Potassium selective calix[4]semitubes

Article abstract of DOI:10.1002/chem.200204518

A new class of ionophore consisting of two calix[4]arene units linked through the lower rim by two ethylene chains, in combination with propyl ether and phenolic functional groups, has been developed. These calix[4]semitube molecules exhibit remarkable se

Full text of DOI:10.1002/chem.200204518

FULL PAPER
Potassium Selective Calix[4]semitubes
Philip R. A. Webber,[a] Andrew Cowley,[a] Michael G. B. Drew,[b] and Paul D. Beer*[a]
Abstract: A new class of ionophore consisting of two calix[4]arene units linked
through the lower rim by two ethylene chains, in combination with propyl ether and
phenolic functional groups, has been developed. These calix[4]semitube molecules
exhibit remarkable selectivity and fast complexation kinetics for potassium over all
Group 1metal cations. Molecular modelling studies, using structural models derived
from crystallographic data, suggest the potassium cation is complexed by a
horizontal, side-on route and not through the calix[4]arene annulus. The length of
the bridging alkylene chain between the respective calix[4]arenes of the semitube
structure dictates the strength and selectivity of alkali metal cation binding.
Keywords: alkali metals ¥ calixar-
enes ¥ ionophores ¥ potassium ¥
supramolecular chemistry
kinetics of potassium cation uptake by calix[4]tubes has lead
us to design less rigid ionophores, the calix[4]semitubes II that
consist of two calix[4]arene moieties connected by two
bridging alkylene chains. In this paper we report the synthesis,
coordination chemistry and modelling investigations of these
novel ionophores that exhibit fast complexation kinetics
whilst retaining remarkable potassium cation selectivity.
Introduction
The design and synthesis of new ionophores for alkali metals
is a continuing area of intense research activity.[1] In partic-
ular, the recent structural determination of the Streptomyces
lividans potassium channel[2] has stimulated further research
into elucidating the essential structural and mechanistic
features of ion channels through the construction of abiotic
systems.[3] The calixarene structural framework has been
modified to produce a series of potassium selective ligands.[4]
Notable examples include the tetraacetates of 1,3-alternate
calix[4]arene[5] and dioxacalix[4]arene,[6] the hexaacetate of
calix[6]arene[7] and a 1,3-alternate calix[4]crown receptor,
‡
‡
which displays higher K /Na selectivity than the natural
ionophore valinomycin.[8] The cation specificity of these
molecules can be attributed to the presence of multiple hard
oxygen donor atoms, a complementary cation-to-cavity size,
the degree of host preorganisation and, for the last receptor at
least, aromatic p cation interactions.
We have recently reported a new class of cryptand-type
ionophore, the calix[4]tube I,[9] based on a bis(calix[4]arene)
scaffold that displays exceptional selectivity for potassium
over all other Group 1metal cations and barium. The slow
[a] Prof. P. D. Beer, Dr P. R. A. Webber, Dr A. Cowley
Department of Chemistry
Inorganic Chemistry Laboratory
University of Oxford
Results and Discussion
South Parks Road, Oxford OX13QR (UK)
Fax : (‡44)1865-272-690
Synthesis: Taking into account the calix[4]tube structure[9] in
which the calix[4]arene units impart a size discriminatory
filter for the entry of metal cations through the calix[4]arene
annulus, our new receptor design features two calix[4]arene
moieties linked by two alkylene groups only, with propyl ether
E-mail: paul.beer@chem.ox.ac.uk
[b] Prof. M. G. B. Drew
Department of Chemistry
University of Reading
Whiteknights, Reading RG6 2AD (UK)
Chem. Eur. J. 2003, 9, 2439 2446
DOI: 10.1002/chem.200204518
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2439

P. D. Beer et al.
FULL PAPER
À
and phenolic OH functional groups making up the eight
oxygen-donor recognition site for alkali metal complexation.
The target calix[4]semitubes were prepared according to
the synthetic route shown in Scheme 1. The reaction of 1,3-
dipropoxy lower-rim substituted calix[4]arene 1 with two
equivalents of the appropriate bromoalkyl ethyl ester in the
presence of sodium hydride gave the ester ether products 2
and 3 in yields of 85% and 46%, respectively.[10] Lithium
aluminium hydride reduction in tetrahydrofuran gave the
corresponding alcohols 4 and 5, which on reaction with tosyl
chloride, gave the tosylates 6 and 7 in good overall yields.
Refluxing a mixture of the appropriate calix[4]arene and
calix[4]arene tosylate in acetonitrile in the presence of excess
potassium carbonate for seven to ten days gave the new
calix[4]semitubes 8, 9 and 10 in yields of 53%, 56% and 71%,
1
respectively. All calix[4]semitubes were characterised by H
Figure 1. The structure of 4 with ellipsoids at 20% probability. Hydrogen
and 13C NMR spectroscopy, mass spectrometry and elemental
analysis (see Experimental Section).
bonds are shown as dotted lines.
nor are there any intermolecular hydrogen bonds. Instead the
À
two OH groups O253 and O453 form hydrogen bonds to the
other oxygens O150 and O350 at distances of 2.99 and 2.74 ä,
respectively.
X-ray crystal structures of 4 and calix[4]semitube 9: Single
crystals of 4 were grown by slow evaporation from a dilute
solution of the compound in dichloromethane/methanol. The
structure of 4 is shown in Figure 1. The calix[4]arene adopts
the well-known distorted cone conformation with approxi-
mate C2 symmetry, in which two of the phenyl rings are
roughly parallel to the cone axis with angles of intersection
with the plane of the four methylene groups of 86.6 and 86.28.
The other two rings are semihorizontal making angles of 43.2
and 46.78 with this plane. There is no solvent in the unit cell,
Crystals of 9, suitable for X-ray diffraction, were grown by
slow evaporation from a solution of the ligand in chloroform/
methanol. The structure of 9 (Figure 2) reveals that both
calix[4]arene moieties in the molecule take up cone con-
formations. The unsubstituted calix[4]arene has a regular
cone structure in which the four phenyl rings intersect the
plane of the four methylene carbon atoms at consecutive
angles of 41.6, 68.4, 55.6 and 62.38. However, the tetra-tert-
K2CO3
Br
NaH
O
O
OH
OH
O
O
O
Br
OH
OH
OH
O
O
OH
CH3CN
OEt
n
n
n
EtO
THF (n = 1)
DMF (n = 3)
O
O
OEt
1
2 (n = 1)
3 (n = 3)
LiAlH4
THF
O
O
O
O
n
n
O
OH
HO
K2CO3
R
O
R
Pyridine or NEt3
p-TsCl
O
O
O
O
O
O
O
O
n
n
n
n
HO
TsO
R
R
R
OH
4
OH
OTs
8 R = tBu, (n = 1)
9 R = H, (n = 1)
6 (n = 1)
7 (n = 3)
4 (n = 1)
5 (n = 3)
10 R = tBu, (n = 3)
Scheme 1. Synthesis of the calix[4]semitubes. Tsˆ toluene-4-sulphonyl.
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2440
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2439 2446

Calix[4]semitubes
2439 2446
Table 1. Stability constant data for the alkali metal complexes.
