure 9) the ion to be threaded through
the crystal, are crosslinked by
À
an additional C H ´´´ p interac-
tion between one of the resor-
cinol phenoxymethylene hydro-
gen atoms in one stack and a
catechol ring in the next (the
H ´´´ p distance is 2.91 Š and
À
the C H ´´´ p angle 1488).
Mass spectrometry: The FAB
mass spectrum of a 1:1 mixture
of BMP25C8 and 4-H ´ PF6 (Fig-
ure 2, R1 ˆ R2 ˆ H) reveals
that, in the gas phase, a strong
‡
Figure 9. The X-ray crystal superstructure of the [2]pseudorotaxane [2 ´ 4-H] formed between BMP25C8 (2) and
1:1 complex is formed between
these two components. Al-
though the base peak corre-
sponds to the uncomplexed di-
‡
the dibenzylammonium cation 4-H . Hydrogen bonding distances and angles [(N ´´´ O), (H ´´´ O) distances [Š],
(N H ´´´ O) angles [8]]: a) 2.91, 2.06, 156; b) 2.94, 2.31, 127; c) 3.08, 2.21, 162.
À
‡
the center of the polyether macrocycle which has an extended
geometry similar to that in its uncomplexed state. The two
phenyl rings in the cation are both fairly steeply inclined (by
41 and 688, respectively) to the near planar all-anti
CCH2NH2 CH2C backbone. Stabilization of the [2]pseudor-
otaxane is via N H ´´´ O hydrogen bonding, supplemented
by a weak p ± p stacking interaction (d in Figure 9) between
the catechol ring in the crown ether macrocycle and one of the
phenyl rings of the cation. These two ring systems are inclined
by about 128 and have a centroid ± centroid separation of
4.02 Š. The [2]pseudorotaxanes are linked by a combination
benzylammonium cation 4-H , a relatively intense signal
‡
(50%) arising from the 1:1 complex [2 ´ 4-H] appears at m/z
646, suggesting[19] that the complex has a [2]pseudorotaxane,
rather than a face-to-face, geometry.
‡
‡
1
À
NMR spectroscopy: The H NMR spectrum of a 1:1 CD2Cl2
solution of BMP25C8 and 4-H ´ PF6 reveals the occurrence of
complexation. However, the spectrum consists of both broad
and sharp peaksÐrather than a series of sharp well-defined
peaks arising from the slow exchange of the free components
and the [2]pseudorotaxane (as is the case with DB24C8)Ð
indicating that, under the experimental conditions, neither a
slow- nor fast-exchange regime is operating. This behavior
precludes the determination of a stability constant by either
the single-point method[20] or dilution/titration[21] techniques.
By employing the bis(4-methoxycarbonylbenzyl)ammonium
À
of p ± p and C H ´´´ p interactions to form sheets (Figure 10).
There are no obvious interactions involving the PF6 ions.
À
The rotaxane 14-H ´ O2CCF3
‡
ion 5-H Ða threadlike molecule (Figure 2, R1 ˆ R2 ˆ
Synthesis: Encouraged by the realization that BMP25C8 (2)
‡
can accommodate an R2NH2 ion within its macrocyclic
CO2Me) with slightly increased steric bulk at its terminiÐ
we have shown[22] that a Ka value can be obtained for the
cavity, the construction of a [2]rotaxane incorporating this
crown ether was our next goal. By employing the same
methodology as that shown to work for DB24C8 (1), the
BMP25C8-containing [2]rotaxane 14-H ´ O2CCF3 was ob-
tained (Scheme 3) in 36% yield.
binding of an R2NH2 ion by BMP25C8. 1H NMR spectra of a
‡
1:1 mixture of BMP25C8 and 5-H ´ PF6 in CD3CN solution
were recorded at 243, 258, 273, and 288 K. At these temper-
atures, equilibration between the [2]pseudorotaxane and its
free components is slow on the 1H NMR time scale, allowing
single-point determinations[20] of Ka values at each one of
these temperatures. Extrapolation of the vanꢀt Hoff plot
obtained using this data gives a value for Ka at 300 K of about
NMR spectroscopy: As in the case of the DB24C8-containing
[2]rotaxane 10-H ´ O2CCF3, the macrocyclic polyether com-
ponent in 14-H ´ O2CCF3 is also desymmetrized facially as a
Chem. Eur. J. 2000, 6, No. 12
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2279 $ 17.50+.50/0
2279

J. F. Stoddart, D. J. Williams et al.
FULL PAPER
+
N
a, b, c
H2
+
NH3
–
2Cl
36 %
8-H2·2Cl
O
O
–
CF3CO2
O
O
+
N
O
H
2 O
O
N
H
N
O
O
H
14-H·O2CCF3
Scheme 3. The synthesis of the [2]rotaxane 14-H ´ O2CCF3: a) NaOH,
H2O/CH2Cl2; b) BMP25C8 (2), CF3CO2H, CH2Cl2; f) 2,6-diisopropylphe-
nylisocyanate (9), room temperature, overall 36%.
Figure 10. The sheet-like superstructure formed by the [2]pseudorotax-
‡
anes [2 ´ 4-H] . The geometries of the p ± p stacking interactions are a)
centroid ± centroid distance 4.02 Š, rings inclined by 128; b) centroid ±
centroid distance 4.52 Š, mean interplanar separation 3.68 Š; c) cent-
roid ± centroid distance 3.99 Š, mean interplanar separation 3.71 Š. The
of there being six sets of chemically inequivalent OCH2
groups (a, b, g, d, e and f) in each polyether arc in BMP25C8,
as opposed to only three (a, b and g) in the case of DB24C8.
Once again, a combination of two-dimensional NMR techni-
quesÐnamely, COSY, T-ROESY and HMQC experimentsÐ
À
C H ´´´ p interaction (d) is characterized by an H ´´´ p distance of 2.84 Š
À
and a C H ´´´ p angle of 1408.
1
consequence of the unsymmetrical nature of the dumbbell-
shaped component. A sharp well-resolved 1H NMR spectrum
(400 MHz, CD2Cl2) was obtained (Figure 11) for 14-H ´
O2CCF3. The separation of the resonances arising from the
crown etherꢀs polyether protons is evident. Additionally, the
spectroscopic characteristics of the BMP25C8 component are
considerably more complex than those of DB24C8 on account
permitted a complete interpretation of the H NMR spec-
trum. Each unique face of the macrocycle could be distin-
guished by inspection of the T-ROESY spectrum, wherein, as
before, the resonances arising from the highly diagnostic H-2
and H-3 protons of the dumbbell only correlated (Figure 12)
with the resonances associated with the OCH2 protons located
on the side of the macrocyclic polyether facing them. Several
Figure 11. The partial 1H NMR spectrum (400 MHz, CD2Cl2) of 14-H ´ O2CCF3. Note that the broad peak at d ˆ 9.4 is a background signal arising from the
NMR probe.
2280
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2280 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 12

Influence of Crown Ether Constitution on Binding
2274±2287
angles of 908 and 08 respective-
ly. All torsion angles were un-
constrained except those being
rotated. After construction of
the model compound within the
INPUT submode of MacroMo-
del 5.5,[25] each set of torsion
angles was varied sequentially,
and the energy was minimized
utilizing MM3*Ðthe Macro-
Model implementation of the
MM3 force field[26] by the
a)
O
O
O
O
H2
H3
H5
+
a
O
b
2 O
N
H4
H6
H
O
O
O
H1
H2
H3
N
N
H
H
b
a
H11
H10
b)
PRCG algorithm[27] using
a
O
φ
maximum of 10000 iterations
or until the final energy gradi-
ent fell below 0.05 kJŠÀ1. Sol-
vation was considered through
use of the GB/SA model[28] for
chloroform. Each energy value
was scaled relative to the lowest
energy point found on the po-
tential surface. The average
height of the energy barrier
corresponding to rotation of
H9
ε
O
O
δ
γ
1
2
β
Figure 13. Partial 13C NMR
spectra (125 MHz, CD3CN:
CD3SOCD3 3:1), recorded at
various temperatures, showing
the coalescence of the signal
arising from the carbon atom of
the methyl groups that consti-
tute the isopropyl groups.
O
α
H7
H8
Figure 12. The structure of the [2]rotaxane 14-H ´ O2CCF3, showing the
labeling scheme for the protons on both a) the dumbbell and b) the crown
ether components, used in describing its NMR-spectroscopic parameters.
The through-space correlations determined by T-ROESY measurements
are also highlighted in a) by double-headed arrows. The schematic
representation (b) highlights the unsymmetrical nature of the dumbbell.
