Anion binding in aqueous media by a tetra-triazolium macrocycle

Article abstract of DOI:10.1039/c2ob25934f

Three tetra-triazole macrocycles were synthesized in good yields by the copper(i)-catalysed cycloaddition of bis-triazole azides and bis-alkynes. One of these was alkylated to give a cyclic tetra-triazolium receptor, which complexes anions strongly in competitive DMSO-water mixtures. In 1:1 DMSO-water, the tetracationic receptor exhibits a preference for the larger halides, bromide and iodide, with all halides associating more strongly than the oxoanion, acetate. The sulfate dianion is complexed far more strongly than any of the monobasic anions (K_a > 10⁴ M^-1). Quantum mechanics/molecular mechanics simulations corroborate the experimentally determined anion binding selectivity trends.

Full text of DOI:10.1039/c2ob25934f

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 6951
www.rsc.org/obc
PAPER
Anion binding in aqueous media by a tetra-triazolium macrocycle†
Nicholas G. White,a Sílvia Carvalho,b Vítor Félixb and Paul D. Beer*a
Received 15th May 2012, Accepted 27th June 2012
DOI: 10.1039/c2ob25934f
Three tetra-triazole macrocycles were synthesized in good yields by the copper(I)-catalysed cycloaddition
of bis-triazole azides and bis-alkynes. One of these was alkylated to give a cyclic tetra-triazolium
receptor, which complexes anions strongly in competitive DMSO–water mixtures. In 1 : 1 DMSO–water,
the tetracationic receptor exhibits a preference for the larger halides, bromide and iodide, with all halides
associating more strongly than the oxoanion, acetate. The sulfate dianion is complexed far more strongly
than any of the monobasic anions (Ka > 104 M−1). Quantum mechanics/molecular mechanics simulations
corroborate the experimentally determined anion binding selectivity trends.
Introduction
Since initial reports of the regioselective synthesis of the 1,2,3-
triazole motif by copper(I)-catalysed cycloaddition of azides and
alkynes (CuAAC),1,2 this heterocycle has found widespread use
across many areas of chemistry. In 2008, Flood,3,4 Craig,5 and
Hecht6 demonstrated that oligomeric or macrocyclic poly-tria-
zole systems are capable of interacting with halide anions
through C–H⋯anion interactions alone in organic solvents, and
in subsequent years the triazole functionality has been incorpo-
rated into a variety of acyclic7–13 and macrocyclic anion
receptors.3,4,14–21
With a few notable exceptions,22–28 the design of anion recep-
Fig. 1 Previously reported tetra-imidazolium and -benzimidazolium
tors that function in competitive aqueous media has relied on the
integration of positive charge into the host structural
framework.29–32 For example, several years ago we reported
alkyl-linked tetra-imidazolium and -benzimidazolium macro-
cycles which exhibit selectivity for fluoride over the other
halides in 9 : 1 CD3CN–D2O solvent mixtures (Fig. 1).33
In spite of the current interest in triazole-based systems, the
exploitation of the triazolium motif for anion recognition appli-
cations has remained largely unexplored.34–40 This is surprising
given that the additional positive charge, resulting from alkyl-
ation, further polarises the heterocycle’s C–H bond, as well as
increasing electrostatic interactions with the anion. Herein we
describe the synthesis of a tetra-triazolium macrocycle, which
macrocyclic anion hosts that bind fluoride selectively in 9 : 1 CD3CN–
D2O.33
binds anionic guest species in competitive aqueous solvent mix-
tures through charge-assisted C–H⋯anion hydrogen bonding.
Results and discussion
Receptor synthesis
We reasoned that rigid tetra-triazole aryl macrocycles, such as
those reported by Flood,3,4,15 would be a challenge to fully
alkylate, both in terms of the difficulty of completely alkylating
a highly conjugated system, and of the possibility of yielding
highly insoluble products. Therefore we targeted the less rigid
macrocycles, 1, 2 and 3 (Scheme 1), in the hope that anion
binding affinity lost by the diminished degree of preorganisation
would be compensated by the increased binding strength result-
ing from greater electrostatic interactions.
aChemistry Research Laboratory, Department of Chemistry, University
of Oxford, South Parks Road, Oxford, OX1 3TA, United Kingdom.
E-mail: paul.beer@chem.ox.ac.uk; Fax: +44 (0)1865 272690;
Tel: +44 (0)1865 285142
bDepartamento de Química, CICECO and Secção Autónoma de
Ciências da Saúde, Universidade de Aveiro, 3810-193 Aveiro, Portugal
†Electronic supplementary information (ESI) available: NMR spectra of
new compounds, titration protocols, details of modelling. CCDC
881646–881649. For ESI and crystallographic data in CIF or other elec-
tronic format see DOI: 10.1039/c2ob25934f
The receptors were synthesized using a general strategy, which
allows access to a wide range of symmetric or non-symmetric
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6951–6959 | 6951

View Online
Scheme 2 Synthesis of 7·BF4. Conditions and reagents: (i) (Me3O)-
(BF4), CH2Cl2, 54%.
Scheme 1 Synthesis of tetra-triazole macrocycles, 1, 2 and 3. Con-
ditions and reagents: (ia) 1,6-heptadiyne, [CuI(CH3CN)4](PF6), TBTA,
DIPEA, CH2Cl2, (ib) NaN3, DMSO, 60 °C, 30% from 1,6-heptadiyne;
(ii) 1,6-heptadiyne, [CuI(CH3CN)4](PF6), TBTA, DIPEA, CH2Cl2, 49%;
(iiia) 1,3-diethynylbenzene, [CuI(CH3CN)4](PF6), TBTA, DIPEA,
CH2Cl2, (iiib) NaN3, DMSO, 76% from 1,3-diethynylbenzene; (iv)
1,3-diethynylbenzene or 3,5-diethynylpyridine, [CuI(CH3CN)4](PF6),
TBTA, DIPEA, CH2Cl2, 67% for 2, 59% for 3.
investigations showed that these alkylated compounds were
poorly soluble, hampering chromatographic purification
procedures.
Pleasingly, 1 has good solubility in a range of organic sol-
vents. A solution of 1 in dry dichloromethane containing five
equivalents (i.e. 1.25 equivalents per triazole group) of (Me3O)-
(BF4) was stirred for two days at room temperature. The resulting
crude solid was recrystallised from ethanol–acetonitrile (6 : 1) to
give the tetra-cationic product, 7·BF4, as white crystals in 54%
yield (Scheme 2).
tetra-triazole macrocycles. The commercially available bis-
alkynes, 1,6-heptadiyne and 1,3-diethynylbenzene were reacted
with 3-azidopropoxytosylate, 4,41 under CuAAC conditions to
give the bis-triazole tosylates. Reacting these with sodium azide
in DMSO gave the bis-triazolyl azides, 5 and 6. The CuAAC
cyclisation reaction of these bis-azides and one equivalent of a
bis-alkyne under pseudo high dilution conditions (1.0 mmol L−1
in dichloromethane) gave the desired macrocycles, 1, 2 and 3 in
good yields (49–67%) after purification by column chromato-
graphy (Scheme 1).