[a]
K [mÀ1
]
8
9
‡
‡
Na
K
Rb
Cs
20
> 105
30
5
> 105
20
‡
‡
0
0
[a] Determined in 1:1 CDCl 3:CD3OD at 291K. The alkali metal cations
were added as their perchlorate, thiocyanate or hexafluorophosphate salts.
mixtures. In comparison, the information from Table 1shows
‡
‡
that 8 and 9 form very weak complexes with Na and Rb and
‡
that no evidence of complexation was observed with Cs .
Figure 3 reveals how the H NMR spectrum of 9 changes
with increasing equivalent amounts of K . The two singlets at
1
‡
around 6.15 and 6.8 ppm in free 9, due to the ArH protons on
the calixarene bearing the propyl groups, give strong evidence
that this calixarene in the receptor adopts the C2-distorted
cone conformation in solution. The chemical shift values of
the remaining ArH protons indicate that the other calixarene
in the molecule adopts a regular cone conformation. Both of
these facts demonstrate that the structure seen in solution is
similar to that in the solid state. Upon complexation, large
structural alterations are apparent from the significant
Figure 2. The structure of 9 in crystals of 9 ¥ CHCl3. Ellipsoids are drawn at
20% probability. The chloroform molecule is not shown.
1
changes seen in the H NMR spectrum. Most notably, the
positioning of the ArH resonances would tend to suggest that
butylcalix[4]arene has the more distorted cone conformation
with the four phenyl rings intersecting the plane of the four
methylene carbon atoms at consecutive angles of 75.6, 46.1,
83.2 and 24.98. The ether linkages between the two calix[4]-
arenes have different conformations as measured by the
torsion angles of the five bonds, which starting from the tert-
butyl end are 116.0, 145.0, 163.4, 43.8, 67.8 and À71.8, 148.1,
86.6, 89.6, À60.88. Thus, the two central O-C-C-O torsion
angles are trans (163.48) and gauche (86.68).
both calixarene moieties in the complex adopt regular cone
À
conformations. Interestingly, the propyl CH3 protons shift
upfield by about 0.6 ppm to À0.4 ppm upon complexation.
This is surprising because in the crystal structure of free 9, this
CH3 group is quite distant from any potential source of
interaction. For such a large upfield shift to occur, these CH3
groups probably move inwards and reside over the faces of the
free phenolic units of the adjacent calixarene moiety.
Analogous 1H NMR studies of Group 1metal-cation
coordination with the butyl-linked calix[4]semitube 10 con-
taining a larger cavity revealed no evidence of complexation
of any of the alkali metal cations in 1:1 CDCl3/CD3OD. This
observation suggests that the length of the bridging alkylene
chain between the respective calix[4]arenes of the calix[4]se-
mitube structure critically determines the strength of com-
plexation and Group 1selectivity preferences for this new
class of ionophore.
Coordination studies
1H NMR complexation investigations: Preliminary experi-
ments were undertaken on calix[4]semitubes 8 and 9 to
‡
establish their kinetic complexation behaviour towards Na ,
‡
‡
K and Rb ions. The addition of one equivalent of NaClO4,
KSCN or RbPF6 to solutions of the respective calix[4]semi-
tubes in 1:1 CDCl3/CD3OD revealed that although the
complexation/decomplexation process was slow on the
NMR timescale, the time required for each complexation to
reach equilibrium was fast, at least shorter than the time taken
to procure the NMR titration experiment, which was less than
4 minutes. This contrasts with calix[4]tube I for which a
heterogeneous mixture of 20 equivalents of KI and I in 4:1
CDCl3:CD3OD took just under 2 h to reach equilibrium.[9]
Stability constants for 8 and 9 with alkali metal cations were
determined by direct integration of host and complex
resonances in the 1H NMR spectrum; these values are shown
in Table 1. As was hoped, both calix[4]semitubes display
Molecular modelling: Molecular modelling was undertaken in
an effort to ascertain the mode of entry of potassium into the
calix[4]semitube, and account for this ionophore×s remarkable
selectivity and fast kinetics of potassium cation binding.
In the X-ray crystal structure of 9, it is seen that the two
ether spacers have different conformations, one being trans
and the other gauche. These differences are rather surprising
as in the calix[4]tube such linkages are equivalent. The
calix[4]tube has four linkages that lead to increased rigidity.
Perhaps, therefore, it is not surprising that the linkages on
opposite sides of the central cavity were always equivalent so
that in the absence of metal ions the four linkages were
alternately gauche and trans (gtgt), while with an encapsulated
metal, the four linkages were all necessarily gauche (gggg), so
that the metal could bond to all eight oxygen atoms.
‡
excellent K selectivity. In fact, both ligands are 100%
‡
complexed upon addition of one equivalent of K . The
solubility properties of 8 and 9 precluded complexation
studies being carried out in more competitive solvent
Chem. Eur. J. 2003, 9, 2439 2446
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2441

P. D. Beer et al.
FULL PAPER
1
‡
Figure 3. Changes in the H NMR spectrum of 9 in the presence of specified equivalents of K (500 MHz, 1:1 CDCl3:CD3OD).
[11 12]
‡
The established mechanism[9]
,
for the insertion of
in an intermediate position. In Ia ¥ K , the metal ion has now
reached the centre of the calix[4]tube. This is accompanied by
a change in conformation of the cage so that all four torsion
angles are now gauche enabling all eight oxygen atoms to
bond to the metal ion. The conformations of the individual
calix[4]arenes in the calix[4]tube both change from C2-
distorted cones to regular cones.
metal ions into calix[4]tubes is illustrated by the crystal
2‡
‡
structures of Ia, Ib ¥ Tl2 and Ia ¥ K shown in Figure 4a c.
This series represents the three stages in the introduction of
metals ions into the cavity. Initially, the calix[4]tube is
uncomplexed (Figure 4a). In the second stage, the metal
enters the top of the calixarene and forms an intermediate
complex interacting with the aromatic rings (Figure 4b). The
metal then proceeds into the centre of the calix[4]tube and
forms bonds to the eight oxygen atoms (Figure 4c).
In previous molecular dynamics simulations[12, 13, 19] we
modelled the formation of metal complexes of the tert-
butyl-calix[4]tube. We now report a study of K within 9 to
‡
explain its remarkable selectivity for potassium and its fast
kinetics of complexation relative to that observed in the
calix[4]tube.
We first calculated the hole size of the macrocycle by
‡
inserting K into the centre of the macrocycle in the gg cone/
À
cone conformation, fixing the K O distances at specific values
(from 2.3 to 3.3 ä at 0.5 ä intervals) and allowing the
macrocycle to relax its conformation by means of molecular
À
mechanics. A resulting plot of energy against K O distance
provided a minimum at 2.84 ä, exactly the same value as was
obtained previously for the calix[4]tube and ideally suited for
‡
the complexation of K .
A conformation analysis on 9 by using molecular dynamics
was then performed. The simulation was carried out at 3000 K
with a stepsize of 1fs. A total of 1000 conformations were
saved at 1ps intervals and subsequently optimised with
molecular mechanics with the Cerius2 software. All low
energy structures showed both O-C-C-O linkages to be trans
and were therefore different from that observed in the crystal
structure of 9 which was located in the analysis but with an
energy higher by about 25 kcalmolÀ1. However, the cone
cone conformation was maintained in all 1000 energy con-
Figure 4. The three stages of metal encapsulation within calix[4]tubes
illustrated by the crystal structures of Ia, Ib ¥ Tl2 and Ia ¥ K .
