Molecular design using electrostatic interactions. Part 2: Synthesis and properties of flexible dipodand di- and tetra-cations with restricted conformations. Charge, shape and size in molecular complexation

Article abstract of DOI:10.1016/S0040-4020(00)00279-9

A number of dipodand dications and tetracations were synthesised with structures based on 1,2,4,5-, penta- and hexa-substituted benzene, DABCO substituents providing the positive charges. The binding of these polycations in water to a variety of polyanions was examined with respect to the effect of the 'spectator' counter cations on the observed association constants. The polycations precipitate ferri- and ferrocyanide, in some cases preferentially for one oxidation state, and an X-ray crystallographic analysis of the complex between the tetracation 5 and ferricyanide reveals the complexity involved even in the molecular recognition of small, charged compounds. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00279-9

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 4501±4509
Molecular Design using Electrostatic Interactions.
Part 2: Synthesis and Properties of Flexible Dipodand Di- and
Tetra-cations with Restricted Conformations. Charge, Shape and
Size in Molecular Complexation
a
b
Peter J. Garratt,a, Yiu-Fai Ng and Jonathan W. Steed
*
aDepartment of Chemistry, University College London, 20 Gordon Street, London WC1H 0AJ, UK.
bDepartment of Chemistry, King's College London, Strand, London WC2R 2LS, UK.
Received 21 January 2000; revised 13 March 2000; accepted 30 March 2000
AbstractÐA number of dipodand dications and tetracations were synthesised with structures based on 1,2,4,5-, penta- and hexa-substituted
benzene, DABCO substituents providing the positive charges. The binding of these polycations in water to a variety of polyanions was
examined with respect to the effect of the `spectator' counter cations on the observed association constants. The polycations precipitate ferri-
and ferrocyanide, in some cases preferentially for one oxidation state, and an X-ray crystallographic analysis of the complex between the
tetracation 5 and ferricyanide reveals the complexity involved even in the molecular recognition of small, charged compounds. q 2000
Elsevier Science Ltd. All rights reserved.
Introduction
gent properties which disrupted the liposome vesicles.14,15
The guanidinium group has also been used to provide the
charge in cation hosts and both cyclic systems16 and acyclic
systems17 with this function have been shown to bind
signi®cantly to phosphate ions with signi®cant binding
constants.
The design of systems that can recognise speci®c anions has
been an active research area since Simmons and Park1 in
1968 reported the encapsulation of halide ions in a charged
molecular framework. A variety of polyammonium macro-
cycles have been subsequently prepared, which can bind
anions with varying degrees of speci®city2±4 with the
guest anion penetrating into the cavity of the host molecule
and being bound by electrostatic and speci®c hydrogen bond
interactions.5±7 These studies were carried out on the
protonated species under acidic conditions and in order to
investigate the process over a wider pH range a variety of
quartenary salts were prepared, mainly by methylation of
the parent tertiary amines, and their binding af®nities
studied.8±11 These hosts are soluble in water but guest
complexion again appears to be by inclusion,12 although
the possibility of hydrogen bonding from nitrogen has
been lost. A series of cyclophanes with disubstituted diaza-
bicyclo[2.2.2]octanes [DABCO] providing the cationic
charge were found to bind large anions on the outside of
the cavity13 and, consequently, the binding constants did not
vary greatly with ring size. The DABCO motif had
previously been used in compounds designed as membrane
transfer agents for ATP, but these suffered from the deter-
The initial interaction of all molecules in the cell is probably
electrostatic since all other types of interaction fall off much
more rapidly with distance.18 Many of these biomolecules
will be charged and the interaction may be with another
charged molecule or with a molecule that can be polarised.
In order to study simple electrostatic interactions in water,
we prepared tripodand molecules based on benzene with
DABCO units providing the cationic charge.19 These were
shown to bind differentially to a number of anions and the
hexasubstituted derivatives, where rotation is restricted,
were found to have higher binding af®nities than the tri-
substituted analogues. The cations could, in some cases,
select between ferri- and ferrocyanide anions and the role
of bound water in these and related polycation complexes
determined by X-ray crystallography.20 Similar systems
using imidazole and pyrazole rather than DABCO, which
involve both electrostatic and hydrogen bonding, have been
described by others.21±23
We now report the synthesis of a series of polycation hosts
based on substituted benzene frameworks with DABCO
substituents again providing the cationic centres. We have
examined the interaction of these hosts with a variety of
Keywords: molecular recognition; polycations; carboxylates; X-ray.
* Corresponding author. Tel.: 120-76-79-46-20; fax: 120-76-79-74-63;
e-mail: p.j.garratt@ucl.ac.uk
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00279-9

4502
P. J. Garratt et al. / Tetrahedron 56 (2000) 4501±4509
Scheme 1.
polyanions, altering both their shape and charge. These
interacting molecules, although much smaller than their
biological counterparts, must undergo the same sequence
of events, coulombic attraction, rearrangement of bound,
surface water and the association into a molecular complex.
We attempted to exclude factors other than coulombic inter-
actions from this association but, of necessity, the occur-
rence of hydrogen bonding in the water and between the
water and the ions must remain.
