Sequence-Dependent Guest Release Triggered by Orthogonal Chemical Signals

Article abstract of DOI:10.1021/jacs.5b13016

Three Zn^II₄L₄ coordination cages, assembled from trisiminopyridine ligands, exhibit differences in their guest-binding selectivities and reactivity with tris(2-aminoethyl)amine (tren), which enabled the design of a molecul

Full text of DOI:10.1021/jacs.5b13016

Article
pubs.acs.org/JACS
Sequence-Dependent Guest Release Triggered by Orthogonal
Chemical Signals
Ana M. Castilla, Tanya K. Ronson, and Jonathan R. Nitschke*
Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, United Kingdom
S
* Supporting Information
ABSTRACT: Three ZnII L4 coordination cages, assembled from
4
trisiminopyridine ligands, exhibit differences in their guest-binding
selectivities and reactivity with tris(2-aminoethyl)amine (tren), which
enabled the design of a molecular network that responded in distinct
ways to different chemical signals. When two of these cages were
present in solution together, one of them was observed to selectively
encapsulate chloroform, and the other was observed to selectively
encapsulate cyclohexane. The two guests could be released
sequentially, in a specified order defined by the input of two separate
chemical signals: tren and perrhenate. Furthermore, the observed
reactivity of tren with the initial cage mixture provided control over the uptake and release of perrhenate within the third cage
formed in situ. One of these tetrahedral cages has been identified as a tight (Ka > 107 M−1) and selective host for perrhenate, an
anion of great physicochemical similarity to pertechnetate, both having uses in nuclear medicine.
overall response to be dependent on the sequence of applied
stimuli, a property that is not characteristic of any one cage
structure but which emerges from the system. This sequence
dependent response confers a further level of complexity on the
system, which would not be attainable by a simple collection of
two receptors that bind two different guests, where release can
be triggered by the addition of other competing guests, for
example.
INTRODUCTION
■
Increasingly fine control over the processes and outcomes of
chemical self-assembly has enabled the development, in recent
years, of complex chemical systems with useful functions that
emerge from the collectivity of their individual components.1,2
In order to shape these functions, studies have been carried out
into designing molecular networks and elucidating how they
behave in response to stimuli.3 These synthetic chemical
networks enable the design of materials able to adapt their
properties to changes in the environment.3a−g,4 Advances in
this area require gaining control over systems in which different
stimuli trigger independent and distinct responses, allowing
different behavior to be engendered.5,6 Selective sequences of
stimuli have been employed to determine the direction of travel
of a molecular walker5b or the successive release of cargos from
silica nanoparticles.5c
The well-defined inner phases of self-assembled metal−
organic polyhedra7,8 have proven useful in a diverse range of
applications,9 from molecular recognition and sensing8a,10 to
gas sequestration,11 stabilization of reactive species,12 and
catalysis.13 These hosts are excellent candidates for incorpo-
ration into molecular networks to explore complex and stimuli-
responsive behaviors14,15 due to their encapsulation abilities
and the dynamic nature of the linkages that hold them together.
Investigating systems that comprise multiple hosts and guests
together may allow for new functions to be designed, beyond
what is achievable with single host−guest systems.14a,15c,16
Here we describe a system composed of self-assembled cages
that has been designed to exhibit complex guest release
behavior in response to two distinct chemical signals. These are
a competing guest and a reagent, tren, which effects a host
transformation. In addition, we demonstrate the system’s
RESULTS AND DISCUSSION
■
To design this system, we selected three face-capped17 ZnII L4
4
tetrahedral capsules, in which tritopic iminopyridine ligands are
formed from either a phenyl-centered tris-aniline16c,17a or a
phenyl-centered tris-formylpyridine.18 A detailed study allowed
us to identify two key features for the implementation of a
stimuli-responsive molecular network: first, contrasting guest
binding preferences and affinities, and second, orthogonal
reactivities of the tris-aniline and tris-formylpyridine based
structures with tris(2-aminoethyl)amine (tren). Consideration
of these features led to the design of the network depicted in
Scheme 1. Two different neutral guests, cyclohexane and
chloroform, were each selectively encapsulated in one of the
two ZnII L4 hosts (1 and 2). Each guest could be selectively
4
released using one of two distinct chemical signals: treatment
with tren released cyclohexane and addition of perrhenate
liberated chloroform. Reversing the order of the signals
reversed the order of guest release. Intriguingly, whereas one
signaling pattern (sequence I) resulted in complete destruction
of the cages and ultimately ejection of perrhenate into solution,
Received: December 16, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 1. Sequence-Selective Release of Guests Triggered by Orthogonal Chemical Signals
a
Signal Sequence I: (i) disassembly of 1 and release of C6H12; (ii) release of CHCl3 by displacement with ReO4− from the cavity of 2; (iii) release of
−
−
ReO4 upon disassembly of 2. Signal Sequence II: (iv) release of CHCl3 by displacement with ReO4 from the cavity of 2; (v) simultaneous
breakdown of 1 releasing C6H12 and transformation of 2 into 3 while maintaining sequestration of ReO4 . A and B denote the face-capping
−
subcomponents 1,3,5-tris(4′-aminophenyl)benzene (for cage 1) and 1,3,5-tris(2′-formylpyridyl-5′)benzene (for cages 2 and 3). Note that the
insoluble products in sequence I contain triamine A and trialdehyde B.
the reverse pattern (sequence II) allowed perrhenate to be
trapped within a stable host (3) formed at the end of the
sequence.
