Polyaza metacyclophanes as ditopic anion receptors

Article abstract of DOI:10.1039/b506828b

Five macrocyclic polyaza metacyclophanes L¹-L⁵ prepared by dipode coupling of the tosylated precursors have been studied. The basicity of the ligands has been measured potentiometrically and their ability to complex halides and perch

Full text of DOI:10.1039/b506828b

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Polyaza metacyclophanes as ditopic anion receptors
Christos A. Ilioudisa and Jonathan W. Steed*b
a Department of Chemistry, King’s College London, Strand, London, UK WC2R 2LS
b Department of Chemistry, University of Durham, South Road, Durham, UK DH1 3LE.
E-mail: jon.steed@durham.ac.uk; Fax: +44 (0)191 384 4737; Tel: +44 (0)191 334 2085
Received 16th May 2005, Accepted 17th June 2005
First published as an Advance Article on the web 20th July 2005
Five macrocyclic polyaza metacyclophanes L1–L5 prepared by dipode coupling of the tosylated precursors have been
studied. The basicity of the ligands has been measured potentiometrically and their ability to complex halides and
perchlorate has been studied in the solid state by X-ray crystallography. The results reveal that the ligands generally
act as ditopic halide receptors with even the largest, L5, being too small to envelop the anion. The ligand’s basicity
−
behaviour parallels that observed for related para-analogues. Despite the ready crystallisation of fluoride, HF2
,
chloride, bromide, iodide and triiodide salts in the solid state, there appears to be little affinity for halides in aqueous
solution in the pH range accessible via potentiometry. The results do give a detailed insight into the role of the aryl
ring in restricting the conformational flexibility of the ligands and, hence, the ability to chelate perching anions.
size and heteroatom placement and examine the protonation and
Introduction
anion binding behaviour of the compounds. Anion coordination
Azacorands (the nitrogen analogues of crown ethers)1 and
geometries are compared to work on model, acyclic systems.27,28
azaoxacorands are possibly the most well studied class of
macrocyclic polyamine receptors for anionic species. They have
Results and discussion
attracted the interest of researchers since the early 1980’s as these
substances are cyclic analogues of biological polyamines such
as histamine, spermidine and putrescine and could therefore
interact with biomolecules. Indeed, the molecular recognition
of nucleotides and other biological phosphates, along with
phosphoryl transfer catalysis, has been the theme for much
of the work regarding protonated azamacrocycles.2–9 In the
case of catalytic dephosphorylation of adenosine triphosphate,
it was found that the ring size plays a crucial role. The 21-
membered polyamine ring was found to be superior to larger
macrocycles. Moreover, rates of dephosphorylation were found
to increase with increasing number of nitrogen atoms in the
ring. In an effort to obtain further insight into the mechanism
of the dephosphorylation reaction, the crystal structure of
the pentahydrobromide salt of [21]N5O2 was determined.10 In
contrast to [22]N6·6HCl,11 the macrocycle ring crystallizes in
a boat form (which is also the case for the tetrachloride salt
of [21]N7,12 as well as [24]N6O213), maintaining an ellipsoidal
shape.14 However, no bromide is incorporated in the macrocyclic
cavity, although Br− does perch on one face of the macrocycle,
Synthesis
Macrocycles L1–L5 (of which only L5 is novel) were all prepared
by K2CO3 templated cyclisation of the appropriate linear
tosylamide precursors with 1,3-bis(bromomethyl)benzene, to
give tosylated macrocycles L1–L5 (R = SO2C6H4Me). A 20-
fold excess of base was required in order to obtain satisfactory
yields, which were in the region of 60–75%. Detosylation to
give L1–L5 (R = H) was achieved by either three day reflux
with aqueous HBr and phenol, or by a milder and faster
procedure involving 2 h reflux of the tosylated macrocycle in
a toluene suspension of metallic sodium with gradual additon
of ethanol (see Experimental). Ligand L4 has been prepared
independently during the course of this work in slightly better
yield by the same procedure.15 Ligands L1 and L3 have also
been prepared by a similar macrocyclisation procedure as ligand
systems for enzyme mimics29 and in CO2 activation and sulfate
binding,30,31 respectively. The interaction of L3 with DNA and
RNA has been studied.32 Ligand L2 has been prepared using
b-trimethylsilylethanesulfonamides.33 The cyclometallated N-
methyl analogue of L2 is also known as a ligand for Cu(III)
and Rh(III) from intramolecular CH activation.34,35
+
interacting with four NH2 units to form a square pyramid.
Incorporation of aromatic rings within the macrocycle in-
creases rigidity and hence potentially preorganisation. A range
of ortho-, meta- and para-azacyclophanes have been studied
for both their cation binding behaviour (as model sites for
metal-containing enzymes) and as anion hosts,15–20 as have
non-cyclophane azamacrocycles.21–23 For example, triproto-
nated 2,6,9,13-tetraaza[14]paracyclophane is an effective host
The tosylated precursor macrocycles L1–L5 (R
=
SO2C6H4Me) were all characterised by X-ray crystallography
in addition to the usual techniques (see Experimental). The
tosylated L1 and L4 include disordered dichloromethane in
pockets formed by aryl rings while the other compounds do
for anionic species as ATP4− and P2O7 at around neutral
4−
not co-crystallise with solvent. The principal crystal packing
36–38
pH.20 With the exception of the picrate complex of 2,5,9,12-
orthocyclophane,15 however, little structural information exists
on azacyclophane–anion complexes. As part of our own work
we have shown that bicyclic azametacyclophanes include anions
in a way that is limited by anion–p repulsion and thus exhibit
sharp selectivity for fluoride.24 A smaller analogue acts as
a highly basic proton host.25 We have also shown that a
large metacyclophane, 2,9,16-triaza[17]metacyclophane, acts as
a ditopic host for polyiodides with anion binding occuring on
each face of the macrocycle.26 We now report the extension of
this later work to a series of metacyclophanes of varying ring
=
interaction appears to be aryl CH · · · O S hydrogen bonds.
The conformations of tosylated L1–L4 are relatively open
while L5 is folded in on itself, minimising empty space. Therefore,
there is very little open space within the macrocyclic cavity in
any of the compounds. Full coordinates have been deposited
with the CCDC.†
† CCDC reference numbers 272024–272042. See http://dx.doi.org/
10.1039/b506828b for crystallographic data in CIF or other electronic
format.
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5
2 9 3 5
©

View Article Online
solutions of the ligands in dilute acid or by diffusion of acetone
into aqueous acid solutions. This resulted in the isolation of the
ten halide complexes of the protonated macrocycles listed below.
Ligand L2: (i) L2·HF·2H[HF2]·H2O, (ii) L2·3HCl and
(iii) L2·3HBr;
ligand L3: (i) L3·3HF·H[HF2]·3H2O and (ii) L3·4HBr;
ligand L4: (i) L4·4HCl·11 H2O;
ligand L5: (i) L5·5HCl·H22O, (ii) L5·5HCl·21 H2O, (iii) L5·5HBr·
2
2H2O and (iv) L5·4HI·HI3.
In addition the hydroperchlorate complex of L2, L2·3HClO4·
H2O was isolated as part of the purification of L2 and a mixed
hydroperchlorate–hydrobromide complex of L3, L3·3HClO4·
HBr·H2O was obtained by serendipitous incorporation of HBr
vapour into a crystallisation of the perchlorate salt. Unfor-
tunately no anion complexes were isolated for the smallest
macrocycle, L1.
Halide complexes. 2,5,8,11,14-Pentaaza[15]metacyclophane
(L5) is the largest macrocycle studied. The crystal struc-
tures of two pseudopolymorphic chloride salts, L5·5HCl·H2O
Free ligand structures
and L5·5HCl·21 H2O, were analyzed and in both cases L5
2
acts as a ditopic receptor for chloride with anions hydrogen
bonding to both faces of the macrocycle via charge assisted
NH+ · · · Cl− interactions. In the monohydrate the macrocycle
adopts a boat conformation, wrapping around the anion, which
is a conformation observed in other systems.11–13,40 This has the
Two of the detosylated free ligands, L1 and L3, were also
characterised by X-ray crystallography. The remaining free
ligands were all waxy solids or oils. Amine L1 forms offset
dimeric pairs comprised of two crystallographically independent
molecules (Fig. 1). These pairs then form further interactions
linking them in helical chains resulting in chiral crystal packing,
space group P212121. The structure of L3 exhibits an open
conformation for the macrocycle with the molecular cavity held
open by intramolecular NH · · · N interactions across the propyl
spacers, giving an S(6) motif.39 Macrocycles are linked together
by two hydrogen bonds on each side giving an infinite stack
(Fig. 1b). The open conformation of the free amine is a marked
contrast to the tosylated precursor in which there is very little
free cavity volume.