Attempts to alkylate 2 and 3 were hampered by their low solu-
bility in common solvents, which ruled out the use of (Me3O)-
(BF4). Partial alkylation of 2 was possible using benzyl bromide
in the microwave at 120 °C, giving a mixture of mono- and di-
cationic species along with unreacted starting material, as evi-
denced by TLC and ESI-MS. Attempts to alkylate further using
higher temperatures caused decomposition of the triazole groups.
It was possible to selectively alkylate 3 at the pyridine group
using methyl iodide or bromohexadecane, but initial
X-ray crystallography
Crystals of 2 suitable for single crystal X-ray structural analysis
were obtained from two different solvent systems. Slow vapour
diffusion of dichloromethane into a d6-DMSO solution of the
macrocycle gave solventless crystals, while the slow evaporation
of a very dilute solution of 2 in 3 : 1 methanol–water gave crys-
tals of the trihydrate 2·3(H2O). The two structures are shown in
Fig. 2. Interestingly, in the solventless crystal, the macrocycle
has a very open structure, while 2·3(H2O) exhibits a folded struc-
ture, presumably due to solvophobic effects. The single crystal
structures of the propyl-linked macrocycles, 1 and 7·BF4 were
also obtained—the neutral macrocycle crystallizes without
solvent, the tetracationic receptor as the acetonitrile adduct
(Fig. 3). No significant hydrogen bonding or anion–π contacts
are observed between the tetrafluoroborate anions and the host.
6952 | Org. Biomol. Chem., 2012, 10, 6951–6959
This journal is © The Royal Society of Chemistry 2012

View Online
Fig. 2 Solid state structures of 2 when grown from DMSO–CH2Cl2
(top) and methanol–water (bottom). Ellipsoids shown at 50% prob-
ability; solvent molecules and some hydrogen atoms are omitted for
clarity.
Anion binding studies
Fig. 3 Solid state structures of 1 (top) and 7·BF4 (bottom). Ellipsoids
shown at 50% probability; some hydrogen atoms are omitted for clarity.
Neutral tetra-triazole macrocycles
Receptors 2 and 3 are insoluble in common organic solvents, but
have sufficient solubility in d6-DMSO to allow investigation of
their anion binding properties via 1H NMR titration experiments.
Addition of aliquots of TBA·Cl to a solution of either 2 or 3
caused perturbation in the triazole signals, as well as smaller
shifts in the phenylene and/or pyridyl protons. Job Plot analysis
Table 1 Association constants, Ka (M−1) for tetra-triazole macrocycles
with chloride anion; a estimated standard errors are given in parentheses
Solvent
1
2
3
d6–DMSO
1 : 1 CDCl3–CD3CN
CD2Cl2
<10
<10
31 (3)
77 (5)
—
—
37 (3)
—
—
indicated
1 : 1
stoichiometric
binding
in
solution.
WINEQNMR242 analysis of the titration data, monitoring the
triazole proton enabled association constants to be determined.
As shown in Table 1, both macrocycles, 2 and 3 bind chloride
weakly in this competitive solvent with the bis-phenylene recep-
tor, 2, displaying relatively stronger halide association than the
pyridyl-substituted host, 3. Association of cyclic tetra-alkyl
receptor 1 and chloride was negligible in d6-DMSO, and in less
competitive 1 : 1 CDCl3–CD3CN. Even in CD2Cl2, the affinity
of 1 for a range of anions (Cl−, I−, C6H5CO2−, SO42−) was
extremely weak (Ka < 60 M−1).43
a Chloride added as tetrabutylammonium salt. Association constants
calculated at 293 K, using WINEQNMR2.42
WINEQNMR242 showed very strong 1 : 1 stoichiometric
binding of both halides (Ka > 104 M−1), and so further titration
experiments were undertaken in the more competitive 1 : 1
d6-DMSO–D2O aqueous mixture. As shown in Table 2 and Fig. 4,
the macrocycle displays a preference for the larger halides,
bromide and iodide, with all halides associating more strongly
than the acetate oxoanion. The sulfate dianion is bound extre-
mely strongly, presumably due to its doubly negative charge.
Interestingly, the anion binding trend displayed by 7·BF4
(sulfate ≫ iodide > bromide > other monobasic anions) is
similar to the trends reported by Kubik et al. for their
neutral cyclic pseudo-peptide systems in methanol–water
Tetra-triazolium macrocycle
Unfortunately, 7·BF4 is insoluble in the 9 : 1 CD3CN–D2O
solvent mixture used in our previous studies of tetra-imidazolium
macrocycles.33 Fluoride and chloride binding studies were
instead conducted in 9 : 1 d6-DMSO–D2O, monitoring the four
equivalent triazolium protons. Analysis of the titration data using
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6951–6959 | 6953

View Online
Table 2 Association constants, Ka (M−1) for 7·BF4 and various
ESI†) were immersed in cubic boxes of a DMSO–H2O (1 : 1)
solvent mixture composed of 458 all atom DMSO molecules49
and 1800 TIP3P water molecules.50 The number of these mol-
ecules were calculated considering an equal volume of water and
DMSO with densities of 1.00 and 1.10 g cm−3, respectively. Fur-
thermore, this total number of molecules led to equilibrated
cubic boxes with side lengths (∼47.7 Å), large enough to simu-
late the solvated complexes using a 10 Å cut-off for long-range
electrostatic interactions and non-bonded van der Waals inter-
actions. The complexes in these solvent boxes were subjected to
classical MD simulation runs at room temperature (300 K) under
periodic conditions following the multistage protocol described
in the ESI.† However, all five anions were quickly solvated by
the water molecules leaving irreversibly the macrocyle during
the first 100 ps of the equilibration runs. In other words, these
results show that the interaction of 7 with different anions is not
appropriately described in this competitive solvent mixture using
only electrostatic and van der Waals force field parameters.
Hence, in an effort to understand the experimental binding data
of 7 summarised in Table 2, we decided to investigate the anion
complexes in the aqueous solvent mixture by QM/MM simu-
lations using the same starting configurations, but with 7 and the
anions described at the PM3 theory level and solvent molecules
with classical force field parameters. The dynamical behaviours
of the halide and sulfate anion complexes were evaluated for 2
ns using the previously equilibrated structures. Two extra repli-
cates were carried out for fluoride and iodide complexes and
equivalent sequences of events were observed as reported below.