2‡
‡
In Ia, both parts of the calix[4]tube have the C2-distorted
cone conformation with links that can be described as tgtg
(signs ignored), with alternate O-C-C-O torsion angles trans
2‡
and gauche. In Ib ¥ Tl2 , metal ions have entered the
calix[4]tube and, while the conformation remains the same,
the metals ions now interact with the phenyl rings at each end
2442
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2439 2446

Calix[4]semitubes
2439 2446
formations; therefore, it seemed appropriate to use the cone
cone conformation with the tt conformation in further
calculations.
We next investigated the likely mode of entry of potassium
ions into the macrocycle. For 9, there are three possible modes
of entry down the axis as shown in Figure 5; from either a) the
tetra-tert-butyl end or b) the unsubstituted end or c) through a
horizontal route perpendicular to the main axis. In the
calix[4]tube the horizontal route was shown to be highly
unlikely but it was far more accessible for 9 as there were only
two linkages between the calix[4]arenes; this led to a more
open cage.
Figure 6. The molecular mechanics energy obtained by moving the metal
ion along the entry path into the macrocycle 9. The distance plotted is that
from the centre of the macrocycle. a) diamonds by the vertical route from
the tert-butyl end, b) squares by the vertical route from the unsubstituted
end and c) triangles by the horizontal route.
carried out at a variety of temperatures with no constraints to
test the likelihood of each path being followed. The horizontal
route was successfully followed by the metal ion at 350 K, a
temperature significantly lower than that which proved
possible for the completion of the other routes. Snapshots of
this simulation are shown in Figure 7. The metal ion moved
around its starting position for the first 210 ps with the
macrocycle remaining in its original conformation (Fig-
ure 7a). After 215.6 ps the metal began to move towards the
cavity (Figure 7b) and it had nearly reached the cavity by
216.3 ps (Figure 7c). By 216.67 ps it had completed its move
(Figure 7d) and was interacting with all eight oxygen atoms.
As the metal moved towards the centre of the cavity the
conformation changed from tt to gg. However, this did not
Figure 5. Three possible modes of entry of a cation into the cavity of 9:
a) vertically from the tert-butyl end, b) vertically from the unsubstituted
end and c) horizontally.
Accordingly, we carried out molecular mechanics calcula-
tions to investigate the three possible pathways. The metal ion
was positioned at the centre of the macrocycle and moved at
0.5 ä intervals in the three possible directions a, b and c. At
each position the macrocycle was allowed to relax around the
metal and the energy of the metal complex was calculated.
The resulting energies are plotted in Figure 6.
The shape of the curve for path b, the vertical route from
the unsubstituted end, is similar in shape to that observed for
the calix[4]tube and shows an energy minimum around 4.5 ä
for the intermediate position whereby the metal associates
with the phenyl rings, although the energy maximum at
around 2 ä is not so significant as in the case of the
calix[4]tube. However, a minimum energy for an intermediate
position is not observed in either curves a or c showing that
the metal can follow a low energy path into the metal cavity.
These calculations show that all three entry paths were
possible.
‡
The mode of insertion of K into the macrocycle was also
investigated by molecular dynamics. We carried out three sets
of simulations to establish by which path the metal was likely
to enter the macrocycle. In each starting model, the metal ion
was positioned at about 8 ä from the centre of the cavity
along each path as shown in Figure 5. The simulations were
Figure 7. Snapshots at various times during the molecular dynamics
simulation of the formation of 9 ¥ K showing how the metal ion enters
the cavity by the horizontal route. Snapshots are taken after a) 210.0,
b) 215.6, c) 216.3 and d) 216.6 ps.
‡
Chem. Eur. J. 2003, 9, 2439 2446
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2443

P. D. Beer et al.
FULL PAPER
were obtained on an elementar vario EL and were performed by the
Inorganic Chemistry Laboratory, University of Oxford.
happen until the metal was within 2 ä of the centre of the
cavity. It is apparent from Figure 7, that as the metal
approaches the cavity, the terminal ether group acts as a flap,
opening to allow the metal ion to enter the cavity and then
closing once the metal is enclosed.
Subsequent simulations showed that the metal ion could
enter the cavity by the two vertical routes, but only when the
temperature was increased to 500 K. It is interesting to note
that when the metal follows the vertical route there is no
specific conformational change as is observed in the calix[4]-
tube. Instead we find in a typical simulation that the metal
reaches the plane at the bottom of the phenyl rings while the
conformation is still tt, one torsion angle changes to gauche as
the metal approaches the plane of the four oxygen atoms at
the bottom of the cone and it is only when the metal enters the
cavity that the conformation changes to gg.
p-tert-Butylcalix[4]arene,[14] calix[4]arene[15] and 1,3-dipropoxy-tert-butyl-
calix[4]arene (1)[16] were prepared according to literature procedures.
5,11,17,23-Tetra-tert-butyl-25,27-bis[(ethoxycarbonyl)methoxy]-26,28-di-
propoxycalix[4]arene (2): A solution of 1 (5.0 g, 6.82 mmol) and NaH
(1.63 g, 40 mmol) in tetrahydrofuran (125 mL) was stirred at room temper-
ature for 1hour then ethyl bromoacetate (6.68 g, 4.44 mL, 40 mmol) was
added. Following a further 72 hours stirring, water (50 mL) was added
dropwise. The organic solvent was then removed in vacuo and the aqueous
layer extracted with dichloromethane (3 Â 50 mL). The combined organic
extracts were dried (MgSO4) and reduced in vacuo yielding a brown oil. By
recrystallising from methanol/dichloromethane in the fridge, pure 2 was
obtained as a white powder (5.23 g, 85%). 1H NMR (500 MHz, [D]CHCl3,
188C): d ˆ 1.00 (t, 3J(H,H) ˆ 7.4 Hz, 6H; OCH2CH2CH3), 1.01 (s, 18H;
(CH3)3C), 1.14 (s, 18H; (CH3)3C), 1.28 (t, 3J(H,H) ˆ 7.1Hz, 6H;
C(O)OCH2CH3), 1.95 (m, 4H; OCH2CH2CH3), 3.17 (d, 2J(H,H) ˆ
12.8 Hz, 4H; ArCH2Ar), 3.82 (t, 3J(H,H) ˆ 7.5 Hz, 4H; OCH2CH2CH3),
4.21(q, 3J(H,H) ˆ 7.1Hz, 4H; C(O)OC H2CH3), 4.62 (d, 2J(H,H) ˆ 12.8 Hz,
4H; ArCH2Ar), 4.79 (s, 4H; OCH2C(O)), 6.68 (s, 4H; ArH), 6.86 ppm (s,
4H; ArH); 13C NMR (75 MHz, [D]CHCl3, 1 8C8): d ˆ 10.44
(OCH2CH2CH3), 14.18 (OCH2CH3), 23.21(OCH 2CH2CH3), 31.32
((CH3)3C), 31.44 ((CH3)3C), 31.62 (ArCH2Ar), 33.70 ((CH3)3C), 33.85
((CH3)3C), 60.33 (OCH2CH3), 70.76 and 77.04 (OCH2CH2CH3 and
It is interesting that with the calix[4]tube, the reaction path
‡
was successfully followed by the K ion at 400 K. These results
confirm that the metal ion readily enters the cavity of 9 by the
horizontal path and the results are consistent with the
experimental results that the metal complexes with 9 more
readily than with the calix[4]tube.