+
effect of preventing one NH2 group, N(5), from interaction
with the perching Cl− anions and it interacts with extra-
cavity Cl− instead. In the more hydrated form the macrocycle
adopts a flatter, more extended conformation in order to span
the portion of the lattice parallel to the included water. This
results in a shorter inter-halide separation across the macrocycle,
and an increased number of hydrogen bonds to the included
chloride (see Fig 3 caption for inter-halide distances). The
structure of the macrocycle in both forms is very different from
the folded conformation observed in the tosylated precursor.
The flatter conformation of the macrocycle in L5·5HCl·21 H2O
allows increased interaction to the included anions and d2eeper
penetration (Fig. 3). The added interactions apparently lower
the symmetry to give two independent macrocycles instead of
one in the monohydrate. In both cases, the macrocycles form
dimers sandwiching a pair of chloride anions (Fig. 2). This
basic structural type comprising a stack of dimers sandwiching
included anions is the fundamental building block of most of
the compounds studied and reflects the polyiodide structures
2−
observed for 2,9,16-triaza[17]metacyclophane in which I4 is
sandwiched as part of a related stack.26
A common feature revealed in these structures is the existence
of several CH · · · X− interactions, depending on the size of the
complexed anion and its proximity to the macrocyclic cavity.
For the crystal structure of L5·5HCl·H2O these interactions vary
˚
between C · · · Cl = 3.457(10) and 3.549(10) A. Interestingly, in
all the structures of polyammonium macrocyclic salts studied
there is an Ar–H · · · X− short contact between the hydrogen
of the 2 position of the aromatic ring and the anion nesting
at the bottom side of the macrocycle, C(1) · · · Cl(1)− 3.457(10)
5
˚
A for L ·5HCl·H2O, while the closer approach of the chloride
anions to the amino groups of the macrocyclic rings leads to
longer Ar–H · · · Cl− interactions in L5·5HCl·21 H2O C · · · Cl−:
2
˚
3.740(6) and 3.763(6) A.
In previous work on acyclic polyammonium species we
have adopted a simple arithmetic approach to the analysis of
the hydrogen bonded network.27,28 In the case of the two L5
hydrochloride hydrates there is a total of 12 or 15 acidic protons,
one or 2.5 water oxygen atom acceptors and five chloride ions.
Thus, each Cl− should accept 2 or, at the most, 3 (in the absence
of water–water interactions) non-bifurcated hydrogen bonds.
The observation of up to 5-coordinate chloride is evidence for
the operation of a macrocyclic enhancement of the binding of
the intra-cavity anions.
Fig. 1 (a) Offset hydrogen bonded dimers in the structure of L1;
(b) intra- and inter-molecular hydrogen bonding in L3. Two molecules
that form an infinite stack are shown in different shades (CH hydrogen
atoms omitted for clarity).
Anion complexes
Attempts were made to structurally characterise each ligand as a
series of hydrohalide salts obtained either by slow evaporation of
2 9 3 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5

View Article Online
Fig. 4 Hydrogen bonding in L5·5HBr·2H2O and L5·4HI·HI3. While
there are fewer NH+ · · · Br− interactions to the macrocycle compared to
the chloride and iodide cases, the environment around Br(2) is completed
by interactions to water and CH2 groups.
complexes, however, the macrocycle adopts a conformation in
which the aromatic ring is much more perpendicular to the plane
of the macrocycle, resulting in relatively short anion · · · centroid
˚
distances of 3.77 and 4.22 A for the bromide and iodide
complexes, respectively. These distances are longer than the
˚
almost constant figure of ca. 3.65 A observed for a series
of anion-encapsulating cryptand structures.24 Work by Kochi
et al.42 has shown that anions form stable charge transfer
complexes with a variety of electron deficient aromatic rings.
The crystal structures reveal that the anion sits in an offset
fashion at the edge of the aromatic rings rather than above
˚
the centroid, with anion–carbon distances as short as 2.93 A for
tetrachloro-o-quinone and Br−, compared to a van der Waals
˚
radius of 3.55 A. In the present case, however, there does not
seem to be a significant anion–p interaction. Surprisingly, and
despite their expected large coordination sphere, I(1) and I(2)
do not form any hydrogen bonds with other amine moieties,
apart from those belonging to the host macrocyclic ring. A short
Fig. 2 (a) Interaction of four independent anions with an independent
pair of macrocycles in L5·5HCl·2 1 H2O; (b) the more symmetrical
2
arrangement in L5·5HCl·H2O.
−
−
˚
Ar–H · · · I contact is also observed (C(1) · · · I(2) 3.820(13) A).
The I3− anion participates in a single NH2 · · · I− hydrogen bond.
+
−
However, there are several short contacts between I3 and the
CH2 moieties of the aliphatic chains.
Conceptually removing one binding site and ethylene linker
from L5 gives the smaller homologue L4, 2,5,8,11-tetraaza[12]-
metacyclophane. Only a hydrochloride salt L4·4HCl·11 H2O was
2
isolated for this material. The complex highlights the steric
constraints on the macrocycle induced by the arene ring, with
only three of the NH2+ groups being able to interact with the Cl−
anion perching on the opposite face of the cavity to the arene
+
ring, (Fig. 5). The remaining NH2 group is turned out from
the cavity and interacts with other chloride anions in the lattice.
This L4 complex is also unique in not acting as a ditopic halide
receptor. There is a second halide anion on the opposite face
of the molecule but it is situated above the aliphatic chain and
only forms long and very non-linear CH · · · Cl− interactions,
suggesting little significant interaction.
Fig. 3 Space filling plots of the Cl− binding in (a) L5·5HCl·2 1 H2O,
2
(b) L5·5HCl·H2O, (c) L5·5HBr·2H2O and (d) L5·4HI·HI3. The Cl− is
more tightly bound in L5·5HCl·2 1 H2O than in the monohydrate. The
2
macrocycle conformation is much more upright in the bromide and
iodide complexes. Inter-halide distances: (a) 3.51, (b) 3.85, (c) 5.01 and
˚
(d) 4.60 A.
The hydrobromide and hydroiodide/hydrotriiodide com-
plexes of L5 have also been isolated; L5·5HBr·2H2O
and L5·4HI·HI3 (Fig. 4). In both cases the macrocycle behaves
as a ditopic receptor as in the chloride complexes. The bromide
case is unusual in that the water, which forms a linear hydrogen
bonded chain41 through the structure, draws the anions out of
the cavity giving the largest inter-halide separation (Fig 3c, see
caption for inter-halide distances). In both bromide and iodide
Fig. 5 Partial structure of L4·4HCl·1 1 H2O showing the perching chlo-
ride anion interacting with three NH2 2 groups. Selected hydrogen bond
+
distances: N(2) · · · Cl(1) 3.148(5), N(1) · · · Cl(1) 3.123(5), N(3) · · · Cl(1)
˚
3.167(5) A.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5
2 9 3 7

View Article Online
The tetraaza[14]metacyclophane L3 is slightly larger than L4
and thus might be expected to be better able to wrap around
individual anions and exhibit a ditopic coordination mode as
observed for all structure of L5. Two complexes of L3 were
isolated, L3·3HF·H(HF2)·3H2O and L3·4HBr. The hydrobro-
mide salt exhibits a similar conformation to the iodide complex
of L5 with two anions bound one to either side of the receptor,
each via two NH · · · Br− hydrogen bonds in this case. Anion
Br(2) is actually three coordinate and forms a cyclic interaction
with an adjacent host molecule (Fig. 6), while Br(1) appears
to be genuinely two-coordinate with no other particularly close
interactions. The structure is of overall poor precision due to
persistent crystal twinning but the key structural features are
unambiguous.
F(4) and thence a discrete, well resolved water trimer in which
each of the water hydrogen atoms interacts with fluoride. On
the opposite, ‘lower’ face of the macrocycle there is a single
interaction to a fluorine atom of an HF2 anion based on F(2),
one of two symmetry independent hydrogen difluoride ions in
−
the structure (the other is based on F(5)). The F · · · F distances
˚
are 2.261(10) and 2.152(10) A, respectively, and in each case the
central proton is located on a two-fold rotation axis in C2/c.