Representative snapshots taken from the QM/MM simulations of
chloride, bromide and iodide complexes showing the anions sur-
rounded by solvent molecules are presented in Fig. 5.
anions; a estimated standard errors are given in parentheses
Anion
9 : 1 d6-DMSO–D2O
1 : 1 d6-DMSO–D2O
F−
>104
>104
—
—
—
226 (28)
228 (5)
388 (29)
463 (24)
148 (30)
>104
Cl−
Br−
I−
OAc−
SO4
2−
—
a Anions added as tetrabutylammonium salts. Association constants
calculated at 293 K, using WINEQNMR2.42
Fig. 4 Experimental titration data (solid points) and fitted binding iso-
therms (lines) for addition of TBA·anion to 1 : 1 d6-DMSO–D2O solu-
tions of 7·BF4. Sulfate binding was too strong to allow determination of
a binding constant using WINEQNMR242 (Ka > 104 M−1).
In these complexes the macrocycle almost preserves the
flattened conformation (see ESI†), establishing with the anion
(A) four intermittent C–H⋯A− hydrogen bonds with average
distances and standard deviations described in Table 3. The
mixtures,22,23,25 where the observed selectivity is attributed to
complementary host/guest size.
It is noteworthy that amongst the halides, the selectivity trend
of 7·BF4 is different from that of tetra-imidazolium and tetra-
benzimidazolium analogues.33 These receptors exhibit a pro-
nounced selectivity for fluoride in 9 : 1 CD3CN–D2O, with other
anions being bound only weakly, whereas 7·BF4 preferentially
binds bromide and iodide.44
Molecular modelling
The binding affinity of 7 for halide and sulfate anions in the
competitive 1 : 1 DMSO–H2O solvent mixture was also investi-
gated by classical molecular dynamics (MD) simulations fol-
lowed by quantum mechanics/molecular mechanics (QM/MM)
simulations. These theoretical studies were performed with the
AMBER 11 software package45 as described in the ESI.† The
starting binding arrangements of the sulfate and halide com-
plexes were obtained in the gas phase by quenched MD simu-
lations with 7 and the polyatomic anion described with the
general amber force field (GAFF) parameters46 and restrained
electrostatic potential (RESP) atomic charges.47 The monoatomic
anions, with atomic charge set to −1, were described with van der
Waals force field parameters developed for the transferable inter-
molecular potential three point (TIP3P) water model (See ESI†).48
Fig. 5 Illustrative snapshots (QM/MD) of 7·Cl (top, left), 7·Br (top,
right) and 7·I (bottom) complexes showing the secondary anion solvent
shell.
2−
The structures of the halide (F−, Cl−, Br−, and I−) and SO4
complexes determined from gas phase MD simulations (see
6954 | Org. Biomol. Chem., 2012, 10, 6951–6959
This journal is © The Royal Society of Chemistry 2012

View Online
larger standard deviations estimated for H⋯I− distances indicate
the charged assisted hydrogen bonding interactions were more
often interrupted in the iodide complex than in the chloride and
bromide complexes. They also reflect the prominent exposition
of the largest halide anion to the water solvent molecules derived
from the positions occupied by the anion, recurrently outside of
the macrocyclic cavity, throughout the course of the QM/MM
simulation. This structural feature will be further discussed
below. In contrast, the smaller standard deviations for H⋯Br−
distances suggest that among these three monoatomic anions,
bromide is the more tightly bonded to 7. Therefore, the under-
standing of the experimental binding preference of macrocycle
for iodide requires the evaluation of other structural parameters
as demonstrated below.
Fig. 6 Evolution of C–H⋯F− hydrogen bond (blue) distances and
Ht⋯Ht (magenta) distances along the QM/MM simulation of the
fluoride complex.
Conformational changes in the macrocycle 7 were also ascer-
tained by measuring the distances between opposite triazolium
C–H protons (Ht⋯Ht) on the macrocyclic backbone during the
course of the QM/MM simulations. The corresponding average
values are listed in Table 4 together with the distances of the
anions to the mass centre of the macrocycle (A⋯CM) defined
here by all non-hydrogen atoms excluding the methyl substitu-
ents on the triazolium motifs.
monitored, the fluoride complex has a drastic change in one of
these distances as shown in Fig. 6 where they are plotted
together with the four C–H⋯F− distances. During the first
898 ps of QM/MM simulation, the macrocycle exhibits an
almost flattened conformation forming three uninterrupted
hydrogen bonds with fluoride with identical average distances of
1.71, 1.69 and 1.71 Å (Table 3 first row). Subsequently, the
complex undergoes a significant conformational change with 7
wrapping the fluoride anion in a folded fashion until the end of
the simulation. This binding event is accompanied by a decrease
of one Ht⋯Ht distance (Table 4) and concomitant increase of the
number of C–H⋯F− hydrogen bonding interactions from three
to four. These structural findings show that the cavity provided
by the flattened macrocyclic conformation is too large to accom-
modate the fluoride anion. Accordingly, 7 undertakes a confor-
mational change with a substantial decrease of the macrocyclic
cavity size in order to bind the smallest halide anion efficiently
in a binding arrangement composed of four H⋯F− distances
with small standard deviations (see Table 3). This conformation-
al rearrangement is illustrated in Fig. 7 with three sequential
snapshots taken from the QM/MM simulation.
In contrast with the chloride, bromide and iodide complexes,
in which small variations in both Ht⋯Ht distances were
Table 3 Average H⋯A distances (Å) with their standard deviationsa
between the anions and the four C–H triazolium groups
H1⋯A
H2⋯A
H3⋯A
H4⋯A
F−b
1.71 0.08 1.69 0.06 4.40 0.58d 1.71 0.07
1.73 0.07 1.72 0.06
2.22 0.52 2.17 0.38
2.64 0.27 2.68 0.31
3.31 0.75 3.31 0.79
1.73 0.08 1.73 0.07
2.28 0.48 2.36 0.55
2.64 0.28 2.63 0.25
3.12 0.53 3.18 0.76
3.32 0.84 2.94 0.97
3.20 1.01 3.74 0.47
3.69 0.68 3.36 0.99
3.00 1.00 3.11 0.98
Cl−
Br−
I−
SO42−, O1c 3.40 0.75 3.71 1.16
SO42−, O2c 3.13 1.00 2.97 1.18
SO42−, O3c 3.46 0.95 3.90 0.96
c
SO42−, O4
3.15 0.97 3.21 1.04
a The standard deviations were calculated for N = 2000 with the
exception of the values quoted for 7·F, which were calculated with N =
898 for the first 898 ps of simulation and N = 1102 for the remaining
simulation length. b The values in italics are for the folded binding
arrangement. c The H⋯O distances listed for sulfate were calculated
between each C–H triazolium group and the four oxygen atoms. d The
large standard deviation for this measurement indicates that H3 is not
involved in hydrogen bonding interactions.