OCH2C(O)), 124.80, 125.34, 133.09, 133.98, 144.28, 144.91, 152.90 and
‡
ˆ
153.57 (Ar), 170.47 ppm (C O); MS (ES): m/z: 928 [M ‡Na], 944
‡
[M ‡K]; elemental analysis calcd (%) for C58H80O8: C 76.95, H 8.91;
found: C 76.75, H 9.81.
5,11,17,23-Tetra-tert-butyl-25,27-bis[(ethoxycarbonyl)propoxy]-26,28-di-
propoxycalix[4]arene (3): Calix[4]arene 1 (2.0 g, 2.73 mmol) and NaH
(0.66 g, 16.4 mmol) were stirred in dimethylformamide (75 mL) at room
temperature for 2 hours. After this time, ethyl-4-bromobutyrate (3.20 g,
16.4 mmol) was added and stirring continued for a further 48 hours.
Following removal of the solvent in vacuo, the mixture was partitioned
between dichloromethane (75 mL) and water (75 mL). The organic layer
was then washed with saturated NH4Cl (75 mL), dried (MgSO4) and
reduced in vacuo. The residue was purified by column chromatography on
silica gel eluting with 3:2 (v/v) chloroform:hexane. The fastest moving
fraction was collected and reduced in vacuo giving 3 as a white powder
(1.20 g, 46%). 1H NMR (500 MHz, [D]CHCl3, 1 88C): d ˆ 1.00 (t,
3J(H,H) ˆ 7.5 Hz, 6H; OCH2CH2CH3), 1.07 (s, 18H; (CH3)3C), 1.12 (s,
Conclusion
The new calix[4]semitube ligand design consisting of two
calix[4]arene units linked by two ethylene groups, together
with 1,3-disubstituted propyl ether and phenolic OH func-
À
tional groups creating eight oxygen donor sites, represents a
new class of ionophore that exhibits remarkable selectivity for
potassium together with fast complexation kinetics. This is
contrary to the slow kinetics of binding exhibited by the
potassium selective calix[4]tube.
Molecular modelling simulations, on structural models
from crystallographic determinations, demonstrate that the
most likely route of entry for the potassium cation into the
calix[4]semitube is through a horizontal or ™side-on∫ route,
which contrasts with the preferred vertical pathway through
the calix[4]arene annulus of the calix[4]tube. This horizontal
mode of potassium cation entry may account for the
calix[4]semitube exhibiting fast kinetics of complexation
relative to those of the calix[4]tube. The length of the bridging
alkylene chain between the respective calix[4]arenes of the
calix[4]semitube crucially dictates the strength and selectivity
of alkali metal cation binding.
3
18H; (CH3)3C), 1.29 (t, J(H,H) ˆ 7.0 Hz, 6H; C(O)OCH2CH3), 2.03 (m,
4H; OCH2CH2CH3), 2.33 (m, 4H; OCH2CH2CH2C(O)), 2.52 (t, 3J(H,H) ˆ
7.5 Hz, 4H; OCH2CH2CH2C(O)), 3.14 (d, 2J(H,H) ˆ 12.5 Hz, 4H; ArCH2-
Ar), 3.82 (t, 3J(H,H) ˆ 7.5 Hz, 4H; OCH2CH2CH3), 3.91(t, 3J(H,H) ˆ
7.5 Hz, 4H; OCH2CH2CH2C(O)), 4.17 (q, 3J(H,H) ˆ 7.0 Hz, 4H;
2
C(O)OCH2CH3), 4.39 (d, J(H,H) ˆ 12.5 Hz, 4H; ArCH2Ar), 6.76 (s, 4H;
ArH), 6.82 ppm (s, 4H; ArH); 13C NMR (125 MHz, [D]CHCl3, 1 88C): d ˆ
10.25 (OCH2CH2CH3), 14.26 (C(O)OCH2CH3), 23.29 (OCH2CH2CH2-
C(O)), 25.54 (OCH2CH2CH3), 31.01 and 31.22 (ArCH2Ar and
OCH2CH2CH2C(O)), 31.40 ((CH3)3C), 31.45 ((CH3)3C), 33.77 ((CH3)3C),
33.81((CH 3)3C), 60.24 (C(O)OCH2CH3), 74.11 (OCH2CH2CH2C(O)),
77.04 (OCH2CH2CH3), 124.86, 124.98, 133.52, 133.87, 144.20, 144.47 and
153.39 (coincident) (Ar), 173.41 ppm (O(CH2)3C(O)); MS (ES): m/z: 979
‡
‡
[M ‡NH4], 984 [M ‡Na]; elemental analysis calcd (%) for C62H88O8: C
77.46, H 9.23; found: C 77.33, H 9.43.
5,11,17,23-Tetra-tert-butyl-25,27-bis(hydroxyethoxy)-26,28-dipropoxyca-
lix[4]arene (4): LiAlH4 (1.24 g, 33 mmol) was added in small portions at
08C to a solution of calix[4]arene bis(ester) 2 (5.95 g, 6.57 mmol) in
tetrahydrofuran (150 mL). The reaction mixture was allowed to warm to
room temperature and then left to stir overnight. Following careful
addition of 2m H2SO4 (50 mL), the mixture was left to stir for a further
hour. The organic solvent was removed in vacuo and the aqueous layer
extracted with dichloromethane (2 Â 125 mL). The combined organic
extracts were then dried (MgSO4) and reduced in vacuo. If necessary, the
product can be recrystallised from methanol/dichloromethane in the fridge
to yield pure 4 as large colourless crystals (4.56 g, 85%). 1H NMR
(500 MHz, [D]CHCl3, 1 88C): d ˆ 0.84 (s, 18H; (CH3)3C), 0.95 (t,
3J(H,H) ˆ 7.5 Hz, 6H; OCH2CH2CH3), 1.36 (s, 18H; (CH3)3C), 1.88 (m,
4H; OCH2CH2CH3), 3.20 (d, 2J(H,H) ˆ 12.5 Hz, 4H; ArCH2Ar), 3.74 (t,
3J(H,H) ˆ 7.5 Hz, 4 H ; OCH2CH2CH3), 4.03 (m, 4H; OCH2CH2OH), 4.12
Experimental Section
General: All chemicals were commercial grade and used without further
purification unless stated otherwise. Solvents were predried, purified by
distillation and stored under nitrogen where appropriate. Acetonitrile and
dichloromethane were distilled from calcium hydride, tetrahydrofuran
from sodium wire with benzophenone as an indicator and triethylamine
was distilled from potassium hydroxide. Prior to use, pyridine was dried
over potassium hydroxide and dimethylformamide was dried over molec-
ular sieves (4 ä).