−
The proton is thus positioned in the exact centre of the HF2
ions by symmetry, although minor unresolved disorder of the
proton cannot be ruled out in X-ray data. The F · · · F distance of
2.15 A is extremely short, even for HF2− ions. A CSD search43,44
˚
˚
reveals the normal range to be ca. 2.20–2.30 A. Examination
of the thermal ellipsoids for F(5) reveal them to be somewhat
enlarged, whereas those for F(2) are normal. It is likely therefore,
that the very short distance is an artefact of a minor unresolved
disorder. In terms of NH · · · F interactions, the N(1) · · · F(2)
interaction (to the HF2 ion as opposed to F−) is significantly
−
−
˚
˚
longer at 2.913(5) A than the distances to F (2.60–2.73 A)
and presumably reflects the lower negative charge density per
fluorine in HF2−, along with the fact that the proton involved
˚
on N(1) also interacts with F(1), N · · · F 2.734(5) A.
Turning to the smallest ligand studied crystallographi-
cally, 2,6,10-triaza[11]metacyclophane (L2) the hydrofluoride,
hydrochloride and hydrobromide salts, L2·HF·2H[HF2]·H2O,
L2·3HCl and L2·3HBr were isolated. Despite the small size of
this ligand’s cavity, the crystal structure of L2·HF·2H[HF2]·H2O
shows that two crystallographically equivalent fluoride anions
approach the ring from the ‘top’ and the ‘bottom’ side. There
+
is only one hydrogen bond between an NH2 moiety and the
fluoride anion at the ‘top’ side of the ring (N(2) · · · F(1)−), as well
as one hydrogen bond between an NH2+ moiety and the fluoride
anion at the ‘bottom’ side of the ring (N(1) · · · F(1)−, Fig. 8).
Numerous CH · · · F− contacts from methylene groups and one
from an aryl CH group complete the coordination sphere
of the ‘bottom’ fluoride anion, possibly because of the large
Fig. 6 Dimeric arrangement and ditopic Br− binding by proto-
nated L3 in the tetrahydrobromide salt. Hydrogen bond distances:
˚
N · · · Br(1)/Br(2) 3.279(16)–3.437(17) A.
negative charge density on the surface of the fluoride. C · · · F−
The hydrogen fluoride salt of L3 also exhibits ditopic anion
binding. Fluoride ion F(1) is held by three hydrogen bonds to
the upper face of the macrocycle (the same side as the aryl
ring) (Fig. 7). The macrocycle conformation is reminiscent of
+
˚
contacts range from 3.174(5) to 3.214(6) A. The NH2 group
N(2) interacts with a single fluoride and one water molecule
while N(1) interacts with F− and HF2 and N(3) binds to an
−
−
that observed for L4 (Fig. 5), in which one NH2 group turns
+
HF2 anion and foms a bifurcated interaction to water and
the second HF2−. There is thus no anion chelate effect. The
out from the macrocycle cavity and interacts in this case with
−
HF2 ions both exhibit slightly longer F · · · F distances (2.32–
3
˚
2.33 A) than in L ·3HF·H(HF2)·3H2O and the X-ray determined
H-atom positions suggest that the ions are slightly bent and
unsymmetrial.
Fig. 8 . Anion coordination environment for L2·HF·2H(HF2)·H2O. Se-
(a) Fluoride F(1) binding on the ‘upper’ face of L3
lected hydrogen bond lengths (A): N(1) · · · F(1) 2.625(5), N(1) · · · F(4)
˚
Fig.
7
in L3·3HF·H(HF2)·3H2O, including details of the water network and
2.652(4), F(3) · · · F(2) 2.330(4), N(2) · · · O(1) 2.814(4), N(2) · · · F1
2.569(5), N(3) · · · F(3) 2.601(4), N(3) · · · O(1) 2.847(5), N(3) · · · F(5)
2.939(4), F(4) · · · F(5) 2.322(4).
−
(b) HF2 inclusion, sandwiched between the lower faces of a pair of
macrocycles.
2 9 3 8
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5

View Article Online
In the case of L2·3HCl, the large size of chloride in comparison
with the small size of the cavity result in no anion being at
the ‘top’ side of the macrocyclic ring. Oddly, however, the
aromatic group of another ring approaches the macrocyclic
ring, thus giving rise to a C–H · · · p interaction45–47 shown in
were isolated as part of this study (vide supra). The imme-
diate environment of the macrocycle in the L3 mixed bro-
mide/perchlorate salt (Fig. 11) is highly reminiscent of the
ditopic bromide salt L3·4HBr (cf. Fig. 6), with one bromide
anion occupying one face of the macrocycle and a perchlorate
anion on the other. The two anions are held to the host by both
NH · · · anion and CH · · · anion interactions. The perchlorate
anion occupies the more sterically hindered face near the aryl
ring, possibly because of the smaller size of the anion oxygen
atom that makes the closest approach, in comparison to Br−.
˚
Fig. 9 (C(11) · · · centroid: 3.89 A). There is a chloride at the
‘bottom’ side of the ring accepting only one hydrogen bond
+
−
˚
from an NH2 moiety (N(2) · · · Cl(2) : 3.183 A). This chloride is
involved in three weak C–H · · · Cl− interactions with distances
˚
typical for this type of binding (3.645(2)–3.794(2)A for the
corresponding C · · · Cl(2) distances). Fig. 9 also shows the
overall NH · · · chloride connectivity.
Fig. 11 Macrocycle–anion interactions in the mixed salt L3·3HClO4·
HBr·H2O.
The structure of L2·3HClO4·H2O is shown in Fig. 12. The
perchlorate anions above and below the macrocycle are held
only by CH · · · anion interactions, with distances similar to
those reported in the literature.38 For example, O(12) comes
in close contact with three C–H protons that belong to the
macrocyclic system. One of the interactions is rather short
Fig. 9 Hydrogen bond network in the proximity of the ligand for the
crystal structure of L2·3HCl.
˚
(C(11) · · · O(12): 3.304(9) A) but the other two are longer
˚
˚
(C(1) · · · O(12): 3.481(10) A, C(9) · · · O(12): 3.654(10) A). All
+
of the NH2 groups are directed away from the cavity. Two NH
For L2·3HBr there is a bromide anion at the bottom of the
macrocyclic cavity forming an NH · · · Br− hydrogen bond as
well as four C–H · · · Br− weak interactions, more than those
seen for the fluoride and the chloride complexes apparently due
to the larger size of the bromide anion. A very short NH · · · Br−
hydrogen bond is also formed between an amine moiety and the
bromide anion at the ‘top’ of the macrocyclic ring (N(2) · · · Br(2):
groups interact with two symmetry equivalent water molecules
which, in turn, hydrogen bond to the perchlorate anion situated
over the cavity on the same side as the aryl ring. The remaining
NH units all interact directly with perchlorate anions, forming
both non-bifurcated and bifurcated hydrogen bonds, Fig. 12.
˚
3.216(4) A). The small size of the cavity as well as the fact that the
aromatic ring ‘faces’ the bromide anion (centroid–Br: 4.200(5)
˚
A) disfavour any further interactions between the bromide and
the macrocycle. The macrocycle cavity appears to be closed by
an intramolecular C–H · · · p interaction. A somewhat longer
contact is also observed in the chloride case (C(12) · · · centroid:
˚
3.819(7) A, Fig. 10).
Fig. 12 View of L2·3HClO4·H2O showing that NH · · · anion interac-
tions are directed away from the cavity. Anions situated above and below
the macrocycle are held solely by CH · · · anion interactions and one
hydrogen bond to the included water molecule.
Potentiometric measurements
Fig. 10 Hydrogen bond network in the proximity of the ligand
for the crystal structure of L2·3HCl showing intramolecular CH · · · p
interaction.
The reaction involving the transfer of a proton from one
atom to another has been described as ‘the most general and
important reaction in chemistry’.48 The complexation properties
of polyazamacrocycles towards anionic and cationic species
depend largely on their basicity behaviour. Hydrogen bonding
Perchlorate complexes. In addition to pure halide complexes,
the perchlorate salts L3·3HClO4·HBr·H2O and L2·3HClO4·H2O
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5
2 9 3 9

View Article Online
Table 1 Logarithms of the stepwise protonation constants for the meta-
cyclophanes synthesized. Conditions: 0.001 M ligand, 0.01 M HCl,
0.1 M NaNO3, 293 K
this compound and it was attributed to the effects of a particular
solvation promoted by the arene ring.52 The third protonation
constant is considerably higher than the corresponding log K
value of L1 that also has ethylene units in its spacer. This
is because the third proton does not enter the macrocycle
between two nitrogen atoms already protonated but between
one nitrogen that is protonated and one nitrogen that is not.