Further understanding of the binding selectivity of 7 for halide
anions can be acquired by the comparison of the Ht⋯Ht dis-
tances for the tetra-triazolium macrocycle in complexes and in
free form. Hence, the free macrocycle was also submitted to a
QM/MM run using the same simulation protocol. The Ht⋯Ht
average values listed in Table 4 for the free macrocycle are far
from those reported for chloride and fluoride complexes. This is
particularly evident for the latter complex, in which the macro-
cycle adopts a folded conformation to accommodate the smallest
anion. In contrast, in the iodide complex, Ht⋯Ht distances are
much closer to the distances reported for the free macrocycle.
This comparison suggests that the energetic cost associated with
the conformational change of the macrocycle to bind the iodide is
less than for the other halides, leading to a more stable complex.
The binding arrangement determined in the gas phase for the
SO42− complex (see ESI†) was not maintained in solution as can
be seen from the snapshot taken at the beginning of the data col-
lection run presented in Fig. 8. In fact, during the equilibration
Table 4 Average Ht⋯Ht and A⋯CM distances (Å) with their standard
deviationsa
H1⋯H3
H2⋯H4
A⋯CM
7
6.18 0.78
5.96 0.63
3.30 0.19
5.84 0.65
5.29 0.34
6.59 0.86
2.96 0.20
3.37 0.11
5.88 0.73
5.35 0.62
—
F−b
1.29 0.25
0.39 0.25
1.66 0.72
2.27 0.29
I−
SO4
2−
a The standard deviations were calculated for N = 2000 with the
exception of the values quoted for 7·F, which were calculated with N =
898 for the first 898 ps of simulation and N = 1102 for the remaining
simulation length. b The values in italics are for the folded binding
arrangement.
2−
stage, the SO4 left the macrocyclic cavity to form, with the
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6951–6959 | 6955

View Online
Table 5 Average number of solvent molecules enclosing the anions
within the 1st and 2nd shells with radii of 3.4 and 5.0 Å respectivelya
DMSO
1st
H2O
1st
2nd
2nd
F−
0.6 (0–3)
0.8 (0–3)
0.7 (0–3)
0.7 (0–3)
0.5 (0–3)
2.1
2.1
2.5
3.0
1.6
0.5 (0–3)
0.7 (0–3)
1.7 (0–4)
1.5 (0–5)
5.9 (2–10)
3.6
3.8
4.7
3.8
9.3
Cl−
Br−
I−
2−
SO4
a The maximum and minimum number found in both coordination
spheres are given in parentheses.
However, the smaller standard deviations obtained for
C–H⋯Br− hydrogen interactions indicate that the strength of the
interaction between Br− and the macrocycle is stronger than with
Cl−, which is agreement with experimental binding data for
these two anions. The average F−⋯CM distance during the first
898 ps where the macrocycle also adopts a flattened confor-
mation is 1.29 Å in agreement with existence of three C–H⋯F−
interactions only.
Fig. 7 Three representative snapshots of fluoride complex taken at the
simulation times 0.160 (top-left), 0.899 (top-right) and 2 (bottom) ns
showing the different binding arrangements observed throughout the
course of the QM/MM simulation length.
Solvent effects
The impact of the competitive 1 : 1 DMSO–H2O solvent mixture
on the binding affinity of 7 for halides and sulfate anions was
ascertained by calculating the number of water and DMSO mo-
lecules around each anion along the QM/MM simulations
(see Table 5).
It is clear that in 7·Br, 7·I, and 7·SO4 complexes, the anion is
preferentially solvated by water molecules in the first solvation
shell while in 7·F and 7·Cl complexes, the anion is surrounded
by an equivalent smaller number of water and DMSO solvent
molecules. This fact is understandable considering that the two
smaller anions are sheltered from both solvent molecules by the
macrocycle. Indeed, the chloride anion is barely outside of the
flattened macrocyclic conformation while the fluoride anion is
wrapped by the folded macrocyclic conformation during most of
the simulation time. In contrast, I− and SO4 anions, as shown
2−
Fig. 8 Snapshot of 7·SO4 complex taken at the beginning of the pro-
duction run illustrating the binding pose adopted by the sulphate anion
along the QM/MM simulation. The anion is exposed to the solvent mo-
lecules establishing O–H⋯O hydrogen bonds with several waters,
drawn as blue dashed lines.
above, are positioned outside of the macrocycle exposed to the
competitive water molecules, in particular the polyatomic anion.
Nonetheless, the average number of water molecules within the
first solvation shell of bromide (1.7) is slightly higher than in the
iodide (1.5) not reflecting the Br⋯CM and I⋯CM distances, as
would be expected. This surprising result corroborates the
experimental binding data of 7·Br and 7·I complexes (see
Table 2), when they are compared alone.
As expected, in all complexes the extent of anion solvation
increases in the second coordination sphere, which is preferen-
tially composed of water molecules. This is consistent with the
cubic box used to simulate the 1 : 1 DMSO–H2O competitive
solvent mixture, which contained predominantly water mole-
cules, and with the cut-off used to compute the number of both
solvent molecules in the second shell of 5.0 Å, which is proxi-
mal to the bulk solvent. The significant number of DMSO mole-
cules surrounding the anions, in particular in the case of halide
complexes, reflects the solvation of the tetra-triazolium
four C–H triazolium protons, multiple C–H⋯O hydrogen
bonding interactions based on the intermittent swapping of all
oxygen atoms, as evident by the large standard deviations deter-
mined for H⋯O distances measured along the simulation length
(see Table 3).
The average A⋯CM distances listed in Table 4 for 7·Cl, 7·Br
and 7·I complexes reflect the anion size order (I− > Br− > Cl−)
being determined by the fitting between the size of each anion
and the cavity size of the flattened macrocyclic conformation.
2−
Thus, the larger anions I− (1.66 Å) and SO4 (2.27 Å) are
clearly positioned outside of the macrocycle whereas the smaller
anions Br− (0.55 Å) and Cl− (0.49 Å) are barely out of 7.
6956 | Org. Biomol. Chem., 2012, 10, 6951–6959
This journal is © The Royal Society of Chemistry 2012

View Online
macrocyle by a large number of DMSO molecules. This is par-
ticularly evident for the iodide complex with an average of three
DMSO molecules protecting the iodide from extensive solvation
by coordinating water molecules.