The protons (a1, b1, g1, d1, e1, f1) on one face of the BMP25C8 macrocycle
are oriented toward the 3,5-di-tert-butylphenyl stopper, whereas those
protons (a2, b2, g2, d2, e2, f2) on the opposite face of the macrocycle are
directed toward the 2,6-diisopropylphenyl stopper.
an isopropyl group is roughly
À1
À
6.5 kcalmol , whereas rotation of the N Ar bond is signifi-
cantly more hindered, showing an average energy barrier of
14 kcalmolÀ1, a value which is in good agreement with that
determined experimentally (15.0 ± 15.2 kcalmolÀ1). The en-
ergy profile is dominated by two large ꢂmountainsꢀ, corre-
sponding to the direct clashing of the ureido oxygen atom with
the methyl groups of the isopropyl unit, and two smaller
ꢂmountainsꢀ corresponding to steric interactions between the
through-space correlations were detected in the T-ROESY
experiment: some of the more important ones are indicated in
Figure 12. Perhaps the most notable interaction is that
observed between the H-9 resonanceÐthe isolated aromatic
proton attached to the resorcinol ring present in the
BMP25C8 componentÐand the two benzylic methylene
‡
groups adjacent to the NH2 center of the dumbbell. Such
an interaction is not possible in a DB24C8 system since there
are no protons oriented toward the center of the macrocyclic
cavity. Another interesting featureÐalso observed in the case
of 10-H ´ O2CCF3Ðis the inequivalence of the methyl groups
that constitute the isopropyl moieties on one of the stoppers.
This heterotopicity is manifest in the 13C NMR spectrum,
wherein two resonances are observed for the carbon atoms of
the isopropyl methyl groups. On warming up an NMR sample
of 14-H ´ O2CCF3 dissolved in CD3CN/CD3SOCD3 (3:1), the
two resonances are observed (Figure 13) to coalesce[24]
between 312 and 317 K, indicating that the inequivalence
arises as a consequence of a slow bond rotation with an energy
³
barrier in the range of DG ˆ 15.0 ± 15.2 kcalmolÀ1. In order to
determine the relative energetics of rotation about the groups
attached to the 2,6-disubstituted stopper, a molecular me-
chanics investigation was undertaken. A Ramachandran-like
plot (Figure 14) was constructed, whereby the torsion angles
of the bonds from the phenyl group to the urea (defined by
atoms Co-Ci-Nu-Hu and referred to as q) and to the isopropyl
group (defined by atoms Ci-Co-Cm-Hm and referred to as f)
were rotated by 3608 in 108 increments from their initial
Figure 14. A Ramachandran-like plot depicting the energy profile of bond
rotations associated with the 2,6-diisopropylphenyl stopper moiety.
Chem. Eur. J. 2000, 6, No. 12
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2281 $ 17.50+.50/0
2281

J. F. Stoddart, D. J. Williams et al.
FULL PAPER
methyl groups and the ureido hydrogen atom. The rest of the
profile is divided into two broad valleys running parallel to the
axis denoted f, reflecting relatively facile rotation of the
isopropyl group with respect to the phenyl ring. These valleys
are lined with fairly high walls in the direction of q, implying
that rotation of the urea group is hindered more than rotation
of the isopropyl units. The energy barrier to rotation about q
of 6.5 kcalmolÀ1 is easily surmountable at room temperature,
whereas the 14 kcalmolÀ1 barrier will be significantly less
readily overcome. It is reasonable to assume that the
isopropyl groups are rotating rapidly about the f bond,
whereas rotation about the urea ± phenyl bond is rapid on the
NMR time scale only at elevated temperatures, as indicated
by the coalescence between 312 and 317 K of the signal arising
from the methyl carbon atom.
two crystallographically independent molecules of the [2]ro-
taxane 14-H ´ O2CCF3 is in the relative orientations of the
resorcinol ring systems with respect to the macrocycle as a
whole. Stabilization of the [2]rotaxane is by a combination of
‡
À
À
N
H ´´´ O and C H ´´´ O hydrogen bonds between the cation
and the polyether macrocycle. As in 10-H ´ O2CCF3, the urea
component of the cation is hydrogen bonded to the trifluor-
oacetate anion. Also, in common with this [2]rotaxane, there
is a similar apparent overlap between a catechol ring in the
BMP25C8 macrocycle and the central p-toluidinyl ring of the
cation. In both independent molecules, the two ring systems
are inclined (by 19 and 248 in molecules i and ii, respectively)
and the centroid ± centroid separations are fairly long (4.22
and 4.29 Š in i and ii, respectively), thus minimizing any
potential p ± p interactions. It is interesting to note that the
geometric relationship observed between these two ring
systems in both independent molecules in 10-H ´ O2CCF3
and 14-H ´ O2CCF3 is probably a consequence of the adoption
of a ꢂpreferredꢀ conformation for the cation and the ensuing
sterically/hydrogen bonded enforced geometric relationship
between the two components. The only inter-[2]rotaxane
X-ray crystallography: Single crystals, suitable for X-ray
crystallographic analysis, were obtained upon layering a
CH2Cl2 solution of the [2]rotaxane 14-H ´ O2CCF3 with
hexanes. The structure (Figure 15) of the [2]rotaxane utilizing
À
feature of note is a C H ´´´ p interaction between H-11 (as
defined in Figure 12) of the resorcinol ring of molecule i and
the 2,6-diisopropylphenyl ring of the cation in molecule ii (the
À
H ´´´ p distance is 2.77 Š and the associated C H ´´´ p angle is
1548).
Bismetaphenylene[26]crown-8 (BMP26C8)
Background: In BMP26C8 (3),[29] both of the catechol rings
present in the parent DB24C8 (1) framework, have been
replaced by resorcinol rings, allowing for symmetrical sub-
stitution of both aromatic moieties. However, as in the case of
the [25]crown-8 analogue, this attribute is only usefulÐin the
context of forming extended threaded arraysÐif this partic-
ular crown ether constitution retains an appreciable affinity
‡
for R2NH2 ions.
Binding properties: This crown etherꢀs ability to act as a host
‡
for R2NH2 ions has been studied by Gibson and co-work-
ers.[13] In agreement with our own independent observations,
association in solution of these two components could not be
1
detected by H NMR spectroscopy. However, in contrast to
our observations,[30] mass spectrometric studies reported by
Gibson et. al.,[13] revealed the presence of intense signals
corresponding to 1:1 complexes.
Figure 15. The molecular structures of the two crystallographically inde-
pendent [2]rotaxanes present in the crystals of 14-H ´ O2CCF3. Hydrogen
À
bonding distances and angles [(X ´´´ O), (H ´´´ O) distances [Š], (X H ´´´
Attempted rotaxanation: Our observations indicate that
BMP26C8 (3) does not form threaded complexes with
R2NH2 ions, either in the solution or the gas phase. As a
O) angles [8]]: for molecule i; a) 2.90, 2.07, 152; b) 2.92, 2.17, 140; c) 3.06,
2.26, 149; d) 3.34, 2.48, 149; e) 2.83, 1.95, 163; f) 2.88, 2.00, 166; for molecule
ii; a) 2.83, 1.94, 170; b) 2.95, 2.34, 125; c) 3.13, 2.26, 163; d*) 3.30, 2.43, 150;
e) 2.93, 2.04, 170; f) 2.89, 2.00, 172; g) 3.22, 2.42, 141. The centroid ± centroid
distances h for molecules i and ii are 4.22 and 4.29 Š, respectively.
‡
consequence of these results, we predicted that subjecting
BMP26C8 (3) to the rotaxanation protocolÐemployed suc-
cessfully in the syntheses of DB24C8- and BMP25C8-con-
taining [2]rotaxanes, 10-H ´ O2CCF3 and 14-H ´ O2CCF3, re-
spectivelyÐshould afford no [2]rotaxane. Moreover, analysis
of the crude reaction mixture by FAB mass spectrometry
revealed no signal for the [2]rotaxane. Rather, the major
products were dumbbell-shaped compounds in which either
one or both of the amino functionalities of the diamine 8 had
BMP25C8 (2) as the encircling macrocycle is very similar to
that of the DB24C8 analog 10-H ´ O2CCF3 in that the crystals
contained two independent molecules in the asymmetric unit.
In both of these molecules, the geometry of the cationic
thread is very similar (Table 1) and differs only slightly from
that observed in 10-H ´ O2CCF3. The difference between the
2282
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2282 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 12

Influence of Crown Ether Constitution on Binding
2274±2287
experiments by sonicating the solution for approximately 30 min in an
ultrasound bath. 1H-13C correlation by heteronuclear zero and double
quantum coherence (HMQC) experiments[33] were performed using BIRD
sequence for presaturation of protons not attached to 13C nuclei, and
processed in a 1 K x 1 K matrix. Fast atom bombardment (FAB) mass
spectra were obtained using a ZAB-SE mass spectrometer, equipped with a
krypton primary atom beam, utilizing a m-nitrobenzyl alcohol matrix.
Cesium iodide or poly(ethylene glycol) were employed as reference
compounds. Microanalyses were performed by either the University of
North London Microanalytical Service (UK) or Quantitative Technologies,
Inc (USA).
reacted with 2,6-diisopropylphenylisocyanate 9. This result is
consistent with our belief that BMP26C8 (3) does not bind
R2NH2 ions in a threaded manner.
