Bis-triazolium containing macrocycles, pseudorotaxanes and interlocked structures for anion recognition

Article abstract of DOI:10.1039/c4ra00615a

Two bis-triazole macrocycles were prepared by the copper(i)-catalysed azide alkyne cycloaddition of a bis-azide and 1,6-heptadiyne or 1,3-diethynylbenzene. Upon alkylation of one of these to give the dicationic bis-triazolium cycle, a potent receptor was generated which was demonstrated to be selective for sulfate. Two acyclic bis-triazolium receptors were also synthesized and the halide anion recognition properties of these systems investigated, as well as their ability to form anion templated pseudorotaxanes with an isophthalamide macrocycle. Interestingly, the propyl-linked bis-triazolium host displayed stronger anion binding affinities than the analogous 1,3-phenyl-linked system, although the phenyl-linked receptor formed a more stable halide anion templated pseudorotaxane. An acyclic bis-triazolium threading component produced a stable sulfate anion templated pseudorotaxane with a bis-triazolium macrocycle, which was used to synthesise a tetra-triazolium catenane. An anion templated stoppering strategy was used to prepare a bis-triazolium rotaxane host system which binds halide anions selectively over dihydrogenphosphate in competitive 1:1 CDCl₃:CD₃OD solvent media.

Full text of DOI:10.1039/c4ra00615a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Bis-triazolium containing macrocycles,
pseudorotaxanes and interlocked structures for
Cite this: RSC Adv., 2014, 4, 12133
anion recognition†
*
Nicholas G. White, Henry G. Lovett and Paul D. Beer
Two bis-triazole macrocycles were prepared by the copper(I)-catalysed azide alkyne cycloaddition of a bis-
azide and 1,6-heptadiyne or 1,3-diethynylbenzene. Upon alkylation of one of these to give the dicationic
bis-triazolium cycle, a potent receptor was generated which was demonstrated to be selective for
sulfate. Two acyclic bis-triazolium receptors were also synthesized and the halide anion recognition
properties of these systems investigated, as well as their ability to form anion templated pseudorotaxanes
with an isophthalamide macrocycle. Interestingly, the propyl-linked bis-triazolium host displayed
stronger anion binding affinities than the analogous 1,3-phenyl-linked system, although the phenyl-
linked receptor formed a more stable halide anion templated pseudorotaxane. An acyclic bis-triazolium
threading component produced a stable sulfate anion templated pseudorotaxane with a bis-triazolium
macrocycle, which was used to synthesise a tetra-triazolium catenane. An anion templated stoppering
strategy was used to prepare a bis-triazolium rotaxane host system which binds halide anions selectively
over dihydrogenphosphate in competitive 1 : 1 CDCl3 : CD3OD solvent media.
Received 21st January 2014
Accepted 17th February 2014
DOI: 10.1039/c4ra00615a
www.rsc.org/advances
Introduction
In 2008, the groups of Craig,1 Flood,2 and Hecht3 independently
demonstrated that oligomeric or macrocyclic hosts containing
the readily-prepared 1,2,3-triazole4 motif bind anions purely
through polarised C–H/anion interactions. More recently, this
heterocycle has been incorporated into a range of receptor
structural frameworks that display efficient anion recognition
properties in organic solvents.5,6 Also in 2008, Pandey's group7
introduced the alkylated cationic triazolium heterocycle, which
proved to be an even more potent anion recognition motif,
benetting from a polarised C–H hydrogen bond donor, as well
as electrostatic attraction for negatively charged species.
Fig. 1 Bis-triazolium motifs used in this work.
acetone : water mixtures.8g These systems demonstrate the
strong binding that can be achieved using triazolium groups in
competitive aqueous solvent media. An alternative approach
has seen the integration of a single triazolium group into the
axle component of rotaxane host systems8a,h,k whose preorgan-
ised three-dimensional binding cavities selectively recognise
halides in competitive organic solvents.
Herein we detail an investigation of acyclic, macrocyclic,
interpenetrated and interlocked hosts containing bis-tri-
azolium motifs (Fig. 1). The ultimate aim was to combine the
impressive binding strength of multiply positively charged poly-
triazolium systems with the high level of degree of selectivity
exhibited by interlocked host structures.
While the development of triazolium containing receptors8
is still in its infancy, a number of notable host systems have
been prepared recently.9 Simple tetra-triazolium macrocycles
capable of binding anions in 1 : 1 d6-DMSO : D2O,8i have been
constructed while
complexes larger oxoanions in 2 : 3 acetone : water,8j and a
tetra-triazolium porphyrin cage senses anions in
a
tetra-triazolium tetra-pyrrole cycle
Chemistry Research Laboratory, Department of Chemistry, University of Oxford,
Manseld Road, Oxford, OX1 3TA, UK. E-mail: paul.beer@chem.ox.ac.uk; Fax: +44
(0)1865 272690; Tel: +44 (0)1865 285142
Results and discussion
† Electronic supplementary information (ESI) available: NMR spectra of new
compounds, synthesis and characterisation of additional compounds, ROESY Acyclic and macrocyclic receptors
NMR spectra of 8$Cl$10, 5$SO4$8 and 22$(PF6)2, details of instrumentation,
Synthesis. A range of bis-alkynes are commercially available,
or readily prepared, so a bis-azide precursor was sought to allow
copper(I)-catalysed azide alkyne cycloaddition (CuAAC) reaction
titration protocols and binding isotherms. CCDC 982137–982143. For ESI and
crystallographic data in CIF or other electronic format see DOI:
10.1039/c4ra00615a
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 12133–12147 | 12133

View Article Online
RSC Advances
Paper
with bis-alkynes to prepare bis-triazole macrocycles. This bis-
azide 1, was prepared from known bis-amine 2, using imid-
azole-1-sulfonylazide10 in methanol in 70% yield (Scheme 1).
CuAAC reactions11 of bis-azide 1 with 1,6-heptadiyne or 1,3-
diethynylbenzene were conducted in dichloromethane using
catalytic [Cu(CH3CN)4](PF6), TBTA and DIPEA12 to give bis-tri-
azole macrocycles 3 and 4 (TBTA ¼ tris[{1-benzyl-1,2,3-triazol-4-
yl}methyl]amine, DIPEA ¼ diisopropylethylamine, Scheme 2).
Yields were reasonably high for the macrocyclisation involving
the diethynyl benzene (45%), but more modest when exible
1,6-heptadiyne was used (25%). The new macrocycles were
1
characterised by H and 13C NMR spectroscopy, as well as by
high resolution ESI mass spectrometry and X-ray crystallog-
raphy (see ESI† for 1H and 13C NMR spectra of all new
compounds).
Single crystals of macrocycles 3 and 4 were obtained, allow-
ing determination of the solid state structures by X-ray crystal-
lography. In both structures (Fig. 2), the triazole groups adopt a
syn–syn conguration, with the triazole protons pointing into
the macrocycle cavity. The bis-triazole macrocycles are signi-
cantly shorter and wider in the solid state than analogous iso-
phthalamide macrocycles.13
Fig. 2 Solid state structures of 3 and 4. Solvent molecules and most
hydrogen atoms are omitted for clarity; only one of the two inde-
pendent macrocycles in the crystallographic asymmetric unit is shown
in each case.
Macrocycle 3 was alkylated using two equivalents of trime-
thyloxonium tetrauoroborate in dry dichloromethane to give
the bis-triazolium cycle 5$(BF4)2 (Scheme 3). The solid state
structure of 5$(BF4)2 (Fig. 3) reveals that one of the tetra-
uoroborate anions is encapsulated within the cycle, forming
Scheme 3 Synthesis of bis-triazolium macrocycle 5$(BF4)2.
Scheme 1 Synthesis of bis-azide 1.
Fig. 3 Solid state structure of 5$(BF4)2. Solvent molecules and most
Scheme 2 Synthesis of bis-triazole macrocycles 3 and 4.
12134 | RSC Adv., 2014, 4, 12133–12147
hydrogen atoms are omitted for clarity.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Table 1 Chloride anion association constants of macrocycles 3 and 4
in d6-acetone
moderate hydrogen bonds with the macrocycle's triazolium
ꢀ
protons [C–H/Fꢀ ¼ 3.232(7), 3.271(7) A; :C–H/F ¼ 161,
˚
169ꢁ]. A strong anion–ꢀp interaction is also present between the
Macrocycle
Kaa (Mꢀ1
)
non-encapsulated BF4 anion and one of the triazolium rings
ꢀ
˚
[F /centroid ¼ 2.803(6) A].
3
4
93(8)
32(3)
Two acyclic bis-triazolium receptors were also prepared, with
the triazolium heterocycles joined by either a phenyl or propyl
group. CuAAC reaction of 1,3-diethynylbenzene or 1,6-hepta-
diyne with octyl azide gave phenyl- and propyl-linked bis-tri-
azole precursors 6 and 7 in good yields (75 and 74%,
respectively, Scheme 4). Alkylation using trimethyloxonium
tetrauoroborate in dry dichloromethane gave the bis-tri-
azolium receptors 8$(BF4)2 and 9$(BF4)2.
a
Chloride anion added as tetrabutylammonium (TBA) salt, estimated
standard errors are given in parentheses. Association constants
calculated using WINEQNMR214 (293 K).
1 : 1 stoichiometric anion association constants were deter-
mined using WINEQNMR2,14 monitoring the macrocycle's tri-
azolium resonance (Table 2, Fig. S50†). A range of anions are
bound by the macrocycle, with anion binding generally
following basicity trends.16 Addition of dihydrogenphosphate or
sulfate anions caused rapid precipitation, preventing determi-
nation of quantitative data.
The binding of sulfate was investigated further in d6-DMSO,
where very large downeld shis in the macrocycle's interior
phenylene and triazolium signals were observed (>1 ppm
movement of triazolium resonance aer 1 equivalent of
oxoanion). Indeed, association of sulfate with 5$(BF4)2 in
d6-DMSO was too strong for WINEQNMR214 to calculate an
association constant (Ka > 104 Mꢀ1), despite the competitive
nature of this solvent.
Anion recognition properties. The chloride anion recogni-
tion properties of the neutral macrocyclic receptors 3 and 4 were
investigated by 1H NMR titration experiments in d6-acetone.