500 MHz. Mass spectra were recorded on a Bruker microTOF
spectrometer. Copies of NMR spectra and titration protocols are
detailed in the ESI.†
The sulfate anion is widely solvated by water molecules in
both solvation shells and consequently, the experimental binding
preference of 7 for the sulfate anion seems to be dictated by the
net charges of both partners, in other words, by the strength of
Phenyl bis-triazolyl azide, 5. 1,3-Diethylnyl benzene
(0.359 mL, 0.341 g, 2.70 mmol) and tosyl-protected azide, 4,
(1.45 g, 5.67 mmol, 2.1 equiv) were dissolved in dry dichloro-
methane (60 mL). DIPEA (1.1 mL, 0.78 g, 6.0 mmol), TBTA
(0.287 g, 0.540 mmol) and [CuI(CH3CN)4(PF6) (0.202 g,
0.540 mmol) were added and the resulting pale yellow solution
stirred at room temperature under a nitrogen atmosphere for 4
days. It was taken to dryness under reduced pressure and purified
by column chromatography (3% methanol in dichloromethane)
to give a pale yellow foam, which was used immediately to
prepare 5.
the electrostatic interactions between charged tetra-imidazolium
2−
macrocycle (+4) and SO4
.
In conclusion the modelling studies show that the affinity of 7
for the different anions is governed by multiple charged assisted
C–H⋯A− interactions in which the steric requirements imposed
by the apparently rigid macrocyclic backbone together with sol-
vation effects seem to have an important role.
1H NMR (CDCl3): 8.20 (s, 2H, 2 × trz-H), 7.75–7.83 (m, 7H,
3
4 × Ts-H, 3 × Ph-H), 7.50 (t, J = 7.7 Hz, 1H, Ph-H), 7.33 (d,
Conclusions
3J = 8.0 Hz, 4H, 4 × Ts-H), 4.51 (t, 3J = 6.6 Hz, 4H, 4 ×
CH2-trz), 4.06 (t, 3J = 5.7 Hz, 4H, 4 × TsO-CH2), 2.31–2.43 (m,
10H, 6 × CH3, 4 × CH2-CH2-CH2). LRESI-MS (pos.): 659.26,
calc. for [C30H32N6O6S2·Na]+ = 659.17.
We have prepared three tetra-triazole macrocycles in good yield
using a CuAAC cyclisation of triazole bis-azides and bis-
alkynes. One of these macrocycles was successfully alkylated to
give the tetra-triazolium host, 7·BF4. This receptor binds anions
strongly by charge-assisted C–H⋯anion hydrogen bonding,
even in the competitive 1 : 1 d6-DMSO–D2O aqueous solvent
mixture. Among the halides, iodide and bromide associate more
strongly than chloride and fluoride, and all halides form stronger
complexes than acetate. The sulfate dianion is bound extremely
strongly, with an association constant greater than 104 M−1. The
observed solution anion binding trends are supported by exten-
sive molecular modelling analysis, which suggest macrocycle
cavity size, solvation effects and strength of electrostatic inter-
actions significantly influence the observed anion recognition
processes. Hence, this paper demonstrates that the incorporation
of multiple triazolium motifs into a cyclic host framework pro-
duces a potent anion receptor that functions in aqueous solvent
mixtures.
The ditosylate, was dissolved in DMSO (25 mL). Sodium
azide (0.478 g, 7.35 mmol, 2.50 equiv) was added and the
mixture heated at 60 °C under a nitrogen atmosphere over the
weekend. It was cooled to room temperature, and partitioned
between water and (100 mL each). The organic layer was taken
and the aqueous layer extracted further with (100 mL). The com-
bined organic fractions were washed with water (100 mL), satu-
rated brine (100 mL) and dried (magnesium sulfate), taken to
dryness under reduced pressure and purified by column
chromatography (2% methanol in dichloromethane) to give 5
as a pale yellow foamy solid. Yield: 0.773 g (76% from
1,3-diethynylbenzene).
1H NMR (CDCl3): 8.29 (s, 2H, 2 × 2 × trz-H), 7.82–7.87 (m,
3
3
3H, 3 × Ph-H), 7.51 (t, J = 7.7 Hz, 1H, Ph-H), 4.53 (t, J =
6.7 Hz, 4H, 4 × CH2-trz), 3.41 (t, 3J = 6.3 Hz, 4H, 4 × CH2-N3),
2.17–2.31 (m, 4H, CH2-CH2-CH2). 13C NMR (CDCl3): 147.3,
131.1, 129.5, 125.3, 122.8, 120.4, 48.0, 47.3, 29.4. LRESI-MS
(pos.): 401.18, calc. for [C16H18N12·Na]+ = 401.17; 779.38,
calc. for [C32H36N24·Na]+, i.e. [M2·Na]+ = 779.35. HRESI-MS
(pos.): 401.1667, calc for [C16H18N12·Na]+ = 401.1670.
Experimental
General remarks
CAUTION: Low molecular weight organic azides are poten-
tially explosive. While no problems were encountered in the
course of this work, they should be handled in small quantities
and with appropriate care.
Propyl bis-triazolyl azide, 6. 1,6-Heptadiyne (0.103 mL,
0.083 g, 0.90 mmol) and tosyl-protected azide, 4, (0.482 g,
1.89 mmol, 2.1 equiv.) were dissolved in dichloromethane
(25 mL). DIPEA (0.35 mL, 0.26 g, 2.00 mmol), TBTA (0.096 g,
0.18 mmol) and [CuI(CH3CN)4(PF6) (0.067 g, 0.18 mmol) were
added and the resulting pale yellow solution stirred at room
temperature under a nitrogen atmosphere overnight. It was taken
to dryness under reduced pressure and purified by column
chromatography (2% methanol in chloroform) to give a yellow
oil. This was used immediately to prepare 6.
TBTA51 and 3-azidopropoxytosylate, 4,41 were prepared as
previously described. 3,5-Diethynylpyridine was prepared by
deprotecting 3,5-bis(trimethylsilylethynyl)-pyridine52 using
KOH in methanol.53 Other chemicals were available commer-
cially and used as received. Where solvents are specified as
“dry”, they were purged with nitrogen and passed through an
MBraun MPSP-800 column. Water was de-ionised and microfil-
tered using a Milli-Q Millipore machine. Tetrabutylammonium
salts were stored in a vacuum dessicator. All chromatography
was performed on silica gel (particle size: 40–63 μm).
1H NMR (CDCl3): 7.75–7.79 (m, 4H, 4 × Ts-H), 7.33–7.37
3
(m, 4H, 4 × Ts-H), 7.29 (s, 2H, 2 × trz-H), 4.40 (t, J = 6.7 Hz,
3
4H, 4 × trzN-CH2), 4.00 (t, J = 5.7 Hz, 4H, 4 × TsO-CH2),
3
Routine NMR spectra were recorded on a Varian Mercury 300
spectrometer with 1H NMR operating at 300 MHz, 13 °C at
75.5 MHz. Spectra for 1H NMR titrations were recorded on a
Varian Unity Plus 500 spectrometer with 1H operating at
2.73 (t, J = 7.5 Hz, 4H, 4 × trzC-CH2), 2.44 (s, 6H, 6 × CH3),
2.23–2.32 (m, 4H, 4 × TsO-CH2-CH2), 1.97–2.07 (m, 2H,
2 × trzC-CH2-CH2). LRESI-MS (pos.): 625.23, calc. for
[C27H34N6O6S2·Na]+ = 625.19.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6951–6959 | 6957

View Online
The ditosylate, was dissolved in dimethylsulfoxide (10 mL).