by single-crystal X-ray diffraction (Figure 1). The four facially
coordinated ZnII centers are bridged by four face-capping
ligands, resulting in a tetrahedral arrangement with approximate
T-symmetry. All ZnII stereocenters within a cage share the same
Δ or Λ stereochemistry; both cage enantiomers are present in
the crystal. The cavity of 2 is almost completely enclosed by the
ligands, with pores of less than 1.3 Å in diameter. The Zn−Zn
distances are in the range 11.278(4)−11.774(3) Å (average
11.5 Å), and the cavity volume was calculated to be 130 Å3
using VOIDOO (see section 7 in the Supporting Informa-
tion).23
Similarly, the reaction in acetonitrile/methanol (1:1) of B,
tren, and Zn(NTf2)2 in a 1:1:1 ratio generated tetrahedral cage
3, isolable as a yellowish crystalline solid. The single-crystal
structure of 3 (Figure 1) closely resembles that of 2, except that
tren residues cap the vertices of the tetrahedron, forming an
extended cryptand-like architecture. The Zn−Zn distances of
11.749(3)−11.775(3) Å fall within the range observed for 2;
the average distance is 11.8 Å. The cavity volume was calculated
to be 111 Å3, marginally smaller than 2 due to the faces of 3
pressing inward slightly relative to those of 2 (see section 7 in
the Supporting Information). The use of a smaller tris-
formylpyridine based ligand thus leads to cages that enclose
less volume than cage 1 (Zn−Zn distance 14.6 Å, volume 188
Å3), formed from the analogous tris-aniline subcomponent A
(Figure 1).16c
The design of this network is grounded upon systematic
investigations of the guest binding properties of hosts 1−3
(Scheme 1), which also revealed the unprecedented affinity of
new hosts 2 and 3 for perrhenate.19,20 This anion is relevant as
a surrogate in the design of receptors for radioactive
pertechnetate and also to applications in nuclear medicine;
the development of selective perrhenate and pertechnetate
receptors has proven particularly challenging.19,21 Furthermore,
significant differences in anion uptake kinetics were uncovered
between tris-formylpyridine-based cages 2 and 3, whose
vertices are capped with three toluidine residues or one tren,
respectively. These differences provided insight into the guest
uptake and exchange mechanisms22 of the face-capped
tetrahedra described herein.
Synthesis and Characterization of Cages 1−3.
Tetrahedra 1−3 (Scheme 2) self-assembled from zinc(II) and
tritopic subcomponents, either 1,3,5-tris(4′-aminophenyl)-
benzene (A,17a cage 1) or 1,3,5-tris(2′-formylpyridyl-5′)-
benzene (B,18 cages 2 and 3). The synthesis of 1 has been
previously described.16c The reaction of B, p-toluidine, and
−
zinc(II) bis(trifluoromethylsulfonyl)imide (triflimide, NTf2 )
in a 1:3:1 ratio in acetonitrile afforded 2, isolable as a greenish
crystalline solid. Vapor diffusion of diethyl ether into an
acetonitrile solution of 2 produced crystals suitable for analysis
B
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
−
NTf2 , confirmed that the cages do not bind this anion in
solution (Figures S3 and S4). Triflimide was, indeed, chosen
specifically to avoid counterion encapsulation, in order to
facilitate host−guest studies.10a Considering the volumes of
Scheme 2. Subcomponent Self-Assembly of Cages 1−3 and
a
the Transformation of 2 into 3
18
−
their FeII-templated analogs, we anticipated that NTf2 (157
Å3) would be too voluminous to fit in the cavity of cages 2 and
3.10a
In similar fashion to their FeII-containing congeners,18 cage 3
could also be prepared through substitution of the 12 p-
toluidine residues incorporated into the periphery of cage 2
with 4 equiv of tren (Scheme 2). The treatment of a solution of
cage 2 in acetonitrile with 4.5 equiv of tren at 70 °C afforded
1
cage 3 as the only product observed by H NMR and ESI-MS
(see section 1.4 in the Supporting Information). We infer this
imine exchange reaction to be driven by the more electron-rich
character of tren and the chelate effect.
Cage 2 was also prepared from the zinc(II) salts of
tetrafluoroborate (2·[BF4]8), perchlorate (2·[ClO4]8), and
triflate (2·[OTf]8). Similarly 3·[OTf]8 was obtained from
Zn(OTf)2 in a CH3CN/CH3OH mixture. In contrast, attempts
to form cage 3 from Zn(BF4)2 or Zn(ClO4)2 resulted in
insoluble products. Cage 3·[BF4]8 could, however, be prepared
through reaction of 2·[BF4]8 with tren, whereas analogous
reactions with 2·[ClO4]8 and 2·[OTf]8 afforded intractable
precipitates.
Anion Binding Studies. The anion encapsulation abilities
of cages 2 and 3 were probed by treating them in solution with
a series of anions having different shapes and volumes (listed in
Tables S1 and S6 in the Supporting Information). Previous
studies determined that tetrahedron 1 does not bind anions in
its cavity.16c,17a Cage 2 was observed to bind the anions (in
−
−
−
−
−
−
order of size) NO3 , BF4 , ClO4 , ReO4 , PF6 , SbF6 , and
TfO−, as confirmed by H and 19F NMR. The addition of the
1
tetrabutylammonium salt of ClO4 , ReO4 , PF6 , or TfO− or
−
−
−
−
the potassium salt of SbF6 (0.5 equiv) to a solution of
2·[NTf2]8 resulted in the appearance of a new set of H NMR
1
a
Only one ligand is drawn per structure for clarity.
signals, assigned to the inclusion complexes in slow exchange
with the free cage on the NMR time scale (Figure S12).