The last protonation constant of L4 is very low, as protonation
takes place between amine moieties already protonated.
Log K1
Log K2
Log K3
Log K4
Log K5
R log Ki
L1
9.58(14) 7.69(11) 3.51(13)
—
—
2.71(4)
3.49(4)
7.35(4)
—
—
—
—
20.78(38)
23.92(10)
26.45(16)
29.44(16)
L2 10.09(4)
8.74(3)
8.65(4)
9.02(4)
9.18(6)
6.55(4)
5.76(3)
7.13(3)
8.81(4)
L4
L3
9.33(5)
9.80(5)
L5 10.97(3)
4.10(5) 40.41(22)
Moving to L3, the trend is very similar to that observed
for L4. This time however, the existence of two propylene units
increases the basicity constants in comparison with L4. The
impact of the propylene units is particularly shown in the third
protonation constant which is larger by 1.37 logarithmic units as
compared with that of L4. It is clear that the introduction of the
propylene units makes protonation of the middle amine groups
much more favourable. The present data for L3 are comparable
to that obtained for the compound in the presence of NMe4Cl.30
Even larger basicity constants were found for L5 because
of the large size of the macrocycle. It is interesting, however,
that the basicity constants of this macrocycle are larger than
the basicity constants of its ‘parent’ amine, TEP. It is also
remarkable that the fourth protonation constant is quite large
despite that the protonation takes place next to at least one
nitrogen atom that is already protonated. Exactly the same trend
was found in the para-analogue of L5 and implies that the fourth
protonation step for this species involves a reorganization of
the protonation sites within the molecule such that the middle
nitrogen remains unprotonated and each protonated nitrogen
has only one adjacent protonated site.53
in anion coordination, although not clearly understood, seems
to be crucial in water.49,50 Also, the positive charge imparted
to protonated macrocyclic polyamines in aqueous solutions
is obviously an important factor in stabilizing a host–anion
guest complexes. Therefore, the basicity of the monocyclic
metacyclophanes was studied in aqueous solutions, in the
presence of 0.1 M NaNO3 as supporting electrolyte, Table 1.
The metacyclophanes reported herein follow trends similar to
those observed for 1 : 1 and 2 : 2 polyazacyclophanes.51 As
observed for paracyclophanes, the overall basicity of these
compounds increases in an almost linear fashion as a function
of the number of atoms in the polyamine chain bridging the
arene unit.52 The stepwise basicity constants depend on the
number of amine moieties present in the macrocycle, as well as
on the aliphatic spacers between the amine moieties. In general,
minimum electrostatic repulsion between charges of the same
sign explains the protonation trends observed.
Of interest are the marked differences in the stepwise
protonation constants between the metacyclophanes studied
and the ‘parent’ linear chain aliphatic amines. These differ-
ences imply a greater conformational freedom of the ‘par-
ent’ amines in comparison with the macrocyclic compounds
and the fact that primary rather than secondary amines are
involved in some cases. Indeed, with the exception of the
pair L5/tetraethylenepentammine, the overall basicities, as well
as each stepwise basicity constant, for the macrocycles studied
are lower than the corresponding basicity of the ‘parent’
aliphatic amine. Similar trends have also been observed for the
paracyclophane analogues.52 For example, the para-analogue
of L5 displays larger basicity constants (10.68, 9.29, 8.66, 7.23,
3.83, R log Ki = 39.7) than its parent amine, tetraethylenepen-
tammine, which is not much different to those observed for L5.
It is also of practical interest to inspect the protonation state of
these molecules at neutral pH. For the first three macrocycles,
the diprotonated form is the predominant species, with relative
concentrations at around 90% or more. For the larger and more
basic species L3 and L5 at neutral pH, the triprotonated ligands
are the predominant species, with relative concentrations 60–
70%. This is indicative of that fact that, despite the strong
basicity of these species, it is rather difficult to protonate two
adjacent amine groups.51,53
The basicity constants of L5 were redetermined in the presence
of TsOH/TsONa (Ts = MeC6H4SO2; values obtained log K1 =
9.63(7), log K2 = 8.19(5), log K3 = 4.65(6) and log (K4 × K5) =
4.91(42)). There is a remarkable difference between the values
observed in this medium and with HNO3/NaNO3, apparently
because the basicity of L5 is enhanced as a result of stronger
anion binding in the presence of NO3−. In order to probe the
macrocycles’ anion binding in aqueous solution, pH titrations
were conducted in this medium in the presence of NaF and
NaCl but no difference was observed with the results of the
pH titration in the presence of TsOH/TsONa only. A possible
explanation for this result could be the fact that pH titrations
is the least sensitive method for the determination of binding
constants, thus making the measurement of low log K values
difficult. It is also likely that the affinity of these macrocycles for
halides is insignificant for pH >2.5. What is observed in the solid
state by means of X-ray crystallography does not necessarily
reflect what happens in solution. All polyammonium halide
crystals were grown from a very acidic solution, whereas pH
titrations are considered to be accurate only for 2.5 < pH <
11. A similar effect is observed in the crystal structures of
polyammonium salts of cryptand analogues.24
Starting with L1, two large stepwise protonation constants
are observed and one much lower. The two NH moieties next
to the aromatic ring are protonated first. The last protonation
steps is a lot more unfavourable and involves the binding of a
proton on a nitrogen atom between two amine moieties that are
already protonated. The situation is different in L2, however,
which is much more basic because of reduced repulsion between
protonated nitrogen atoms due to the presence of the propylenic
chains. The increased log K1 value may relate to inductive effect
and decreased intramolecular hydrogen bonding. These simple
trends are also reflected in the marked difference to the overall
basicity of L1.
Conclusions
A total of twelve X-ray structure determinations of hydrohalide
and perchlorate salts of ligands L2–L5 have been undertaken.
In the case of the larger macrocycles, particularly L5, ditopic
binding is normal, with anions positioned both above and
below the macrocyclic cavity engaging in multiple chelate
NH · · · anion interactions with the same macrocycle. As the
macrocycle gets smaller the tendency towards ditopic binding
and towards anion chelation diminish. They are replaced by
CH · · · p and intermolecular NH · · · anion interactions. Even
the largest macrocycle, L5, is insufficiently large to encapsulate
halides in a 1 : 1 fashion, as observed for fluoride complexes
of sapphyrin, for example.54–56 Macrocyclic meta-azaphanes
studied by pH potentiometry possess a protonation behaviour
not different from that observed for macrocyclic para-azaphanes
and related compounds.51 Unfortunately, we have not been able
The same considerations explain the basicity constants of L4.
Again, the first two protonation constants are attributed to
the nitrogen atoms close to the arene ring. This time the first
protonation constant is particularly low compared to the other
macrocycles. A similar trend was found for the para-analogue of
2 9 4 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5

View Article Online
to detect the binding constants of any halide species probably
due to the low affinity of these species for halides at pH >2.5.
In the crystal structue of L3·3HF·H[FHF]·3H2O the HF2
−
ion based on F(5) proved only partially occupied (58%) with
disordered atom F(5A) substituted into the H-atom special
position in 42% of unit cells. Thus the true formula of the crystal
studied is L3·3.21HF·0.79H[FHF]·3H2O.
Experimental
In the crystal structure of L5·5HCl·H2O atom O(1) that
belongs to a partially occupied solvated water molecule site,
was refined isotropically.
Instrumental
1
NMR Spectra. H and 13C NMR spectra were measured with
a Bruker Avance NMR spectrometer, operating at 360 MHz
and 90 MHz, respectively, and the chemical shifts are reported
in ppm relative to tetramethylsilane. Fast atom bombardment
(low resolution) mass spectra were obtained with a Kratos
MS 890 Mass Spectometer. High resolution mass spectra
were obtained with a Bruker Apex III Mass Spectrometer by
electrospray ionization. Elemental analysis for carbon, hydrogen
and nitrogen was carried out by the Elemental Analysis Service
at the London Metropolitan University.