Results and Discussion
Synthesis
The DABCO motif is a simple method of introducing a
positive charge into a designed cationic host. It is readily
alkylated24,25 and, since the two nitrogen atoms are
effectively insulated from each other, double differential
alkylation is readily accomplished.26 The DABCO
moiety can thus be used to provide one or two charges on
the selected skeletal framework if this is suitably substi-
tuted. Initially, a series of bis-bromomethylbenzenes was
prepared by bromomethylation of the appropriate poly-
methylbenzene by van der Made's method,27 as shown in
Scheme 1.
Since we had found previously that the preferred route to di-
substituted DABCO hosts was by reaction of the alkylated
DABCO derivative with the precursor host framework
molecule,19 we prepared N-alkyl-DABCO bromides from
DABCO and the required alkyl bromide and treated these
with the appropriate bromomethylbenzene as shown in
Scheme 2.
The bis-benzyl-DABCO derivatives 1, 2 and 3 were
prepared in this way as the dibromide salts. The FAB
mass spectra all showed ions corresponding to the loss of
The tetrabromide salts of the benzene derivatives 4, 5 and 6
were prepared in this way, utilising 2-bromopropane to
alkylate DABCO. The FAB mass spectra all showed an
1
one bromide ion. The compounds had the expected H and
13C NMR spectra except that the proton spectrum of 1 was
more complex. This was thought to arise from restricted
rotation and to investigate this we converted the dibromide,
which was only soluble in water to any appreciable extent,
to the di-hexa¯uorophosphate, which is soluble in dimethyl
sulfoxide. In the variable temperature 1H NMR spectrum in
DMSO the benzylic protons appear as two signals at 258C
which coalesce to a singlet at 908C with a rotational barrier
1
ion corresponding to the loss of one bromide ion. The H
NMR spectra were in agreement with the assigned
structures and the 13C NMR spectra showed the expected
number of carbons. The H NMR spectrum of 4 showed
1
multiple signals for the methylene groups in a way similar
to that of the dicationic analogue 1. The tetracation 4 was
converted to the tetra-hexa¯uorophosphate which dissolved
in DMSO, but attempts to determine the coalescence
of ca. 80 kJ mol21
.
Scheme 2.

P. J. Garratt et al. / Tetrahedron 56 (2000) 4501±4509
4503
temperature were frustrated by decomposition of the
compound at ca. 808C.
Binding studies
The association constants of these polycations with a variety
1
of polyanion guests were determined by H NMR titration
studies29 and the stoichiometry of association by the
method of continuous variation,30 as we have described
previously.19 The titrations were carried out in water and,
in order to investigate the importance of spectator ions on
the apparent association constants, anionic salts of lithium,
sodium, potassium and calcium were used. The polycations
were all in the form of their bromide salts. Dication receptor
1 showed a similar association constant with the dilithium,
disodium and dipotassium 1,2-benzenedicarboxylate
(Table 1). This suggests that the association constants of
these complexes are all similar, in agreement with the
literature association constants that show only a small
variation.31 All three combinations had 1:1 stoichiometries.
Dication 2 has a lower dissociation constant with sodium
1,2-benzenedicarboxylate but the lithium and potassium
salts did not have 1:1 stochiometries, as was also found
for the sodium and potassium salts of dication 3. The alkali
salts of 1,3-benzenedicarboxylate gave somewhat larger
association constants and all had 1:1 stoichiometries.
Calcium 1,3-benzenedicarboxylate was also investigated
and this had a similar association constant to the alkali
salts, suggesting that the calcium ion does not bind to
both anionic centres of the 1,3-dicarboxylate simul-
taneously. We also investigated the association of 1 with
disodium and calcium dipicolinate, since dipicolinic acid is
a bacterial spore component responsible for the chelation of
calcium.32 In this case the association constant with calcium
dipicolinate (log K 1.18) was signi®cantly smaller than with
disodium dipicolinate (log K 1.66) which was similar to that
for calcium 1,3-benzenedicarboxylate (log K 1.78). This
difference is consistent with the much larger association
constant for calcium dipicolinate (log K 4.61) than for
calcium 1,3-benzenedicarboxylate (log K 0.93)33 with the
polycation 1 having dif®culty in competing with the calcium
ion for the dipicolinate. With calcium dipicolinate, binding
of the calcium ion to both anionic centres is probably
facilitated by the nitrogen lone pair and the absence of the
intervening proton.
A related series of compounds, 7, 8 and 9, were synthesised
from 11-N-DABCO-undecanethiol, prepared by alkylation
of DABCO with 11-bromo-undecanethiol, and the appro-
priate benzyl bromide. The FAB spectra all show ions corre-
sponding to the loss of one bromide ion and the 1H and 13
C
NMR spectra are consistent with the assigned structures.
These compounds have been used to assemble cationic
monolayers on gold surfaces.28
Dication 1 with trisodium 1,3,5-benzenetricarboxylate and
tetrasodium 1,2,4,5-benzene-tetracarboxylate showed
signi®cant changes in the NMR chemical shift but the
complexes did not have 1:1 stoichiometries. Attempts to
determine the association constants of the tetracations 4±6
with the tetrasodium 1,2,3,4-benzenetetracarboxylate failed,
1
the H NMR titration plot showing at ®rst an increasing
value of the change in chemical shift with increasing
host±guest ration but, after a maximum value, a further
increase in the host caused a decrease in the change of the
chemical shift. A more complex equilibrium situation is
most probably involved in these cases, possibly with 2:1
complexes at higher cation concentrations.