Sodium azide (0.234 g, 3.60 mmol) was added and the mixture
heated at 60 °C under a nitrogen atmosphere over the weekend.
It was cooled to room temperature, and partitioned between
water and diethyl ether (100 mL each). The organic layer was
taken and the aqueous layer extracted further with diethyl ether
(100 mL). The combined organic fractions were dried (mag-
nesium sulfate), taken to dryness under reduced pressure and
purified by column chromatography (1.5% methanol in dichloro-
methane) to give 6 as a pale yellow oil. Yield: 0.093 g (30%
from 1,6-heptadiyne).
[C25H23N13·Na]+ = 528.21; 1033.55, calc. for [C50H46N26·Na]+,
i.e. [M2·Na]+ = 1033.43. LRESI-MS (neg.): 540.19, calc. for
[C25H23N13·Cl]− = 540.19; 1045.42, calc. for [C50H46N26·Cl]−,
i.e. [M2·Cl]− = 1045.41. HRESI-MS (pos.): 528.2081, calc. for
[C25H23N13·Na]+ = 528.2092.
Propyl-linked tetra-triazole macrocycle, 1. The bis-azide, 6
(0.069 g, 0.20 mmol) and 1,6-heptadiyne (0.023 mL, 0.018 g,
0.20 mmol) were dissolved in dichloromethane (200 mL).
DIPEA (0.035 mL, 0.026 g, 0.20 mmol), TBTA (0.021 g,
0.040 mmol) and [CuI(CH3CN)4](PF6) (0.015 g, 0.040 mmol)
were added and the yellow solution stirred at room temperature
under a nitrogen atmosphere for 3 days. It was concentrated to
∼50 mL under reduced pressure and filtered, with the precipitate
washed thoroughly with 10% methanol in dichloromethane. The
combined filtrates were taken to dryness under reduced pressure
and purified by column chromatography (gradient: 5–15%
methanol in chloroform) to give 1 as a white powder. Yield:
0.043 g (49%).
3
1H NMR (CDCl3): 7.37 (s, 2H, 2 × trz-H), 4.47 (t, J = 6.6
3
Hz, 4H, 4 × trzN-CH2), 3.41 (t, J = 6.4 Hz, 4H, 4 × CH2-N3),
2.73 (t, 3J = 8.8 Hz, 4H, 4 × trzC-CH2), 1.95–2.25 (m, 6H,
4 × N3-CH2-CH2, 2 × trzC-CH2-CH2). 13C NMR (CDCl3):
147.8, 121.5, 48.2, 47.1, 29.6, 29.2, 25.0. LRESI-MS (pos.):
367.19, calc. for [C13H20N12·Na]+ = 367.18; 711.40, calc. for
[C26H40N24·Na]+, i.e. [M2·Na]+ = 711.38. HRESI-MS (pos.):
367.1826, calc. for [C13H20N12·Na]+ = 367.1826.
1H NMR (CDCl3): 7.41 (s, 4H, 4 × trz-H), 4.29 (t, 3J =
6.2 Hz, 8H, 8 × trzN-CH2), 2.64–2.73 (m, 12H, 8 × trzC-CH2,
Phenyl-linked tetra-triazole macrocycle, 2. The bis-azide, 5,
(0.151 g, 0.400 mmol) and 1,3-diethynylbenzene (0.0532 mL,
0.0505 g, 0.400 mmol) were dissolved in dichloromethane
(400 mL). DIPEA (0.17 mL, 0.13 g, 1.0 mmol), TBTA (0.042 g,
0.080 mmol) and [CuI(CH3CN)4](PF6) (0.030 g, 0.080 mmol)
were added and the pale yellow solution stirred for 4 days under
a nitrogen atmosphere, during which time a fluffy yellow precipi-
tate developed. The reaction mixture was taken to dryness under
reduced pressure and purified by column chromatography (gradi-
ent: 5–10% methanol in dichloromethane) to give 2 as a white
powder. Yield: 0.136 g (67%).
3
4 × trzN-CH2), 2.09 (qn, J = 6.9 Hz, 4H, 4 × trzC-CH2). 13C
NMR (CDCl3): 147.3, 122.9, 46.5, 29.5, 28.3, 23.9. LRESI-MS
(pos.): 459.28, calc. for [C20H28N12·Na]+ = 459.25; 895.59,
calc. for [C40H56N24·Na], i.e. [M2·Na]+ = 895.50. LRESI-MS
(neg): 471.23, calc. for [C20H28N12·Cl]− = 471.23. HRESI-MS
(pos.): 459.2455, calc. for [C20H28N12·Na]+ = 459.2452.
Propyl-linked tetra-triazolium macrocycle,
7 (BF4)4. The
neutral macrocycle, 1 (0.022 g, 0.050 mmol) was dissolved in
dry dichloromethane (20 mL). Trimethyloxonium tetrafluoro-
borate (0.037 g, 0.25 mmol) was added and the mixture stirred at
room temperature under a nitrogen atmosphere for 2 days.
Methanol (1 mL) was added and the solution taken to dryness
under reduced pressure. The crude solid was stirred in dichloro-
methane (15 mL); the supernatant fluid was decanted and the
resulting white powder washed with further dichloromethane
(10 mL). The solid was recrystallised from boiling 6 : 1 ethanol–
acetonitrile to give 7·4(BF4)4 as large white needle crystals.
Yield: 0.023 g (54%).
4
1H NMR (d6-DMSO): 8.11 (s, 4H, 4 × trz-H), 7.98 (t, J =
3
4
1.4 Hz, 2H, 2 × Ph-H), 7.56 (dd, J = 7.5 Hz, J = 1.4 Hz, 4H,
3
3
4 × Ph-H), 7.30 (t, J = 7.5 Hz, 2H, 2 × Ph-H), 4.53 (t, J =
3
5.3 Hz, 8H, 8 × trz-CH2), 2.68 (qn, J = 5.3 Hz, 4H, 4 × CH2-
CH2). 13C NMR (d6-DMSO): 146.1, 130.8, 128.9, 123.9, 122.4,
121.3, 47.9, 29.2. LRESI-MS (neg.): 539.18, calc. for
[C26H24N12·Cl]− = 539.19. HRESI-MS (neg.): 539.1936, calc.
for [C26H24N12·Cl]− = 539.1941.