‡
Conclusion
We have shown that, upon replacing one of the two catechol
rings in the well-known DB24C8 receptor with a resorcinol
moiety, the resulting crown etherÐpossessing a [25]crown-8
constitutionÐis still capable of recognizing and binding
Bis(4-methoxycarbonylbenzyl)ammonium Hexafluorophosphate (5-H ´
PF6):
A
solution of bis(4-methoxycarbonylbenzyl)amine (5)[7b] (1.0 g,
3.2 mmol) in MeOH (50 mL) was treated with 5n HCl (10 mL) and the
solvents were removed in vacuo. The resulting white solid was dissolved in
H2O (50 mL) and an excess of NH4PF6 was added until no further
precipitation occurred. Filtration afforded the title compound as a white
solid (1.35 g, 92%); M.p. 240 ± 2428C; 1H NMR (400 MHz, CD3CN): d ˆ
3.88 (s, 6H; CH3), 4.31 (s, 4H; CH2-N-CH2), 7.55 ± 7.59 (m, 4H; ArH); 8.04-
8.08 (m, 4H; ArH); 13C NMR (100 MHz, CD3CN): d ˆ 52.2 (CH2-N-CH2),
‡
R2NH2 ions, not only in solution, but also in the gas phase
and in the solid state. However, replacing both catechol with
resorcinol rings perturbs the recognition elements so severely
‡
that R2NH2 ion binding is no longer observed in solution. By
utilizing a kinetic stoppering methodologyÐemploying the
reaction of isocyanates with amines to generate ureasÐwe
have demonstrated, initially with DB24C8, and then sub-
sequently with BMP25C8, that [2]rotaxanes incorporating
these crown ethers can be made and isolated. Moreover, the
isolation of these compounds is a testament to the binding
ability of the crown ethers employed during the synthesis, that
is, the immediate precursor of the [2]rotaxane must be the
corresponding [2]pseudorotaxane. The fact that no [2]rotax-
ane is generated when BMP26C8 is the macrocyclic compo-
nent added to the reaction mixture suggests that this crown
53.1 (CH3), 130.9 and 131.5 (aromatic CH), 132.6 and 136.1 (quaternary
‡
ˆ
aromatic), 167.2 (C O); MS (FAB): m/z: 314 [M À PF6] ; C18H20F6NO4P
(459.3): calcd C 47.07, H 4.39, N 3.05; found C 47.12, H 4.28, N 2.84.
4-Aminobenzyl-3,5-di-tert-butylbenzylamine dihydrochloride (8-H2 ´ 2Cl):
4-Azidobenzylamine 7[15] (1.70 g, 11.5 mmol) and 3,5-di-tert-butylbenzal-
dehyde (6)[14] (2.50 g, 11.5 mmol) were dissolved in C6H6 (100 mL), and the
reaction mixture was heated to reflux over a 20 h period during which time
the H2O formed in the reaction was collected by means of a Dean ± Stark
trap. After the reaction mixture was allowed to cool down to ambient
temperature, MeOH (50 mL) was added to it. The subsequent portionwise
addition of NaBH4 (4.35 g, 115 mmol) was completed in 30 min, and the
reaction mixture was left to stir for 4 d under ambient conditions.
Hydrochloric acid (5n, 100 mL) was added and the solvents were removed
in vacuo. The residue was partitioned between NaOH (2n, 250 mL) and
CH2Cl2 (250 mL) and, upon separation, the aqueous portion was washed
further with CH2Cl2 before the organic extracts were combined, dried
(MgSO4), and evaporated to dryness to afford a thick yellow oil. The oil was
dissolved in CH2Cl2 (100 mL) and HCl gas was bubbled through this
solution for 25 min. After stirring for a further 30 min, Et2O (100 mL) was
added and the solution became cloudy. It was reduced in volume to about
100 mL, before adding more Et2O (100 mL). Filtration afforded the title
compound as a pale yellow solid (3.85 g, 67%); m.p. >2008C (decomp);
1H NMR (400 MHz, CD3SOCD3): d ˆ 1.29 (s, 18H; C(CH3)3), 4.07 ± 4.18
(m, 4H; CH2-N-CH2), 7.35 (d of an AA'BB', J ˆ 8.4 Hz, 2H; ArH), 7.40 (m,
1H; ArH), 7.44 (m, 2H; ArH), 7.65 (d of an AA'BB', J ˆ 8.4 Hz, 2H; ArH),
‡
ether does not allow for the threading of R2NH2 ions through
its macrocyclic cavity. In summary, the ability to add
substituents at C-5 on the resorcinol ring, without reducing
the symmetry of the BMP25C8 macrocycleÐsomething that
is not possible with DB24C8Ðin conjunction with its affinity
‡
for R2NH2 ions, suggests that BMP25C8 is a potentially
useful building block for the construction of interlocked
molecular architectures.
‡
9.84 (br s, 2H; NH2 ); 13C NMR (100 MHz, CD3SOCD3): d ˆ 31.2 (CH3),
34.6 (C(CH3)3), 49.3 and 50.4 (CH2-N-CH2), 122.2, 122.7, 124.4, 130.8, 131.1,
Experimental Section
‡
131.6, 133.7 and 150.6 (aromatic C); MS (FAB): m/z: 326 [M À 2Cl] ;
C22H34N2Cl2 ´ 0.5H2O (406.4): calcd C 65.01, H 8.68, N 6.89; found C 65.12,
H 8.39, N 6.93.
General: ChemicalsÐincluding DB24C8 (1)Ðwere purchased from Al-
drich and used as received unless indicated otherwise.BMP26C8 (3),[29] 4-
H ´ PF6,[4b] 3,5-di-tert-butylbenzaldehyde (6)[14] and 4-azidobenzylamine
(7)[15] were prepared according to literature procedures. Anhydrous CH2Cl2
and MeCN were obtained by distillation from CaH2 under an inert
atmosphere (N2 or Ar). Thin-layer chromatography was carried out using
aluminium sheets precoated with silica gel 60F (Merck 5554). The plates
were inspected by UV light and, if required, developed in I2 vapor. Column
chromatography was carried out using silica gel 60F (Merck 9385, 0.040 ±
0.063 mm). Melting points were determined on an Electrothermal 9100
apparatus and are uncorrected. 1H and 13C NMR spectra were recorded on
either a Bruker AC300 (300 and 75 MHz, respectively), Bruker ARX400
(400 and 100 MHz, respectively), or Bruker ARX500 (500 and 125 MHz,
respectively) spectrometer, using residual solvent as the internal standard.
All chemical shifts are quoted on the d scale, and all coupling constants are
expressed in Hertz (Hz). All 2D NMR experiments were recorded with 1 K
data points and 256 time increments in phase sensitive mode using TPPI,
with the sample nonspinning. Double quantum filtered 1H-1H COSY
experiments[31] were processed in a 1 K x 1 K matrix, and a forward linear
prediction was performed in the F1 dimension. T-ROESY experiments[32]
were performed with a spin lock mixing time of 300 ms, and processed in a
1 K Â 1 K matrix with a forward linear prediction in the F1 dimension. All
samples were degassed immediately before performing the T-ROESY
1,3-Bis(2-{2-[2-(2-p-tolylsulfonyloxy)ethoxy]ethoxy}ethoxy)benzene (12):
1,3-Bis(2-{2-[2-(2-hydroxy)ethoxy]ethoxy}ethoxy)benzene (11)[18] (10.0 g,
26.7 mmol), Et3N (27.0 g, 267 mmol), and a catalytic quantity of DMAP
were dissolved in anhydrous CH2Cl2 (100 mL). This solution was cooled
down to 08C in an ice bath and TsCl (25.5 g, 134 mmol) dissolved in dry
CH2Cl2 (100 mL) was added dropwise over a 3 h period. The reaction
mixture was allowed to warm up to ambient temperature before being
stirred overnight. Hydrochloric acid (5n, 200 mL) was added carefully to
the reaction mixture before the organic phase was recovered and washed
with 2n HCl (200 mL) and saturated brine (200 mL). This solution was
then dried (MgSO4) and the solvents were removed in vacuo to afford a
dark yellow oil, which was purified by chromatography (SiO2: gradient
elution with CH2Cl2 to 1:1 CH2Cl2/EtOAc) to yield the title compound as a
pale yellow oil (11.8 g, 64%); 1H NMR (300 MHz, CDCl3): d ˆ 2.42 (s, 6H;
CH3), 3.56 ± 3.87 (m, 16H; OCH2), 4.04 ± 4.19 (m, 8H; OCH2), 6.45 ± 6.54
(m, 3H; ArH), 7.14 (t, J ˆ 8.3 Hz, 1H; ArH), 7.32 (d of an AA'BB', J ˆ
8.2 Hz, 4H; tosyl ArH), 7.79 (d of an AA'BB', J ˆ 8.2 Hz, 4H; tosyl ArH);
13C NMR (75 MHz, CDCl3): d ˆ 21.7 (CH3), 67.5, 68.8, 69.4, 69.9, 70.9
(OCH2), 101.9, 107.2, 128.1, 130.0, 133.1, 145.0, 160.1 (aromatic C); MS
‡
(FAB): m/z: 683 [M‡H] ; C32H42O12S2 (682.8): calcd C 56.29, H 6.20; found
C 56.12, H 6.24.