Solutions containing PF6− or TfO− showed two new 19F NMR
ESI-MS and NMR analyses reflect solution structures of 2
and 3 analogous to what is observed in the solid state. Their
simple 1H NMR spectra, with only one set of ligand resonances,
are consistent with the formation of a single diastereomer with
T point symmetry. Their 19F NMR spectra, with only one sharp
signal having a chemical shift corresponding to unencapsulated
−
signals (in addition to the NTf2 resonance) attributed to free
and encapsulated anions (Figures S8 and S15). The 19F NMR
−
spectrum of the solution containing SbF6 showed a
broadened, extended multiplet assigned to this anion due to
overlapping signals of free and encapsulated species.
Figure 1. X-ray crystal structures of cages 1,16c 2, and 3. Anions and solvent molecules are omitted for clarity.
C
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
In contrast, the addition of tetrabutylammonium salts of the
−
−
smaller anions BF4 or NO3 to a solution of “empty” 2
1
provided evidence for anion binding in fast exchange. The H
NMR signals of the host were observed to shift, with the
resonances due to the central phenyl and inward-facing
pyridine protons undergoing the greatest shifts (Figure S12).
−
In the case of BF4 , broadening of the 1H and 19F NMR spectra
was also observed. Encapsulation of BF4− was further supported
1
by a H−19F HOESY spectrum, in which correlations were
observed between the anion resonance and signals correspond-
ing to the protons of the ligand pointing toward the inside of
the cavity (Figure S18). Other anions screened, such as Cl−,
Br− or I−, gave rise to a color change and precipitation
following their addition to a solution of 2·[NTf2]8, consistent
with cage decomposition.
1
−
Figure 2. (left) Curve fit for the H NMR titration of ReO4 into a
1
solution of TfO−⊂2 in CD3CN to a competitive binding model.10a
Anion-binding strengths were quantified through H NMR
titrations, and the results are given in Table 1 (see section 2.2 in
1
(right) Imine region of selected H NMR spectra showing formation
−
of ReO4 ⊂2 (red) and consumption of TfO−⊂2 (black) upon
−
addition of increasing amounts of ReO4 . See Figure S27 for further
Table 1. Summary of Binding Constants (Ka) for Anions in
a
details on the data fitting.
Cages 2 and 3
b
Ka (M−1); NMR exchange
Strikingly, cage 3 was found to exhibit substantially different
guest binding abilities from cage 2, despite their structural
V
c
d
guest
(Å3)
2
3
similarities.29 The addition of NO3 , BF4 , ClO4 , PF6 , or
−
−
−
−
(1.5 0.3) × 104; fast
(7.1 0.2) × 102; fast
(3.0 0.2) × 103; slow
(2.2 0.4) × 107; slow
(1.4 0.1) × 104; slow
(2.5 0.6) × 106; slow
nonbinding
nonbinding
nonbinding
>105; slow
−
NO3
40.7
53.3
54.8
59.8
74.7
84.7
85.0
TfO− to a solution of 3·[NTf2]8 in acetonitrile caused only
−
BF4
1
slight (<0.08 ppm) shifts in the H NMR spectra, even after
−
ClO4
equilibration at room temperature for several hours (see
Figures S33 and S47), in marked contrast with the behavior of
the cage 2. We attribute these changes to a weak interaction of
the anions with the exterior of the cage rather than
encapsulation.30
−
ReO4
e
f
−
PF6
21 3; slow
e
f
−
SbF6
115 8; slow
(3.6 0.3) × 105; slow 41 3; no exchange observed
e
TfO−
a
b
Cage 1 does not encapsulate anionic guests.16c,17a Full details of how
Previous work has shown that the incorporation of electron-
rich or electron-poor aniline residues into the periphery of
Ka values and corresponding errors were calculated are given in the
c
Supporting Information sections 2.2 and 2.3. Addition of Cl−, Br−, or
related FeII L6 capsules did not affect their anion-binding
d
I− to solutions of 2 or 3 induced cage decomposition. Calculated van
4
preferences.10a We had therefore not anticipated that the
exchange of p-toluidine for the more electron-rich tren, in going
from 2 to 3, would have such an impact on the anion-binding
preferences. We reasoned the different behavior of cage 3 may
be attributed to the covalent locking effect of tren preventing
partial cage opening during anion exchange (discussed
below).18,10h
e
der Waals volumes, see the Supporting Information. Estimated values.
f
Not observed below 70 °C.
the Supporting Information for experimental details). The
affinity of 2 for SbF6 , ReO4 , or TfO− was found to be too
high for an accurate direct determination of their Ka values,
which were instead derived through competitive binding
−
−
In order to probe whether the failure to observe anion
binding within 3 is due to a thermodynamic or a kinetic effect,
we performed three different sets of experiments followed by
NMR, illustrated in Scheme 3. In the first (Scheme 3i), cage 3
was prepared from subcomponents in the presence of different
prospective anionic guests. In the second (Scheme 3ii), the
fates of anions encapsulated within 2 were charted during the
course of a 2 to 3 transformation. In the third (Scheme 3iii),
preformed 3·[NTf2]8 was treated with the same series of anions
at 70 °C during a time course of many days. Cage 3 was
−
−
experiments: titration of SbF6 and TfO− against PF6 ⊂2
and titration of ReO4 against TfO−⊂2 (Figure 2).