Addition of the halide anion caused downeld shis in the
triazole proton resonances, as well as the interior phenylene
proton of the aryl-substituted host. Quantitative 1 : 1 stoichio-
metric association constants were determined using
WINEQNMR2,14 monitoring the triazole resonance. In both
cases, chloride association was found to be weak (Table 1,
Fig. S49†), and of a similar magnitude to previously reported
acyclic bis-triazole receptors.15
The anion recognition properties of 5$(BF4)2 were investi-
gated using 1H NMR titration experiments in CD3CN [initial
experiments showed that chloride anion binding was too strong
to quantify (Ka > 104 Mꢀ1) in d6-acetone]. Downeld shis in the
triazolium and interior phenylene proton resonances were
typically observed upon the addition of anions. Quantitative
Single crystals of the chloride, iodide and azide salts of 52+
were obtained, suitable for X-ray diffraction studies (Fig. 4). In
each case, one anion associates within the macrocycle cavity,
while the other is located peripherally. In the case of the halide
anion salts, the anion located in the cavity forms a strong
hydrogen bond to one triazolium group, but is not involved in
signicant hydrogen bonding to the other triazolium hetero-
cycle or the macrocycle's interior phenylene proton17 [C–H/Clꢀ
ꢀ
ꢀ
¼ 3.298(7) A, :C–H/Cl ¼ 173ꢁ; C–H/I ¼ 3.699(8) A, :C–
H/Iꢀ ¼ 165ꢁ]. The other halide anion forms weak hydrogen
bonds to solvent molecules and a methyl proton from the tri-
azolium-methyl group. In the structure of 5$(N3)2, an unusual
hydrogen bonding arrangement occurs, where the encapsulated
azide anion bonds to both triazolium groups, with one C–H/N
˚
˚
Table 2 Anion association constants of bis-triazolium macrocycle
5$(BF4)2 in CD3CN
Anion
Kaa (Mꢀ1
)
Clꢀ
Brꢀ
Iꢀ
457(17)
210(5)
129(6)
187(3)
467(62)
0b
ꢀ
N3
BzOꢀ
ꢀ
PF6
ꢀ
H2PO4
SO4
Precipitation
Precipitation
2ꢀ
a
Anions added as TBA salts, estimated standard errors are given in
parentheses. Association constants calculated using WINEQNMR214
(293 K). No movement of the triazolium resonance was observed,
b
Scheme 4 Synthesis of acyclic bis-triazolium receptors 8$(BF4)2 and
9$(BF4)2.
implying no binding event.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 12133–12147 | 12135

View Article Online
RSC Advances
Paper
Table 3 Halide anion association constants Ka (Mꢀ1),a of acyclic bis-
triazolium receptors 8$(BF4)2 and 9$(BF4)2 in d6-acetone
Receptor
Clꢀ
Brꢀ
8$(BF4)2
9$(BF4)2
4365(257)
>104b
1593(49)
2421(124)
a
Anions added as TBA salts, estimated standard errors are given in
parentheses. Association constants calculated using WINEQNMR214
b
(293 K). Binding too strong for WINEQNMR214 to determine an
association constant.
delocalised into the aryl substituent, reducing the electropositive
character of the triazolium proton. We suggest a similar effect
occurs here, explaining the greater recognition strength of the
alkyl-substituted bis-triazolium receptor 9$(BF4)2.
Pseudorotaxane formation. Having demonstrated the ability
of both acyclic bis-triazolium receptors to form stable 1 : 1
stoichiometric complexes with halide anions, attention was
turned to investigating the ability of these compounds to form
anion templated pseudorotaxanes with an isophthalamide
macrocycle, using 1H NMR spectroscopy. Addition of 8$(BF4)2 or
9$(BF4)2 and one equivalent of either TBA$Cl or TBA$Br to iso-
phthalamide macrocycle 10 resulted in upeld movement of the
macrocycle's hydroquinone resonances, indicative of donor–
acceptor interactions between these electron-rich rings and the
electron-decient bis-triazolium motifs in an interpenetrated
assembly (Fig. 5).
Fig. 4 Solid state structures of 5$(Cl)2, 5$(I)2 and 5$(N3)2. Some anions,
solvent molecules and most hydrogen atoms are omitted for clarity.
interaction involving a terminal nitrogen atom, and the other
the central nitrogen. Both hydrogen bonds are moderately
ꢁ
˚
strong [C–H/N ¼ 3.233(11) A, :C–H/N ¼ 142 ; C–H/N ¼
ꢁ
˚
3.297(11) A, :C–H/N ¼ 154 for interactions involving the
central and terminal nitrogen atom, respectively].
The halide anion recognition properties of acyclic bis-tri-
azolium receptors 8$(BF4)2 and 9$(BF4)2 were investigated using
1H NMR titration experiments in d6-acetone. Addition of halide
anions caused large downeld shis in the triazolium protons
of both receptors, as well as smaller shis in the triazolium-
methyl resonances. In the case of phenyl-linked 9$(BF4)2, the
central phenylene proton also underwent a signicant down-
eld perturbation. Quantitative 1 : 1 stoichiometric association
constants were determined using WINEQNMR2,14 monitoring
the triazolium resonance (Table 3, Fig. S51†).
Both receptors bind chloride anion more strongly than
bromide anion, presumably due to the smaller halide's greater
basicity. Interestingly, the alkyl-substituted host 9$(BF4)2 has a
signicantly higher affinity for both halide anions than 8$(BF4)2.
We have previously observed that mono-triazolium axle compo-
nents have very similar anion binding affinities whether they
contain alkyl or aryl substituents, despite the fact that aryl-
substitution would be expected to give a more acidic triazolium
group.8k This was explained by modelling studies, which demon-
Fig. 5 Formation of pseudorotaxanes between 82+ and 10, and
between 92+ and 10. Middle: truncated 1H NMR spectra of 10; top:
1
truncated H NMR spectrum of a 1 : 1 : 1 mixture of 8$(BF4)2, 10 and
1
TBA$Cl; bottom: truncated H NMR spectrum of a 1 : 1 : 1 mixture of
strated that some of the triazolium group's positive charge was 9$(BF4)2, 10 and TBA$Cl (d6-acetone, 500 MHz, 293 K).
12136 | RSC Adv., 2014, 4, 12133–12147
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Protons a and b from the triazolium receptor move upeld upon
RSC Advances
The anion templated assembly of pseudorotaxanes incor-
addition to macrocycle 10, while protons 2 and 3 move downeld, porating bis-triazolium macrocycle 5$(BF4)2 and bis-triazolium
as the chloride anion is now being shared between the threading threading component 8$(BF4)2 was investigated. Addition of
component and macrocycle, again consistent with pseudorotaxane chloride anion to a 1 : 1 mixture of 5$(BF4)2 and 8$(BF4)2 in
formation. The interpenetrated nature of the assembly was also CD3CN did not result in upeld shis of the macrocycle's
conrmed by 2D ROESY NMR spectroscopy (Fig. S46†). Impor- hydroquinone resonance, suggesting no anion templated
tantly, addition of 8$(BF4)2 or 9$(BF4)2 to 10 without a templating interpenetration. Given that 5$(BF4)2 displays only moderately
halide anion resulted in no signicant movement of the macrocycle strong chloride association in this solvent [Ka ¼ 457(17) Mꢀ1], it
or threading components' proton resonances, demonstrating the would appear that the favourability of chloride binding is not
crucial role of the halide anion template.
sufficient to overcome the electrostatic repulsion between the
The pseudorotaxane titration data were analysed using two dicationic components.
WINEQNMR2,14 monitoring the upeld shi of the hydroqui-
The investigation of sulfate as a templating anion was not
none resonance, to give apparent association constants18 for possible in CD3CN as addition of TBA2$SO4 to 5$(BF4)2 and
pseudorotaxane formation (Table 4, Fig. S52†). Signicantly 8$(BF4)2 in this solvent caused immediate precipitation, and so
stronger anion templated assemblies are observed with phenyl- these studies were carried out in d6-DMSO. Addition of sulfate
substituted threading component 8$(BF4)2 than with 9$(BF4)2, to a 1 : 1 mixture of 5$(BF4)2 and 8$(BF4)2 in this solvent
presumably due to the greater rigid aromatic portion of the
molecule that can undergo donor–acceptor interactions with
the macrocycle's electron-rich hydroquinone rings. Chloride
anion leads to more stable pseudorotaxanes than bromide, as
has been previously observed for related systems.18a,19
Table 4 Apparent association constants, Kapp (Mꢀ1),a for halide tem-
plated pseudorotaxane formation between 10 and bis-triazolium
threading components in d6-acetone
Threading
component
Clꢀ
Brꢀ
8$(BF4)2
9$(BF4)2
3214(317)
465(9)
544(3)
154(12)
a
Equimolar mixtures of threading component and TBA$halide were
added, estimated standard errors are given in parentheses. Apparent
association constants calculated using WINEQNMR214 (293 K).
Fig. 6 Formation of sulfate templated pseudorotaxane between 52+
and 82+. Bottom: truncated 1H NMR spectrum of 1 : 1 mixture of
5$(BF4)2 and 8$(BF4)2; top: truncated 1H NMR spectrum of 1 : 1 : 1
mixture of 5$(BF4)2 and 8$(BF4)2 and TBA2$SO4 (d6-DMSO, 300 MHz, Scheme 5 Synthesis of bis-triazolium axle components 14$(BF4)2 and
293 K).
15$(BF4)2.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 12133–12147 | 12137

View Article Online
RSC Advances
Paper
resulted in large downeld shis of the interior phenylene and with the formation of a stable pseudorotaxane in solution.
triazolium protons of both components (Fig. 6), consistent with Attempts to determine the stability of the pseudorotaxane
hydrogen bonding from both macrocycle and threading assembly using quantitative NMR titration experiments were
component to the anion. Upeld shis of the macrocycle's hampered as the threading component did not have sufficient
hydroquinone resonances20 suggest that the threading compo- solubility in d6-DMSO.21
nent is penetrating the annulus of the macrocycle, with the
Bis-triazolium axle containing rotaxane synthesis. Given the
resulting aromatic donor–acceptor interactions causing these potent anion recognition properties exhibited by acyclic and
shis. A qualitative study was undertaken to determine if any macrocyclic bis-triazolium receptors, the synthesis of inter-
other anions could template interpenetration in d6-DMSO. locked structures containing this motif was investigated.