Chem. Eur. J. 2000, 6, No. 12
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2283 $ 17.50+.50/0
2283

J. F. Stoddart, D. J. Williams et al.
FULL PAPER
Benzometaphenylene[25]crown-8 (2): A solution of the ditosylate 12
(5.0 g, 7.3 mmol) and catechol 13 (0.81 g, 7.3 mmol) in dry MeCN (500 mL)
was added dropwise over a 3 d period to a suspension of Cs2CO3 (11.9 g,
36.6 mmol) in refluxing MeCN (500 mL). After 4 d of heating under reflux,
the reaction mixture was allowed to cool down. Subsequently, it was
filtered to remove inorganic salts. The filtrate was evaporated to dryness
and the residue partitioned between CH2Cl2 (250 mL) and K2CO3 solution
(10% w/v, 250 mL). The organic phase was then washed with a further
aliquot of aqueous K2CO3, before being dried (MgSO4). The solvents were
removed in vacuo and the crude product was subjected to chromatography
(SiO2: EtOAc/Hexane 9:1), yielding BMP25C8 (2) as a white solid (2.73 g,
83%); m.p. 66 ± 688C; 1H NMR (400 MHz, CDCl3): d ˆ 3.71 ± 3.75 (m, 8H;
OCH2), 3.81 ± 3.87 (m, 8H; OCH2), 4.13-4.16 (m, 8H; OCH2), 6.50 (dd, J ˆ
2.4 and 8.0 Hz, 2H; ArH), 6.70 (t, J ˆ 2.4 Hz, 1H; ArH); 6.89-6.92 (m, 4H;
catechol ArH), 7.12 (t, J ˆ 8.0 Hz, 1H; ArH); 13C NMR (100 MHz, CDCl3):
d ˆ 68.1, 69.1, 70.0, 71.0 and 71.1 (OCH2), 102.8, 107.9, 115.3, 121.8, 129.8,
corresponding [2]rotaxane could be gathered. However, FAB mass
spectrometric analysis of the crude reaction mixture revealed signals for
1) the dumbbell-shaped compound formed in the reaction between
diamine 8 and 2,6-diisopropylphenyl isocyanate, and 2) the bis-urea
derivative formed upon reaction of both amino groups present in 8 with
the isocyanate.
X-ray crystallography: The X-ray crystal structure of 10-H ´ O2CCF3 has
been discussed briefly in a preliminary communication.[7g]
Crystal data for 2: C24H32O8, Mr ˆ 448.5, monoclinic, space group P21/c (no.
14), a ˆ 17.974(2), b ˆ 5.140(1), c ˆ 24.851(3) Š, b ˆ 92.93(1)8, Vˆ
2292.8(4) Š3, Z ˆ 4, 1calcd ˆ 1.299 gcmÀ3
,
m(MoKa) ˆ 0.97 cmÀ1
F(000) ˆ
,
960, Tˆ 293 K; clear prismatic needles, 0.83  0.40  0.27 mm, Siemens
P4/PC diffractometer, graphite-monochromated MoKa radiation, w scans,
4028 independent reflections. The structure was solved by direct methods
À
and the non-hydrogen atoms were refined anisotropically. The C H
‡
hydrogen atoms were placed in calculated positions, assigned isotropic
thermal parameters, U(H) ˆ 1.2Ueq(C), and allowed to ride on their parent
atoms. Refinements were by full-matrix least-squares based on F 2 to give
R1 ˆ 0.062, wR2 ˆ 0.151 for 2493 independent observed reflections [jFo j>
4s(jFo j ), 2q ꢀ 508]and 289 parameters. All computations were carried out
149.1 and 160.1 (aromatic C); MS (FAB): m/z: 449 [M ‡ H] ; C24H32O8
(448.5): calcd C 64.27, H 7.19; found C 64.33, H 7.12.
General procedure for rotaxanation: The dichloride salt 8-H2 ´ 2Cl (ca.
1.0 g) was partitioned between CH2Cl2 (100 mL) and NaOH (100 mL).
After the mixture had been stirred for 10 min, the organic phase was
collected, dried (MgSO4), and evaporated to dryness. A portion of the
residue (340 mg, 1.05 mmol) was dissolved in anhydrous CH2Cl2 and the
crown ether (1, 2 or 3) (1.41 g, 3.15 mmol) was added. Upon dissolution of
the crown ether, trifluoroacetic acid (120 mg, 1.05 mmol) was added, and,
after 5 min, 2,6-diisopropylphenyl isocyanate (1.07 g, 5.25 mmol) was
introduced into the reaction mixture which was stirred under ambient
conditions for a further 5 d. The solvents were removed in vacuo and the
residue subjected to chromatography (SiO2: gradient elution with CH2Cl2/
MeOH, 99:1 to 9:1) affording pure rotaxanes when DB24C8 (1) and
BMP25C8 (2) were employed as the crown ethers. Data for the DB24C8-
containing [2]rotaxane 10-H ´ O2CCF3, isolated as an off-white powder
(0.73 g, 64%); m.p. 117 ± 1198C; 1H NMR (400 MHz, CD2Cl2): d ˆ 1.19 (s,
30H; C(CH3)3 and CH(CH3)2), 3.32 (sept, J ˆ 7.0 Hz, 2H; CH(CH3)2),
3.42 ± 3.47 (m, 4H; H-g1), 3.52 ± 3.58 (m, 4H; H-g2), 3.67 ± 3.73 (m, 4H;
H-b1), 3.78 ± 3.85 (m, 4H; H-b2), 4.07 ± 4.17 (m, 8H; H-a1 and H-a2), 4.45 ±
4.49 (m, 2H; CH2(b)), 4.68 ± 4.71 (m, 2H; CH2(a)), 6.79 ± 6.85 (m, 4H;
H-7), 6.88 ± 6.95 (m, 4H; H-8), 7.10 (d of an AA'BB', J ˆ 8.5 Hz, 2H; H-3),
7.14 (d, J ˆ 7.5 Hz, 2H; H-5), 7.21 ± 7.26 (m, 1H; H-6), 7.28 (d, J ˆ 1.5 Hz,
2H; H-2), 7.35 (t, J ˆ 1.5 Hz, 1H; H-1), 7.44 (d of an AA'BB', J ˆ 8.5 Hz,
using the SHELXTL PC program system.[34]
.