NMR spectra were recorded by using either a 300 MHz Varian VXWorks
spectrometer or a 500 MHz Varian Unity spectrometer. Electrospray mass
spectra were recorded by using Micromass LCT equipment. Microanalyses
2444
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2439 2446

Calix[4]semitubes
2439 2446
2
(m, 4H; OCH2CH2OH), 4.36 (d, J(H,H) ˆ 12.5 Hz, 4H; ArCH2Ar), 5.09
(brs, 2H; OH), 6.50 (s, 4H; ArH), 7.16 ppm (s, 4H; ArH); 13C NMR
(75 MHz, [D]CHCl3, 1 8C8): d ˆ 10.15 (OCH2CH2CH3), 22.52
(OCH2CH2CH3), 30.77 (ArCH2Ar), 31.06 ((CH3)3C), 31.69 ((CH3)3C),
33.59 ((CH3)3C), 34.09 ((CH3)3C), 61.65, 76.90 and 78.46 (OCH2CH2CH3,
OCH2CH2OH and OCH2CH2OH), 124.71, 125.80, 131.75, 135.55, 144.62,
4H; ArCH2Ar), 3.74 (t, 3J(H,H) ˆ 7.5 Hz, 4H; OCH2CH2CH3), 3.82 (t,
3J(H,H) ˆ 7.5 Hz, 4H; O(CH2)3CH2OTs), 4.10 (t, 3J(H,H) ˆ 6.5 Hz, 4H;
OCH2(CH2)3OTs), 4.30 (d, 2J(H,H) ˆ 12.5 Hz, 4H; ArCH2Ar), 6.73 (s, 4H;
ArH), 6.80 (s, 4H; ArH), 7.36 (d, 3J(H,H) ˆ 7.5 Hz, 4H; Ts-ArH), 7.81ppm
(d, 3J(H,H) ˆ 7.5 Hz, 4H; Ts-ArH); 13C NMR (75 MHz, [D]CHCl3, 1 88C):
d ˆ 10.40 (OCH2CH2CH3), 21.70 (Ts-CH3), 23.44, 25.84 and 26.24
(OCH2CH2CH3, OCH2CH2(CH2)2OTs and O(CH2)2CH2CH2OTs), 31.09
(ArCH2Ar), 31.46 ((CH3)3C), 31.50 ((CH3)3C), 33.81((CH 3)3C), 33.88
((CH3)3C), 70.44 (O(CH2)3CH2OTs), 74.03 (OCH2(CH2)3OTs), ca.77 under
solvent (OCH2CH2CH3), 124.79, 124.93, 127.79, 129.78, 133.07, 133.32,
133.74, 144.17, 144.48, 144.60 and 153.24 ppm (2 coincident) (Ar and Ts-
‡
145.58, 151.16 and 153.76 ppm (Ar); MS (ES): m/z: 844 [M ‡Na], 860
‡
[M ‡K]; elemental analysis calcd (%) for C54H76O6: C 78.98, H 9.33;
found: C 78.73, H 10.32.
5,11,17,23-Tetra-tert-butyl-25,27-bis(hydroxybutoxy)-26,28-dipropoxyca-
lix[4]arene (5): LiAlH4 (0.16 g, 4.16 mmol) was added in small portions at
08C to a solution of calix[4]arene bis(ester) 3 (1.0 g, 1.04 mmol) in
tetrahydrofuran (40 mL). The mixture was then allowed to stir overnight at
room temperature. After careful addition of 2m H2SO4 (40 mL) and a
further hour of stirring, the tetrahydrofuran was removed in vacuo. The
aqueous layer was then extracted with dichloromethane (2 Â 50 mL). The
combined organic extracts were subsequently dried (MgSO4) and reduced
in vacuo. Pure 5 could be precipitated from methanol/water as a white
‡
‡
Ar); MS (ES): m/z: 1209 [M ‡Na], 1225 [M ‡K]; elemental analysis
calcd (%) for C72H96O10S2: C 72.94, H 8.16; found: C 72.66, H, 8.22.
tert-Butyl ethyl semitube (8): A suspension of p-tert-butylcalix[4]arene
(0.58 g, 0.89 mmol) and K2CO3 (1.23 g, 8.9 mmol) was refluxed in
acetonitrile (50 mL) for 2.5 hours. After this time, the bis(tosyl) substituted
calix[4]arene 6 (1.0 g, 0.89 mmol) was added and the mixture was refluxed
for a further 7 10 days. Following removal of the solvent in vacuo, the
mixture was partitioned between chloroform (50 mL) and water (50 mL).
The organic layer was dried (MgSO4) and reduced in vacuo. Ethanol
(75 mL) was added to this crude material, which after briefly refluxing, was
hot-filtered giving 8 as a white powder (0.68 g, 53%). 1H NMR (500 MHz,
1
powder (0.79 g, 87%). H NMR (500 MHz, [D]CHCl3, 1 88C): d ˆ 0.99 (s,
18H; (CH3)3C), 1.03 (t, 3J(H,H) ˆ 7.5 Hz, 6H; OCH2CH2CH3), 1.19 (s,
18H; (CH3)3C), 1.67 (m, 4H; O(CH2)2CH2CH2OH), 1.98 (m, 4H;
OCH2CH2CH3), 2.15 (m, 4H; OCH2CH2(CH2)2OH), 3.13 (d, 2J(H,H) ˆ
13.0 Hz, 4H; ArCH2Ar), 3.76 (m, 8H; OCH2CH2CH3 and O(CH2)3CH2OH
coincident), 3.96 (t, 3J(H,H) ˆ 7.5 Hz, 4H; OCH2(CH2)3OH), 4.41(d,
2J(H,H) ˆ 13.0 Hz, 4H; ArCH2Ar), 6.66 (s, 4H; Ar), 6.91ppm (s, 4H; Ar);
13C NMR (75 MHz, [D]CHCl3, 1 88C): d ˆ 10.65 (OCH2CH2CH3), 23.51,
3
[D]CHCl3, 1 88C): d ˆ 0.62 (t, J(H,H) ˆ 7.2 Hz, 6H; OCH2CH2CH3), 0.84
(s, 18H; (CH3)3C), 1.21 (s, 36H; 2Â [(CH3)3C)] (coincident)), 1.36 (s, 18H;
(CH3)3C), 1.71 (m, 4H; OCH2CH2CH3), 3.25 (d, 2J(H,H) ˆ 12.9 Hz, 4H;
2
3
ArCH2Ar), 3.43 (d, J(H,H) ˆ 13.5 Hz, 4H; ArCH2Ar), 3.95 (t, J(H,H) ˆ
7.8 Hz, 4H; OCH2CH2CH3), 4.56 (d, 2J(H,H) ˆ 12.3 Hz, 4H; ArCH2Ar),
26.55
and
29.43
(OCH2CH2CH3,
OCH2CH2(CH2)2OH
and
2
3
O(CH2)2CH2CH2OH), 31.14 (ArCH2Ar), 31.39 ((CH3)3C), 31.62
((CH3)3C), 33.75 ((CH3)3C), 33.95 ((CH3)3C), 62.96 (O(CH2)3CH2OH),
74.83 (OCH2(CH2)3OH), 77.17 (OCH2CH2CH3), 124.62, 125.05, 132.91,
134.43, 143.99, 144.38, 153.09 and 153.86 ppm (Ar); MS (ES): m/z: 895
4.60 (d, J(H,H) ˆ 12.3 Hz, 4H; ArCH2Ar), 5.10 (t, J(H,H) ˆ 7.8 Hz, 4H;
ROCH2CH2OR'), 5.28 (t, 3J(H,H) ˆ 7.8 Hz, 4H; ROCH2CH2OR'), 6.42 (s,
4H; ArH), 6.99 (s, 4H; ArH), 7.07 (s, 4H; ArH), 7.12 (s, 4H; ArH),
9.29 ppm (s, 2H; OH); 13C NMR (75 MHz, [D]CHCl3, 1 88C): d ˆ 10.07
(OCH2CH2CH3), 22.77 (OCH2CH2CH3), 31.18, 31.29, 31.56 and 31.72
((CH3)3C), 32.26, 33.56, 33.87, 33.92, 34.01and 34.02 ((CH 3)3C and 2 Â
[ArCH2Ar]), 71.05, 73.54 and 77.49 (OCH2), 124.12, 125.16, 125.68, 126.45,
128.45, 131.56, 132.81, 134.96, 141.56, 143.70, 144.58, 146.26, 150.41, 150.76,
‡
‡
[M ‡NH4], 900 [M ‡Na]; elemental analysis calcd (%) for C62H88O8: C
79.41, H 9.65; found: C 79.20, H 9.68.