Syntheses
Materials were obtained from standard commercial sources.
N,Nꢀ,Nꢀꢀ-Tritosyl-1,5,9-triazanonane. Dipropylenetriamine
(5.25 g, 40 mmol), and K2CO3 (11.06 g, 80 mmol) were
suspended in water (600 mL) at 80 ◦C. To this mixture, tosyl
chloride (23.00 g, 121 mmol) was added in batches over a period
of 1 h. After addition was complete, vigorous stirring and
heating were continued overnight. The tosylated macrocycle
precipitated as a white solid which was filtered under a reduced
pressure, washed thoroughly with 500 mL of water and quickly
with 50 mL of methanol and dried under a high vacuum to
Potentiometric titrations
All potentiometric titrations were performed at rt, using
carbonate-free NaOH. A Titrino model 736 GP along with a
Metrohm combined glass electrode was used. The protonation
constants were determined from titrations of an approximately
10−3 M ligand solution containing an excess of HCl or HNO3 or
TsOH (0.01 M) in the presence of NaNO3, Me4NCl or TsONa
to maintain ionic strength at 0.1 M. The range of accurate pH
measurements was considered to be 2.5–11. Stability constants
were calculated with the program HYPERQUAD.57
1
give a white solid (16.16 g, 27.2 mmol, 68% yield). H NMR
(CDCl3): 7.72 (d, J = 8.2, 4H), 7.63 (d, J = 8.2, 2H), 7.29
(d, J = 8.2, 4H), 7.27 (d, J = 8.2, 4H), 3.10 (t, J = 6.7, 4H),
2.95 (pt, J = 6.2, 4H), 2.42 (s, 3H), 2.41 (s, 6H), 1.71 (m, J =
6.4, 4H); 13C NMR (CDCl3): 144.16, 144.83, 137.15, 135.77,
130.31, 130.16, 127.48, 127.41, 47.18, 40.46, 29.60, 21.96; MS
m/z (FAB) 594 ([M + H]+); anal. calcd for C27H35S3O6N3: C,
54.61%; H, 5.94%; N, 7.08%. Found: C, 54.75%; H, 5.83%; N,
6.94%.
N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ-Tetratosyl-1,5,8,12-tetraazadodecane. 1,2-Bis-
(3-aminopropyl)diaminoethane (6.97 g, 40 mmol), tosyl
chloride (23.00 g, 121 mmol) and K2CO3 (11.06 g, 80 mmol)
were reacted by following the same procedure as that for
the synthesis of N,Nꢀ,Nꢀꢀ-tritosyl-1,5,9-triazanonane. After
work-up, the product was isolated as a white solid (23.41 g,
29.6 mmol, 74% yield). 1H NMR (DMSO-d6): 7.74 (d, J = 8.2,
4H), 7.68 (d, J = 8.2, 4H), 7.34 (d, J = 8.2, 4H), 7.31 (d, J =
8.2, 4H), 5.35 (b, 2H), 3.24 (s, 4H), 3.17 (t, J = 6.8, 4H), 2.99
X-Ray crystallography
Crystal data and data collection parameters are summarised
in Table 1. Crystals were mounted using silicone grease on a
thin glass fibre. All crystallographic measurements were carried
out with a Nonius KappaCCD diffractometer, equipped with
graphite monochromated Mo–Ka radiation using wide φ and
x-scans. Data collection temperature was 120 K, maintained
using an Oxford Cryosystem low temperature device. Integration
was carried out by the Denzo-SMN package.58 Data sets were
corrected for Lorentz and polarization effects and for the effects
of absorption (Scalepack58) and crystal decay where appropriate.
Structures were solved using the direct methods option of
SHELXS-9759 and developed using conventional alternating cy-
cles of least-squares refinement (SHELXL-97)60 and difference
Fourier synthesis with the aid of the program XSeed.61 In all
cases non-hydrogen atoms were refined anisotropically except
for some disordered, while C–H hydrogen atoms were fixed in
idealised positions and allowed to ride on the atom to which they
were attached. Hydrogen atom thermal parameters were tied to
those of the atom to which they were attached. Where possible,
non C–H hydrogen atoms were located experimentally and
their positional and isotropic displacement parameters refined.
Otherwise a riding model was adopted. All calculations were
carried out on an IBM–PC compatible personal computer. Some
structural data is not of high precision due to frequent crystal
twinning or generally poor crystal quality, but does establish
the gross structural features of the compounds. Specific issues
relating to each refinement are detailed below.
1
(b, 4H), 2.46 (s, 6H), 2.43 (s, 6H), 2.26 (pt, 4H); 13C-{ H} NMR
(CDCl3): 144.32, 143.83, 137.12, 135.47, 130.39, 130.16, 127.62,
127.41, 49.29, 47.52, 40.36, 29.39, 21.96, 21.92; MS m/z (FAB)
791 ([M + H]+); anal. cCalcd for C36H46S4O8N4: C, 54.66%; H,
5.86%; N, 7.08%. Found: C, 54.60%; H, 5.85%; N, 6.88%.
N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀꢀ-Pentatosyl-1,4,7,10,13-pentaazadecatriane.
Tosyl chloride (40 g, 0.211 mol), H2O (40 mL) and diethyl ether
were stirred and cooled to 0 ◦C in an ice bath. To this mixture,
a solution of tetraethylenepentamine (7.97 g, 42 mmol) and
NaOH (10 g, 0.25 mol) in H2O (80 mL) was added dropwise over
a period of 1 h. The reaction mixture was stirred further for 3 h
at rt. The precipitate was filtered and then washed with diethyl
ether and water. Recrystallization from hot CHCl3–MeOH
afforded the product as white powder (16.55 g, 17 mmol, 41%
1
yield). H NMR (CDCl3): 7.79 (d, J = 8.3, 2H), 7.75 (d, J =
8.3, 4H), 7.72 (d, J = 8.3, 4H), 7.38 (d, J = 8.3, 2H), 7.34 (d,
J = 8.3, 4H), 7.29 (d, J = 8.3, 4H), 5.54 (b, 2H), 3.38 (b, 8H),
1
3.20 (b, 8H), 2.48 (s, 3H), 2.45 (s, 6H), 2.42 (s, 6H); 13C-{ H}
NMR (CDCl3): 144.41, 144.34, 143.87, 136.92, 135.13, 134.88,
130.46, 130.38, 130.16, 127.92, 127.86, 127.55; MS m/z (FAB)
961 ([M + H]+); anal. calcd for C43H53S5O10N5: C, 53.79%; H,
5.56%; N, 7.29%. Found: C, 53.85%; H, 5.45%; N, 7.18%.
In the crystal structure of L1–Ts3 a disordered chloroform
molecule was refined by assigning a site occupancy factor of
0.333 to the atoms C(1S), Cl(1S), Cl(2S), Cl(3S).
In the crystal structure of L4–Ts4 there are two disordered
dichloromethane molecules. In one of them, C(43), C(43A),
Cl(2) and Cl(2A) were assigned a site occupancy factor of 0.5
each. In the other, Cl(3), Cl(3A), Cl(3B), Cl(4), Cl(4A), and
Cl(4B) were assigned a site occupancy factor of 0.333 each.
In the crystal structure of L3 the atoms H(31A), H(31B),
H(41A), and H(41B) were assigned with an occupancy factor
of 0.5 each.
N,Nꢀ,Nꢀꢀ-Tritosyl-2,5,8-triaza[9]metacyclophane
(L1–Ts3).
N,Nꢀ,Nꢀꢀ-Tritosyl-1,4,7-triazaheptane62 (10.70 g, 18.9 mmol)
and K2CO3 (52.24 g, 378 mmol) were suspended in refluxing
CH3CN (700 mL). To this mixture, a solution of 1,3-
bis(bromomethyl)-benzene (5.00 g, 18.9 mmol) in CH3CN
(700 mL) was added dropwise. After the addition was complete,
the suspension was refluxed and stirred for 36 h and then
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5
2 9 4 1

View Article Online
filtered. The solvent was removed and the crude product was
purified by column chromatography on silica (toluene–AcOEt,
85 : 15). The product was obtained as a white solid (9.36 g,
14.0 mmol, 74% yield). Suitable crystals for X-ray analysis were
obtained as colourless blocks after slow diffusion of hexane
into a solution of the product in chloroform for a few days. 1H
NMR (CDCl3): 7.72 (d, J = 8.2, 4H), 7.64 (d, J = 8.2, 2H),
7.14–7.40 (m, 4H), 7.35 (d, J = 8.2, 4H), 7.28 (d, J = 8.2, 2H),
4.21 (s, 4H), 3.04 (t, 4H), 2.58 (t, 4H), 2.46 (s, 6H), 2.42 (s,
K2CO3 (52.24 g, 378 mmol) and 1,3-bis(bromomethyl)-benzene
(5.00 g, 18.9 mmol) yielded the product as a white solid (8.77 g,
8.25 mmol, 69% yield). Suitable crystals for X-ray analysis were
obtained as colorless blocks in the same manner as for L1–Ts3.