Crystal data for [2 ´ 4-H][PF6] : [C H48NO8][PF6], Mr ˆ 791.7, triclinic, space
38
Å
group P1 (no. 2), a ˆ 11.265(1), b ˆ 12.853(3), c ˆ 14.495(2) Š, a ˆ
104.12(2), b ˆ 102.72(1), g ˆ 96.53(2)8, Vˆ 1954.1(6) Š3, Z ˆ 2, 1calcd
ˆ
1.346 gcmÀ3, m(MoKa) ˆ 1.50 cmÀ1, F(000) ˆ 832, Tˆ 293 K; clear prisms,
0.57 Â 0.33 Â 0.13 mm, Siemens P4/PC diffractometer, graphite-monochro-
mated MoKa radiation, w scans, 5696 independent reflections. The structure
was solved by direct methods. Disorder was found in one of the polyether
arms and this was resolved into two partial occupancy orientations with the
non-hydrogen atoms of the major occupancy orientation being refined
anisotropically (those of the minor occupancy orientation were refined
isotropically). The remaining non-hydrogen atoms were refined anisotropi-
À
cally. The N H hydrogen atoms were located from a DF map and allowed
À
À
to refine isotropically subject to an N H distance constraint. The C H
hydrogen atoms were placed in calculated positions, assigned isotropic
thermal parameters, U(H) ˆ 1.2Ueq(C), and allowed to ride on their parent
atoms. Refinements were by full-matrix least-squares based on F 2 to give
R1 ˆ 0.083, wR2 ˆ 0.201 for 2613 independent observed reflections [jFo j>
4s(jFo j ), 2q ꢀ 478]and 480 parameters. All computations were carried out
using the SHELXTL PC program system.[34]
‡
2H; H-4), 7.53 (br s, 2H; NH2 ), 8.97 (s, 1H; NH(b)CO), 10.16 (s, 1H;
Crystal data for 14-H ´ O2CCF3: [C59H82N3O9][CF3CO2]´ 0.25CH 2Cl2 ´
NH(a)CO); 13C NMR (100 MHz, CDCl3): d ˆ 23.5 and 24.4 (CH(CH3)2),
28.6 (CH(CH3)2), 31.5 (C(CH3)3), 34.9 (C(CH3)3), 52.9 (CH2(a)), 53.1
(CH2(b)), 68.1 (C-a), 70.2 (C-b), 70.5 (C-g), 112.8 (C-7), 118.4 (C-4), 122.0
(C-8), 122.4 (quaternary aromatic), 122.9 (C-5), 123.3 (C-1), 123.7 (C-2),
127.1 (C-6), 129.7 (C-3), 131.6, 133.2, 143.2, 147.4, 147.6 and 151.4
Å
0.25MeCN, Mr ˆ 1121.8, triclinic, space group P1 (no. 2), a ˆ 14.966(2),
b ˆ 22.879(3), c ˆ 22.915(2) Š, a ˆ 64.64(1), b ˆ 72.82(1), g ˆ 76.20(1)8,
Vˆ 6717(2) Š3, Z ˆ 4 (there are two crystallographically independent
molecules in the asymmetric unit), 1calcd ˆ 1.109 gcmÀ3
,
m(CuKa) ˆ
8.41 cmÀ1, F(000) ˆ 2400, Tˆ 193 K; clear rhombs, 1.00  0.37  0.13 mm,
Siemens P4/RA diffractometer, graphite-monochromated CuKa radiation,
w scans, 16840 independent reflections. The structure was solved by direct
methods. In each of the independent molecules disorder was found in one
of the polyether arms and in both of the tert-butyl groups; disorder was also
present in one of the CF3CO2À ions. In each case this was resolved into two
partial occupancy orientations with the non-hydrogen atoms of the major
occupancy orientations being refined anisotropically (those of the minor
occupancy orientations were refined isotropically). The remaining non-
hydrogen atoms of the [2]rotaxanes and the anions were refined anisotropi-
cally, whilst those of the solvent molecules were refined isotropically. The
ˆ
(quaternary aromatic), 155.7 (C O); MS (FAB): m/z: 976 [M À
‡
O2CCF3] ; C61H82F3N3O11 (1090.3): calcd C 67.20, H 7.58, N 3.85; found C
66.68, H 7.55, N 3.58. Data for the BMP25C8-containing [2]rotaxane 14-H ´
O2CCF3, isolated as an off-white powder (0.41 g, 36%); m.p. 102-1058C;
1H NMR (400 MHz, CD2Cl2): d ˆ 1.19 (br s, 12H; CH(CH3)2), 1.24 (s, 18H;
C(CH3)3) 3.31 (sept, J ˆ 7.0 Hz, 2H; CH(CH3)2), 3.35 ± 3.70 (m, 12H; H-b,
H-g and H-d), 3.73 ± 3.79 (m, 2H; H-e1), 3.80 ± 3.86 (m, 2H; H-e2), 3.91 ±
3.98 (m, 2H; H-a1), 4.06 ± 4.13 (m, 2H; H-a2), 4.15 ± 4.26 (m, 4H; H-f),
4.27 ± 4.32 (m, 2H; CH2(b)), 4.39 ± 4.45 (m, 2H; CH2(a)), 6.58 (dd, J ˆ 2.5
and 8.5 Hz, 2H; H10), 6.64 (t, J ˆ 2.5 Hz, 1H; H-9), 6.72 ± 6.76 (m, 2H; H-7),
6.90 ± 6.95 (m, 2H; H-8), 7.07 (d of an AA'BB', J ˆ 8.5 Hz, 2H; H-3), 7.14
(d, J ˆ 7.5 Hz, 2H; H-5), 7.21 ± 7.27 (m, 2H; H-6 and H-11), 7.30 (d, J ˆ
1.5 Hz, 2H; H-2), 7.43 (t, J ˆ 1.5 Hz, 1H; H-1), 7.48 (d of an AA'BB', J ˆ
À
À
N H and C H hydrogen atoms were placed in calculated positions,
assigned isotropic thermal parameters, U(H) ˆ 1.2Ueq(C/N) [U(H) ˆ
À
1.5Ueq(C Me)], and allowed to ride on their parent atoms. Refinements
‡
8.5 Hz, 2H; H-4), 7.73 (br s, 2H; NH2 ), 8.94 (s, 1H; NH(b)CO), 10.23 (s,
were by full-matrix least-squares based on F 2 to give R1 ˆ 0.162, wR2 ˆ
0.420 for 7798 independent observed reflections [jFo j> 4s(jFo j ), 2q ꢀ
1158]and 1548 parameters. All computations were carried out using the
SHELXTL PC program system.[34]
1H; NH(a)CO); 13C NMR (100 MHz, CDCl3): d ˆ 23.5 and 24.4
(CH(CH3)2), 28.6 (CH(CH3)2), 31.5 (C(CH3)3), 35.0 (C(CH3)3), 52.7
(CH2(a)), 53.2 (CH2(b)), 67.9 (C-f), 68.1 (C-a), 69.7 (C-b), 70.4 (C-e),
70.7 (C-d), 71.1 (C-g), 103.9 (C-9), 107.6 (C-10), 112.5 (C-7), 118.5 (C-4),
121.3 (quaternary aromatic), 122.3 (C-8), 122.9 (C-5), 123.8 (C-1 and C-2),
127.2 (C-6), 129.8 (quaternary aromatic), 130.6 (C-3), 131.0 (C-11), 133.1,
Crystallographic data (excluding structure factors) for the structures
reported in this paper have been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication no. CCDC-138348 (10-
H ´ O2CCF3), CCDC-138346 (2), CCDC-138347 ([2 ´ 4-H][PF6]), and
CCDC-138349 (14-H ´ O2CCF3). Copies of the data can be obtained free
of charge on application to CCDC, 12 Union Road, Cambridge CB21EZ,
UK (fax: (‡44)1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
ˆ
143.5, 146.5, 147.5 and 151.8 (quaternary aromatic), 155.6 (C O), 159.7
(quaternary aromatic); MS (FAB): m/z: 976 [M À O2CCF3] ;
‡
C61H82F3N3O11 (1090.3): calcd C 67.20, H 7.58, N 3.85; found C 67.66, H
7.70, N 3.78. When the crown ether component employed in an attempted
rotaxanation was BMP26C8 (3), no evidence for the formation of the
2284
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2284 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 12

Influence of Crown Ether Constitution on Binding
2274±2287
C. Schiavo, N. Spencer, J. F. Stoddart, P. A. Tasker, A. J. P. White,
D. J. Williams, Chem. Eur. J. 1996, 2, 709 ± 728; c) P. T. Glink, C.
Schiavo, J. F. Stoddart, D. J. Williams, Chem. Commun. 1996, 1483 ±
1490; d) P. R. Ashton, M. C. T. Fyfe, P. T. Glink, S. Menzer, J. F.
Stoddart, A. J. P. White, D. J. Williams, J. Am. Chem. Soc. 1997, 119,
12514 ± 12524; e) P. R. Ashton, M. C. T. Fyfe, M.-V. Martínez-Díaz, S.
Menzer, C. Schiavo, J. F. Stoddart, A. J. P. White, D. J. Williams, Chem.
Eur. J. 1998, 4, 1523 ± 1534; f) M. Montalti, L. Prodi, Chem. Commun.
1998, 1461 ± 1462; g) N. Yamaguchi, L. M. Hamilton, H. W. Gibson,
Angew. Chem. 1998, 110, 3463 ± 3466; Angew. Chem. Int. Ed. 1998, 37,
3275 ± 3279; h) M. C. T. Fyfe, J. F. Stoddart, Adv. Supramol. Chem.
1999, 5, 1 ± 53; i) E. Ishow, A. Credi, V. Balzani, F. Spadola, L.
Mandolini, Chem. Eur. J. 1999, 5, 984 ± 989; j) D. Fitzmaurice, S. N.
Rao, J. A. Preece, J. F. Stoddart, S. Wenger, N. Zaccheroni, Angew.
Chem. 1999, 111, 1220 ± 1224; Angew. Chem. Int. Ed. 1999, 38, 1147 ±
1150.
Acknowledgement
This research was sponsored by the University of California, Los Angeles
in the USA, and by the Engineering and Physical Sciences Research
Council in the UK.
[1]For in-depth and historical analyses of cation/crown ether complex-
ation, see:a) J. S. Bradshaw, R. M. Izatt, A. V. Bordunov, C. Y. Zhu,
J. K. Hathaway in Comprehensive Supramolecular Chemistry, Vol. 1
(Eds.: J. L. Atwood, J. E. D. Davies, D. D. MacNicol, F. Vögtle, G. W.
Gokel), Pergamon, Oxford, 1996, pp. 35 ± 95; b) R. M. Izatt, K.
Pawluk, J. S. Bradshaw, R. L. Bruening, Chem. Rev. 1995, 95, 2529 ±
2586; c) R. M. Izatt, K. Pawluk, J. S. Bradshaw, R. L. Bruening, Chem.