−
In combination, these experiments show that cage 2 is
capable of accommodating in its interior monocharged anions
with volumes ranging from 40 to 85 Å3 with the following
hierarchy: ReO4− > SbF6− > TfO− > PF6− ≈ NO3− > ClO4
>
−
−
BF4 . These relative affinities deviate from what would be
predicted from Rebek’s 55% occupancy optimum.24,25 We infer
that a subtle interplay of size and shape complementarities
between host cavity and guest, solvation effects, and electro-
static interactions determine together the observed hierarchy,
with no single factor predominating.20,26 Within a series of
anions with the same geometry, larger anions are more strongly
−
−
−
−
observed to bind ReO4 , PF6 , SbF6 , and TfO− but not NO3 ,
−
−
BF4 , or ClO4 during its formation (Scheme 3i,ii); the same
set of anions were encapsulated following prolonged heating
(Scheme 3iii), with the exception of triflate. Experimental
details of anion encapsulation studies are provided in the
Supporting Information, section 2.3.
−
−
−
bound, such as ReO4 and SbF6 . Despite TfO− and SbF6
having the same molecular volume (Table 1), we infer that the
better symmetry match between octahedral SbF6 and the
Binding of perrhenate inside tetrahedron 3 was confirmed by
X-ray crystallography (Figure 3). The encapsulated ReO4 is
−
−
tetrahedral cavity renders it a better guest. The high association
located close to the center of the tetrahedral cavity with the
oxygen atoms oriented toward the zinc centers. The Zn−Zn
distances and volume are similar to the empty cage. In addition,
−
constant determined for the trigonal planar NO3 , five times
−
greater than that of the larger tetrahedral ClO4 , may be
attributed to the lower hydrophobicity of nitrate.27,20
the complex ReO4 ⊂3 was found to be stable in water. The
−
D
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
−
excess of the selected anion (Table 1). The binding of ReO4
Scheme 3. Experiments Probing the Anion-Binding
Properties of Cage 3
a
to 3 was found to be too strong for estimation of its association
constant by this method, but a lower limit of 105 M−1 could be
established by NMR (see Supporting Information, section
2.3.3).
−
−
−
In summary, with the exception of BF4 , NO3 , and ClO4 ,
which have been found to bind only to 2, both tris-
formylpyridine-based ZnII L4 structures 2 and 3 showed similar
4
trends in anion-binding preferences: ReO4− ≫ SbF6− > TfO− >
−
PF6 .
The quantification of anion-binding strengths revealed an
−
outstanding selectivity of cages 2 and 3 for ReO4 . Cage 2 has
−
−
10 and 60 times greater affinity for ReO4 than for SbF6 or
TfO−, respectively, the next most strongly bound anions (Table
1). As discussed above, this is likely due to a combination of
symmetry match between host and guest and optimal volume
occupation ratio. To the best of our knowledge, cage 2
represents the strongest 1:1 perrhenate binding host (Ka = (2.2
0.4) × 107 M−1) reported to date in either organic or
aqueous media.19,21 The combination of water stability and
−
exceptional affinity for ReO4 suggests that 3 might show
promise in pertechnetate binding, of relevance in the context of
radiopharmaceuticals and nuclear waste treatment, as discussed
in the Supporting Information, section 2.3.4.19,20,21c,28
Kinetics and Mechanism of Anion Uptake. Despite
showing similar anion binding preferences, very different guest
exchange kinetics were observed for cages 2 and 3. This
observation led us to carry out a brief kinetic study, the results
of which shed light upon the mechanisms of guest exchange.
a
(i) When formed from subcomponents, 3 is observed to encapsulate
−
−
−
−
−
−
ReO4 , PF6 , SbF6 , and TfO−, but not NO3 , BF4 , or ClO4 ; (ii) the
same anion selectivity was observed during the formation of 3 from 2;
(iii) the same anions were observed to be taken up within 3 following
lengthy equilibration, except triflate.
−
−
The smallest anions, NO3 and BF4 , exchanged between free
and encapsulated states within 2 at a rate more rapid than the
NMR time scale. We estimate a lower limit of 30 s−1 for the
exchange rates of these anions, considering a difference of
1
about 27 Hz between H NMR resonances of the empty and
guest-containing cage.31 The guest exchange kinetics of ClO4
−
were examined by 1H−1H exchange spectroscopy (EXSY)
NMR,22b,32,33 providing an uptake rate constant (kin) of (3.0
0.5) × 103 M−1 s−1 at 25 °C. Rate constants for the guest
−
−
−
exchange of ReO4 , PF6 , SbF6 , and TfO− could not be
determined by EXSY because the uptake rates were too slow
for the time scale of this technique (even at 70 °C) but also too
1
fast to be followed by H NMR: in all cases the system had
already reached equilibrium by the time the first H NMR
1
spectrum could be acquired following addition of anion to the
cage solution. Considering the time scale of the EXSY
experiment, we infer the kin values for these guests to be
lower than 103 M−1 s−1 (see Supporting Information, section
2.4 and Table S7).
The slower anion uptake rates exhibited by the tren-
−
Figure 3. Crystal structure of ReO4 ⊂3. Only one of the two
crystallographically distinct cages is shown. The encapsulated ReO4− is
shown in space-filling mode and nonencapsulated anions are omitted
for clarity.
containing tetrahedron 3 allowed encapsulation to be followed
by 1H NMR (PF6 ) or UV−vis (ReO4 ), following the
addition of excess anion to a solution of empty cage under
pseudo-first order conditions. These experiments were
−
−
−
performed at 70 °C since exchange of PF6 was not observed
nitrate salt of host−guest complex, although of modest
solubility (ca. 0.2 mM), showed no degradation following 24
h at room temperature in D2O (Figure S41).
at lower temperatures. At concentrations suitable for NMR
−
analysis, the addition of any excess of ReO4 to 3 in solution
caused precipitation. To circumvent this practical problem, we
The slow uptake of anions into tetrahedron 3 prevented
determination of their association constants through titration
−
followed ReO4 uptake at lower concentrations by UV−vis.