1H NMR (CDCl3): 7.76 (d, J = 8.3, 4H), 7.70 (d, J = 8.3, 4H),
7.64 (d, J = 8.3, 2H), 7.48 (d, J = 8.2, 2H), 7.36 (d, J = 8.3,
4H), 7.34 (d, J = 8.3, 4H), 7.24 (m, 1H), 7.17 (d, J = 8.5, 2H),
7.04 (s, 1H), 4.25 (s, 4H), 3.18 (b, 4H), 3.13 (b, 4H), 3.02 (b,
1
8H), 2.47 (s, 6H), 2.46 (s, 3H), 2.45 (s, 6H); 13C-{ H} NMR
1
3H); 13C-{ H} NMR (CDCl3): 142.55, 142.35, 134.93, 134.45,
(CDCl3): 144.49, 144.21, 144.14, 136.95, 135.89, 135.31, 134.14,
130.32, 130.01, 128.68, 128.60, 128.00, 127.89, 127.82, 53.91,
51.17, 50.26, 49.99, 47.75, 21.97; MS m/z (FAB) 1063 ([M +
H]+); anal. calcd for C51H59S5O10N5: C, 57.66%; H, 5.60%; N,
6.59%. Found: C, 57.81%; H, 5.73%; N, 6.70%.
133.83, 129.87, 129.52, 129.07, 128.71, 128.60, 125.96, 125.88,
53.19, 52.22, 49.17, 46.06, 20.32, 20.29; MS m/z (FAB) 668
([M + H]+); anal. calcd for C33H37S3O6N3: C, 59.37%; H, 5.59%;
N, 6.29%. Found: C, 59.50%; H, 5.47%; N, 6.26%.
N,Nꢀ,Nꢀꢀ-Tritosyl-2,6,10-triaza[11]metacyclophane (L2–Ts3).
By following a procedure similar to that described for
the synthesis of L1–Ts3, N,Nꢀ,Nꢀꢀ-tritosyl-1,5,9-triazanonane
(11.23 g, 18.9 mmol), K2CO3 (52.24 g, 378 mmol) and 1,3-
bis(bromomethyl)-benzene (5.00 g, 18.9 mmol) yielded the
product as a white solid (9.87 g, 14.2 mmol, 75% yield). Suitable
crystals for X-ray analysis were obtained as colorless blocks in
2,6,9,13-Tetraaza[14]metacyclophane (L3). A mixture of to-
sylated amine L3–Ts3 (2.00 g, 2.24 mmol), phenol (4.0 g,
42.50 mmol) and 60 mL of 48% aqueous HBr was stirred and
heated to reflux for 72 h. After cooling to rt, the mixture was
repeatedly washed with chloroform. The aqueous phase was
cooled to 0 ◦C and sodium hydroxide was added slowly until the
pH of the solution became at least 12. The product was extracted
in chloroform which was removed under a high vacuum to afford
the free amine as a waxy solid (0.32 g, 1.16 mmol, 52% yield).
Suitable cystals for X-ray analysis of the fluoride, chloride and
bromide salts of L3 were obtained in two ways: Either from a
solution of this material in the corresponding diluted acid after a
few days by slow evaporation or by slow diffusion of acetone into
a concentrated solution of the macrocycle in the corresponding
diluted acid. The species L3·3HClO4·HBr·H2O was synthesized
after a mixture of L3 and perchloric acid was accidentally treated
with fumes of hydrobromic acid inside a fume cubpoard (for
crystal data, see Table 2). Suitable cystals for X-ray analysis
of the free amine L3 were obtained by slow evaporation of a
concentrated solution of the compound in chloroform. 1H NMR
(CDCl3): 7.34 (s, 1H), 7.18 (t, J = 7.4, 1H), 7.04 (d, J = 7.6,
2H), 3.73 (s, 4H), 2.62–2.66 (m, 4H), 2.61 (s, 4H), 1.73 (s, b, 4H),
1
the same manner as for L1–Ts3. H NMR (CDCl3): 7.72 (d,
J = 8.2, 4H), 7.58 (d, J = 8.2, 2H), 7.14–7.47 (m, 4H), 7.35
(d, J = 8.2, 4H), 7.16 (d, J = 8.2, 2H), 4.18 (s, 4H), 3.07 (t,
J = 6.8, 4H), 2.86 (t, J = 7.3, 4H), 2.45 (s, 6H), 2.41 (s, 3H),
1
1.35 (m, 4H); 13C-{ H} NMR (CDCl3): 143.87, 143.53, 137.57,
135.59, 130.12, 129.93, 129.76, 129.71, 129.69, 129.24, 127.41,
127.27, 54.46, 47.97, 47.94, 29.34, 21.74, 21.68; MS m/z (FAB)
696 ([M + H]+); anal. calcd for C35H36S3O6N3: C, 60.42%; H,
5.94%; N, 6.04%. Found: C, 60.37%; H, 5.88%; N, 5.94%.
N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ-Tetratosyl-2,5,8,11-tetraaza[12]metacyclophane
(L4–Ts4). By following a procedure similar to that described
for the synthesis of L1–Ts3, N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ-tetratosyl-1,4,7,10-
tetraazadecane62 (14.42 g, 18.9 mmol), K2CO3 (52.24 g,
378 mmol) and 1,3-bis(bromomethyl)benzene (5.00 g,
18.9 mmol) yielded the product as a white solid (11.11 g,
12.9 mmol, 68% yield). Suitable crystals for X-ray analysis were
obtained as colorless blocks in the same manner as for L1–Ts3.
1H NMR (CDCl3): 7.75 (d, J = 8.2, 4H), 7.69 (d, J = 8.2, 4H),
7.11–7.38 (m, 4H), 7.37 (d, J = 8.2, 4H), 7.31 (d, J = 8.2, 4H),
4.13 (s, 4H), 2.98 (t, J = 5.3, 4H), 2.89 (t, J = 5.6, 4H), 2.68
1
1.60–1.67 (m, 4H); 13C-{ H} NMR (CDCl3): 140.78, 128.79,
127.70, 127.30, 54.48, 50.38, 50.00, 48.79, 29.87; HRMS calcd
for C16H29N4 [M]+ 277.2387, found 277.2363. Ligand L3 has
previously been prepared ina similar way.
2,5,8,11,14-Pentaaza[15]metacyclophane (L5). By following
a procedure similar to that described for the synthesis of L3, L5–
Ts5 (2.00 g, 1.88 mmol), phenol (4.0 g, 42.50 mmol) and 60 mL of
48% aqueous HBr afforded the product as a waxy solid (0.43 g,
1.47 mmol, 78% yield). Suitable cystals for X-ray analysis of
the chloride (two structures), bromide and iodide salts of the
1
(s, 4H), 2.47 (s, 6H), 2.44 (s, 6H); 13C-{ H} NMR (CDCl3):
144.37, 144.12, 137.03, 135.90, 134.80, 130.40, 130.25, 129.99,
129.31, 127.88, 127.81, 54.82, 50.15, 49.21, 48.28, 21.98, 21.97;
MS m/z (FAB) 865 ([M + H]+); anal. calcd for C42H48S4O8N4:
C, 58.31%; H, 5.59%; N, 6.48%. Found: C, 58.39%; H, 5.54%;
N, 6.43%.