Rev. 1991, 91, 1721 ± 2085; d) G. W. Gokel, Crown Ethers and
Cryptands, The Royal Society of Chemistry, Cambridge, 1991,
pp. 64 ± 98; e) Cation Binding by Macrocycles, (Eds.: Y. Inoue, G. W.
Gokel), Marcel Dekker, New York, 1990; f) F. Vögtle, E. Weber in
Crown Ethers and Analogs (Eds.: S. Patai, Z. Rappoport), Wiley, 1989,
pp. 207 ± 304; g) I. Goldberg in Inclusion Compounds, Vol. 2 (Eds.:
J. L. Atwood, J. E. D. Davies, D. D. MacNicol), Academic Press,
London, 1984, pp. 261 ± 335.
[2]For some recent examples, see: a) A. Drljaca, M. J. Hardie, C. L.
Raston, L. Spiccia, Chem. Eur. J. 1999, 5, 2295 ± 2299; b) A. P.
Marchand, H.-S. Chong, S. Alihodzic, W. H. Watson, S. G. Bodige,
Tetrahedron 1999, 55, 9687 ± 9696; c) H. Liu, S. Liu, L. Echegoyen,
Chem. Commun. 1999, 1493 ± 1494; d) S. Flink, F. C. J. M. van Veggel,
D. N. Reinhoudt, J. Phys. Chem. B 1999, 103, 6515 ± 6520; e) R. A.
Bartsch, H.-S. Hwang, V. S. Talanov, G. G. Talanova, D. W. Purkiss,
R. D. Rogers, J. Org. Chem. 1999, 64, 5341 ± 5349.
[3]In Pedersenꢀs seminal paper (C. J. Pedersen, J. Am. Chem. Soc. 1967,
89, 7017 ± 7036), he noted that primary alkylammonium ions form 1:1
complexes with dibenzo[18]crown-6 (DB18C6). This ground-breaking
discovery subsequently spawned much research in the areas of host ±
guest and supramolecular chemistry. For examples spanning from
Pedersenꢀs day until the present, see: a) E. B. Kyba, K. Koga, L. R.
Sousa, M. G. Siegel, D. J. Cram, J. Am. Chem. Soc. 1973, 95, 2692 ±
2693; b) D. J. Cram, J. M. Cram Science 1974, 183, 803 ± 809; c) T. L.
Tarnowski, D. J. Cram, J. Chem. Soc., Chem. Commun. 1976, 661 ±
663; d) J. M. Timko, S. S. Moore, D. M. Walba, P. C. Hiberty, D. J.
Cram, J. Am. Chem. Soc. 1977, 99, 4207 ± 4219; e) E. P. Kyba, J. M.
Timko, L. J. Kaplan, F. de Jong, G. W. Gokel, D. J. Cram, J. Am. Chem.
Soc. 1978, 100, 4558 ± 4568; f) D. J. Cram, J. M. Cram Acc. Chem. Res.
1978, 11, 8 ± 14; g) J. F. Stoddart, Chem. Soc. Rev. 1979, 8, 85 ± 142;
h) M. J. Bovill, D. J. Chadwick, M. R. Johnson, N. F. Jones, I. O.
Sutherland, R. F. Newton, J. Chem. Soc., Chem. Commun. 1979,
1065 ± 1066; i) J.-M. Lehn in IUPAC Frontiers of Chemistry, (Ed. K. J.
Laidler), Pergamon, Oxford, 1982, pp. 265 ± 272; j) I. Goldberg, J. Am.
Chem. Soc. 1980, 102, 4106 ± 4113; k) F. De Jong, D. N. Reinhoudt,
Adv. Phys. Org. Chem. 1980, 17, 279 ± 433. l) D. J. Cram, K. N.
Trueblood, Top. Curr. Chem. 1981, 98, 43 ± 106; m) J. S. Bradshaw,
S. L. Baxter, J. D. Lamb, R. M. Izatt, J. J. Christensen, J. Am. Chem.
Soc. 1981, 103, 1821 ± 1827; n) K. N. Trueblood, C. B. Knobler, D. S.
Lawrence, R. V. Stevens, J. Am. Chem. Soc. 1982, 104, 1355 ± 1362;
o) R. Aldag, G. Schröder, Liebigs Ann. Chem. 1984, 1036 ± 1045;
p) I. O. Sutherland, Chem. Soc. Rev. 1986, 15, 63 ± 91; q) J. F. Stoddart,
Top. Stereochem. 1987, 17, 205 ± 288; r) D. J. Cram, Angew. Chem.
1988, 100, 1041 ± 1052; Angew. Chem. Int. Ed. Engl. 1988, 27, 1009 ±
1020; s) S. Misumi, Pure Appl. Chem. 1990, 62, 493 ± 498; t) I. O.
Sutherland, Pure Appl. Chem. 499 ± 504; u) R. M. Izatt, T. Wang, J. K.
Hathaway, X. X. Zhang, J. C. Curtis, J. S. Bradshaw, C. Y. Zhu, P.
Huszthy, J. Incl. Phenom. 1994, 17, 157 ± 175; v) M. T. Reetz, J. Huff, J.
Rudolph, K. Töllner, A. Deege, R. Goddard, J. Am. Chem. Soc. 1994,
116, 11588 ± 11589; w) B. L. Williamson, C. S. Creaser, Int. J. Mass
Spectrom. 1998, 188, 53 ± 61; x) A. P. Hansson, P.-O. Norrby, K.
Wärnmark, Tetrahedron Lett. 1998, 39, 4565 ± 4568; y) H. Tsukube,
M. Wada, S. Shinoda, H. Tamiaki, Chem. Commun. 1999, 1007 ±
1008.
[5]P. R. Ashton, P. J. Campbell, E. J. T. Chrystal, P. T. Glink, S. Menzer,
D. Philp, N. Spencer, J. F. Stoddart, P. A. Tasker, D. J. Williams,
Angew. Chem. 1995, 107, 1997 ± 2001; Angew. Chem. Int. Ed. Engl.
1995, 34, 1865 ± 1869.
[6]Complexes formed between secondary dialkylammonium ions and
crown ethers having less than 24 atoms in their macrorings have been
observed to occur in a face-to-face manner. See: a) J. C. Metcalfe, J. F.
Stoddart, G. Jones, J. Am. Chem. Soc. 1977, 99, 8317 ± 8319; b) J.
Krane, O. Aune, Acta Chem. Scand. Sect. B 1980, 34, 397 ± 401; c) J. C.
Metcalfe, J. F. Stoddart, G. Jones, A. Atkinson, I. S. Kerr, D. J.
Williams, J. Chem. Soc. Chem. Commun. 1980, 540 ± 543; d) S. S.
Abed-Ali, B. J. Brisdon, R. England, J. Chem. Soc. Chem. Commun.
1987, 1565 ± 1566.
[7]a) A. G. Kolchinski, D. H. Busch, N. W. Alcock, J. Chem. Soc. Chem.
Commun. 1995, 1289 ± 1291; b) P. R. Ashton, P. T. Glink, J. F. Stod-
dart, P. A. Tasker, A. J. P. White, D. J. Williams, Chem. Eur. J. 1996, 2,
729 ± 736; c) M.-V. Martínez-Díaz, N. Spencer, J. F. Stoddart, Angew.
Chem. 1997, 109, 1991 ± 1994; Angew. Chem. Int. Ed. Engl. 1997, 36,
1904 ± 1907; d) P. R. Ashton, I. Baxter, M. C. T. Fyfe, F. M. Raymo, N.
Spencer, J. F. Stoddart, A. J. P. White, D. J. Williams, J. Am. Chem.
Soc. 1998, 120, 2297 ± 2307; e) P. R. Ashton, R. Ballardini, V. Balzani,
Â
Â
I. Baxter, A. Credi, M. C. T. Fyfe, M. T. Gandolfi, M. Gomez-Lopez,
M.-V. Martínez-Díaz, A. Piersanti, N. Spencer, J. F. Stoddart, M.
Venturi, A. J. P. White, D. J. Williams, J. Am. Chem. Soc. 1998, 120,
11932 ± 11942; f) A. G. Kolchinski, N. W. Alcock, R. A. Roesner,
D. H. Busch, Chem. Commun. 1998, 1437 ± 1438; g) S. J. Cantrill, D. A.
Fulton, M. C. T. Fyfe, J. F. Stoddart, A. J. P. White, D. J. Williams,
Tetrahedron Lett. 1999, 40, 3669 ± 3672; h) S. J. Rowan, S. J. Cantrill,
J. F. Stoddart, Org. Lett. 1999, 1, 129 ± 132 i) S. J. Cantrill, M. C. T.
Fyfe, A. M. Heiss, J. F. Stoddart, A. J. P. White, D. J. Williams, Chem.