The second-order rate constants, kin, for ReO4− and PF6− were
determined to be 47 2 M−1 s−1 and (1.7 0.4) × 10−3 M−1
experiments. The binding strengths of PF6 , SbF6 , and TfO−
were estimated by measuring the relative integration of signals
due to free and bound host in the 1H NMR spectra of samples
following the transformation of 2 into 3 in the presence of an
−
−
s−1, respectively, at 70 °C. The kinetics of inclusion for SbF6
−
and TfO− into 3 could not be determined because of their very
E
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
slow and nonobserved uptakes, respectively (see section 2.4.2
in the Supporting Information).
Table 2. Comparison of Neutral Guest-Binding Properties of
Cages 1 and 2
a
The time scales for anion exchange given in Figure 4
illustrate the large differences in uptake rates between 2 and 3.
b
Ka (M−1
)
c
d
guest
V (Å3)
1
2
e
CH2Cl2
CHCl3
CCl4
60.9
74.7
low
1.2 0.2
11 0.2
(1.5 0.1) × 102
(1.2 0.1) × 103
88.7
15
3
e
tBuOH
95.4
low
3.5 0.8
10 0.8
6.6 0.6
2.8 0.4
nonbinding
nonbinding
6.8 0.3
5.2 0.3
nonbinding
g
cyclopentane
95.3
(6.7 0.4) × 102
e
cyclopentanol
methylcyclopentane
1-methylcyclopentanol
cyclohexane
102.6
113.2
120.5
111.9
116.5
120.2
138.3
146.5
99.5
low
(1.6 0.1) × 102
Figure 4. Relative time scales of anion uptake by cages 2 and 3. Half-
e
low
lives are based on apparent rate constants at a 1 mM guest
f
(4.9 0.1) × 102
(6.9 0.4) × 102
(1.1 0.1) × 103
(49 3)
−
concentration. Note that for PF6 and 3 no exchange was observed
below 70 °C.
norbornene
norbornane
7-bromonorbornane
cyclooctane
The incorporation of chelating tren in 3 was thus observed to
slow encapsulation dramatically. Both of the plausible anion
uptake mechanisms, diffusion of guest through the structure’s
portals or partial disassembly to create transient larger
portals,22b are expected to be more energetically costly in
cage 3. The covalent bonds of 3 must be distorted or cleaved in
order for the cage to open, whereas 2 may be opened through
the stretching or rupture of weaker coordinative linkages.
The enclosed and rigid structure of the face-capped
tetrahedra 2 and 3, which appeared to leave no access for
guest diffusion through the small portals on the edges (Figures
S95 and S96), led us to hypothesize that the exchange of any
guest would require N→Zn bond breakage. The observed
marked dependence of anion uptake rates upon the size and
shape of the guest, however, suggests that more than one
mechanism may be at work. The fast exchange of the smallest
e
low
e
benzene
low
nonbinding
nonbinding
nonbinding
nonbinding
nonbinding
nonbinding
g
1,3,5-trifluorobenzene
1,3,5-trimethoxybenzene
naphthalene
113.1
180.1
151.0
106.8
125.2
159.1
177.3
nonbinding
nonbinding
nonbinding
e
n-pentane
low
n-hexane
nonbinding
(59 4)
adamantane
1-bromoadamantane
nonbinding
g
a
b
Cage 3 showed no evidence for binding neutral guests. Full details
of how Ka values and corresponding errors were calculated are given in
Supporting Information section 2.5. Calculated van der Waals
volumes, see the Supporting Information for details. From Krel
values determined in competitive experiments with cyclohexane. The
c
d
reported error for each Ka value was estimated by error propagation
e
analysis (see Table S8 and section 2.5.1). Binding too weak to
−
anions, BF4− and NO3 , in and out of cage 2 seems unlikely to
f
g
displace cyclohexane. Determined by 1H NMR titration. Not
involve bond-breaking.34 We infer that these anions may be
undergoing exchange via a through-portal mechanism, whereby
the ligands distort sufficiently to allow anion exchange without
coordinative bond cleavage.35,15b The slower exchange
examined for binding to 2 due to its large size
1
host−guest complexes were inferred to form, the H NMR
exhibited by the largest anions PF6 , SbF6 , and TfO− appears
likely to involve partial cage opening and N→Zn bond rupture,
which we infer to incur a considerably higher energetic penalty
for cage 3.22b Perchlorate, showing an uptake rate intermediate
between these two classes of anions, may exchange via a more
energetically costly cage deformation or partial vertex
decoordination or both. In addition, the higher association
spectrum of an equilibrated mixture of an excess of the selected
guest and the cage in CD3CN showed two sets of host peaks,
attributed to “empty” ZnII L4 and guest⊂ZnII L4 in slow
−
−
4
4
exchange, and also two sets of signals for the guest, assigned to
the free and encapsulated guests (Figures S64−S80).