1
macrocycle were obtained in a similar manner as for L3. H
NMR (CDCl3): 7.54 (s, 1H), 7.12 (t, J = 7.4, 1H), 7.00 (d, J =
N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ-Tetratosyl-2,6,9,13-tetraaza[14]metacyclophane
(L3–Ts4). By following a procedure similar to that described
for the synthesis of L1–Ts3, N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ-tetratosyl-1,5,8,12-
tetraazadodecane (14.95 g, 18.9 mmol), K2CO3 (52.24 g,
378 mmol) and 1,3-bis(bromomethyl)-benzene (5.00 g,
18.9 mmol) yielded the product as a white solid (11.98 g,
13.42 mmol, 71% yield). Suitable crystals for X-ray analysis
were obtained as colorless blocks in the same manner as
1
7.3, 2H), 3.73 (s, 4H), 2.65–2.76 (m, 16H), 2.32 (b, 5H); 13C-{ H}
NMR (CDCl3): 141.16, 128.26, 127.23, 126.84, 53.96, 49.65,
49.53, 49.52, 48.99; HRMS calcd for C16H30N5 [M]+ 292.2496,
found 292.2491.
2,5,8-Triaza[9]metacyclophane (L1). By following a proce-
dure similar to that described for the synthesis of L3, L1–Ts3
(2.00 g, 2.99 mmol), phenol (4.0 g, 42.50 mmol) and 60 mL of
48% aqueous HBr afforded the product as a waxy solid (0.40 g,
1.94 mmol, 65% yield). Suitable crystals for X-ray analysis of
this material were obtained by slow diffusion of n-hexane into
a solution of the compound in chloroform after a few days. 1H
NMR (CDCl3): 8.20 (s, 1H), 7.08 (t, J = 7.5, 1H), 6.95 (d, J =
7.3, 2H), 3.83 (s, 4H), 2.66 (t, J = 5.1, 4H), 2.45 (s, b, 3H), 2.06 (s,
1
for L1–Ts3. H NMR (CDCl3): 7.72 (d, J = 8.2, 4H), 7.65 (d,
J = 8.2, 4H), 7.11–7.54 (m, 4H), 7.37 (d, J = 8.2, 4H), 7.33
(d, J = 8.2, 4H), 4.17 (s, 4H), 3.11 (t, J = 7.2, 4H), 2.94 (t,
J = 7.0, 4H), 2.66 (s, 4H), 2.46 (s, 6H), 2.44 (s, 6H), 1.41 (m,
1
4H); 13C-{ H} NMR (CDCl3): 144.24, 144.10, 138.09, 135.65,
135.05, 130.45, 130.27, 129.46, 128.58, 127.92, 127.72, 54.68,
48.66, 48.59, 48.50, 29.50, 22.03; MS m/z (FAB) 893 ([M +
H]+); anal. calcd for C44H52S4O8N4: C, 59.17%; H, 5.87%; N,
6.27%. Found: C, 59.06%; H, 5.63%; N, 6.05%.
1
b, 4H); 13C-{ H} NMR (CDCl3): 142.28, 127.60, 126.31, 125.80,
53.40, 48.27, 47.67; HRMS calcd for C12H19N3 [M]+ 206.1652,
found 206.1654. Ligand L1 has previously been prepared by a
very similar method.29
N,Nꢀ,Nꢀꢀ,NꢀꢀꢀNꢀꢀꢀꢀ -Pentatosyl-2,5,8,11,14-pentaaza[15]meta-
cyclophane (L5–Ts5). By following a procedure similar to
that described for the synthesis of L1–Ts3, N,Nꢀ,Nꢀꢀ,Nꢀꢀꢀ,Nꢀꢀꢀꢀ-
pentatosyl-1,4,7,10,13-pentaazadecatriane (18.16 g, 18.9 mmol),
2,6,10-Triaza[11]metacyclophane (L2). By following a pro-
cedure similar to that described for the synthesis of L3, L2–Ts3
(2.00 g, 2.87 mmol), phenol (4.0 g, 42.50 mmol) and 60 mL of
2 9 4 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5

View Article Online
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5
2 9 4 3

View Article Online
48% aqueous HBr afforded the product as a viscous oil (0.48 g,
2.07 mmol, 72% yield). Suitable cystals for X-ray analysis of
the chloride, bromide and iodide salts of the macrocycle were
obtained in a similar manner as for L3. Ligand L2 is difficult
to handle because of its viscous nature. Thus, for analytical
purposes its perchlorate salt was prepared by dropwise addition
of perchloric acid into a concentrated solution of the compound
in ethanol. The resulting solution was left in the refrigerator
overnight and the resulting precipitate filtered and dried under
18 M. T. Albelda, J. Aguilar, S. Alves, R. Aucejo, P. Diaz, C. Lodeiro,
J. C. Lima, E. Garcia-Espan˜a, F. Pina and C. Soriano, Helv. Chim.
Acta, 2003, 86, 3118.
19 A. Bencini, M. I. Burguete, E. Garcia-Espan˜a, S. V. Luis, J. F. Miravet
and C. Soriano, J. Org. Chem., 1993, 58, 4749.
20 A. Andres, M. I. Burguete, E. Garcia-Espan˜a, S. V. Luis, J. F.
Miravet and C. Soriano, J. Chem. Soc., Perkin Trans. 2, 1993,
749.
21 A. C. Warden, M. Warren, M. T. W. Hearn and L. Spiccia, Inorg.
Chem., 2004, 43, 6936.
22 A. C. Warden, M. Warren, M. T. W. Hearn and L. Spiccia,
New J. Chem., 2004, 28, 1301.
1
a high vacuum. H NMR (CDCl3): 7.62 (s, 1H), 7.13 (t, J =
7.5, 1H), 6.97 (d, J = 7.5, 2H), 3.81 (s, 4H), 2.71 (t, J =
5.6, 4H), 2.47 (t, J = 5.9, 4H), 1.79 (s, b, 3H), 1.64 (m, 4H);
23 A. C. Warden, M. Warren, A. R. Battle, M. T. W. Hearn and L.
Spiccia, CrystEngComm, 2004, 6, 522.
24 C. A. Ilioudis, D. A. Tocher and J. W. Steed, J. Am. Chem. Soc., 2004,
126, 12395.
25 C. A. Ilioudis, M. J. Bearpark and J. W. Steed, New J. Chem., 2005,
29, 64.
26 C. A. Ilioudis and J. W. Steed, CrystEngComm, 2004, 6, 239.
27 C. A. Ilioudis, K. S. B. Hancock, D. G. Georganopoulou and J. W.
Steed, New J. Chem., 2000, 24, 787.
1
13C-{ H} NMR for L2·3HClO4·3H2O (D2O): 131.92, 131.76,
131.66, 49.93, 42.58, 42.33, 20.84; MS m/z (FAB) 234 ([M +
H]+); anal. calcd for the perchlorate salt C14H28N3Cl3O15: C,
28.76%; H, 4.83%; N, 7.19%. Found: C, 28.64%; H, 4.87%; N,
7.17%. Ligand L2 has been previously prepared by a modified
Richman–Atkins method.33
28 C. A. Ilioudis, D. G. Georganopoulou and J. W. Steed, CrystEng-
Comm, 2002, 4, 26.
29 V. Fusi, A. Llobet, J. Mahia, M. Micheloni, P. Paoli, X. Ribas and P.
Rossi, Eur. J. Inorg. Chem., 2002, 987.
30 P. Arranz, A. Bencini, A. Bianchi, P. Diaz, E. Garcia-Espana, C.
Giorgi, S. V. Luis, M. Querol and B. Valtancoli, J. Chem. Soc., Perkin
Trans. 2, 2001, 1765.
31 S. Andres, B. Escuder, A. Domenech, E. Garcia-Espana, S. V. Luis,
V. Marcelino, J. M. Llinares, J. A. Ramirez and C. Soriano, J. Phys.
Org. Chem., 2001, 14, 495.
32 D. K. Chand, H. J. Schneider, J. A. Aguilar, F. Escarti, E.
Garcia-Espana and S. V. Luis, Inorg. Chim. Acta, 2001, 316,
71.
33 R. C. Hoye, J. E. Richman, G. A. Dantas, M. F. Lightbourne and
L. S. Shinneman, J. Org. Chem., 2001, 66, 2722.
34 X. Ribas, D. A. Jackson, B. Donnadieu, J. Mahia, T. Parella, R.
Xifra, B. Hedman, K. O. Hodgson, A. Llobet and T. D. P. Stack,
Angew. Chem., Int. Ed., 2002, 41, 2991.