3
1H NMR (d6-DMSO): 8.47 (s, 4H, 4 × trz+-H), 4.69 (t, J =
Pyridyl-linked tetra-triazole macrocycle, 3. The bis-azide, 5,
(0.265 g, 0.700 mmol) and 3,5-diethynylpyridine (0.0890 g,
0.700 mmol) were dissolved in dichloromethane (700 mL).
DIPEA (0.31 mL, 0.24 g, 1.0 mmol), TBTA (0.074 g,
0.14 mmol) and [CuI(CH3CN)4](PF6) (0.052 g, 0.14 mmol)
were added and the pale yellow solution stirred for three days
under a nitrogen atmosphere at room temperature, during which
time a pale precipitate developed. It was taken to dryness under
reduced pressure and purified by column chromatography (gradi-
ent: 10–15% dichloromethane in methanol) to give 3 as a white
powder. Yield: 0.208 g (59%).
7.0 Hz, 8H, 8 × trzN-CH2), 4.22 (s, 12H, 12 × CH3), 2.96 (t,
3J = 7.0 Hz, 8H, 8 × trzC-CH2), 2.48–2.54 (m, obscured by
DMSO solvent peak, 4 × trzN-CH2-CH2), 2.02–2.08 (m, 4H,
4
×
trzC-C2-CH2). 19F NMR (d6-DMSO): −148.4 (m).
13C NMR (d6-DMSO): 143.8, 128.0, 49.5, 37.5, 28.8, 23.8,
21.6. LRESI-MS (pos.): 335.18, calc. for [(C24H40N12)·2
(BF4)]2+ = 335.18; 757.38, calc. for [(C24H40N12)·3(BF4)]+
=
757.36. HRESI-MS (pos.): 757.3593, calc. for [(C24H40N12)·
3(BF4)]+ = 757.3581.
1H NMR (d6-DMSO): 8.56 (br. s, 2H, 2 × py-H2,6), 8.10 (t,
4J = 1.9 Hz, 1H, py-H4), 8.05 (s, 2H, 2 × trz-H), 7.91 (s, 2H,
2 × trz-H), 7.84 (t, 4J = 1.5 Hz, 1H, Ph-H), 7.36 (dd, 3J =
X-ray crystallography
Single crystal X-ray diffraction data for 2 from DMSO–CH2Cl2
and from methanol–water were collected using graphite mono-
chromated Mo Kα radiation (λ = 0.71073 Å) on a Nonius Kappa
CCD diffractometer equipped with a Cryostream N2 open-flow
cooling device,54 and the data were collected at 150(2) K. Series
of ω-scans were performed in such a way as to collect all unique
4
3
7.7 Hz, J = 1.5 Hz, 2H, 2 × Ph-H), 7.20 (t, J = 7.7 Hz, 2 ×
Ph-H), 4.58–4.64 (m, 8H, 8 × trz-CH2, 4 × Hg), 2.66–2.76 (m,
4H, 4 × trz-CH2-CH2). 13C NMR (d6-DMSO): 146.0, 144.8,
143.2, 130.6, 128.7, 128.3, 126.2, 124.0, 121.6, 121.3, 120.9,
48.8, 48.6, 28.8. LRESI-MS (pos.): 528.25, calc. for
6958 | Org. Biomol. Chem., 2012, 10, 6951–6959
This journal is © The Royal Society of Chemistry 2012

View Online
21 Y. Hua, R. Ramabhadran, J. Karty, K. Raghavachari and A. Flood, Chem.
Commun., 2011, 47, 5979–5981.
22 S. Kubik, R. Goddard, R. Kirchner, D. Nolting and J. Seidel, Angew.
Chem., Int. Ed., 2001, 40, 2648–2651.
23 S. Kubik and R. Goddard, Proc. Natl. Acad. Sci. U. S. A., 2002, 99,
5127–5132.
24 I. E. Drubi Vega, S. Camiolo, P. A. Gale, M. B. Hursthouse and M.
E. Light, Chem. Commun., 2003, 1686–1687.
25 S. Kubik, R. Kirchner, D. Nolting and J. Seidel, J. Am. Chem. Soc.,
2002, 124, 12752–12760.
26 Z. Rodriguez-Docampo, S. Pascu, S. Kubik and S. Otto, J. Am. Chem.
Soc., 2006, 128, 11206–11210.
27 R. Yang, W.-X. Liu, H. Shen, H.-H. Huang and Y.-B. Jiang, J. Phys.
Chem. B, 2008, 112, 5105–5110.
28 C. Caltagirone, J. R. Hiscock, M. B. Hursthouse, M. E. Light and
P. A. Gale, Chem.–Eur. J., 2008, 14, 10236–10243.
reflections to a maximum of 0.80 Å. Cell parameters and inten-
sity data (including inter-frame scaling) were processed using the
DENZO-SMN package.55 Single crystal X-ray diffraction data
for 1 and 7·BF4 were collected using graphite monochromated
Cu Kα radiation (λ = 1.54184 Å) on an Oxford Diffraction
SuperNova diffractometer. The diffractometer was equipped with
a Cryostream N2 open-flow cooling device,54 and the data were
collected at 150(2) K. Series of ω-scans were performed in such
a way as to collect all unique reflections to a maximum of
0.80 Å. Cell parameters and intensity data (including inter-frame
scaling) were processed using CrysAlis Pro.56 The structures
were solved by charge-flipping methods using SUPERFLIP57
and refined using full-matrix least-squares on F2 within the
CRYSTALS suite.58 All non-hydrogen atoms were refined with
anisotropic displacement parameters. Hydrogen atoms were
generally visible in the difference map and their positions and
displacement parameters were refined using restraints prior to
inclusion into the model using riding constraints.59 Crystallo-
graphic data for the structures have been deposited with
the Cambridge Crystallographic Data Centre, CCDC:
881646–881649.†
29 R. Vilar, Eur. J. Inorg. Chem., 2008, 357–367.
30 P. Ballester, Chem. Soc. Rev., 2010, 39, 3810–3830.
31 S. Kubik, Chem. Soc. Rev., 2010, 39, 3648–3663.
32 D. Mercer and S. Loeb, Chem. Soc. Rev., 2010, 39, 3612–3620.
33 W. W. H. Wong, M. S. Vickers, A. R. Cowley, R. L. Paul and P. D. Beer,
Org. Biomol. Chem., 2005, 3, 4201–4208.
34 A. Kumar and P. S. Pandey, Org. Lett., 2008, 10, 165–168.
35 K. M. Mullen, J. Mercurio, C. J. Serpell and P. D. Beer, Angew. Chem.,
Int. Ed., 2009, 48, 4781.
36 B. Schulze, C. Friebe, M. Hager, W. Günther, U. Köhn, B. Jahn, H. Görls
and U. Schubert, Org. Lett., 2010, 12, 2710–2713.
37 K. Ohmatsu, M. Kiyokawa and T. Ooi, J. Am. Chem. Soc., 2011, 133,
1307–1309.