Host 1 was reported in a preliminary study to accommodate
cyclohexane and tBuOH within its cavity.16c We screened an
extended series of neutral molecules, including those observed
to bind inside cage 2 (see below), and also explored their
relative binding strengths. The association constant (Ka) of
−
−
constants of ClO4 and ReO4 , having the same shape and
−
slightly larger volumes than BF4 , can also hamper exchange,
accounting for why the observed exchange time scale for ReO4
−
1
is on the same order as for the larger anions.
cyclohexane in 1 was calculated through a H NMR titration
Collectively, the insights gained from these anion binding
studies enables the design of systems incorporating the
responsive behavior of tris-formylpyridine based cages 2 and
3 and anions: guest release on treatment of an anion⊂2
complex with an anion with higher affinity and treatment with
tren to form 3 (Scheme 3ii) with concomitant guest release
experiment to be (4.9 0.3) × 102 M−1. For all other guests,
affinities relative to cyclohexane were obtained by NMR on the
basis of their ability to displace cyclohexane from the cavity of
cage 1 (Supporting Information, section 2.5.1). Host 1 was thus
revealed to show similar guest-binding abilities to those of its
FeII congener,17a although 1 was able to bind larger guests than
the latter, such as cyclooctane and adamantane, due to its larger
cavity.16c,17a The most strongly bound guests for 1 are CCl4 >
norbornane > norbornene > cyclopentane > cyclohexane.
Host 2 was found to accommodate small hydrophobic
molecules with volumes from 61 Å3 (dichloromethane) to 120
Å3 (norbornane). Notably, certain molecules, such as benzene,
n-pentane, or cyclohexane, with calculated volumes within the
above range, did not bind within 2, reflecting the necessity of a
shape fit. In all cases, the measured affinities were too weak to
−
−
−
(NO3 , BF4− and ClO4 ) or guest trapping in its cavity
(ReO4 , PF6 , SbF6 and TfO¯).
−
−
Neutral Guest Binding. The ability of tetrahedral cages 1−
3 to act as hosts for neutral molecules was also investigated in
solution by NMR. To first establish the scope of guest binding,
we screened a series of neutral molecules, listed in Table 2,
selected with different sizes and molecular volumes, distributed
around the optimal guest volume for each cage predicted using
Rebek’s 55% optimum occupancy rule.24 In all cases where
F
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
1
allow for determination of the binding constant via H NMR
titration. Instead, they were estimated by measuring the relative
integration of signals for free and bound host in slow exchange
at different guest concentrations (see section 2.5.2 in the
Supporting Information). The strongest binders are CCl4 >
cyclopentane > CHCl3, suggesting the volume range 75−95 Å3
to be optimal for encapsulation within cage 2.
integration), which we infer to be due to the reaction between
liberated A and 2 (as discussed below), yet no signals
corresponding to cage 3 or free A were identified. Subsequent
addition of tren (5 equiv) did not result in the expected
transformation of 2 into 3, resulting instead in the formation of
insoluble material. Only the mononuclear complex and p-
toluidine were observed in solution after heating the mixture
overnight to 70 °C. We infer the precipitate to result from the
reaction of subcomponents A and B, which are only sparingly
soluble in acetonitrile.39
None of the prospective neutral guests showed evidence for
binding to cage 3, even following heating to 60 °C for 5 days or
assembly of 3 in the presence of excess prospective guest.
Selective Guest Binding within Mixtures. Table 2
provides an overview of the neutral-guest-binding properties
of tetrahedra 1 and 2. From the data presented in Tables 1 and
2, it is possible to draw the following conclusions: (i) tetrahedra
2 and 3 encapsulate anions with medium-to-high affinities; (ii)
tetrahedron 3 binds a subset of the anions found to bind to 2,
with lower affinities; (iii) 2 binds weakly a series of neutral
molecules with volumes ranging 60−120 Å3; (iv) tetrahedron 1
encapsulates a wider range of neutral guests, including all of
those observed to bind within 2; (v) in all cases 1 shows a
higher affinity than 2 for each neutral guest, and for both cages
the most strongly bound neutral guest is CCl4. From these
observations, several three-guest systems can be selected
wherein two of the guests (C6H12 and CHCl3 in Scheme 1)
are selectively bound to 1 and 2, respectively, in a 1:1 cage
mixture, and a third anionic guest (ReO4− in Scheme 1) may be
added to the mixture in order to selectively trigger the release
of the first guest from the cavity of cage 2, thus acting as a
selective chemical stimulus to the system.
In a separate experiment, we also tested the reaction of the
mixture of 1 and 2 with excess tren (10 equiv: more than the
amount required to break down 1 and convert 2 into 3) in a
1
single addition (Figure S84). The H NMR spectrum of the
mixture after heating to 70 °C for 12 h confirmed complete
disassembly of 1 accompanied by formation of the mono-
nuclear complex and release of A, as well as the formation of
cage 3 with release of p-toluidine, while no precipitate was
observed.
This pathway-dependent reaction outcome may be a
consequence of the ability of tren to induce the partial
disassembly of 2 by first extracting the metal template from the
structure. Such extraction has been observed to occur during
the substitution reaction with tren of FeII-containing cages,18
and we infer it to be more favorable in a system based on ZnII, a
more labile metal ion. Following the tren-mediated partial
disassembly of 2, the free tris-aniline A present in the mixture
may interfere with the reaction pathway leading ultimately to
the formation of 3. We infer the reaction between tris-aniline A
and tris-formylpyridine B to result in the formation of cross-
linked oligomeric material that precipitates, thus removing both
subcomponents from solution during the disassembly of cage 2
in the presence of A. Indeed, the addition of tren (4.5 equiv) to
a solution containing 2 and tris-aniline A (4.5 equiv) resulted in
precipitation (Figure S87) and not the formation of 3. We thus
infer this process to occur on the second addition of tren to the
cage mixture, once 1 has disassembled. A single addition of the
amount of tren required to react with both cages in the initial
mixture, in contrast, suppresses the formation of insoluble
oligomeric material. In this case, we hypothesize that a broader
range of more flexible and soluble intermediate products may
be generated, in which tren has partially reacted with the
frameworks of both 1 and 2. The excess tren thus serves as a
buffer by preventing A and B from reacting directly in these
intermediate states, thus keeping B in solution long enough for
3 to form.