2,5,8,11-Tetraaza[12]metacyclophane (L4). By following a
procedure similar to that described for the synthesis of L3, L4–
Ts4 (2.00 g, 2.31 mmol), phenol (4.0 g, 42.50 mmol) and 60 mL of
48% aqueous HBr afforded the product as a waxy solid (0.40 g,
1.62 mmol, 70% yield). Suitable cystals for X-ray analysis of the
chloride and bromide salts were obtained in a similar manner
as for L3. 1H NMR (CDCl3): 7.75 (s, 1H), 7.22 (t, J = 7.4, 1H),
7.09 (d, J = 7.8, 2H), 3.84 (s, 4H), 2.77–2.80 (m, 4H), 2.71–
1
2.73 (m, 8H), 1.99 (s, b, 4H);13C-{ H} NMR (CDCl3): 141.59,
128.32, 127.23, 127.11, 53.12, 49.03, 49.01, 48.23; HRMS calcd
for C14H25N4 [M]+ 249.2074, found 249.2077. Ligand L4 has been
prepared independently by asimilar method.15
35 K. Hiraki, Y. Fuchita, Y. Ohta, J. Tsutsumida and K. I. Hardcastle,
J. Chem. Soc., Dalton Trans., 1992, 833.
36 G. R. Desiraju, Acc. Chem. Res., 1996, 29, 441.
37 G. R. Desiraju and T. Steiner, The Weak Hydrogen Bond,
OUP/IUPAC, Oxford, 1999.
38 L. Jiang and L. Lai, J. Biol. Chem., 2002, 277, 37732.
39 J. Bernstein, R. E. Davis, L. Shimoni and N.-L. Chang, Angew.
Chem., Int. Ed. Engl., 1995, 34, 1555.
40 G. Papoyan, K. Gu, J. Wiorkiewicz-Kuczera, K. Kuczera and K.
Bowman-James, J. Am. Chem. Soc., 1996, 118, 1354.
41 L. Infantes, J. Chisholm and S. Motherwell, CrystEngComm, 2003,
5, 480.
References
1 G. Gokel, Crown Ethers and Cryptands, Royal Society of Chemistry,
Cambridge, 1991.
2 M. W. Hosseini, Supramolecular catalysis of phosphoryl anion transfer
processes, in Supramolecular Chemistry of Anions, ed. A. Bianchi,
K. Bowman-James and E. Garcia-Espan˜a, Wiley, New York, 1997,
p. 421.
3 B. Dietrich, M. W. Hosseini, J.-M. Lehn and R. B. Sessions, J. Am.
Chem. Soc., 1981, 103, 1282.
4 M. W. Hosseini and J.-M. Lehn, Helv. Chim. Acta, 1987, 70,
1312.
5 M. W. Hosseini and J.-M. Lehn, J. Am. Chem. Soc., 1987, 109, 7047.
6 A. Domene´ch, E. Garcia-Espan˜a, J. A. Ramirez, B. Celda, M. C.
Martinez, D. Monleon, R. Tejero, A. Bencini and A. Bianchi,
J. Chem. Soc., Perkin Trans. 2, 1999, 23.
7 P. Atkinson, Y. Bretonniere and D. Parker, Chem. Commun., 2004,
438.
42 Y. S. Rosokha, S. V. Lindeman, S.V. Rosokha and J. K. Kochi, Angew.
Chem., Int. Ed., 2004, 43, 4650.
43 CSD, Version 5.25, Nov 2003 + 1 update (Jan 2004).
44 F. H. Allen, Acta Crystallogr., Sect. B, 2002, 58, 380.
45 C. A. Hunter, K. R. Lawson, J. Perkins and C. J. Urch, J. Chem. Soc.,
Perkin Trans. 2, 2001, 651.
8 R. A. Poole, G. Bobba, M. J. Cann, J. C. Frias, D. Parker and R. D.
Peacock, Org. Biomol. Chem., 2005, 3, 1013.
9 P. Atkinson, Y. Bretonniere, D. Parker and G. Muller, Helv. Chim.
Acta, 2005, 88, 391.
10 L. Qian, Z. Sun, J. Gao, B. Movassagh, L. Morales and K. B. Mertes,
J. Coord. Chem., 1991, 23, 155.
11 S. Boudon, A. Decian, J. Fischer, M. W. Hosseini, J.-M. Lehn and G.
Wipff, J. Coord. Chem., 1991, 23, 113.
12 A. Bencini, A. Bianchi, E. Garcia-Espan˜a, E. C. Scott, L. Morales,
B. Wang, T. Deffo, F. Takusagama, M. P. Mertes, K. B. Mertes and
P. Paoletti, Bioorg. Chem., 1992, 20, 8.
13 A. V. Kirsanov and N. A. Kirsanov, Zh. Obshch. Khim., 1962, 32,
887.
46 C. A. Hunter and J. K. M. Sanders, J. Am. Chem. Soc., 1990, 112,
5525.
47 H. Adams, K. D. M. Harris, G. A. Hembury, C. A. Hunter,
D. Livingstone and J. F. McCabe, Chem. Commun., 1996,
2531.
48 Proton-Transfer Reactions, ed. E. F. Caldin and V. Gold, Chapman
and Hall, London, 1975.
49 A. Bianchi, M. Micheloni and P. Paoletti, Pure Appl. Chem., 1988,
60, 525.
50 C. Bazzicalupi, A. Bencini, A. Bianchi, V. Fusi, C. Giorgi, A.
Granchi, P. Paoletti and B. Valtancoli, J. Chem. Soc., Perkin Trans.
2, 1997, 775.
51 A. Bencini, A. Bianchi, E. Garcia-Espan˜a, M. Micheloni and J. A.
Ramirez, Coord. Chem. Rev., 1999, 188, 97.
52 A. Bianchi, B. Escuder, E. Garcia-Espan˜a, S. V. Luis, V. Marcelino,
J. F. Miravet and J. A. Ramirez, J. Chem. Soc., Perkin Trans. 2, 1994,
1253.
53 A. Andre´s, C. Bazzicalupi, A. Bianchi, E. Garcia-Espan˜a, S. V. Luis,
J. F. Miravet and J. A. Ramirez, J. Chem. Soc., Dalton Trans., 1994,
2995.
54 J. L. Sessler, N. A. Tvermoes, J. Davis, P. Anzenbacher, K. Jursikova,
W. Sato, D. Seidel, V. Lynch, C. B. Black, A. Try, B. Andrioletti, G.
Hemmi, T. D. Mody, D. J. Magda and V. Kral, Pure Appl. Chem.,
1999, 71, 2009.
14 L. Qian, Z. Sun, M. P. Mertes and K. B. Mertes, J. Org. Chem., 1991,
56, 4904.
15 M. Chadim, P. Diaz, E. Garcia-Espan˜a, J. Hodacova, P. C. Junk,
J. Latorre, J. M. Llinares, C. Soriano and J. Zavada, New J. Chem.,
2003, 27, 1132.
16 J. Aguilar, A. Bencini, E. Berni, A. Bianchi, E. Garcia-Espan˜a, L.
Gil, A. Mendoza, L. Ruiz-Ramirez and C. Soriano, Eur. J. Inorg.
Chem., 2004, 4061.
17 J. Aguilar, M. G. Basallote, L. Gil, J. C. Hernandez, M. A. Manez,
E. Garcia-Espan˜a, C. Soriano and B. Verdejo, J. Chem. Soc., Dalton
Trans., 2004, 94.
2 9 4 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5

View Article Online
55 J. L. Sessler, S. Camiolo and P. A. Gale, Coord. Chem. Rev., 2003,
240, 17.
56 M. Shionoya, H. Furuta, V. Lynch, A. Harriman and J. L. Sessler,
J. Am. Chem. Soc., 1992, 114, 5714.
57 P. Gans, A. Sabatini and A. Vacca, Talanta, 1996, 43, 1739.
58 Z. Otwinowski and W. Minor, Methods in Enzymology, in Methods
in Enzymology, ed. C. W. Carter and R. M. Sweet, Academic Press,
London, 1997, p. 307.
59 G. M. Sheldrick, SHELXS-97, Program for solution of crystal
structures, University of Go¨ttingen, Germany, 1997.
60 G. M. Sheldrick, SHELXL-97, Program for refinement of crystal
structures, University of Go¨ttingen, Germany, 1997.
61 L. J. Barbour, J. Supramol. Chem., 2001, 1, 189.
62 M. A. Bernardo, J. A. Guerrero, E. Garcia-Espan˜a, S. V. Luis, J. M.
Llinares, F. Pina, J. A. Ramirez and C. Soriano, J. Chem. Soc., Perkin
Trans. 2, 1996, 2335.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 9 3 5 – 2 9 4 5
2 9 4 5