ꢀ
Addition of Fꢀ, Clꢀ, Brꢀ, Iꢀ, OAcꢀ, ClO4ꢀ, H2PO4ꢀ, HSO4 or
Attempted synthesis via clipping. Initially we sought to
C2O42ꢀ to a 1 : 1 mixture of 5$(BF4)2 and 8$(BF4)2 did not result prepare bis-triazolium rotaxanes using a clipping strategy,
in upeld shis of the macrocycle's hydroquinone resonances, analogous to our previous syntheses of mono-triazolium rotax-
suggesting that only sulfate templates the interweaved anes.8a,h,k Two new neutral bis-triazole axle components 11 and
assembly.
12 were prepared in good yield by CuAAC reaction of stoppered
The sulfate templated interpenetrated assembly was further azide 13 and either 1,6-heptadiyne or 1,3-diethynylbenzene in
investigated by 2D ROESY NMR spectroscopy of a 1 : 1 : 1 dichloromethane using catalytic [Cu(CH3CN)4](PF6), TBTA and
mixture of 5$(BF4)2 and 8$(BF4)2 and TBA2$SO4 (Fig. S47†). DIPEA (Scheme 5). These were alkylated using trimethyloxo-
Several inter-component couplings are observed, consistent nium tetrauoroborate to give the dicationic bis-triazolium axle
components 142+ and 152+ as their tetrauoroborate salts.22
Scheme 6 Attempted synthesis of bis-triazolium rotaxanes via clip- Scheme 7 Attempted synthesis of bis-triazolium rotaxanes via amide
ping reactions.
condensation.
12138 | RSC Adv., 2014, 4, 12133–12147
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Preliminary qualitative 1H NMR anion recognition studies
RSC Advances
suggested these bis-triazolium axle components recognise
anions in d6-acetone and CD3CN, although their limited solu-
bility in these solvents precluded quantitative titration experi-
ments. Chloride anion templated rotaxane formation was
attempted using Grubbs' II catalysed RCM reaction of both axle
components, macrocycle precursor 16 and TBA$Cl in dichloro-
methane. In neither case was any evidence for interlocked
1
product formation observed (Scheme 6). Furthermore, the H
NMR spectra of the crude reaction mixture suggested that no
alkene metathesis had occurred. Repetition of the reactions
using different batches of catalyst similarly resulted in no
alkene metathesis. It therefore appears that the bis-triazolium
axle components deactivate the catalyst, preventing rotaxane
formation.
Attempted synthesis via amide condensation. We have
previously demonstrated that the condensation of bis-amine 2
and substituted isophthaloyl chlorides in the presence of a
pyridinium bis-amide axle component gives high yields (up to
60%) of rotaxane anion receptor systems.23 This synthetic
methodology was therefore explored as a potential route to bis-
triazolium rotaxane hosts. Bis-amine
2 and isophthaloyl
dichloride was added to a solution of the bis-triazolium axle
components 14$(BF4)2 and 15$(BF4)2 in dichloromethane
Scheme 8 Synthesis of bis(triazolium xylyl azide) 20$(BF4)2.
Scheme 9 Synthesis of bis-triazolium rotaxane 22$(PF6)2.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 12133–12147 | 12139

View Article Online
RSC Advances
Paper
containing triethylamine (Scheme 7). In both cases, the reac-
tion was shown to contain the target rotaxane products 17$(X)2
and 18$(X)2, as evidenced by high resolution ESI mass spec-
trometry and upeld-shied hydroquinone environment inꢀthe
1H NMR spectrum of the crude reaction mixture (X ¼ BF4 or
Clꢀ). Initial purication, followed by integration of the 1H NMR
spectrum suggested a yield of no more than 2% interlocked
product in both cases, and so further purication was not
attempted.
Synthesis via stoppering. Due to the lack of success of clip-
ping or amide condensation methodologies, it was decided to
attempt the synthesis of bis-triazolium rotaxanes by stoppering
an anion templated pseudorotaxane (vide supra) having appro-
priate pendant functionality. The CuAAC reaction is a powerful
tool for the synthesis of interlocked structures due to its mild
reaction conditions and wide functional group tolerance.
Therefore, the synthesis of an azide-appended analogue of bis-
triazolium threading component 8$(BF4)2 was targeted, to allow
rotaxane formation with a suitable bulky alkyne containing
synthon.
Fig.
7
Truncated 1H NMR spectrum of 22$(PF6)2 (bottom), and
22$(PF6)2 and one equivalent of TBA$Cl (top) (1 : 1 CDCl3 : CD3OD,
500 MHz, 293 K).
This was prepared by reacting a large excess of 1,4-bis(azi-
domethyl)benzene with 1,3-diethynylbenzene under CuAAC
conditions to give the bis triazole-azide 19 in good yield
(Scheme 8). Alkylation with (Me3O)(BF4) in dichloromethane
gave the bis-triazolium threading component 20$(BF4)2, also in
good yield.24
Anion templated stoppering rotaxane formation was
attempted as shown in Scheme 9: a pseudorotaxane was
prepared by mixing one molar equivalent each of iso-
phthalamide macrocycle 10, azide-appended threading
component 20$(BF4)2 and TBA$Cl in dichloromethane (cf.
Fig. 5). Two equivalents of alkyne stopper 21,
[Cu(CH3CN)4](PF6), TBTA and DIPEA were added to effect the
stoppering reaction. The rotaxane was difficult to purify, but
preparative TLC and recrystallisation, followed by anion
exchange using NH4PF6(aq) gave rotaxane 22$(PF6)2 in 20%
overall yield (Scheme 9).25 The new rotaxane was characterised
Fig. 8 Movement of triazolium resonance of 22$(PF6)2 upon addition of
anions; circles represent data points, lines represent binding isotherms
calculated using WINEQNMR214(1 : 1 CDCl3 : CD3OD, 293 K).26
1
by H, 13C, 19F and 31P NMR spectroscopy, as well as by high
resolution ESI mass spectrometry. The interlocked nature of the
product was conrmed by 2D ROESY NMR spectroscopy
(Fig. S48†), which revealed several inter-component couplings,
consistent with formation of the target rotaxane.
A possible reason for the lack of selectivity for halide anions
over acetate is provided by monitoring the hydroquinone reso-
nance: addition of chloride anion causes a signicant upeld
Rotaxane anion recognition properties. Addition of chloride
anion to rotaxane 22$(PF6)2 in competitive 1 : 1 CDCl3 : CD3OD
caused downeld shis of the axle's component triazolium and
axle phenylene protons (Fig. 7) indicative of hydrogen bonding
between these groups and the halide anion. Concomitant
upeld movement of the hydroquinone resonances was also
observed, consistent with enhanced donor–acceptor interac-
tions upon anion complexation.
Table
5 Anion association constants of bis-triazolium rotaxane
22$(PF6)2 in 1 : 1 CDCl3 : CD3OD
Anion
Kaa (Mꢀ1
)
Clꢀ
Brꢀ
Iꢀ
283(20)
407(21)
293(23)
428(70)b
131(12)
1H NMR titration experiments were conducted in 1 : 1
CDCl3 : CD3OD, with 1 : 1 stoichiometric anion association
constants determined using WINEQNMR2 (Fig. 8 and Table 5).
The rotaxane exhibits a selectivity preference for the halide
anions over more basic dihydrogenphosphate, but interestingly
not over acetate. Within the halide anions, a modest preference
for bromide over chloride and iodide is observed.
OAcꢀ
ꢀ
H2PO4
a
Anions added as TBA salts, estimated standard errors are given in
parentheses. Association constants calculated using WINEQNMR214
b
monitoring the triazolium resonance (293 K). The triazolium peak
disappeared during the titration due to deuterium/hydrogen
exchange, and so this value was obtained by monitoring the axle
component's interior phenylene resonance.26
12140 | RSC Adv., 2014, 4, 12133–12147
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
shi in the hydroquinone resonance, suggesting that chlo- tetrauoroborate yielded the desired bis-triazolium
ride anion is complexed in the host's interlocked cavity and precursor 25$(BF4)2 (Scheme 10).
strengthens donor–acceptor interactions between the mac-
The synthesis of hetero-catenane 264+ was attempted as
rocycle's hydroquinone rings and the positively-charged core shown in Scheme 11: CuAAC reaction of one molar
of the axle component. When acetate anion is bound, only a equivalent each of 5$(BF4)2 and 25$(BF4)2 in the presence of
very minor downeld shi in the hydroquinone resonance is one equivalent of TBA2$SO4 in acetonitrile gave a crude
observed, suggesting little strengthening of these interac- reaction mixture, which was shown by 1H NMR analysis
tions, possibly due to acetate associating primarily with the to contain signicant amounts (ꢂ30%) of the target
axle component with little or no interaction with the catenane. Purication was extremely difficult, but at least
macrocycle.
partial separation of the catenane and macrocyclic impuri-
The anion recognition properties of 22$(PF6)2 are ties could be achieved using preparative TLC, eluting
surprisingly modest, with this host exhibiting weaker halide with 17 : 2 : 1 acetonitrile : H2O : NaNO3(aq) and precipi-
binding in 1 : 1 CDCl3 : CD3OD than any of a series of related tating the organic materials from the aqueous phase by
mono-triazolium rotaxanes,8k despite the increased positive addition of sodium tetraphenylborate. Aer several
charge of 22$(PF6)2 and the additional triazolium hydrogen repeats of this process,
a very small amount of a
bond donor group. This is consistent with a previous study reasonably pure sample of the target catenane compound
demonstrating the importance of inter-component pre- 26$(BPh4)4 was obtained [see Fig. S36† for the assigned
organisation for strong anion recognition in interlocked 1H NMR spectrum of 26$(BPh4)4], however this was insuffi-
receptor systems.8k It would appear that the large, poorly cient for anion recognition studies. The synthesis of
preorganised axle component leads to relatively modest the homo-catenane was also attempted via the sulfate
anion recognition, despite the presence of the potent bis- anion templated double cyclisation of 25$(BF4)2 (Scheme
triazolium recognition motif.