Commun. 1999, 1251 ± 1252; j) T. Takata, H. Kawasaki, S. Asai, Y.
Furusho, N. Kihara, Chem. Lett. 1999, 223 ± 224; k) S. J. Cantrill, S. J.
Rowan, J. F. Stoddart, Org. Lett. 1999, 1, 1363 ± 1366.
[8]a) P. R. Ashton, P. T. Glink, M.-V. Martínez-Díaz, J. F. Stoddart,
A. J. P. White, D. J. Williams, Angew. Chem. 1996, 108, 2058 ± 2061;
Angew. Chem. Int. Ed. Engl. 1996, 35, 1930 ± 1933; b) P. R. Ashton,
A. N. Collins, M. C. T. Fyfe, S. Menzer, J. F. Stoddart, D. J. Williams,
Angew. Chem. 1997, 109, 760 ± 763; Angew. Chem. Int. Ed. Engl. 1997,
36, 735 ± 739; c) P. R. Ashton, M. C. T. Fyfe, S. K. Hickingbottom, S.
Menzer, J. F. Stoddart, A. J. P. White, D. J. Williams, Chem. Eur. J.
1998, 4, 577 ± 589; d) M. C. T. Fyfe, J. F. Stoddart, Coord. Chem. Rev.
1999, 183, 139 ± 155; e) N. Yamaguchi, H. W. Gibson, Angew. Chem.
1999, 111, 195 ± 199; Angew. Chem. Int. Ed. 1999, 38, 143 ± 147; f) N.
Yamaguchi, H. W. Gibson, Chem. Commun. 1999, 789 ± 790.
[9]a) D. Philp, J. F. Stoddart, Synlett 1991, 445 ± 458; b) G. M. Whitesides,
J. P. Mathias, C. T. Seto, Science, 1991, 154, 1312 ± 1319; c) D. S.
Lawrence, T. Jiang, M. Levett, Chem. Rev. 1995, 95, 2229 ± 2260; d) D.
Philp, J. F. Stoddart, Angew. Chem. 1996, 108, 1242 ± 1286; Angew.
Chem. Int. Ed. Engl. 1996, 35, 1154 ± 1196; e) P. J. Stang, B. Olenyuk,
Acc. Chem. Res. 1997, 30, 502 ± 518; f) L. Cusack, S. N. Rao, J. Wenger,
D. Fitzmaurice, Chem. Mater. 1997, 9, 624 ± 631; g) M. M. Conn, J.
Rebek, Jr., Chem. Rev. 1997, 97, 1647 ± 1668; h) B. Linton, A. D.
Hamilton, Chem. Rev. 1997, 97, 1669 ± 1680; i) M. Fujita, Chem. Soc.
Rev. 1998, 27, 417 ± 425; j) T. L. Breen, J. Tien, S. R. J. Oliver, T.
Hadzic, G. M. Whitesides, Science 1999, 284, 948 ± 951; k) D. A.
Tomalia, Z. G. Wang, M. Tirrel, Curr. Opin. Colloid Interface Sci.
[4]a) P. R. Ashton, P. J. Campbell, E. J. T. Chrystal, P. T. Glink, S.
Menzer, C. Schiavo, J. F. Stoddart, P. A. Tasker, D. J. Williams, Angew.
Chem. 1995, 107, 2001 ± 2004; Angew Chem. Int. Ed. Engl. 1995, 34,
1869 ± 1871; b) P. R. Ashton, E. J. T. Chrystal, P. T. Glink, S. Menzer,
Chem. Eur. J. 2000, 6, No. 12
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2285 $ 17.50+.50/0
2285

J. F. Stoddart, D. J. Williams et al.
FULL PAPER
Â
1999, 4, 3 ± 5; l) T. Emrick, J. M. J. Frechet, Curr. Opin. Colloid
Interface Sci. 1999, 4, 15 ± 23; m) R. P. Sijbesma, E. W. Meijer, Curr.
Opin. Colloid Interface Sci. 1999, 4, 24 ± 32.
ether, which is encircling the carbanilide derivative with a pseudo-
rotaxane geometry. This phenomenon contrasts with the results of a
similar experiment performed employing carbanilide (N,N-diphenyl-
urea) as the thread compound, in which the global minimum structure
involves the formation of hydrogen bonds from both ureido protons to
one of the ethereal oxygen atoms. The carbonyl oxygen atom of the
urea group constitutes an unfavorable electrostatic interaction, which
repels the crown ether. Thus, in this case, the complete superstructure
involves the crown ether interacting attractively with one side of the
urea moiety, and repulsively with the other side. These results support
the empirical hypothesisÐdeduced from some observations made on
CPK space-filling molecular models and a consideration of the
relative pKa valuesÐthat the crown ether prefers to reside around the
dialkylammonium center, rather than around the urea moiety.
[17]Unfortunately, the resonances arising from a1 and a2 are isochronous.
Cross-peaks between this multiplet and the signals corresponding to
both H-2 and H-3 are observed. As a consequence of this overlap,
it is impossible to deduce that the H-2 resonance only gives a cross-
peak with the a1 resonance, and the H-3 resonance only gives a cross-
peak with the a2 resonance. However, based upon the patterns
observed for the b and g protons, it is not unreasonable to suppose, by
analogy, that selective through-space interactions also occur for the a
protons.
[10]a) P. R. Ashton, A. N. Collins, M. C. T. Fyfe, P. T. Glink, S. Menzer,
J. F. Stoddart, D. J. Williams, Angew. Chem. 1997, 109, 59 ± 62; Angew.
Chem. Int. Ed. Engl. 1997, 36, 59 ± 62; b) M. C. Feiters, M. C. T. Fyfe,
M.-V. Martínez-Díaz, S. Menzer, R. J. M. Nolte, J. F. Stoddart, P. J. M.
van Kan, D. J. Williams, J. Am. Chem. Soc. 1997, 119, 8119 ± 8120; c) F.
Â
Â
Diederich, L. Echegoyen, M. Gomez-Lopez, R. Kessinger, J. F.
Stoddart, J. Chem. Soc. Perkin Trans. 2 1999, 1577 ± 1586.
[11]Our investigations (P. R. Ashton, I. Baxter, S. J. Cantrill, M. C. T. Fyfe,
P. T. Glink, J. F. Stoddart, A. J. P. White, D. J. Williams, Angew. Chem.
1998, 110, 1344 ± 1347, Angew. Chem. Int. Ed. 1998, 37, 1294 ± 1297) on
a self-complementary ꢂdaisy-chainꢀ systemÐin which a secondary
dialkylammonium ion-containing side chain is appended to one of the
catechol residues of a DB24C8 frameworkÐrevealed that, upon
oligo/polymerization, numerous stereoisomeric super-structures can
be formed. It was immediately apparent that this complication could
be circumvented by designing a compound in which the ammonium
ion substituted catechol ring of the DB24C8 macrocycle is replaced by
a resorcinol ring which carries the ammonium ion side-chain attached
to its C-5 position. Such a ꢂmutationꢀ of the monomer, although
simplifying symmetry situation, has the implication of expanding the
macroring from 24 to 25 atoms, thus raising the question - are crown
[18]I. J. A. Mertens, R. Wegh, L. W. Jenneskens, E. J. Vlietstra, A. van der
Kerk-van Hoof, J. W. Zwikker, T. J. Cleij, W. J. J. Smeets, N. Veldman,
A. L. Spek, J. Chem. Soc. Perkin Trans. 2 1998, 725 ± 735.
ethers with
dialkylammonium ions?
a 25C8 constitution effective hosts for secondary
[12]For examples of interlocked macromolecules, see:a) F. M. Raymo,
J. F. Stoddart, Chem. Rev. 1999, 99, 1643 ± 1663; b) A. Harada, M.
Okada, Y. Kawaguchi, M. Kamachi, Polym. Advan. Technol. 1999, 10,
3 ± 12; c) J.-L. Weidmann, J.-M. Kern, J.-P. Sauvage, D. Muscat, S.
Mullins, W. Kohler, C. Rosenauer, H. J. Räder, K. Martin, Y. Geerts,
Chem. Eur. J. 1999, 5, 1841 ± 1851; d) D. Muscat, W. Köhler, H. J.
Räder, K. Martin, S. Mullins, B. Müller, K. Müllen, Y. Geerts,
Macromolecules 1999, 32, 1737 ± 1745; e) K. M. Park, J. Heo, S. G.
Roh, Y. M. Jeon, D. Whang, K. Kim, Mol. Cryst. Liq. Crys. A 1999,
327, 65 ± 70; f) T. Takata, H. Kawasaki, A. Asai, N. Kihara, Y. Furusho,
Chem. Lett. 1999, 111 ± 112; g) C. Hamers, O. Kocian, F. M. Raymo,
J. F. Stoddart, Adv. Mater. 1998, 10, 1366 ± 1369; h) C. G. Gong, H. W.
Gibson, Angew. Chem. 1998, 110, 323 ± 327; Angew. Chem. Int. Ed.