5,11,17,23-Tetra-tert-butyl-25,27-bis(tosylethoxy)-26,28-dipropoxycalix[4]-
arene (6): para-Toluenesulphonyl chloride (10.7 g, 56 mmol) was added in
one portion to a stirred solution of calix[4]arene bis(alcohol) 4 (4.6 g,
5.6 mmol) in pyridine (60 mL) and the mixture was stirred for 30 minutes.
The flask was then refrigerated at around 48C for 5 days after which time
the red solution was poured into iced 2m HCl (200 mL). The precipitate
was filtered, taken up in chloroform (150 mL) and washed with water
(200 mL). The organic layer was dried (MgSO4) and reduced in vacuo. Pure
6 was obtained as large colourless crystals (5.49 g, 87%) by crystallisation
from methanol (75 mL). 1H NMR (300 MHz, [D]CHCl3, 1 88C): d ˆ 0.85 (s,
18H; (CH3)3C), 0.93 (t, 3J(H,H) ˆ 7.2 Hz, 6H; OCH2CH2CH3), 1.30 (s,
18H; (CH3)3C), 1.79 (m, 4H; OCH2CH2CH3), 2.48 (s, 6H; Ts-CH3), 3.08 (d,
2J(H,H) ˆ 12.9 Hz, 4H; ArCH2Ar), 3.62 (t, 3J(H,H) ˆ 7.5 Hz, 4H;
OCH2CH2CH3), 4.23 (d, 2J(H,H) ˆ 12.9 Hz, 4H; ArCH2Ar), 4.24 (t,
3J(H,H) ˆ 6.0 Hz, 4H; OCH2CH2OTs), 4.63 (t, 3J(H,H) ˆ 6.0 Hz, 4H;
OCH2CH2OTs), 6.48 (s, 4H; ArH), 7.02 (s, 4H; ArH), 7.37 (d, 3J(H,H) ˆ
8.1Hz, 4H; Ts-ArH), 7.81ppm (d, 3J(H,H) ˆ 8.1Hz, 4H; Ts-ArH);
13C NMR (75 MHz, [D]CHCl3, 1 88C): d ˆ 10.50 (OCH2CH2CH3), 21.72
(Ts-CH3), 23.29 (OCH2CH2CH3), 31.09 (ArCH2Ar), 31.17 ((CH3)3C), 31.68
((CH3)3C), 33.62 ((CH3)3C), 34.09 ((CH3)3C), 69.03 and 70.62 (OCH2-
CH2OTs and OCH2CH2OTs), 77.63 (OCH2CH2CH3), 124.52, 125.37,
127.78, 129.74, 131.89, 133.41, 134.90, 144.14, 144.47, 145.42, 152.13 and
‡
‡
153.33 and 154.61 ppm (Ar); MS (ES): m/z: 1457 [M ‡Na], 1473 [M ‡K];
elemental analysis calcd (%) for C98H128O8 ¥ 1/3(CHCl3): C 80.13, H 8.78;
found: C 80.53, H 8.61.
De-tert-butyl ethyl semitube (9): A suspension of calix[4]arene (0.38 g,
0.89 mmol) and K2CO3 (1.23 g, 8.9 mmol) was refluxed in acetonitrile
(50 mL) for 2 hours. After this time, the bis(tosyl) substituted calix[4]arene
6 (1.0 g, 0.89 mmol) was added and the mixture was refluxed for a further
9.5 days. Following removal of the solvent in vacuo, the mixture was
partitioned between chloroform (100 mL) and water (100 mL). The organic
layer was dried (MgSO4) and reduced in vacuo. Ethanol (50 mL) was added
to this pink/brown solid which after briefly refluxing, was hot-filtered giving
9 as a white powder (0.52 g). A further small crop of desired product
(0.08 g) could be obtained by loading the remaining residue onto a silica gel
column and gradient eluting from dichloromethane to 10% ethyl acetate to
10% methanol. Combined yield (0.60 g, 56%). 1H NMR (500 MHz,
3
[D]CHCl3, 1 88C): d ˆ 0.57 (t, J(H,H) ˆ 7.5 Hz, 6H; OCH2CH2CH3), 0.82
(s, 18H; (CH3)3C), 1.34 (s, 18H; (CH3)3C), 1.70 (m, 4H; OCH2CH2CH3),
3.26 (d, 2J(H,H) ˆ 13.0 Hz, 4H; ArCH2Ar), 3.47 (d, 2J(H,H) ˆ 13.0 Hz,
4H; ArCH2Ar), 3.91(t, 3J(H,H) ˆ 8.0 Hz, 4H; OCH2CH2CH3), 4.55 (d,
2J(H,H) ˆ 13.0 Hz, 4H; ArCH2Ar), 4.60 (d, 2J(H,H) ˆ 12.5 Hz, 4H;
ArCH2Ar), 5.14 (t, 3J(H,H) ˆ 7.3 Hz, 4H; ROCH2CH2OR'), 5.31(t,
3J(H,H) ˆ 7.3 Hz, 4H; ROCH2CH2OR'), 6.44 (s, 4H; ArH), 6.61(t,
‡
‡
153.22 ppm (Ar and Ts-Ar); MS (ES): m/z: 1153 [M ‡Na], 1169 [M ‡K];
elemental analysis calcd (%) for C68H88O10S2 ¥ H2O: C 71.17, H 7.90; found:
C 70.96, H 8.11.