A tetra-triazolium catenane. Given that sulfate templates a able yield, this was even more difficult to purify than the
stable interweaved tetra-triazolium pseudorotaxane hetero-catenane, and no interlocked structure could be
S4†). While the catenane appeared to be formed in reason-
assembly, the synthesis of either a hetero- or homo-[2]cate- isolated.
nane was attempted. Bis-azide appended macrocycle
precursor 24 was prepared by reacting 1,3-diethynylbenzene
with a large excess of bis-azide 23 to give the triazole bis-azide
in good yield. Alkylation using trimethyloxonium
Scheme 10 Synthesis of bis-triazolium bis-azide 25$(BF4)2.
Scheme 11 Synthesis of tetra-triazolium hetero-catenane 26$(BPh4)4.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 12133–12147 | 12141

View Article Online
RSC Advances
Paper
again taken to dryness to give a pale yellow oil. This was tritu-
rated with ice-cold diethyl ether (12 mL). The resulting ne
white powder was isolated by ltration, washed with cold
diethyl ether (2 ꢃ 5 mL) and dried in vacuo. Yield: 1.80 g (70%).
1H NMR (CDCl3): 6.84 (s, 8H, hydroquinone-H), 4.05–4.10
(m, 8H, hydroquinone-CH2), 3.83 (t, 3J ¼ 4.6 Hz, 4H, CH2), 3.67–
Conclusions
Two bis-triazole macrocycles were prepared; while the chloride
anion binding properties of these receptors were modest,
alkylation of one to the bis-triazolium analogue gave the potent
cyclic host 5$(BF4)2. Acyclic bis-triazolium compounds were also
prepared and shown to complex halide anions, with propyl-
linked system 9$(BF4)2 displaying stronger anion recognition
than the phenyl-linked analogue 8$(BF4)2. Conversely, 8$(BF4)2
forms a more stable halide anion templated pseudorotaxane
with an isophthalamide macrocycle than 9$(BF4)2. Strong
sulfate templated pseudorotaxane formation between 5$(BF4)2
and bis-triazolium macrocycle 8$(BF4)2 was observed in
competitive d6-DMSO. A chloride anion templated stoppering
strategy allowed the synthesis of a novel bis-triazolium axle
containing rotaxane host 22$(PF6)2, which was shown to bind
halide anions selectively over dihydrogenphosphate in
competitive 1 : 1 CDCl3 : CD3OD solvent media. A tetra-tri-
azolium hetero-catenane was prepared using sulfate anion
templation, however, purication of this quadruply-charged
interlocked host proved problematic.
3
3.74 (m, 8H, CH2), 3.56 (t, J ¼ 4.5 Hz, 4H, CH2–N3). 13C NMR
(CDCl3): 153.6, 152.6, 115.8, 115.7, 70.9, 70.8, 70.0, 68.2, 67.8,
50.4. HRESI-MS (pos.): 555.1965, calc. for [C24H32N6O7$K]+
555.1964.
¼
Phenyl bis-triazole macrocycle 3. The bis-azide 1, (0.103 g,
0.200 mmol) and 1,3-diethynylbenzene (0.027 mL, 0.025 g, 0.20
mmol) were dissolved in CH2Cl2 (200 mL); DIPEA (0.045 mL,
0.032 g, 0.25 mmol), TBTA (0.023 g, 0.040 mmol) and
[Cu(CH3CN)4](PF6) (0.015 g, 0.040 mmol) were added and the
reaction stirred at room temperature for 3 days under a nitrogen
atmosphere. It was taken to dryness under reduced pressure
and puried by preparative TLC (6% CH3OH in CH2Cl2) to give 3
as a white powder. Yield: 0.058 g (45%).
1H NMR (CDCl3): 8.04 (s, 2H, trz-H), 7.99 (t, 4J ¼ 1.6 Hz, 1H,
ph-H), 7.95 (dd, 3J ¼ 7.8 Hz, 4J ¼ 1.6 Hz, 2H, ph-H), 7.51 (t, 3J ¼
7.8 Hz, 1H, ph-H), 6.74–6.81 (m, 8H, hydroquinone-H), 4.79 (t, 3J
3
¼ 4.7 Hz, 4H, CH2-trz), 4.32 (t, J ¼ 4.7 Hz, 4H, hydroquinone-
Experimental
General remarks
3
3
CH2), 4.02 (t, J ¼ 4.7 Hz, 4H, hydroquinone-CH2), 3.79 (t, J ¼
4.7 Hz, 4H, CH2), 3.62–3.71 (m, 8H, CH2). 13C NMR (CDCl3):
TBTA,27 bis-amine 2,23a bis-azide 23,28 stopper compounds 13 153.8, 152.2, 147.7, 131.3, 129.7, 125.6, 123.2, 121.3, 115.9,
(ref. 29) and 21,30 and isophthalamide macrocycle 10 (ref. 23b) 115.8, 70.9, 70.9, 69.8, 68.2, 67.6, 50.3. HRESI-MS (pos.):
were prepared as previously described. Octyl azide and 1,4- 665.2696, calc. for [C34H38N6O7$Na]+ ¼ 665.2694.
bis(azidomethyl)benzene were prepared from the correspond-
Propyl bis-triazole macrocycle 4. The bis-azide 1, (0.155 g,
ing bromides using NaN3 in DMSO.31 Other chemicals are 0.300 mmol) and 1,6-heptadiyne (0.034 mL, 0.028 g, 0.30 mmol)
available commercially, and used as received. Where solvents were dissolved in CH2Cl2 (300 mL); DIPEA (0.17 mL, 0.13 g, 1.0
are specied as “dry”, they were purged with nitrogen and mmol), TBTA (0.032 g, 0.060 mmol) and [Cu(CH3CN)4](PF6)
passed through an MBraun MPSP-800 column. NEt3 was (0.022 g, 0.060 mmol) were added and the reaction stirred at
distilled from, and stored over KOH pellets. Water was deion- room temperature under a nitrogen atmosphere for 3 days. It
ised and microltered using a Milli-Q Millipore machine. TBA was taken to dryness under reduced pressure and puried by
salts, TBTA and [Cu(CH3CN)4](PF6) were stored in a vacuum preparative TLC (5% CH3OH in CH2Cl2) to give 4 as a white
dessicator. Chromatography was performed on silica gel powder. Yield: 0.047 g (25%).
(particle size: 40–63 mm) or preparative TLC plates (20 ꢃ 20 cm,
1H NMR (CDCl3): 7.46 (s, 2H, trz-H), 6.71–6.81 (m, 8H,
silica thickness: 1 mm). Details of instrumentation, copies of hydroquinone-H), 4.68 (t, 3J ¼ 4.9 Hz, 4H, CH2-trz), 4.25 (t, 3J ¼
NMR spectra including 2D ROESY spectra, and details of titra- 4.9 Hz, 4H, hydroquinone-CH2), 4.01 (t, 3J ¼ 4.8 Hz, 4H,
tion protocols are given in the ESI.†
hydroquinone-CH2), 3.80 (t, 3J ¼ 4.8 Hz, 4H, CH2), 3.64–3.71 (m,
Polyether bis-azide 1. Caution: while the hydrochloride salt 8H, CH2), 2.77 (t, 3J ¼ 7.5 Hz, 4H, trz-CH2), 2.06 (qn, 3J ¼ 7.5 Hz,
of imidazole sulfonylazide10a was used in this preparation, it has 2H, trz-CH2–CH2). 13C NMR (CDCl3): 153.8, 152.2, 147.7, 122.3,
subsequently been reported that there are potential hazards 115.9, 115.7, 70.9, 70.9, 69.8, 68.3, 67.5, 49.9, 29.0, 24.9. HRESI-
associated with this and that other salts offer safer MS (pos.): 609.3019, calc. for [C31H40N6O7$H]+ ¼ 609.3031.
alternatives.10b
Bis-triazolium macrocycle 5$(BF4)2. The neutral macrocycle 3
The diamine 2, (2.32 g, 5.00 mmol), potassium carbonate (0.193 g, 0.300 mmol) was dissolved in dry CH2Cl2 (40 mL) and
(3.50 g, 25.0 mmol), imidazole sulfonylazide$HCl (2.52 g, 12.0 (Me3O)(BF4) (0.098 g, 0.66 mmol) was added. This solution was
mmol) and copper(II) sulfate pentahydrate (0.025 g, 0.10 mmol) stirred under a nitrogen atmosphere at room temperature for 16
were dissolved in CH3OH (75 mL), and the colourless solution hours. The reaction was quenched by addition of CH3OH
stirred at room temperature under a nitrogen atmosphere (1 mL), and taken to dryness under reduced pressure. The
overnight. The mixture was concentrated to ꢂ10 mL under resulting crude solid was stirred for 2 hours in CH2Cl2 (10 mL),
reduced pressure, diluted with water (50 mL), acidied with then ltered and washed with further CH2Cl2 (2 mL). Purica-
conc. HCl(aq) (10 mL) and extracted with ethyl acetate (3 ꢃ 50 tion by recrystallization from hot 49 : 1 CHCl3 : CH3OH (50 mL)
1
mL). The organic layer was dried (MgSO4) and taken to dryness gave 5$(BF4)2 as white crystals. Yield: 0.141 g (55%). H NMR
under reduced pressure. It was ltered through a short pad of (CD3CN): 8.67 (s, 2H, trz+-H), 7.88–7.91 (m, 2H, ph-H), 7.74–7.75
silica gel, eluting with 4% methanol in dichloromethane and (m, 1H, ph-H), 7.59 (s, 1H, ph-H), 6.80–6.88 (m, 8H,
12142 | RSC Adv., 2014, 4, 12133–12147
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
hydroquinone-H), 5.00 (t, 3J ¼ 4.6 Hz, 4H, trz+-CH2), 4.39 (t, 3J ¼
4.6 Hz, 4H, hydroquinone-CH2), 4.22 (s, 6H, trz+-CH3), 3.98–4.01
(m, 4H, CH2), 3.71–3.74 (m, 4H, CH2), 3.54–3.61 (m, 8H, CH2).
13C NMR (CD3CN): 154.8, 152.8, 142.9, 133.5, 131.8, 130.8,
124.8, 116.7, 116.3, 111.0, 71.4, 71.3, 70.4, 68.9, 66.7, 54.8, 39.7.
19F NMR (CD3CN): ꢀ151.5 (m). HRESI-MS (pos.): 759.3294, calc.
for [C36H44N6O7$BF4]+ ¼ 759.3301.