Selective crystallisation
Although the cations had not given large association
constants with the polyanions investigated, there did appear
to be some shape preference in that the 1,3-benzene-
dicarboxylate salts had larger association constants than

4504
P. J. Garratt et al. / Tetrahedron 56 (2000) 4501±4509
Table 1. Maximum change in chemical shift and the association constant for a series of polycation hosts and guests with varying counter cations
Cation
Anion
Counter cation
Maximum Dd,
(error) Anion
protons
log K
1
1
1
2
3
1
1
1
1
2
2
2
2
3
3
3
3
1
2
3
1
1
10
10
10
10
10
11
11
11
11
11
11
11
11
11
11
11
11
12
12
12
13
13
Li1
0.055 (^0.001)a
0.060 (^0.004)a
0.056 (^0.001)a
0.064 (^0.008)a
0.041 (^0.002)a
0.210 (^0.0025)b
0.229 (^0.001)b
0.252 (^0.003)b
0.231 (^0.004)b
0.131 (^0.000)b
0.133(^0.000)b
0.151 (^0.002)b
0.160 (^0.001)b
0.1475 (^0.002)b
0.149 (^0.002)b
0.152 (^0.006)b
0.160 (^0.003)b
0.279 (^0.001)
0.139 (^0.002)
0.165 (^0.001)
0.258 (^0.003)a
0.076 (^0.001)a
1.69
1.64
1.57
1.34
1.88
1.99
1.73
1.83
1.78
1.95
1.96
1.82
1.86
1.92
1.92
1.86
1.85
1.78
1.89
1.73
1.66
1.18
Na1
K1
Na1
Li1
Li1
Na1
K1
Ca21
Li1
Na1
K1
Ca21
Li1
Na1
K1
Ca21
Na1
Na1
Na1
Na1
Ca21
a Average Dd of protons Ha and Hb.
b Dd of H2.
their 1,2-benzendicarboxylate analogues. We therefore
investigated changing the charge on the counterion while
maintaining the same geometry. For this purpose we chose
the octahedral ferro- (Fe(CN)462 and ferricyanide
(Fe(CN)362) ions as their potassium salts. Three experiments
were done, one involving ferricyanide, one ferrocyanide and
the other an equimolar mixture of the two ions, each with a
dilute solution of the polycation. Where precipitates
occurred they usually did so immediately, but in all cases
the solutions were then left to stand for some weeks after
which time some of the precipitates were crystalline. The
precipitates were removed and investigated by infra-red
spectroscopy, the ferrocyanide ion having an absorption at
ca. 2050 cm21 and the ferricyanide ion at 2110 cm21. The
dication 1 gave a precipitate with ferrocyanide, ferricyanide
and with the mixture, whereas the dication 2 gave precipitates
Figure 1. Partial structure of ferricyanide complex of 5, showing two tetracations and three ferricyanide ions.

P. J. Garratt et al. / Tetrahedron 56 (2000) 4501±4509
4505
Figure 2. Partial structure of ferricyanide complex of 5, showing tetracations, ferricyanide and potassium ions but omitting water molecules.
only with ferrocyanide and the mixture. While the precipi-
tate from 1 with the mixture showed the presence of bands at
2045 and 2109 cm21, that of 2 with the mixture showed only
one band at 2056 cm21. It thus appears that the dication 2
can selectively distinguish ferro- from ferricyanide.
Dication 3 did not form a precipitate in any of the experi-
ments, which suggests that either the two cationic centres
cannot readily form a cavity on the same side because of
coulombic and steric repulsion or that such a cavity is too
small to accommodate the ion in a suitable manner.
The tetracation 4 also gave a precipitate in all three cases,
the precipitate from the mixture again showing IR bands for
both anions. The tetracation 5 also showed precipitates in all
three cases but those for the ferricyanide and mixture
produced crystals, and both of the crystalline materials
had only an absorption at ca 2110 cm21. The tetracation 6
did not form a precipitate with ferrocyanide but did with
both ferricyanide and the mixture, both precipitates
having absorptions at ca. 2110 cm21. These differences in
behaviour of the three tetracations may re¯ect the

4506
P. J. Garratt et al. / Tetrahedron 56 (2000) 4501±4509
differences in the rotational barriers for the DABCO
substituents, 4 having the most and 5 the least hindered
rotation.
Experimental
Melting points were determined on a Reichert melting point
apparatus and are uncorrected. Optical rotations were
measured on an Optical Activity Ltd Polar 2000 automatic
polarimeter. Mass spectra were recorded on a VG-ZAB
An X-ray crystallographic analysis of the complex between
and FeIII(CN)6 showed that it had the formula
5
1
C74H128N26O14Fe3K, revealing the presence of both a spec-
tator ion, K1, and 14 molecules of water in the structure.
There are two tetracations and three anions in the unit cell
and the extra positive charge is provided by the potassium
ion. The unit cell for the complex, without the water
molecules, is shown in Fig. 1 and part of the crystal struc-
ture, again without water molecules, is shown in Fig. 2. Both
tetracation units are in the syn conformation, one with a
ferricyanide unit partially between the pendant groups
while the other shares in two external ferricyanide units.