3
3J(H,H) ˆ 7.5 Hz, 2H; ArH), 6.80 (t, J(H,H) ˆ 7.5 Hz, 2H; ArH), 7.03 (d,
3
3J(H,H) ˆ 7.5 Hz, 2H; ArH), 7.07 (d, J(H,H) ˆ 7.5 Hz, 2H; ArH), 7.16 (s,
5,11,17,23-Tetra-tert-butyl-25,27-bis(tosylbutoxy)-26,28-dipropoxycalix[4]-
arene (7): Triethylamine (3 mL, 21mmol) and para-toluenesulphonyl
chloride (0.76 g, 4.0 mmol)was added to
4H; ArH), 9.44 ppm (s, 2H; OH); 13C NMR (75 MHz, [D]CHCl3, 1 88C):
d ˆ 10.06 (OCH2CH2CH3), 22.91(OCH 2CH2CH3), 31.16 ((CH3)3C), 31.71
((CH3)3C), 32.18, 33.04, 33.56 and 34.04 ((CH3)3C and 2?[ArCH2Ar]),
71.37, 73.78 and 77.62 (OCH2), 119.52, 124.18, 124.53, 125.76, 128.22, 128.88,
129.56, 131.49, 133.27, 134.92, 143.80, 144.72, 152.36, 152.77, 153.27 and
a solution of calix[4]arene
bis(alcohol) (0.70 g, 0.80 mmol) in dichloromethane (20 mL). The
5
mixture was stirred at room temperature for 42 hours. Water (20 mL)
was then added and stirring continued for a further 0.5 hours. The organic
layer was washed with 2m HCl (2 Â 50 mL) and saturated NaCl solution
(50 mL), dried (MgSO4) and then reduced in vacuo. Reprecipitation from
ethanol/water gave pure 7 as a white powder (0.58 g 61%). 1H NMR
(500 MHz, [D]CHCl3, 1 8C8 ): d ˆ 0.94 (t, 3J(H,H) ˆ 7.5 Hz, 6H;
OCH2CH2CH3), 1.05 (s, 18H; (CH3)3C), 1.11 (s, 18H; (CH3)3C), 1.80 (m,
4H; OCH2CH2(CH2)2OTs), 1.92 (m, 4H; OCH2CH2CH3), 2.02 (m, 4H;
O(CH2)2CH2CH2OTs), 2.45 (s, 6H; Ts-CH3), 3.09 (d, 2J(H,H) ˆ 12.5 Hz,
‡
154.59 ppm (Ar); MS (ES): m/z: 1249 [M ‡K]; elemental analysis calcd
(%) for C82H96O8.1/3(CHCl3): C 79.17, H 7.77; found: C 79.21, H 7.47.
tert-Butyl butyl semitube (10): A suspension of p-tert-butylcalix[4]arene
(0.091g, 0.14 mmol) and K 2CO3 (0.077 g, 0.56 mmol) was refluxed in
acetonitrile (50 mL) for 2 hours. After this time, bis(tosyl) substituted
calix[4]arene 7 (0.20 g, 0.17 mmol) was added and the mixture refluxed for
a further 5 days. Following removal of the solvent in vacuo, the mixture was
Chem. Eur. J. 2003, 9, 2439 2446
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2445

P. D. Beer et al.
FULL PAPER
partitioned between chloroform (75 mL) and water (75 mL). The organic
layer was dried (MgSO4) and reduced in vacuo. The residue was heated in
acetonitrile (20 mL) and the solid hot-filtered giving pure 10 as a white
solid (0.149 g, 71%). 1H NMR (500 MHz, [D]CHCl3, 1 88C): d ˆ 0.60 (t,
3J(H,H) ˆ 7.5 Hz, 6H; OCH2CH2CH3), 0.83 (s, 18H; (CH3)3C), 1.10 (s,
18H; (CH3)3C), 1.25 (s, 18H; (CH3)3C), 1.36 (s, 18H; (CH3)3C), 1.68 (m,
4H; OCH2CH2CH3), 2.12 and 2.66 (m, 8H; ROCH2CH2(CH2)2OR' and
final R values were R1 ˆ 0.0728 and wR2 ˆ 0.0857 for 11977 data with I >
2s(I).
CCDC 195350 and 195351 contain(s) the supplementary crystallographic
data for this paper. These data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crys-
tallographic Data Centre, 12 Union Road, Cambridge CB21EZ, UK; fax:
(‡44)1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk).
2
RO(CH2)2CH2CH2OR'), 3.15 (d, J(H,H) ˆ 13.0 Hz, 4H; ArCH2Ar), 3.36
(d, 2J(H,H) ˆ 13.0 Hz, 4H; ArCH2Ar), 3.71(t, 3J(H,H) ˆ 7.5 Hz, 4H;
OCH2CH2CH3), 4.07 (t, 3J(H,H) ˆ 5.0 Hz, 4H; RO(CH2)3CH2OR'), 4.32 (t,
3J(H,H) ˆ 7.5 Hz, 4H; ROCH2(CH2)3OR'), 4.37 (d, 2J ˆ 13.0 Hz, 4H;
ArCH2Ar), 4.52 (d, 2J(H,H) ˆ 13.0 Hz, 4H; ArCH2Ar), 6.43 (s, 4H;
ArH), 6.97 (s, 4H; ArH), 7.03 (s, 4H; ArH), 7.14 (s, 4H; ArH), 8.46 ppm (s,
2H; OH); 13C NMR (75 MHz, [D]CHCl3, 1 88C): d ˆ 9.95 (OCH2CH2CH3),
22.90 (OCH2CH2CH3), 27.25, 28.17, 29.76 and 32.23 (2 Â [ArCH2Ar] and
ROCH2CH2CH2CH2OR'), 31.20, 31.26, 31.72 and 31.82 ((CH3)3C), 33.61,
33.82, 34.09 and 34.11 ((CH3)3C)), 75.32 and 77.24 (OCH2CH2CH3,
ROCH2(CH2)3OR' and RO(CH2)3CH2OR' (coincident)), 124.08, 125.05,
125.46, 125.56, 127.62, 131.82, 133.11, 135.53, 141.18, 143.51, 144.30, 146.72,
149.86, 150.82, 152.48 and 155.12 ppm (Ar); MS (ES): m/z: 1508
Acknowledgement
We thank the EPSRC for a studentship (PRAW) and the EPSRC and
University of Reading for funds for the crystallographic image plate
system.
[1] A. Ikeda, S. Shinkai, Chem. Rev. 1997, 97, 1713; P. D. Beer, P. A. Gale,
Z. Chen, Adv. Phys. Org. Chem. 1998, 31, 1; X. X. Zhang, R. M. Izatt,
J. S. Bradshaw, K. E. Krakowiak, Coord. Chem. Rev. 1998, 174, 179.
[2] D. A. Doyle, J. M. Cabral, R. A. Pfuetzner, A. Kuo, J. M. Gulbis, S. L.
Cohen, B. T. Chait, R. MacKinnon, Science 1998, 280, 69.
[3] G. W. Gokel, Chem. Commun. 2000, 1; J. de Mendoza, F. Cuevas, P.
Prados, E. S. Meadows, G. W. Gokel, Angew. Chem. 1998, 110, 1650;
Angew. Chem. Int. Ed. 1998, 37, 1534; Y. Tanake, Y. Kobuke, M.
Sokabe, Angew. Chem. 1995, 107, 71 7;Angew. Chem Int. Ed. Engl.
1995, 34, 693.