Propyl bis-triazolium thread 9$(BF4)2. The bis-triazole thread
7 (0.201 g, 0.500 mmol) was dissolved in dry CH2Cl2 (40 mL).
(Me3O)(BF4) (0.221 g, 1.50 mmol) was added and the reaction
stirred at room temperature under a nitrogen atmosphere for 2
days. CH3OH (1 mL) was added, and the reaction taken to
dryness under reduced pressure. Purication by preparative
TLC (10% CH3OH in CH2Cl2) gave 9$(BF4)2 as a pale yellow oil.
Yield: 0.158 g (52%).
Phenyl bis-triazole thread 6. Diethynylbenzene (0.133 mL,
0.126 g, 1.00 mmol) and octyl azide (0.342 g, 2.20 mmol) were
dissolved in CH2Cl2 (30 mL). DIPEA (0.51 mL, 0.39 g, 3.0 mmol),
TBTA (0.12 g, 0.22 mmol) and [Cu(CH3CN)4](PF6) (0.075 g, 0.20
mmol) were added and the reaction stirred under a nitrogen
atmosphere for 4 days. The reaction was taken to dryness under
reduced pressure and then puried by column chromatography
(gradient: 0–6% acetone in CH2Cl2) to give 6 as a white powder.
Yield: 0.329 g (75%).
1H NMR (d6-acetone): 8.80 (s, 2H, trz+-H), 4.71 (t, 3J ¼ 7.3 Hz,
4H, trz+-CH2), 4.38 (s, 6H, trz+-CH3), 3.22 (t, 3J ¼ 7.8 Hz, 4H, trz+-
CH2), 2.32–2.38 (m, 4H, trz+-CH2–CH2), 2.17 (qn, 3J ¼ 7.8 Hz, 2H,
trz+-CH2–CH2), 1.22–1.43 (m, 20H, CH2), 0.86 (t, 3J ¼ 7.0 Hz, 6H,
CH3). 13C NMR (CDCl3): 143.9, 128.5, 54.1, 37.6, 31.8, 29.2, 29.0,
28.9, 26.2, 23.8, 22.7, 22.0, 14.2. 19F NMR (d6-acetone): ꢀ151.4
(m). HRESI-MS (pos.): 519.3969, calc. for [C25H48N6$BF4]+
519.3964.
¼
1H NMR (d6-acetone): 8.44 (s, 2H, trz-H), 8.43 (t, 4J ¼ 1.7 Hz,
Phenyl bis-triazole axle 11. Stopper azide 13 (0.617 g, 1.05
mmol) and 1,3-diethynylbenzene (0.063 g, 0.50 mmol) were
stirred in CH2Cl2 (30 mL). [Cu(CH3CN)4](PF6) (0.075 g, 0.20
mmol), TBTA (0.117 g, 0.22 mmol) and DIPEA (0.34 mL, 0.26 g,
2.0 mmol) were added, and the pale yellow solution stirred for 5
days under a nitrogen atmosphere. The reaction mixture was
stirred with aqueous NH4OH/EDTA solution (25 mL), and the
aqueous layer extracted further with CH2Cl2 (2 ꢃ 25 mL). The
combined organic fractions were dried (MgSO4) and taken to
dryness under reduced pressure. Column chromatography
(silica, gradient: 1–2% CH3OH in CH2Cl2) gave 11 as a white
powder. Yield: 0.584 g (90%).
1H, ph-H), 7.84 (dd, 3J ¼ 7.9 Hz, 4J ¼ 1.7 Hz, 2H, ph-H), 7.49 (t, 3J
3
¼ 7.9 Hz, 1H, ph-H), 4.48 (t, J ¼ 7.1 Hz, 4H, trz-CH2), 1.99–?
(obscured by residual acetone solvent peak, 4H, trz-CH2–CH2),
1.24–1.41 (m, 20H, CH2), 0.86 (t, 3J ¼ 6.8 Hz, 6H, CH3). 13C NMR
(CDCl3): 147.5, 131.4, 129.5, 125.3, 122.9, 119.9, 50.6, 31.8, 30.5,
29.2, 29.1, 26.6, 22.7, 14.2. HRESI-MS (pos.): 459.3216, calc. for
[C26H40N6$Na]+ ¼ 459.3207.
Propyl bis-triazole thread 7. 1,6-Heptadiyne (0.11 mL, 0.092
g, 1.00 mmol) and octyl azide (0.342 g, 2.20 mmol) were dis-
solved in dry CH2Cl2 (30 mL). TBTA (0.234 g, 0.440 mmol),
DIPEA (0.51 mL, 0.39 g, 3.00 mmol) and [Cu(CH3CN)4](PF6)
(0.15 g, 0.40 mmol) were added and the mixture stirred at room
temperature under a nitrogen atmosphere for 6 days. It was
stirred vigorously with aqueous NH4OH/EDTA(20 mL) and the
aqueous layer extracted with CH2Cl2 (20 mL). The combined
organic layers were washed with H2O (2 ꢃ 30 mL), dried
(MgSO4) and taken to dryness. Column chromatography (3 : 2
acetone : petrol) gave 7 as a white powder. Yield: 0.299 g (74%).
1H NMR (CDCl3): 7.31 (s, 2H, trz-H), 4.30 (t, 3J ¼ 7.3 Hz, 4H, trz-
CH2), 2.77 (t, 3J ¼ 7.5 Hz, 4H, trz-CH2), 2.00–2.10 (m, 4H, CH2),
1.87 (qn, 3J ¼ 7.5 Hz, 2H, trz-CH2–CH2), 1.20–1.37 (m, 20H,
1H NMR (CDCl3): 8.29 (s, 1H, ph-H), 7.88 (s, 2H, trz-H), 7.77–7.83
(m, 2H, ph-H), 7.43–7.50 (m, 1H, ph-H), 7.20–7.29 (m, 12H, Ar-H),
7.04–7.14 (m, 16H, Ar-H), 6.74–6.81 (m, 4H, Ar-H), 4.64 (t, 3J ¼ 6.9
Hz, 4H, Ar-O–CH2), 3.99 (t, 3J ¼ 6.9 Hz, 4H trz-CH2), 2.38–2.50 (4H,
t
m, Ar-O–CH2–CH2), 1.30 (s, 54H, Bu-H), 13C NMR (CDCl3): 156.2,
148.3, 144.0, 140.1, 132.3, 130.7, 129.4, 125.3, 124.0, 123.1, 120.4,
113.0, 63.9, 63.0, 55.8, 47.4, 43.8, 34.3, 31.4, 30.1, 29.3. HRESI-MS
(pos.) 1323.8117, calc. for [C90H105N6O2$Na]+ ¼ 1323.8113.
Propyl bis-triazole axle 12. 1,6-Heptadiyne (0.057 mL, 0.046
g, 0.50 mmol) and stopper-azide 13 (0.647 g, 1.10 mmol) were
dissolved in dry CH2Cl2 (40 mL). TBTA (0.318 g, 0.600 mmol),
DIPEA (0.27 mL, 0.19 g, 1.5 mmol) and [Cu(CH3CN)4](PF6)
(0.205 g, 0.550 mmol) were added and the resulting solution
stirred under a nitrogen atmosphere for 4 days. The solvent was
removed under reduced pressure, and the resulting off-white
solid puried by column chromatography (2% CH3OH in
CH2Cl2) to give 12 as a white powder. Yield: 0.547 g (86%).
1H NMR (d6-acetone): 7.73 (s, 2H, trz-H), 7.31–7.34 (m, 12H,
Ar-H), 7.09–7.15 (m, 16H, Ar-H), 6.81–6.85 (m, 4H, Ar-H), 4.57 (t,
3J ¼ 6.9 Hz, 4H, trz-CH2), 3.99 (t, 3J ¼ 6.0 Hz, 4H, Ar-O–CH2), 2.68
(t, 3J ¼ 6.9 Hz, 4H, trz-CH2), 2.32–2.37 (m, 4H, 4 Ar-O–CH2–CH2),
1.96 (qn, 3J ¼ 6.9 Hz, 2H, trz-CH2–CH2), 1.29 (s, tBu-H). 13C NMR
(CDCl3): 156.4, 148.5, 144.2, 140.2, 132.5, 130.8, 129.2, 124.2,
113.0, 64.1, 63.2, 47.1, 34.4, 31.5, 30.2, 29.2, 25.0 (one resonance
not observed/coincident). HRESI-MS (pos.): 1267.8429, calc. for
[C87H106O2N6$H]+ ¼ 1267.8450.
3
CH2), 0.86 (t, J ¼ 6.7 Hz, 6H, CH3). 13C NMR (CDCl3): 147.5,
120.8, 50.3, 31.8, 30.4, 29.3, 29.1, 29.0, 26.6, 25.0, 22.7, 14.1.
HRESI-MS(pos.): 425.3366, calc. for [C23H42N6$Na]+
425.3363.
¼
Phenyl bis-triazolium thread 8$(BF4)2. The bis-triazole
thread 6 (0.175 g, 0.400 mmol) was dissolved in dry CH2Cl2 (30
mL). (Me3O)(BF4) (0.129 g, 0.88 mmol) was added and the
reaction stirred under a nitrogen atmosphere for 4 days. It was
quenched by the addition of CH3OH (1 mL), and taken to
dryness under reduced pressure. Preparative TLC (4% CH3OH
in CH2Cl2) gave 8$(BF4)2 as a white powder. Yield: 0.168 g (66%).
1H NMR (d6-acetone): 9.00 (s, 2H, trz+-H), 8.21 (t, 4J ¼ 1.5 Hz,
1H, ph-H), 8.09 (dd, 3J ¼ 7.9 Hz, 4J ¼ 1.5 Hz, 2H, ph-H), 7.91 (t, 3J
¼ 7.9 Hz, 1H, ph-H), 4.76 (t, 3J ¼ 7.3 Hz, 4H, trz-CH2), 4.43 (s, 6H,
trz-CH3), 2.07–2.16 (m, 4H, trz-CH2–CH2), 1.29–1.52 (m, 20H,
3
CH2), 0.88 (t, J ¼ 6.8 Hz, 6H, CH3). 13C NMR (CDCl3): 141.9,
132.4, 131.0, 130.6, 129.2, 123.8, 54.5, 38.9, 31.8, 29.1, 29.1, 28.9,
26.3, 22.7, 14.2. 19F NMR (d6-acetone): ꢀ151.5 (m). HRESI-MS
(pos.): 233.1887, calc. for [C28H46N6]2+ ¼ 233.1886.