Unlike the case of the tripodand trication we have
described previously,19 it would appear that the tetra-
cation 5 is unable to accommodate the ferricyanide
ion in its internal cavity and, presumably, this also applies
to the ferrocyanide ion where charge matching would
occur. The charge in the tripodand is more dispersed, a
single charge residing on each arm, and with its larger cavity
the ferricyanide is probably not required to approach as
close to the benzene ring as it would in the case of the
tetracation 5. We were unable to resolve the 14 water
molecules in this structure but the complexity of the unit
cell suggests that they probably have a more intimate `intra-
molecular' role than the water molecules in the tripodand
structure.20
mass spectrometer with Finnigan Incos II data system. H
NMR spectra were recorded at 400 and 300 MHz and 13C
NMR spectra were recorded at 100 and 75 MHz on a Varian
VXR-400 and a Bruker AC300 instrument, respectively.
Elemental analyses were carried out by the microanalytical
section of the Chemistry Department, University College
London. The polycation bromide salts, which are deliques-
cent, were extensively dried in vacuo but water of crystal-
lisation remained.
Chemical reagents were purchased from Aldrich Chemical
Co., Lancaster Synthesis, ACROS and BDH. Solvents and
reagents were used as received except for tetrahydrofuran
which was distilled over sodium and benzophenone under
an atmosphere of nitrogen immediately prior to use.
Analytical thin layer chromatography (TLC) was performed
on pre-coated aluminium backed plates (Merck Kieselgel 60
F254) and visualised using ultraviolet light (254 nm), iodine
or potassium permanganate solution as appropriate.
4,6-Bis(bromomethyl)-1,2,3-trimethylbenzene. A solu-
tion of HBr in acetic acid (49% w/v, 11.5 mL) was added
in one portion to a solution of paraformaldehyde (3.00 g,
100.0 mmol) and hemellitol (5 mL, 4.47 g, 40.0 mmol) in
glacial acetic acid (250 mL). The resulting mixture was
stirred at 100±1108C for 6 h and was then poured into
water. A brown precipitate formed which was removed by
®ltration and washed exhaustively with water. The solid was
then dissolved in acetone and re-precipitated by the addition
of water. The mixture was then stirred for 2 h, and the solid
was removed by ®ltration and dried under vacuum to
give 4,6-bis(bromomethyl)-1,2,3-trimethylbenzene as a
white solid (4.38 g, 14.3 mmol, 36%); mp 147±1498C,
lit.34 153±1548C.
Conclusions
Rather disappointingly, the association of the polycations
with a range of polyanions showed very similar binding
constants. This may be due to the effect of water solvating
the ions and reducing the importance of charge and shape
matching. The nature of the anion counter ions also had a
rather small but genuine effect on the observed association
constant and such an effect is likely to be signi®cantly larger
in organic solvents. The largest effect of the counter ion was
found with the dipicolinate anion, where the calcium salt is
much less readily complexed than the sodium salt, which
re¯ects the high association constant of calcium dipicoli-
nate. Dication 1 is not, unfortunately, a superior co-ordina-
tor of the calcium salt. Although examples of charge
matched ions can be found that lead to precipitation of the
host±guest complex, for example 2 with ferrocyanide, the
situation with these small polycations is complex, with
water and spectator ions probably playing a role in both
recognition and the subsequent processes of aggregation,
precipitation and crystallisation. In the charge matched
complexes the water, at least in some cases, appears to be
involved mainly in aggregation and crystal formation,20
whereas in the non-matched complexes it may play a
more intimate role in the docking of the host and guest.
Spectator ions, as in the complex of 5 and ferricyanide,
may also be involved in the recognition phase of the process
and they may also exert an in¯uence on host±guest binding
if they are not, as is commonly assumed, completely disso-
ciated from the guest.
N-(11-Thioundecacyl)-DABCO bromide. 11-Bromo-1-
undecanethiol35 (3.87 g, 14.5 mmol) was added to a solution
of DABCO (2.91 g, 25.9 mmol) in acetonitrile (50 mL). The
resulting mixture was stirred for 72 h and then poured into
diethyl ether (600 mL) and stirred for 30 min. The white
precipitate was removed by ®ltration, washed with diethyl
ether, and dried under vacuum to give N-(11-thioun-
decacyl)-DABCO bromide as a white solid (5.23 g,
1
13.8 mmol, 95%); mp 260±2628C dec.; H NMR d (D2O)
3.31 (t, 6H, Jˆ7.4 Hz), 3.16 (m, 2H), 3.10 (t, 6H,
Jˆ7.5 Hz), 2.45 (t, 2H, Jˆ7.1 Hz), 1.70±1.62 (br.m, 2H),
1.51 (m, 2H, Jˆ7.3 Hz), 1.26, 1.21 (2 br.s, 15H); 13C NMR
d 97.7, 65.4, 52.9, 45.0, 34.0, 29.5, 29.4, 29.3, 29.1, 28.4,
26.5, 24.7, 22.0; IR 2918, 2850, 2450, 2346, 1466, 1317,
1099, 1057 cm21; MS m/e 299 (100, M2Br1), 266 (5);