Two sets of guests were selected to demonstrate sequence-
selective release from an initial 1:1 mixture of 1·[NTf2]8 and
2·[NTf2]8 in CD3CN. The first set of guests, shown in the
system of Scheme 1, consists of the two neutral molecules
C6H12 and CHCl3. The H NMR spectrum after addition of
C6H12 and CHCl3 (130 equiv each) showed selective binding
1
of cyclohexane to 1 and of CHCl3 to 2 (Figure S81). The
subsequent addition of ReO4 (1.1 equiv) to the mixture
−
−
showed selective formation of ReO4 ⊂2.
−
The second set of guests comprises two anions (PF6 and
ReO4 ) and a neutral molecule (C6H12). H and 19F NMR
spectra taken of a mixture of 1 (1 equiv), 2 (1 equiv), PF6− (1.7
equiv), and C6H12 (88 equiv) showed exclusive formation of
−
1
−
−
C6H12⊂1 and PF6 ⊂2. Subsequent addition of ReO4 (1.3
equiv) displaced PF6 from 2 to form the ReO4 ⊂2 complex
−
−
(Figure S82).
Reaction of Cage Mixtures with tren. Next we set out to
explore tren as a selective chemical stimulus, taking advantage
of the differential reactivity of cages 1 and 2 toward this
triamine. As discussed above, the reaction of cage 2 with tren
affords cage 3. In contrast, tren is observed to induce
disassembly of cage 1 by extracting its constituents ZnII and
2-formylpyridine to form the mononuclear complex zinc(II)
tris(pyridyliminoethyl)amine and release free A (Figure S86).36
Remarkably, the outcome of the reaction of a mixture of 1 and
2 with tren was observed to be pathway dependent.15g,37
The addition of tren (4 equiv) to a mixture of 1 and 2 (1:1)
in CD3CN resulted in the selective disassembly of cage 1
(Figure S83). After 10 min at 25 °C, 60% of 1 had already been
consumed, whereas 2 remained intact. After equilibration of
this mixture at 60 °C for 12 h, cage 1 had been totally
consumed and the mononuclear complex formed (ca. 4 equiv
relative to the initial amount of 1).38 A decrease in the total
Control of Sequential Guest Release through Orthog-
onal Chemical Signals. The studies described above enabled
us to devise a system displaying complex stimuli-responsive
guest release behavior (Scheme 1). Each step of the sequence
was monitored by NMR (Figures 5 and S88−S92).
Starting from a mixture of C6H12⊂1, CHCl3⊂2, and “empty”
−
2 (1:0.5:0.5), the sequential addition of tren and then ReO4
brought about the release of cyclohexane and chloroform in
that order, as shown in sequence I of Scheme 1. (i) The
selective release of cyclohexane upon disassembly of cage 1
occurred following the addition of tren (4 equiv relative to the
total amount of 1) and heating at 60 °C for 12 h. This process
1
was tracked by following the disappearance of the H NMR
resonances corresponding to encapsulated cyclohexane and
cage 1 (Figure S88). As described in the analogous experiment
in the absence of guests, a small amount of cage 2 had also been
consumed (ca. 20%) after heating. (ii) The subsequent addition
1
amount of cage 2 was also observed (ca. 20% by H NMR
G
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
investigated. The insights gained have enabled the design of a
chemical system with complex guest encapsulation behavior, in
which three guests are individually released in response to
distinct chemical signals. As a result, sequence-selective guest
release triggered by the specific order of applied stimuli was
demonstrated, while the identification of a pathway-dependent
reaction of the cage mixture with tren brought about control
over the system’s overall response, release or capture of the
third guest at the end of the sequence.
These findings provide new means for the rational design of
more complex systems. Control over the order in which guests
are released on demand might be exploited in the development
of multidrug delivery systems, to control the sequential
reactivity of multiple catalysts in a reaction mixture, or the
release of guests that act as signals to activate subsequent
processes. This work thus demonstrates how the study of
systems composed of multiple molecular containers with
different properties and stimuli-responsive behavior may allow
new complex properties and functions, such as pathway-
dependent reactivity, to emerge.
Additionally, cage 2 was found to be an outstanding host for
perrhenate, which can be permanently trapped by in situ
transformation into cage 3 by the addition of tren. This slow
guest exchange kinetics observed for tren-containing 3 may be
relevant for the construction of new more stable capsules for
trapping and storage of perrhenate, pertechnetate, or other
guests.
1
Figure 5. Stacked plots of H NMR spectra corresponding to the
stimulus/response sequences shown in Scheme 1, starting from a 1:1
mixture of 1 and 2 selectively encapsulating cyclohexane and
chloroform, respectively. Only the imine signals of the different
species are labeled as follows: red dot = C6H12⊂1, blue open circle =
−
“empty” 2, blue dot = CHCl3⊂2, blue diamond = ReO4 ⊂2, blue
−
triangle = ReO4 ⊂3, and green star = mononuclear complexes.