38 R. Chhatra, A. Kumar and P. S. Pandey, J. Org. Chem., 2011, 76,
9086–9089.
Acknowledgements
We thank Oxford Chemical Crystallography for the use of instru-
ments, and Dr Amber L. Thompson for helpful discussions.
NGW thanks the Clarendon Fund and Trinity College, Oxford
for funding. SC thanks the FCT for the post-doctoral fellowship
SFRH /BPD/42357/2007.
39 L. C. Gilday, N. G. White and P. D. Beer, Dalton Trans., 2012, 41,
7092–7097.
40 N. G. White and P. D. Beer, Beilstein J. Org. Chem., 2012, 8, 246–252.
41 V. Aucagne, K. Hanni, D. Leigh, P. Lusby and D. Walker, J. Am. Chem.
Soc., 2006, 128, 2186–2187.
42 M. J. Hynes, J. Chem. Soc., Dalton Trans., 1993, 311–312.
43 Presumably this is at least partly due to the tight ion-pairing of TBA salts
in chlorinated solvents. See ref. 16 and S. Alunni, A. Pero and
G. Reichenbach, J. Chem. Soc., Perkin Trans. 2, 1998, 1747–1750.
44 Kim has recently reported a larger tetra-imidazolium receptor, which
binds iodide in preference to the smaller halides or acetate in 9 : 1
CD3CN–D2O. V. Suresh, N. Ahmed, I. S. Youn and K. S. Kim, Chem.–
Asian J., 2012, 7, 658–663.
45 D. A. Case, T. A. Darden, T. E. Cheatham, III, C. L. Simmerling,
J. Wang, R. E. Duke, R. Luo, M. Crowley, R. C. Walker, W. Zhang,
K. M. Merz, B. Wang, S. Hayik, A. Roitberg, G. Seabra, I. Kolossváry,
K. F. Wong, F. Paesani, J. Vanicek, X. Wu, S. R. Brozell, T. Steinbrecher,
H. Gohlke, L. Yang, C. Tan, J. Mongan, V. Hornak, G. Cui,
D. H. Mathews, M. G. Seetin, C. Sagui, V. Babin and P. A. Kollman,
AMBER11, 2010.
46 J. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and D. Case,
J. Comput. Chem., 2004, 25, 1157–1174.
47 R. Bayly, P. Cieplak, W. Cornell and P. Kollman, J. Phys. Chem., 1993,
97, 10269–10280.
48 I. S. Young and T. E. Cheatham, III, J. Phys. Chem. B, 2008, 112,
9020–9041.
49 T. Fox and P. A. Kollman, J. Phys. Chem. B, 1998, 102, 8070–8079.
50 W. L. Jorgensen, J. Chandrasekhar, J. D. Madura, R. Impey and
M. L. Klein, J. Chem. Phys., 1983, 79, 926–935.
References
1 V. Rostovtsev, L. Green, V. Fokin and K. Sharpless, Angew. Chem., Int.
Ed., 2002, 41, 2596–2599.
2 C. Tornoe, C. Christensen and M. Meldal, J. Org. Chem., 2002, 67,
3057–3064.
3 Y. Li and A. H. Flood, Angew. Chem., Int. Ed., 2008, 47, 2649–2652.
4 Y. Li and A. H. Flood, J. Am. Chem. Soc., 2008, 130, 12111–12222.
5 H. Juwarker, J. M. Lenhardt, D. M. Pham and S. L. Craig, Angew.
Chem., Int. Ed., 2008, 47, 3740–3743.
6 R. M. Meudtner and S. Hecht, Angew. Chem., Int. Ed., 2008, 47,
4926–4930.
7 H. Juwarker, J. M. Lenhardt, J. C. Castillo, E. Zhao, S. Krishnamurthy,
R. M. Jamiolkowski, K. H. Kim and S. L. Craig, J. Org. Chem., 2009,
74, 8924–8934.
8 Y. Wang, F. Bie and H. Jiang, Org. Lett., 2010, 12, 3630–3633.
9 S. Lee, Y. Hua and A. Flood, Org. Lett., 2010, 12, 2100–2102.
10 Y. Hua and A. Flood, J. Am. Chem. Soc., 2010, 132, 12838–12840.
11 Q.-Y. Cao, T. Pradhan, S. Kim and J. Kim, Org. Lett., 2011, 13,
4386–4389.
12 Y. Wang, J. Xiang and H. Jiang, Chem.–Eur. J., 2011, 17, 613–619.
13 K. McDonald, R. Ramabhadran, S. Lee, K. Raghavachari and A. Flood,
Org. Lett., 2011, 13, 6260–6263.
14 Y. Li, M. Pink, J. A. Karty and A. H. Flood, J. Am. Chem. Soc., 2008,
130, 17293–17295.
51 B.-Y. Lee, S. R. Park, H. B. Jeon and K. S. Kim, Tetrahedron Lett., 2006,
47, 5105–5109.
52 H. Goto, J. Heemstra, D. Hill and J. Moore, Org. Lett., 2004, 6,
889–892.
53 T. X. Neenan and G. M. Whitesides, J. Org. Chem., 1988, 53,
2489–2496.
15 R. O. Ramabhadran, Y. Hua, Y.-j. Li, A. H. Flood and K. Raghavachari,
Chem.–Eur. J., 2011, 17, 9123–9129.
54 J. Cosier and A. M. Glazer, J. Appl. Crystallogr., 1986, 19, 105–107.
55 Z. Otwinowski and W. Minor, in Methods Enzymol., ed. C. W. Carter and
R. M. Sweet, Academic Press, 1997, ch. 276.
16 Y. Hua, R. O. Ramabhadran, E. O. Uduchi, J. A. Karty, K. Raghavachari
and A. H. Flood, Chem.–Eur. J., 2011, 17, 312–321.
17 V. Haridas, S. Sahu and P. Venugopalan, Tetrahedron, 2011, 67, 727–733.
18 M. R. Krause, R. Goddard and S. Kubik, J. Org. Chem., 2011, 76,
7084–7095.
19 Y. Lau, P. Rutledge, M. Watkinson and M. Todd, Chem. Soc. Rev., 2011,
40, 2848–2866.
20 C.-H. Lee, S. Lee, H. Yoon and W.-D. Jang, Chem.–Eur. J., 2011, 17,
13898–13903.
56 Agilent Technologies, CrysAlisPro, 2011.
57 L. Palatinus and G. Chapuis, J. Appl. Crystallogr., 2007, 40, 786–790.
58 P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout and
D. J. Watkin, J. Appl. Crystallogr., 2003, 36, 1487.
59 R. I. Cooper, A. L. Thompson and D. J. Watkin, J. Appl. Crystallogr.,
2010, 43, 1017–1022.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 6951–6959 | 6959