C17H35BrN2S.H2O requires C, 53.81; H, 9.83; N, 7.38; Br,
21.06. Found: C, 52.45; H, 9.24; N, 7.11; Br, 20.95.
3,6-Bis(DABCO-N-methyl)-1,2,4,5-tetramethylbenzene
dibromide (1). A solution of 3,6-bis(bromomethyl)-1,2,4,5-
tetramethylbenzene27 (1.45 g, 4.5 mmol) in acetonitrile

P. J. Garratt et al. / Tetrahedron 56 (2000) 4501±4509
4507
(50 mL) was added to a solution of DABCO (1.55 g,
13.8 mmol) in acetonitrile (100 mL) and the mixture was
then stirred for 48 h. It was then poured into diethyl ether
(300 mL) and the resulting mixture stirred for 30 min. A
precipitate formed which was removed by ®ltration, washed
with diethyl ether and dried under vacuum to give 1 a white
solid (2.12 g, 3.9 mmol, 86%); mp 256±259 dec.; 1H NMR
d (D2O) 4.93 (s, 4H), 3.48 (t, 12H, Jˆ7.3 Hz), 3.16 (t, 12H,
Jˆ7.8 Hz), 2.42 (s, 12H); 13C NMR (D2O) d 140.8, 128.6,
65.1, 54.4, 47.1, 21.4; IR 3010, 2961, 2889, 1653, 1497
cm21; MS m/e 463 (100, M2Br1), 383 (7, M22Br1),
271 (33), 160 (29), 112 (71); C24H40Br2N4 requires C,
52.95; H, 7.35; N, 10.29. Found: C, 52.76; H, 7.14; N, 10.18.
2,5-Bis(N0-isopropyl-DABCO-N-methyl)-1,4-dimethyl-
benzene tetrabromide (5). Isopropyl-DABCO bromide26
(2.00 g, 8.5 mmol) was added to a solution 2,5-bis(bromo-
methyl)-1,4-dimethylbenzene27 (1.00 g, 3.4 mmol) in aceto-
nitrile (150 mL) and the mixture was stirred for 48 h. A
solid formed which was removed by ®ltration, washed
with acetonitrile and dried under vacuum to give 5 as a
white solid (2.51 g, 3.1 mmol, 91%); mp 232±2358C dec.;
1H NMR d (D2O) 7.58 (s, 2H), 4.93 (s, 4H), 4.14 (br. t,
12H), 3.98 (br. t, 14H), 2.52 (s, 12H), 1.49 (d, 12H,
Jˆ6.5 Hz); 13C NMR (D2O) 141.5, 140.3, 130.0, 71.4,
68.2, 53.9, 51.4, 21.6, 18.2; IR 3022, 2053, 1625, 1488,
1473, 1122, 839 cm21; MS m/e 683 (26, M2Br1), 603 (4,
M22Br1), 155 (100), 112 (4); C28H50Br4N4´3H2O requires
C, 41.19; H, 6.91; N, 6.86. Found: C, 41.09; H, 7.16; N,
6.69.
2,5-Bis(DABCO-N-methyl)-1,4-dimethylbenzene dibro-
mide (2). 2,5-Bis(bromomethyl)-1,4-dimethylbenzene27
(1.00 g, 3.4 mmol) was added to a solution of DABCO
(0.88 g, 7.8 mmol) in acetonitrile (100 mL) and the result-
ing mixture was stirred for 48 h. The mixture was then
poured into Et2O (300 mL) and stirred for 30 min. A pre-
cipitate was formed which was removed by ®ltration,
washed with Et2O, and dried under vacuum to give 2 as a
4,6-Bis(N0-isopropyl-DABCO-N-methyl)-1,2,3-trimethyl-
benzene tetrabromide (6). Isopropyl-DABCO bromide26
(1.56 g, 6.6 mmol) was added to a solution 4,6-bis(bromo-
methyl)-1,2,3-trimethylbenzene34 (0.50 g, 1.6 mmol) in
acetonitrile (150 mL) and the mixture was stirred for 48 h.
A solid formed which was removed by ®ltration, washed
with acetonitrile and dried under vacuum to give 6 as a
white solid (1.22 g, 1.5 mmol, 94%); mp 240±2428C dec.;
1H NMR d (D2O) 7.52 (s, 1H), 5.00 (s, 4H), 4.06 (br. t,
12H), 3.94 (br. t, 14H), 2.47 (s, 6H), 2.36 (s, 3H, 1.47 (d,
12H, Jˆ6.2 Hz); 13C NMR (D2O) 145.4, 144.2, 139.4,
125.0, 71.3, 69.1, 53.8, 51.4, 20.3, 19.5, 18.2; IR 2991,
2903, 2050, 1635, 1476, 1443 cm21; MS m/e 697 (13,
M2Br1), 616 (3, M22Br1), 155 (100), 112 (3);
C29H52Br4N4´3H2O requires C, 41.95; H, 7.04: N, 6.75.
Found: C, 42.23; H, 6.95; N, 6.62.