Phenyl bis-triazolium axle 14$(BF4)2. The bis(triazole) axle 11
(0.260 g, 0.200 mmol) was dissolved in dry CH2Cl2 (50 mL).
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 12133–12147 | 12143

View Article Online
RSC Advances
Paper
3
(Me3O)(BF4) (0.118 g, 0.800 mmol) was added, and the reaction Hz, 4H, xy-H), 7.50 (d, J ¼ 8.2 Hz, 4H, xy-H), 6.03 (s, 4H, trz+-
stirred under a nitrogen atmosphere overnight. CH3OH (1 mL) CH2), 4.51 (s, 4H, CH2–N3), 4.46 (s, 6H, trz+-CH3). 13C NMR
was added and the reaction taken to dryness. Preparative TLC (CDCl3): 143.2, 138.5, 133.5, 133.3, 131.7, 131.5, 130.6, 130.3,
(silica, 4% CH3OH in CH2Cl2) gave 14$(BF4)2 as a white powder. 130.0, 125.1, 57.6, 54.5, 39.6. 19F NMR (d6-acetone): ꢀ151.4 (m).
Yield: 0.261 g (87%).
1H NMR (CDCl3): 8.59 (s, 2H, trz+-H), 7.97 (1H, s, ph-H), 7.71– 619.2584.
7.76 (m, 2H, ph-H), 7.55–7.62 (m, 1H, ph-H), 7.16–7.24 (m, 12H, Bis-triazolium rotaxane 22$(PF6)2. The bis-triazolium
HRESI-MS (pos.): 619.2591, calc. for [C28H28N12$BF4]+
¼
Ar-H), 7.02–7.10 (m, 16H, Ar-H), 6.65–6.71 (m, 4H, Ar-H), 4.72– threading component 20$(BF4)2 (0.030 g, 0.050 mmol) and
4.81 (m, 4H, trz+-H), 4.14 (s, 6H, trz+-CH3), 3.99–4.06 (m, 4H, Ar- macrocycle 10 (0.033 g, 0.050 mmol) were dissolved in dry
O–CH2), 2.42–2.54 (m, 4H, Ar-O–CH2–CH2), 1.28 (s, 54H, tBu-H); CH2Cl2 (50 mL). TBA$Cl (0.014 g, 0.050 mmol) was added, fol-
13C NMR (CDCl3): 156.0, 148.3, 144.0, 141.9, 140.2, 132.4, 132.2, lowed by stopper-alkyne 21 (0.060 g, 0.11 mmol), DIPEA (0.017
130.6, 129.3, 124.1, 123.6, 112.9, 63.7, 63.0, 55.3, 51.7, 38.6, 34.3, mL, 0.013 g, 0.1 mmol), TBTA (0.0053 g, 0.010 mmol) and
31.4, 28.8, 18.6, 17.2; 19F NMR (CDCl3): ꢀ151.7 (m). HRESI-MS [Cu(CH3CN)4](PF6) (0.0037 g, 0.010 mmol), and the clear yellow
(pos.): 665.9371, calc. for [C92H110N6O2]2+ ¼ 665.9357.
solution stirred at room temperature under a nitrogen atmo-
Propyl bis-triazolium axle 15$(BF4)2. The axle component 12 sphere for 7 days. It was taken to dryness under reduced pres-
(0.254 g, 0.200 mmol) was dissolved in dry CH2Cl2 (40 mL). sure and puried by preparative TLC (6% CH3OH in CH2Cl2) to
(Me3O)(BF4) (0.074 g, 0.50 mmol) was added and the reaction give essentially pure rotaxane. This was taken up in CH2Cl2 (10
mixture stirred at room temperature under a nitrogen atmo- mL), and stirred vigorously with NH4PF6(aq) (0.1 M, 10 mL) for
sphere for 16 hours. The reaction was quenched by the addition 30 minutes. The organic phase was taken and washed with
of CH3OH (1 mL), taken to dryness and puried by preparative further NH4PF6(aq), (0.1 M, 6 ꢃ 10 mL) and H2O (3 ꢃ 10 mL) and
TLC (89 : 10 : 1 CH2Cl2 : CH3OH : NH4OH(aq)) to give the bis- then thoroughly dried in vacuo. The white powder was dissolved
triazolium 15$(BF4)2 as a white powder. Yield: 0.188 g (64%).
in hot CH3OH : CH2Cl2 (19 : 1, 20 mL), and the volume of the
1H NMR (d6-acetone): 8.84 (s, 2H, trz+-H), 7.32–7.36 (m, 12H, solution reduced to ꢂ10 mL by boiling. Cooling the solution
Ar-H), 7.12–7.16 (m, 16H, Ar-H), 6.82–6.84 (m, 4H, Ar-H), 4.94 (t, caused precipitation of 22$(PF6)2 as a white powder. Yield: 0.026
3J ¼ 7.0 Hz, 4H, trz+-CH2), 4.32 (s, 6H, trz+-CH3), 4.14 (t, 3J ¼ 5.8 g (20%). 1H NMR (1 : 1 CDCl3 : CD3OD): 8.85 (s, 2H, trz+-H), 8.13
Hz, 4H, Ar-O–CH2), 3.23 (t, 3J ¼ 7.7 Hz, 4H, trz+-CH2), 2.51–2.56 (br. s, 1H, ph-H), 8.10 (d, 4J ¼ 1.5 Hz, 2H, ph-H), 7.85 (s, 2H, trz-
(m, 4H, Ar-O–CH2–CH2), 2.39 (qn, 3J ¼ 7.7 Hz, 2H, trz+-CH2– H), 7.73–7.80 (m, 3H, ph-H), 7.47–7.51 (m, 5H, ph-H, Ar-H), 7.37
t
3
CH2), 1.29 (s, 54H, Bu-H). 13C NMR (d6-DMSO): 155.8, 147.8, (d, J ¼ 8.2 Hz, 4H, Ar-H), 7.17–7.20 (m, 12H, Ar-H), 7.01–7.05
143.9, 143.1, 139.3, 131.5, 130.0, 128.5, 124.2, 113.3, 64.1, 62.6, (m, 16H, Ar-H), 6.72 (d, 3J ¼ 9.0 Hz, 4H, Ar-H), 6.37 (d, 3J ¼ 9.1
50.7, 37.4, 34.1, 31.2, 28.5, 23.5, 21.6. 19F NMR (CDCl3): ꢀ151.7 Hz, 4H, hydroquinone-H), 6.05 (d, J ¼ 9.1 Hz, 4H, hydroqui-
3
(m). HRESI-MS (pos.): 1383.8914, calc. for [C89H112N6O2$BF4]+ ¼ none-H), 5.71 (s, 4H, trz+-CH2), 5.51 (s, 4H, Ar-O–CH2), 4.94 (s,
3
1383.8884.
4H, trz-CH2-xy), 3.97–4.00 (m, 10H, CH2, trz+-CH3), 3.71 (t, J ¼
Bis(triazole xylyl azide) 19. To a solution of 1,3-dieth- 4.8 Hz, 4H, CH2), 3.36–3.50 (m, 16H, CH2), 1.26 (s, 54H, tBu-H).
ynylbenzene (0.067 mL, 0.063 g, 0.50 mmol) and 1,4-bis(azido- 13C NMR (1 : 1 CDCl3 : CD3OD): 169.1, 156.8, 153.7, 153.3,
methyl)benzene (1.41 g, 7.50 mmol) in CH2Cl2 (50 mL) was 152.9, 149.1, 145.0, 144.9, 142.3, 141.1, 137.8, 134.9, 133.0,
added DIPEA (0.27 mL, 0.19 g, 1.5 mmol), TBTA (0.053 g, 0.10 132.5, 131.5, 131.4, 131.3, 131.0, 130.4, 130.1, 129.9, 128.9,
mmol) and [Cu(CH3CN4)](PF6) (0.037 g, 0.10 mmol). The solu- 125.0, 124.7, 124.3, 122.6, 115.8, 115.2, 113.9, 71.4, 71.1, 70.5,
tion was stirred at room temperature under a nitrogen atmo- 68.3, 67.0, 63.8, 61.9, 57.7, 54.1, 40.6, 39.3, 35.7, 34.8, 31.7, 31.5.
sphere for 6 days, taken to dryness under reduced pressure and 19F NMR (1 : 1 CDCl3 : CD3OD): ꢀ71.6 (d, JP,F ¼ 712 Hz). 31P
puried by column chromatography (1% CH3OH in CHCl3) to NMR (1 : 1 CDCl3 : CD3OD): ꢀ144.7 (septet, JP,F ¼ 712 Hz).
give 19 as a white powder. Yield: 0.20 g (78%). 1H NMR (CDCl3): HRESI-MS (pos.): 1133.639, calc. for [C144H166N14O11
]
¼
2+
8.19 (t, 4J ¼ 1.7 Hz, 1H, ph-H), 7.78 (dd, 3J ¼ 7.7 Hz, 4J ¼ 1.7 Hz, 1133.643.