[4] For general reviews on calixarenes, see: C. D. Gutsche, Calixarenes,
The Royal Society of Chemistry, Cambridge, (UK), 1989; V. Bˆhmer,
Angew. Chem. 1995, 34, 785; Angew. Chem. Int. Ed. Engl. 1995, 34,
713.
[5] K. Iwamoto, S. Shinkai, J. Org. Chem. 1992, 116, 3102.
[6] A. Cadogan, D. Diamond, S. Cremin, M. A. McKervey, S. J. Harris,
Anal. Proc. 1991, 28, 1 3.
‡
‡
[M ‡NH4], 1513 [M ‡Na]; elemental analysis calcd (%) for
C102H136O8.1/2(CHCl3): C 79.43, H 8.88; found: C 79.49, H 9.00.
Computational methods: All simulations were carried out in the gas phase.
Molecular dynamics were carried out at a variety of temperatures by using
the Cerius2 software package[17] with parameters from the Universal Force
Field.[18] Partial atomic charges for atoms in the calix[4]semitube were
calculated by using the Gasteiger method implemented in Cerius2. The
atoms in the O-CH2-CH2-O links were given charges taken from ref. [19]
and the charges on the rest of the molecule were then adjusted to give a
neutral tube. The step size was 1.0 fs and the simulations were carried out
for 500 ps. The set of atomic positions were saved every 100 time steps
(0.10 ps) leading to trajectory files with 5000 conformations collected over
500 ps. A constant NVE thermostat with default parameters was used.
‡
Simulations were carried out with K at several different temperatures. In
[7] W. H. Chan, A. W. M. Lee, D. W. J. Kwong, W. L. Tam, K.-M Wang,
Analyst 1996, 121, 531.
these simulations the metals were given a ‡1charge and the charges on the
calix[4]semitube were kept unchanged.
[8] A. Casnati, A. Pochini, R. Ungaro, C. Bocchi, F. Ugozzoli, R. J. M.
Egberink, H. Struijk, R. Lugtenberg, F. de Jong, D. N. Reinhoudt,
Chem. Eur. J. 1996, 2, 436.
[9] S. E. Matthews, P. Schmitt, V. Felix, M. G. B. Drew, P. D. Beer, J. Am.
Chem. Soc. 2002, 124, 1341; S. E. Matthews, N. H. Rees, V. Felix,
M. G. B. Drew, P. D. Beer, Inorg. Chem. 2003, 42, 729.
[10] P. L. H. M. Cobben, R. J. M. Egberink, J. G. Bomer, P. Bergveld, W.
Verboom, D. N. Reinhoudt, J. Am. Chem. Soc. 1992, 114, 10573.
[11] S. E. Matthews, V. Felix, M. G. B. Drew, P. D. Beer, New J. Chem.
2001, 11, 1355.
[12] P. Schmitt, P. D. Beer, M. G. B. Drew, P. D. Sheen, Angew. Chem, 1997,
36, 1399; Angew. Chem. Int. Ed. Engl. 1997, 36, 1840.
[13] S. E. Matthews, V. Felix, M. G. B. Drew, P. D. Beer, Phys. Chem.
Chem. Phys. 2002, 4, 3849.
[14] C. D. Gutsche, M. Iqbal, Org. Synth. 1990, 68, 234.
[15] C. D. Gutsche, J. A. Levine, J. Am. Chem. Soc. 1982, 104, 2652.
[16] K. Iwamoto, K. Araki, S. Shinkai, Tetrahedron 1991, 47, 4325.
[17] Cerius2 v3.5, Molecular Simulations Inc., San Diego (U.S.A.), 1999.
[18] A. K. Rappe, C. J. Casewit, K. S. Colwell, W. A. Goddard III, W. M.
Skiff, J. Am. Chem. Soc. 1992, 114, 10024.
Stability constant determinations: As cation complexation/decomplexation
with 8 and 9 was slow on the NMR timescale, stability constants could be
assessed by direct integration of the host and complex resonances. For both
ligands, spectra were recorded in the presence of various equivalents of
each cation (in the range of 0 to 5 equivalents). A K value was calculated
for each spectrum recorded. The quoted stability constant for each ligand/
cation system is the average of these K values.
Crystallographic measurements: Crystal data for 4: C54H76O6, Mr ˆ 821.15,
monoclinic, P21/c, a ˆ 15.84(2), b ˆ 15.74(2), c ˆ 20.22(3) ä, b ˆ 94.94(1)8,
Vˆ 5021ä 3, Z ˆ 4, 1calcd ˆ 1.086 MgmÀ3. Intensity data were collected with
MoKa radiation using the MARresearch image plate system. The crystal was
positioned at 70 mm from the image plate. A total of 100 frames were
measured at 28 intervals with a counting time of 4 mins to give 8934 in-
dependent reflections. Data analysis was carried out with the XDS
program.[20] The structure was solved by using direct methods with the
SHELX86 program.[21] All non-hydrogen atoms were refined with
anisotropic thermal parameters. The hydrogen atoms were included in
geometric positions and given thermal parameters equivalent to 1.2 times
those of the atom to which they were attached. The structure was refined on
F 2 using SHELXL.[22] The final R values were R1 ˆ 0.0912 and wR2 ˆ
0.2564 for 4459 data with I > 2s(I).
[19] A. Varnek, G. Wipff, J. Comput. Chem. 1996, 17, 1520.
[20] W. Kabsch, J. Appl. Crystallogr. 1988, 21, 916.
Crystal data for 9 ¥ CHCl3: C83H97Cl3O8, Mr ˆ 1329.04, monoclinic, P21/c,
a ˆ 21.4788(1), b ˆ 16.8308(1), c ˆ 21.8377(1) ä, b ˆ 113.3191(3)8, Vˆ
7249.6 ä3, Z ˆ 4, 1calcd ˆ 1.218 MgmÀ3. Data was obtained by using MoKa
radiation on an Enraf-Nonius KappaCCD diffractometer. Crystals were
mounted on a glass fibre and cooled rapidly to 150 K in a stream of cold N2
using an Oxford Cryosystems CRYOSTREAM unit. Intensity data were
processed using the DENZO-SMN package.[23] Structures were solved by
direct methods by using the SIR92 program.[24] Full-matrix least-squares
refinement on F was carried out using the CRYSTALS program suite.[25]
Hydrogen atoms were positioned geometrically after each cycle of
refinement. A Chebychev polynomial weighting scheme was applied. The
[21] SHELX86: G. M. Sheldrick, Acta. Crystallogr. Sect. A 1990, 46, 467.
[22] G. M. Sheldrick, SHELXL, Program for Crystal Structure Refine-
ment, University of Gˆttingen, Gˆttingen (Germany), 1993.
[23] Z. Otwinowski, W. Minor in Methods Enzymol. 1997, 276, 307.
[24] A. Altomare, G. Cascarano, G. Giacovazzo, A. Guagliardi, M. C.
Burla, G. Ploidori, M. Camalli, J. Appl. Crystallogr. 1994, 27, 435.
[25] D. J. Watkin, C. K. Prout, J. R. Carruthers, P. W. Betteridge, R. I.
Cooper, CRYSTALS issue 11, Chemical Crystallography Laboratory,
Oxford (UK), 2001.
Received: October 21, 2002 [F4518]
2446
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2439 2446