1
white solid (0.96 g, 1.8 mmol, 53%); 238±2408C dec.; H
NMR d (D2O) 7.48 (s, 2H), 4.59 (s, 4H), 3.53 (t, 12H,
Jˆ7.4 Hz), 3.20 (t, 12H, Jˆ7.2 Hz), 2.46 (s, 6H); 13C
NMR (D2O) 140.9, 140.2, 130.3, 67.9, 55.0, 47.2, 21.7; IR
2957, 2892, 1646, 1611 cm21; MS m/e 435 (100, M2Br1),
355 (7, M22Br1), 243 (26), 132 (20); C22H36Br2N4. H2O
requires C, 49.45; H, 7.17; N, 10.48. Found: C, 50.03; H,
7.12; N, 10.63.
4,6-Bis(DABCO-N-methyl)-1,2,3-trimethylbenzene di-
bromide (3). 4,6-Bis(bromomethyl)-1,2,3-trimethylbenzene
(1.34 g, 4.4 mmol) was added to a solution of DABCO
(1.23 g, 11.0 mmol) in acetonitrile (180 mL), and the result-
ing mixture was stirred for 48 h. The mixture was then
poured into Et2O (400 mL) and stirred for 30 min. The
resulting precipitate was removed by ®ltration and dried
under vacuum to give 3 as a white solid (1.61 g,
3,6-Bis(1-undecanethiol-N0-11-DABCO-N-methyl)-1,2,4,5-
tetramethylbenzene tetrabromide (7). 3,6-Bis(bromo-
methyl)-1,2,4,5-tetramethylbenzene27 (0.25 g, 0.8 mmol)
was added to a solution of N-(11-thioundecacyl)-DABCO
bromide (0.72 g, 1.9 mmol) in acetonitrile (100 mL) and
the resulting mixture was stirred for 48 h. A solid formed
which was removed by ®ltration, washed with acetonitrile,
and dried under vacuum to give 7 as a white solid (0.50 g,
0.5 mmol, 57%); mp 265±2698C dec.; 1H NMR d
(D6-DMSO) 5.19 (s, 4H), 3.92 (br. m, 28H), 3.49 (br. m,
4H), 2.41 (s, 12H), 1.62 (br. m, 4H), 1.51 (br. m, 4H), 1.24
(br. s, 30H); 13C NMR (D6-DMSO) 138.2, 126.1, 50.5, 49.8,
33.3, 28.9, 28.8, 28.7, 28.5, 28.3, 27.7, 25.5, 23.7, 21.3,
19.0; IR 2923, 2852, 2473, 2369, 1466, 1390, 1105,
1059 cm21; MS m/e 998 (,1, M2Br1), 917 (,1,
M22Br1), 299 (100), 112 (17); C46H86Br4N4S2´H2O
requires C, 50.37; H, 8.09; N, 5.11. Found: C, 50.46; H,
8.16; N, 4.93.
1
2.8 mmol, 64%); mp 245±2478C dec.; H NMR d (D2O)
7.36 (s, 1H), 4.66 (s, 4H), 3.45 (t, 12H, Jˆ6.7 Hz), 3.16
(t, 12H, Jˆ6.7 Hz), 2.41 (s, 6H), 2.32 (s, 3H); 13C NMR
(D2O) 144.3, 143.1, 140.1, 125.5, 68.9, 54.8, 47.2, 20.3,
19.3; IR 2987, 2961, 2947, 2886, 1459, 1363 cm21; MS
m/e 449 (100, M2Br1), 369 (16, M22Br1), 258 (61),
146 (56), 112 (49); C23H38Br2N4´2H2O requires C, 48.77;
H, 7.47; N, 9.89. Found: C, 49.16; H, 7.18; N, 10.02.
3,6-Bis(N0-isopropyl-DABCO-N-methyl)-1,2,4,5-tetra-
methylbenzene tetrabromide (4). Isopropyl-DABCO
bromide26 (4.00 g, 17.0 mmol) was added to a solution of
3,6-bis(bromomethyl)-1,2,4,5-tetramethylbenzene27 (2.00 g,
6.2 mmol) in acetonitrile (200 mL). The mixture was stirred
for 72 h and the resulting solid was removed by ®ltration,
washed with acetonitrile and dried under vacuum to give 4
as a white solid (4.15 g, 5.0 mmol, 81%); mp 249±2518C
dec.; 1H NMR d (D2O) 5.25 (s, 4H), 4.06 (br. t, 12H), 3.93±
3.87 (br. t, 14H), 2.46 (s, 12H), 1.46 (d, 12H, Jˆ6.5 Hz); 13C
NMR (D2O) 141.6, 128.5, 71.3, 65.8, 53.7, 51.4, 21.6, 18.2;
IR 2995, 2074, 1636, 1496, 1475 cm21; MS m/e 711 (8,
M2Br1), 155 (100), 112 (3); C30H54Br4N4.2H2O requires
C, 43.60; H, 7.07; N, 6.78. Found: C, 43.96; H, 7.11; N,
6.84.