3
2H, ph-H), 7.74 (s, 2H, trz-H), 7.44 (t, J ¼ 7.7 Hz, 1H, ph-H),
Bis-triazole bis-azide 24. The bis-azide 23 (7.45 g, 30.0 mmol)
7.31–7.37 (m, 8H, xy-H), 5.58 (s, 4H, trz-CH2), 4.36 (s, 4H, CH2– and 1,3-diethynylbenzene (0.319 mL, 0.303 g, 2.40 mmol) were
N3). 13C NMR (CDCl3): 148.0, 136.4, 134.8, 131.1, 129.6, 129.1, dissolved in CH2Cl2 (75 mL). [Cu(PPh3)3]Br (0.223 g, 0.240
128.7, 125.5, 123.0, 120.0, 54.4, 54.0. HRESI-MS (pos.): 525.1988, mmol) was added and the yellow solution stirred at room
calc. for [C26H22N12$Na]+ ¼ 525.1983.
temperature under a nitrogen atmosphere for 3 days. It was
Bis(triazolium xylyl azide) 20$(BF4)2. The bis-azide 19 (0.101 taken to dryness under reduced pressure and puried by
g, 0.200 mmol) was dissolved in dry CH2Cl2 (10 mL). column chromatography (gradient: 0–2% CH3OH in CHCl3) to
(Me3O)(BF4) (0.065 g, 0.44 mmol) was added and the reaction give 24 as an off-white solid. Yield: 1.15 g (77%).
stirred at room temperature under a nitrogen atmosphere for 16
1H NMR (CDCl3): 8.29 (t, 4J ¼ 1.6 Hz, 1H, ph-H), 8.05 (s, 2H,
hours. CH3OH (1 mL) was added, and the reaction taken to trz-H), 7.84 (dd, 3J ¼ 7.8 Hz, 4J ¼ 1.6 Hz, 2H, ph-H), 7.50 (t, 3J ¼
dryness under reduced pressure. Purication by preparative 7.8 Hz, 1H, ph-H), 6.85 (app. s, 8H, hydroquinone-H), 4.81 (t, 3J
3
TLC (10% CH3OH in CH2Cl2) gave 20$(BF4)2 as a pale yellow ¼ 4.8 Hz, 4H, trz-CH2), 4.37 (t, J ¼ 4.8 Hz, 4H, hydroquinone-
1
3
3
foam. Yield: 0.113 g (80%). H NMR (d6-acetone): 9.11 (s, 2H, CH2), 4.09 (t, J ¼ 5.1 Hz, 4H, hydroquinone-CH2), 3.56 (t, J ¼
trz+-H), 8.23 (t, 4J ¼ 1.6 Hz, 1H, ph-H), 8.09 (dd, 3J ¼ 7.8 Hz, 4J ¼ 5.1 Hz, 4H, CH2–N3). 13C NMR (CDCl3): 153.3, 152.6, 147.7,
1.6 Hz, 2H, ph-H), 7.92 (t, 3J ¼ 7.8 Hz, 1H, ph-H), 7.67 (d, 3J ¼ 8.2 131.3, 129.6, 125.6, 123.1, 121.2, 115.9, 115.9, 67.7, 67.3, 50.3,
12144 | RSC Adv., 2014, 4, 12133–12147
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
50.2. HRESI-MS (pos.): 645.2405, calc. for [C30H30N12O4$Na]+ ¼ and rened using full-matrix least-squares on F2 within the
645.2405.
CRYSTALS suite.38 All non-hydrogen atoms were rened with
Bis-triazolium bis-azide 25$(BF4)2. The bis-triazole bis-azide anisotropic displacement parameters. Hydrogen atoms were
24 (0.093 g, 0.15 mmol) was dissolved in dry CH2Cl2 (40 mL). generally visible in the difference map and were initially rened
(Me3O)(BF4) (0.049 g, 0.33 mmol) was added and the reaction with restraints on bond lengths and angles, aer which the
stirred at room temperature under a nitrogen atmosphere for 16 positions were used as the basis for a riding model.39 Further
hours. The reaction was quenched by the addition of CH3OH details, including full crystallographic data in CIF format are
(1 mL), taken to dryness under reduced pressure, and puried provided in the ESI.†
by preparative TLC (16 : 3 : 1 CH2Cl2 : acetone : CH3OH) to give
1
25$(BF4)2 as a waxy yellow solid. Yield: 0.090 g (72%). H NMR
Acknowledgements
(CD3CN): 8.67 (s, 2H, trz+-H), 7.87–7.91 (m, 3H, ph-H), 7.78–7.80
(m, 1H, ph-H), 6.87–6.95 (m, 8H, hydroquinone-H), 5.00 (t, 3J ¼ We thank Diamond Light Source for an award of beamtime on
4.8 Hz, 4H, trz+-H), 4.45 (t, 3J ¼ 4.8 Hz, 4H, hydroquinone-CH2), Beamline I19. NGW thanks the Clarendon Fund, Trinity College
3
4.22 (s, 6H, trz+-CH3), 4.10 (t, J ¼ 5.1 Hz, 4H, hydroquinone- Oxford, and the Vice-Chancellors' Fund for nancial support.
CH2), 3.56 (t, 3J ¼ 5.1 Hz, 4H, CH2–N3). 13C NMR (CD3CN): 154.3,
153.2, 142.9, 133.7, 131.8, 131.3, 130.7, 124.8, 116.9, 116.6, 68.6,
Notes and references
66.7, 54.7, 51.0, 39.6. 19F NMR (CD3CN): ꢀ151.6 (m). HRESI-MS
(pos.): 362.1476, calc. for [C32H36N12O4]2+ ¼ 362.1486.
1 H. Juwarker, J. M. Lenhardt, D. M. Pham and S. L. Craig,
Angew. Chem., Int. Ed., 2008, 47, 3740.
Hetero-catenane 26$(BPh4)4. The bis-triazolium macrocycle
5$(BF4)2 (0.042 g, 0.050 mmol), bis-triazolium bis-azide
precursor 25$(BF4)2 (0.041 g, 0.050 mmol) and TBA2$SO4 (0.029
g, 0.050 mmol) were dissolved in dry CH3CN (200 mL) and the
solution stirred under a nitrogen atmosphere for 20 minutes. To
this was added 1,3-diethynylbenzene (0.0063 g, 0.0067 mL,
0.050 mmol), DIPEA (0.0078 g, 0.010 mL, 0.0060 mmol), and
TBTA (0.0021 g, 0.0040 mmol) and the solution deoxygenated
with bubbling nitrogen for 2 minutes. [Cu(CH3CN)4](PF6)
(0.0015 g, 0.0040 mmol) was added and the solution stirred
under a nitrogen atmosphere for 8 days. The reaction mixture
was taken to dryness under reduced pressure, and puried by
iterative preparative TLC (17 : 2 : 1 CH3CN : H2O : sat. aq.
NaNO3). The CH3CN was removed under reduced pressure and
the remaining solution stirred with NaBPh4 (0.086 g, 0.25
mmol) in H2O (1 mL). The resulting white precipitate was iso-
lated by ltration and washed with cold H2O (5 ꢃ 2 mL). The
solid was dried thoroughly to give slightly impure 26$(BPh4)4 as
2 (a) Y. Li and A. H. Flood, Angew. Chem., Int. Ed., 2008, 47,
2649; (b) Y. Li and A. H. Flood, J. Am. Chem. Soc., 2008,
130, 12111.
3 R. M. Meudtner and S. Hecht, Angew. Chem., Int. Ed., 2008,
47, 4926.
4 (a) V. V. Rostovtsev, L. G. Green, V. V. Fokin and
K. B. Sharpless, Angew. Chem., Int. Ed., 2002, 41, 2596; (b)
¨
C. W. Tornoe, C. Christensen and M. Meldal, J. Org. Chem.,
2002, 67, 3057.
5 (a) 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; (b) S. Lee, Y. Hua
and A. H. Flood, Org. Lett., 2010, 12, 2100; (c) Y. Wang,
F. Bie and H. Jiang, Org. Lett., 2010, 12, 3630; (d) Y. Hua
and A. H. Flood, J. Am. Chem. Soc., 2010, 132, 12838; (e)
Y. Hua, R. O. Ramabhadran, E. O. Uduchi, J. A. Karty,
K. Raghavachari and A. H. Flood, Chem.–Eur. J., 2011, 17,
312; (f) Y. Wang, J. Xiang and H. Jiang, Chem.–Eur. J., 2011,
17, 613; (g) Q.-Y. Cao, T. Pradhan, S. Kim and J. Kim, Org.
Lett., 2011, 13, 4386; (h) Y. Hua, R. O. Ramabhadran,
J. A. Karty, K. Raghavachari and A. H. Flood, Chem.
Commun., 2011, 47, 5979; (i) K. P. McDonald,
R. O. Ramabhadran, S. Lee, K. Raghavachari and
A. H. Flood, Org. Lett., 2011, 13, 6260; (j)
R. O. Ramabhadran, Y. Hua, Y. Li, A. H. Flood and
K. Raghavachari, Chem.–Eur. J., 2011, 17, 9123; (k) G. Yu,
Z. Zhang, C. Han, M. Xue, Q. Zhou and F. Huang, Chem.
Commun., 2012, 48, 2958; (l) S. C. Picot, B. R. Mullaney and
P. D. Beer, Chem.–Eur. J., 2012, 18, 6230; (m) S. Beckendorf,
1
a white solid. Yield: 0.007 g (ꢂ5%). H NMR (d6-acetone): 9.07
(s, 2H), 8.95 (s, 2H), 8.47 (s, 2H), 8.27 (s, 1H), 7.94–8.01 (m, 2H),
7.85–7.87 (m, 2H), 7.77 (s, 1H), 7.68–7.70 (m, 2H), 7.60 (s, 1H),
7.53–7.55 (m, 2H), 7.47–7.51 (m, 1H), 7.32–7.41 (m, 33H), 6.92–
6.95 (m, 24H), 6.77–6.87 (m, 16H), 6.55 (d, J ¼ 9.1 Hz, 4H), 6.13
(d, J ¼ 9.1 Hz, 4H), 5.02 (t, J ¼ 4.7 Hz), 4.97 (t, J ¼ 4.7 Hz), 4.82 (t,
J ¼ 4.7 Hz), 4.45 (t, J ¼ 4.7 Hz), 4.39 (t, J ¼ 4.7 Hz), 4.37 (t, J ¼ 4.7
Hz), 4.17 (s, 6H), 4.13 (s, 6H), 3.39–3.58 (m, 16H). See Fig. S22†
for assignment of spectrum. HRESI-MS (pos.): 362.6665, calc.
4+
for [C78H86N18O11
]
¼ 362.6675.
X-ray crystallography. X-ray crystal diffraction data were
˚
˚
collected using Mo Ka (l ¼ 0.71073 A), Cu Ka (l ¼ 1.5418 A) or
synchrotron32 Ka (l ¼ 0.6889 A) radiation. A Cryostream N2
S. Asmus, C. Muck-Lichtenfeld and O. G. Mancheno,
Chem.–Eur. J., 2013, 19, 1581; (n) L. C. Gilday, N. G. White
and P. D. Beer, Dalton Trans., 2013, 42, 15766.