1998, 37, 310 ± 314; i) W. Herrmann, M. Schneider, G. Wenz, Angew.
Chem. 1997, 109, 2618 ± 2621; Angew. Chem. Int. Ed. Engl. 1997, 36,
2511 ± 2514; j) H. W. Gibson, M. C. Bheda, P. T. Engen, Prog. Polym.
Sci. 1994, 19, 843 ± 945.
[19]Previously (see ref. [7d)], mass spectrometric analysis of an equimolar
mixture of DB24C8 and bis(4-tert-butylbenzyl)ammonium hexafluor-
ophosphate revealed the base peak in the spectrum to be that
corresponding to the uncomplexed ammonium compound (m/z 310).
However, a signal arising from the 1:1 complex formed between these
two components was observed at m/z 758, with a relative intensity of
6%. The significance of this result lies in the fact that the bis(4-tert-
butylbenzyl)ammonium cation is too large to thread through the
cavity of DB24C8. This restriction means that any association of these
two components must occur with a face-to-face geometry. In this case,
‡
the peak arising from the 1:1 complex [2 ´ 4-H] (m/z 646) has an
‡
intensity corresponding to 50% of that observed for 4-H (the base
peak), implying that there is an interaction more enduring than a
simple face-to-face association, that is, the threading of the cation
through the macroring of DB24C8, resulting in the formation of a
[2]pseudorotaxane in the gas phase.
[20]For leading references related to this method, see: J. C. Adrian, C. S.
Wilcox, J. Am. Chem. Soc. 1991, 113, 678 ± 680.
[13]Gibson and co-workers (W. S. Bryant, I. A. Guzei, A. L. Rheingold,
J. S. Merola, H. W. Gibson, J. Org. Chem. 1998, 63, 7634 ± 7639)
comment on the symmetrical nature of bis-meta-phenylene-contain-
ing crown ethers when they are incorporated into polymeric inter-
locked systems.
[21]For leading references, see: a) K. A. Connors, Binding Constants,
Wiley, New York, 1987 ; b) H. Tsukube, H. Furuta, A. Odani, Y.
Takeda, Y. Kudo, Y. Inoue, Y. Liu, H. Sakamoto, K. Kimura in
Comprehensive Supramolecular Chemistry, Vol. 8 (Eds.: J. L. Atwood,
J. E. D. Davies, D. D. MacNicol, F. Vögtle, J. A. Ripmeester), Perga-
mon, Oxford, 1996, pp. 425 ± 482 and references therein.
[22]P. R. Ashton, R. A. Bartsch, S. J. Cantrill, R. E. Hanes, Jr., S. K.
Hickingbottom, J. N. Lowe, J. A. Preece, J. F. Stoddart, V. S. Talanov,
Z.-H. Wang, Tetrahedron Lett. 1999, 40, 3661 ± 3664.
[14]M. S. Newman, L. F. Lee, J. Org. Chem. 1972, 37, 4468 ± 4469
[15]H. G. McFadden, J. N. Phillips, Z. Naturforsch. C: Biosci. 1990, 45,
196 ± 206.
‡
[16]The protons of the R 2NH2 center are far more acidic than those of the
urea moiety (pKa values of about 10 and >20, respectively), thus
bestowing upon them a much greater capacity for hydrogen bonding.
Additionally, inspection of CPK space-filling molecular models
reveals that insertion of the urea portion of the dumbbell into the
cavity of the DB24C8 ring is not favored sterically as a consequence of
the proximal 2,6-diisopropylphenyl group. Also, this geometry would
result in an unfavorable electrostatic interaction between the carbonyl
oxygen atom on the urea group and the ethereal oxygen atoms lining
the perimeter of the macrocyclic cavity. Indeed, a brief Monte Carlo
conformational search within MacroModel (1000 steps, all variable
torsions considered, Amber* force field, GB/SA CHCl3, extended
nonbonded cutoffs of 4, 8, and 20 Š for van der Waals, hydrogen
bonding and electrostatic interactions, respectively) of the complexes
formed between 2,6-diisopropylcarbanilide and DB24C8 furnished a
global minimum conformation in which the ureido proton closest to
the diisopropylphenyl unit is sterically prevented from participating in
hydrogen bonding interactions by virtue of its proximity to the two
isopropyl moieties. The structure corresponding to the global mini-
mum involves a lone hydrogen bond emanating from the sterically
unhindered ureido proton to an ethereal oxygen atom of the crown
[23]However, a report has appeared in the literature (ref. [8e)]describing
the inability of a derivative of 2 to form a [2]pseudorotaxane in
CD3COCD3/CDCl3 (1:1) solution with 4-H ´ PF6.
[24]The poor signal to noise ratio obtained when recording these 13C NMR
spectra preclude the calculation of accurate thermodynamic and
kinetic parameters for this dynamic process. However, assuming that
the coalescence of the two signalsÐcorresponding to the iPr groupsÐ
does occur in the range 312 ± 317 K, estimates for the kinetic and
thermodynamic parameters for this process can be obtained as
described in I. O. Sutherland, Annu. Rep. NMR Spectrosc. 1971, 4,
71 ± 235. The rate constant for this bond rotation (kc) at the
coalescence temperature (Tc) was calculated by employing the
approximate expression kc ˆ p(Dn)/(2)1/2, where Dn is the limiting
frequency difference (96 Hz) between the coalescing peaks. The
³
relationship DGc ˆ ÀRTcln(kch/kBTc)Ðwhere R, h and kB corre-
spond, respectively, to the gas, Planck and Boltzmann constantsÐwas
³
used to determine the free activation barriers (DGc ) of this bond
rotation process at 312 and 317 K, respectively, thus giving the range
15.0 ± 15.2 kcalmolÀ1
.
2286
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2286 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 12

Influence of Crown Ether Constitution on Binding
2274±2287
[25]F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M. Lipton,
C. Caulfield, G. Chang, T. Hendrickson, W. C. Still, J. Comput. Chem.
1990, 11, 440 ± 467.
[26]N. L. Allinger, Y. H. Yuh, J. H. Lii, J. Am. Chem. Soc. 1989, 111, 8551 ±
8566.
2458 ± 2464, ii) G. W. Buchanan, M. Lefort, A. Moghimi, C. Bensimon,
J. Mol. Struct. 1997, 415, 267 ± 275, and iii) H. W. Gibson et. al. see ref.
[13]. Interestingly, in neither of the latter papers is any mention made
of the earlier X-ray crystallographic analyses. In only one paper,
namely iii), is an earlier report of the synthesis of BMP26C8
acknowledged!
Á
[27]E. Polak, G. Ribie re, Rev. Fr. Informat. Rech. Oper. 1969, 16, 35 ± 43.
[28]W. C. Still, A. Tempczyk, R. C. Hawley, T. Hendrickson, J. Am. Chem.
Soc. 1990, 112, 6127 ± 6129.
[30]The FAB mass spectrum of a 1:1 mixture of 3 and 4-H ´ PF6 contains
only signals for the ꢂfreeꢀ components, that is, the dibenzylammonium
‡
[29]One of the earliest syntheses of bismetaphenyl-ene[26c]rown-8
(sometimes referred to as dibenzo[26]-crown-8 (DB26C8)) was
reported (G. Lindsten, O. Wennerström, R. Isaksson, J. Org. Chem.
1987, 52, 547 ± 554) well over 10 years ago now. Subsequently, this
compound was employed in the construction of an ion-selective
electrode for the potentiometric sensing of guanidinium ions. See:
F. N. Assubaie, G. J. Moody, J. D. R. Thomas, Analyst 1989, 114,
1545 ± 1550. However, during the 1990s, three research groups have
independently solved the X-ray crystal structure of this crown ether:
the structural parameters reported by each group are, within
experimental error, the same. See: i)E. Luboch, M. S. Fonar, Y. A.
Siminov, V. K. Bel'sky, J. Biernat, Izv. Akad. Nauk, Ser. Khim. 1995,
cation 4-H and BMP26C8. No peaks were observed for any
complexes formed between these two components, suggesting that a
[2]pseudorotaxane does not exist under the conditions employed
during this particular analysis.
[31]A. Derome, M. Williamson, J. Mag. Reson. 1990, 88, 177 ± 185.
[32]a) A. Bax, D. G. Davis, J. Mag. Reson. 1985, 63, 207 ± 213; b) T.
Hwang, A. J. Shaka, J. Am. Chem. Soc. 1992, 114, 3157 ± 3159.
[33]A. Bax, S. Subramanian, J. Mag. Reson. 1986, 67, 565 ± 569.
[34]SHELXTL PC version 5.03, Siemens Analytical X-Ray Instruments,
Inc., Madison, WI, 1994.
Received: January 10, 2000 [F2234]
Chem. Eur. J. 2000, 6, No. 12
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0612-2287 $ 17.50+.50/0
2287