2,5-Bis(1-undecanethiol-N0-11-DABCO-N-methyl)-1,4-
dimethylbenzene tetrabromide (8). 2,5-Bis(bromo-
27
methyl)-1,4-dimethylbenzene
(0.20 g, 0.7 mmol) was
added to a solution of N-(11-thioundecacyl)-DABCO
bromide (0.77 g, 2.0 mmol) in acetonitrile (100 mL) and
the resulting mixture was stirred for 48 h. A solid formed
which was removed by ®ltration, washed with acetonitrile,
and dried under vacuum to give 8 as a white solid (0.38 g,
0.3 mmol, 50%); mp 243±2478C dec.; 1H NMR d
(D6-DMSO) 7.53 (s, 2H), 4.96 (s, 4H), 4.07, 3.91 (2 br. s,

4508
P. J. Garratt et al. / Tetrahedron 56 (2000) 4501±4509
28H), 3.51 (br. m, 4H), 2.44 (s, 6H), 1.63 (br. m, 4H), 1.51
(br. m, 4H), 1.25 (br. s, 30H); 13C NMR (D6-DMSO) 137.7,
127.7, 63.4, 50.5, 50.0, 33.3, 28.9, 28.8, 28.7, 28.5, 28.4,
27.7, 25.5, 23.7, 19.3; IR 2994, 2924, 2852, 2372, 1466,
1389, 1107 cm21; MS m/e 972 (4, M2Br1), 677 (6), 591
(10), 299 (85), 154 (100), 112 (57); C44H82Br4N4S2.2H2O
requires C, 48.62; H, 7.97; N, 5.15. Found: C, 48.39; H,
8.06; N, 5.15.
Hydrogen atoms were generated in idealised positions,
assigned an isotropic displacement parameter 1.2 times
the atom to which they were attached (1.5 times for methyl
groups) and a riding model adopted. Hydrogen atoms for
disordered water molecules were not located. The K1 ion
was found to be disordered over two sites, each of 50%
occupancy. As a result extensive disorder was found for
the hydrated portion of the structure. The 14 water mole-
cules were modelled as occupying a total of 26 sites, with
occupancies ranging from 1.0±0.25. A clear division in the
difference Fourier map existed between the last partial water
molecule and background. The CIF ®le of crystallographic
data has been deposited with the Cambridge Crystallo-
graphic Data Service.
Polyanions
These were either obtained commercially or were prepared
from the commercially available acids by treatment with the
appropriate metal hydroxide followed by precipitation with
acetone, isolation and drying under vacuum.
Crystallisation with ferro- and ferricyanide
Acknowledgements
A half saturated solution of the potassium ferro- or ferri-
cyanide was added to a dilute solution of the polycation
(concentration ca. 10 mg mL21) and, after standing, the
precipitate was removed by ®ltration.
We thank Professor J. H. Ridd (UCL) for provision of the
program for non-linear curve ®tting to determine the K
values, Dr A. J. Ibbett for valuable discussions, the
EPSRC and King's College London for the provision of
the X-ray diffractometer and the Nuf®eld Foundation for
the provision of computing equipment. P. J. G. thanks the
Leverhulme Trust for the award of an emeritus fellowship.
For the competitive experiment, a solution was made from
equal volumes of saturated potassium ferricyanide and satu-
rated ferrocyanide which was then added to a dilute solution
of the polycation (concentration ca. 10 mg mL21). After
standing the precipitate was removed by ®ltration.
References
NMR titrations and job plots
1. Simmons, H. E.; Park, C. H. J. Am. Chem. Soc. 1968, 90, 2428.
2. Dietrich, B.; Hosseini, M. W.; Lehn, J. M.; Sessions, R. B.
J. Am. Chem. Soc. 1981, 103, 1282±1283.
These were carried out as previously described.19
X-Ray structure analyses
3. Kimura, E.; Sakonaka, A.; Kodama, M. J. Am. Chem. Soc.
1982, 104, 4984±4985.
4. Bencini, A.; Bianchi, A.; Burguete, M. I.; Domenech, A.;
Garcia-EspanÄa, E.; Luis, S. V.; Nino, M. A.; Ramirez, J. A.
J. Chem. Soc., Perkin Trans. 2 1991, 1445±1451.
5. Dietrich, B.; Guilhem, J.; Lehn, J. M.; Pascard, C.; Sonveaux, E.
Helv. Chim. Acta 1984, 67, 91±104.
Crystal data for 5: C74H128Fe3KN26O14 M 1812.67 g mol21
,
Ê
triclinic, P1, aˆ16.2507(6), bˆ17.5519(6), cˆ18.0737(6) A,
Ê
aˆ79.734(1), bˆ63.885(1), gˆ74.677(1)8, Uˆ4452.9(3) A,
Zˆ3, data 14694, parameters 1180, R1 [F2.2s(F2)] 0.0642,
wR2 (all data) 0.1187.
6. Hosseini, M. W.; Lehn, J. M. Helv. Chim. Acta 1988, 71, 749±
756.
Crystals were mounted on a thin glass ®bre using a fast
setting epoxy resin and cooled on the diffractometer to
100 K using an Oxford Cryostream low temperature attach-
ment. A total of 90 oscillation frames each of width 2o in f
and of 30 s exposure time were recorded using a Nonius
KappaCCD diffractometer, with a detector to crystal
distance of 25 mm. Crystals were indexed from the ®rst
ten frames using the DENZO package36 and positional
data were re®ned along with diffractometer constants to
give the ®nal unit cell parameters. Integration and scaling
(DENZO, Scalepack36) resulted in unique data sets
corrected for Lorentz and polarisation effects and for the
effects of crystal decay and absorption by a combination
of averaging of equivalent re¯ections and an overall volume
and scaling correction. Structures were solved using
shelxs-9737 and developed via alternating least squares
cycles and difference Fourier synthesis (shelxl-9737) with
the aid of the program XSeed.38 All calculations were
carried out with either a Silicon Graphics Indy workstation
or an IBM compatible PC. From systematically absent data
and subsequent re®nement the space group was found to be
P-1. All non-hydrogen atoms were re®ned anisotropically.