˚
ˇ
¨
open-ow cooling device was used to cool the samples to 100 K
or 150 K. Series of scans were performed in such a way as to
collect a complete set of unique reections to a maximum
6 Kubik and Flood have reported cyclic and foldamer triazole
receptors that bind anions in aqueous media through a
mixture of hydrogen bonding interactions and solvophobic
effects:(a) M. R. Krause, R. Goddard and S. Kubik, J. Org.
Chem., 2011, 76, 7084; (b) Y. Hua, Y. Liu, C.-H. Chen and
A. H. Flood, J. Am. Chem. Soc., 2013, 135, 14401.
˚
resolution of 0.80 A. Raw frame data (including data reduction,
inter-frame scaling, unit cell renement and absorption
corrections) were processed using Denzo-SMN,33 CrystalClear34
or CrysAlis Pro.35 The structures were solved by direct methods
using SIR92,36 or charge-ipping methods using SUPERFLIP37
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 12133–12147 | 12145

View Article Online
RSC Advances
Paper
7 A. Kumar and P. S. Pandey, Org. Lett., 2008, 10, 165.
8 (a) K. M. Mullen, J. Mercurio, C. J. Serpell and P. D. Beer,
Angew. Chem., Int. Ed., 2009, 48, 4781; (b) B. Schulze,
atom provided by the triazole cycle compared with the
amide group. C. J. Serpell, O. Thomas, B. Odell,
T. D. W. Claridge, R. Chall, A. L. Thompson and
P. D. Beer, CrystEngComm, 2011, 13, 4586.
¨
¨
C. Friebe, M. D. Hager, W. Gunther, U. Kohn, B. O. Jahn,
¨
H. Gorls and U. Schubert, Org. Lett., 2010, 12, 2710; (c) 14 M. J. Hynes, J. Chem. Soc., Dalton Trans., 1993, 311.
R. Chhatra, A. Kumar and P. S. Pandey, J. Org. Chem., 2011, 15 N. G. White and P. D. Beer, Supramol. Chem., 2012, 24, 473.
76, 9086; (d) K. Ohmatsu, M. Kiyokawa and T. Ooi, J. Am. 16 Although surprisinglythe anion binding affinities were lower
Chem. Soc., 2011, 133, 1307; (e) N. G. White and P. D. Beer,
Beilstein J. Org. Chem., 2012, 8, 246; (f) Q.-Y. Cao,
than those for a ferrocene bis-triazolium macrocycle in the
same solvent, ref. 8e.
T. Pradhan, M. H. Lee, K. No and J. S. Kim, Analyst, 2012, 17 A signicant hydrogen bond being dened as one where the
137, 4454; (g) L. C. Gilday, N. G. White and P. D. Beer,
Dalton Trans., 2012, 41, 7092; (h) G. T. Spence, M. B. Pitak
H/anion distance was less than the sum of the van der
Waals radii of the two atoms.
and P. D. Beer, Chem.–Eur. J., 2012, 18, 7100; (i) 18 These are apparent 1 : 1 constants for more complex
N. G. White, S. Carvalho, V. Felix and P. D. Beer, Org.
Biomol. Chem., 2012, 10, 6951; (j) J. Cai, B. P. Hay,
N. J. Young, X. Yang and J. L. Sessler, Chem. Sci., 2013, 4,
equilibria. For a discussion of other processes involved
and the justications behind this approach, see:(a)
M. R. Sambrook, P. D. Beer, J. A. Wisner, R. L. Paul,
A. R. Cowley, F. R. Szemes and M. G. B. Drew, J. Am. Chem.
Soc., 2005, 127, 2292; (b) G. T. Spence, N. G. White and
P. D. Beer, Org. Biomol. Chem., 2012, 10, 7282.
´
1560; (k) N. G. White, P. J. Costa, S. Carvalho, V. Felix and
P. D. Beer, Chem.–Eur. J., 2013, 19, 17751.
9 Halotriazolium species have also proven to be effective anion
receptors:(a) N. L. Kilah, M. D. Wise, C. J. Serpell, 19 J. A. Wisner, P. D. Beer and M. G. B. Drew, Angew. Chem., Int.
A. L. Thompson, N. G. White, K. E. Christensen and Ed., 2001, 40, 3606.
P. D. Beer, J. Am. Chem. Soc., 2010, 132, 11893; (b) 20 These upeld shis are small, but do not occur when sulfate
N. L. Kilah, M. D. Wise and P. D. Beer, Cryst. Growth Des.,
2011, 11, 4565; (c) F. Kniep, L. Rout, S. M. Walter,
H. K. V. Bensch, S. H. Jungbauer, E. Herdtweck and
S. M. Huber, Chem. Commun., 2012, 48, 9299.
is added to the macrocycle in the absence of the threading
component, or when the threading component is added to
the macrocycle in the absence of sulfate.
21 A concentrated solution (100 mM) of threading component
is needed for quantitative pseudorotaxane titrations, and
the thread does not have sufficient solubility to permit
this. The stability of this pseudorotaxane assembly seems
to be very dependent on the batch of TBA2$SO4 used; solid
10 (a) E. D. Goddard-Borger and R. V. Stick, Org. Lett., 2007, 9,
3797; (b) N. Fischer, E. D. Goddard-Borger, R. Greiner,
¨
T. M. Klapotke, B. W. Skelton and J. Stierstorger, J. Org.
Chem., 2012, 77, 1760.
t
11 Pyridyl and Bu-phenyl substituted macrocycles were also
TBA2$SO4 is prepared by thoroughly drying
a
prepared, from 1 and 3,5-diethynylpyridine or 1 and 3,5-
diethynyl-tbutylbenzene, respectively (Scheme S1†).
12 These reaction conditions are based on those of Chouhan
commercially-available aqueous solution, and the success
or otherwise of pseudorotaxane formation seems to rely on
obtaining extremely dry TBA2$SO4.
and
James,
who
showed
that
they
favoured 22 We also attempted to prepare axle components having the
macrocyclisation vs. oligomerisation in the intramolecular
cyclisation of an azido-alkyne: G. Chouhan and K. James,
Org. Lett., 2011, 13, 2754.
bulky stopper groups directly attached to the triazolium
groups, but these appeared to be almost completely
insoluble in common organic solvents.
13 Serpell et al. have reported eight different forms of the 23 (a) L. M. Hancock and P. D. Beer, Chem.–Eur. J., 2009, 15, 42;
crystal structure of an analogous isophthalamide (b) L. M. Hancock, L. C. Gilday, S. Carvalho, P. J. Costa,
macrocycle, of which six structures have syn–syn
arrangement of the amide groups. In these structures the
´
a
V. Felix, C. J. Serpell, N. L. Kilah and P. D. Beer, Chem.–
Eur. J., 2010, 16, 13082.
distances between the amide nitrogen atoms ranges from 24 We initially prepared a bis-triazolium threading component S4
˚
5.10–5.24 A, the distances between the CH2 groups
containing octyl azide chains (Scheme S2†); rotaxanes could be
prepared from this using anion templated stoppering (Scheme
S3†) in approximately 25% yield, however, purication was very
difficult, and no pure rotaxanes were isolated.
˚
attached to the amide atoms ranges from 7.56–7.70 A, and
the distance between the interior phenylene carbon atom
and central polyether oxygen atom averages 13.7 A. For
˚
comparison purposes, in the two independent macrocycles 25 Stoppering rotaxanation reactions were also attempted
in the structure of 3 (Z0 ¼ 2), the distances between the
using isophthalamide macrocycles with different
t
˚
protic triazole carbon atoms are 5.39 and 5.41 A, the
substituents in place of the butyl group in 10 (specically,
t
distances between the CH2 groups attached to these are
having a phenyl or 5-pyridyl in place of the butyl-phenyl
9.57 and 9.61 A, and the distances between the interior
motif). In all cases, 1H NMR spectroscopy of the crude
reaction mixture suggested that the rotaxane was formed
in 25–30% yield, however, it was only possible to isolate
pure rotaxane when 10 was used.
˚
phenylene carbon atom and central polyether oxygen atom
are 12.6 and 12.9 A. There is no structurally characterised
analogue of 4 to permit a similar analysis, but 4 also
˚
seems to have a wider and squatter shape than similar 26 Data for acetate binding are not shown as the triazolium
isophthalamide macrocycles presumably due to the extra
proton disappeared during the titration due to hydrogen/
12146 | RSC Adv., 2014, 4, 12133–12147
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
deuterium exchange, and so the association constant for 32 H. Nowell, S. A. Barnett, K. E. Christensen, S. J. Teat and
acetate was determined following the interior phenylene D. R. Allan, J. Synchrotron Radiat., 2012, 19, 435.
proton. Association constants calculated using the 33 Z. Otwinowski and W. Minor, in Methods Enzymol, Academic
triazolium are considered to be the most accurate as this Press, 1997, p. 276.
proton displayed the largest shis, however calculating 34 CrystalClear, Rigaku Corporation, 2010.
association constants using the movement of the 35 CrysAlis Pro, Oxford Diffraction, 2011.
phenylene resonance gave very similar values.
27 B.-Y. Lee, S. R. Park, H. B. Jeon and K. S. Kim, Tetrahedron
Lett., 2006, 47, 5105.
36 A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi,
M. C. Burla, G. Polidori and M. Camalli, J. Appl.
Crystallogr., 1994, 27, 435.
28 D. E. Phipps and P. D. Beer, Tetrahedron Lett., 2009, 50, 3454. 37 L. Palatinus and G. Chapuis, J. Appl. Crystallogr., 2007, 40,
29 H. Zheng, W. Zhou, J. Lv, X. Yin, Y. Li, H. Liu and Y. Li,
786.
Chem.–Eur. J., 2009, 15, 13253.
38 P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout
and D. J. Watkin, J. Appl. Crystallogr., 2003, 36,
1487.
¨
30 V. Aucagne, K. D. Hanni, D. A. Leigh, P. J. Lusby and
D. B. Walker, J. Am. Chem. Soc., 2006, 128, 2186.
31 E. J. O'Neil, K. M. DiVittorio and B. D. Smith, Org. Lett., 2007, 39 R. I. Cooper, A. L. Thompson and D. J. Watkin, J. Appl.
9, 199.
Crystallogr., 2010, 43, 1100.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 12133–12147 | 12147