7. Dietrich, B.; Dilworth, B.; Lehn, J. M.; Sauvage, J.-P.; Cesario,
M.; Guilhem, J.; Pascard, C. Helv. Chim. Acta 1996, 79, 569±587.
8. Schmidtchen, F. P. Chem. Ber. 1980, 113, 864±874.
9. Schmidtchen, F. P. Chem. Ber. 1981, 114, 597±607.
10. Ichikawa, K.; Yamamoto, A.; Hossain, M. A. Chem. Lett.
1993, 2175±2178.
11. Ichikawa, K.; Hossain, M. A.; Tamura, T.; Kamo, N. Supra-
mol. Chem. 1995, 5, 219±224.
12. Schmidtchen, F. P.; Mueller, G. J. Chem. Soc., Chem.
Commun. 1984, 1115±1116.
13. Menger, F. M; Catlin, K. K. Angew. Chem., Int. Ed. Engl.
1995, 34, 2147±2150.
14. Li, T.; Krasne, S. J.; Persson, B.; Kaback, H. R.; Diederich, F.
J. Org. Chem. 1993, 58, 380±384.
15. Riggs, J. A.; Hossler, K. A.; Smith, B. D.; Karpa, M. J.; Grif®n,
G.; Duggan, P. J. Tetrahedron Lett. 1996, 37, 6303±6306.
16. Dietrich, B.; Fyles, D. L.; Fyles, T. M.; Lehn, J.-M. Helv.
Chim. Acta 1979, 62, 2763±2787.
17. Jubian, V.; Dixon, R. P.; Hamilton, A. D. J. Am. Chem. Soc.
1992, 114, 1120±1121.
18. Perutz, M. Science 1978, 201, 1187±1191.
19. Garratt, P. J.; Ibbett, A. J.; Ladbury, J. E.; O'Brien, R.;

P. J. Garratt et al. / Tetrahedron 56 (2000) 4501±4509
Hursthouse, M. B.; Malik, K. M. A. Tetrahedron 1998, 54, 949± 29. Macomber, R. S. J. Chem. Ed. 1992, 69, 375±378.
4509
968.
20. Christo®, A. M.; Garratt, P. J.; Hogarth, G.; Ibbett, A. J.; Ng,
Y.-F.; Steed, J. W. In preparation.
30. Gil, V. M. S.; Oliveira, N. C. J. Chem. Ed. 1990, 67, 473±478.
31. Daniele, P. G.; De Robertis, A.; De Stefano, C.; Sammartano,
S.; Rigano, C. J. Chem. Soc., Dalton Trans. 1985, 2353±2361.
32. Russell, A. D.; Chopra, I. Understanding Antibacterial Action
and Resistance, 2nd ed.; Ellis Horwood, 1996.
21. Metzger, A.; Lynch, V. M.; Anslyn, E. V. Angew. Chem., Int.
Ed. Engl. 1997, 36, 862±865.
22. Sato, K.; Arai, S.; Yamagishi, T. Tetrahedron Lett. 1999, 40,
5219±5222.
33. Lewis, J. C. J. Biol. Chem. 1972, 247, 1861±1868.
34. Tomioka, H.; Okada, H.; Watanabe, T.; Banno, K.; Komatsu,
K.; Hirai, K. J. Am. Chem. Soc. 1997, 119, 1582±1593.
35. Bain, C. D.; Troughton, E. B.; Tao, Y.-T.; Eval, J.; Whitesides,
G. M.; Nuzzo, R. G. J. Am. Chem. Soc 1989, 111, 321±335.
36. Otwinowski, Z.; Minor, W. In Methods in Enzymology; Carter,
C. W., Sweet, R. M., Eds.; Academic Press: New York, 1996; Vol.
276, pp 307±326.
23. Chin, J.; Walsdorff, C.; Stranix, B.; Oh, J.; Chung, H. J.; Park,
S.-M.; Kim, K. Angew. Chem., Int. Ed. Engl. 1999, 38, 2756±2759.
24. Mann, F. G.; Mukherjee, D. P. J. Chem. Soc. 1949, 2298±
2302.
25. Mann, F. G.; Baker, F. C. J. Chem. Soc. 1957, 1881±1899.
26. Oae, S.; Hovarth, B.; Zalut, C.; Harris, R. J. Org. Chem. 1959,
47, 578±579.
È
37. Sheldrick, G. M. shelxl-97, University of Gottingen, 1997.
27. van der Made, A. W.; van der Made, R. H. J. Org. Chem. 1993,
58, 1262±1263.
38. Barbour, L. J. XSeedÐa program for the interpretation and
re®nement of crystallographic data, University of Missouri,
Columbia, USA, 1999.
28. Ng, Y.-F.; Strutwolf, J.; Garratt, P. J.; Williams, D. E.
J. Electroanal. Chem. 1999, 464, 263±267.