Intensities have been scaled for clarity.
of ReO4− (1.1 equiv) brought about the complete displacement
−
of CHCl3 from 2 to form the ReO4 ⊂2 inclusion complex after
equilibration of the mixture at 70 °C for 2 h. (iii) Finally, the
−
liberation of ReO4 was achieved upon disassembly of 2 and
precipitation of subcomponents A and B on addition of a third
signal, tren (4 equiv), to the previous mixture and heating at 70
°C for 12 h, as confirmed by H NMR.
40
1
ASSOCIATED CONTENT
* Supporting Information
Crystallographic data have been deposited with the Cambridge
Crystallographic Data Centre (entries CCDC 1410005−
1410007). The Supporting Information is available free of
charge on the ACS Publications website at DOI: 10.1021/
jacs.5b13016.
■
When the sequence of signals applied was reversed, so was
the order of guests released, as shown in sequence II in Scheme
1. (iv) Chloroform was selectively displaced from the cavity of
S
−
cage 2 following the addition of ReO4 (1.6 equiv) to the
starting host−guest system and equilibration of the mixture at
1
70 °C for 2 h, as confirmed by H NMR (Figures 5 and S89).
(v) Addition of tren (10 equiv) to the previous mixture
triggered disassembly of cage 1, thus releasing cyclohexane, and
the transformation of cage 2 into 3 with concomitant
General synthetic methods, revised synthesis of B,
synthesis and characterization of cages 2 and 3, NMR
and ESI-MS spectra, full experimental procedures for the
determination of association constants, kinetics of anion
uptake studies, selective-sequence guests release experi-
ments and volume calculations (PDF)
−
entrapment of ReO4 inside the latter. After equilibration of
the sample at 70 °C for 12 h, the 1H NMR spectrum confirmed
formation of mononuclear complexes, disappearance of the
−
resonances due to C6H12⊂1, formation of ReO4 ⊂3, and the
Structure of 2 (CIF)
presence of free p-toluidine and tris-aniline A in solution. The
mixture remained soluble throughout the experiment.
−
Structure of ReO4 ⊂3 (CIF)
Structure of 3 (CIF)
Chloroform, cyclohexane, and perrhenate were used as a
representative guest set. Additionally, we have demonstrated
the same orthogonal control over the guest release sequence
AUTHOR INFORMATION
Corresponding Author
*jrn34@cam.ac.uk
Notes
■
−
−
with PF6 (in place of CHCl3), cyclohexane, and ReO4 (see
Supporting Information, section 5.2). Other mixtures are
predicted to behave similarly, as long as the first two guests
are chosen to bind selectively within 1 and 2, and the third
guest has a higher affinity for 2 than its initial guest.
Alternatively, in keeping with the differential anion affinities
The authors declare no competing financial interest.
−
−
−
ACKNOWLEDGMENTS
of 2 and 3, anions such as BF4 , NO3 , or ClO4 , could be
incorporated in place of ReO4 in this network, which would
■
−
This work was supported by the European Research Council.
We thank Diamond Light Source (U.K.) for synchrotron
beamtime on I19 (MT8464) and Dr. Rana A. Bilbeisi for a
preliminary screening of guests for cage 1.
result in their release upon transformation of 2 into 3 in step v
of sequence II.
CONCLUSIONS
■
The guest binding properties of two new ZnII L4 tetrahedra
REFERENCES
4
■
based on a tris-formylpyridine subcomponent have been
studied in detail. The differing reactivity of tris-formylpyridine
and tris-aniline based structures with tren has also been
(1) (a) Ludlow, R. F.; Otto, S. Chem. Soc. Rev. 2008, 37, 101.
(b) Mattia, E.; Otto, S. Nat. Nanotechnol. 2015, 10, 111. (c) Lehn, J.-
M. Angew. Chem., Int. Ed. 2013, 52, 2836.
H
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(2) (a) del Amo, V.; Philp, D. Chem. - Eur. J. 2010, 16, 13304.
(b) Chung, M.-K.; Severin, K.; Lee, S. J.; Waters, M. L.; Gagne, M. R.
Chem. Sci. 2011, 2, 744. (c) You, L.; Berman, J. S.; Anslyn, E. V. Nat.
Chem. 2011, 3, 943. (d) Ritson, D.; Sutherland, J. D. Nat. Chem. 2012,
(10) (a) Hristova, Y. R.; Smulders, M. M. J.; Clegg, J. K.; Breiner, B.;
Nitschke, J. R. Chem. Sci. 2011, 2, 638. (b) Neelakandan, P. P.;
Jimenez, A.; Nitschke, J. R. Chem. Sci. 2014, 5, 908. (c) Ronson, T. K.;
́
League, A. B.; Gagliardi, L.; Cramer, C. J.; Nitschke, J. R. J. Am. Chem.
Soc. 2014, 136, 15615. (d) Turega, S.; Whitehead, M.; Hall, B. R.;
Meijer, A. J. H. M.; Hunter, C. A.; Ward, M. D. Inorg. Chem. 2013, 52,
1122. (e) Yazaki, K.; Kishi, N.; Akita, M.; Yoshizawa, M. Chem.
4, 895. (e) Wilson, A.; Gasparini, G.; Matile, S. Chem. Soc. Rev. 2014,
̌
43, 1948. (f) Hsu, C.-W.; Miljanic,
́
O. S. Angew. Chem., Int. Ed. 2015,
54, 2219. (g) van der Zwaag, D.; de Greef, T. F. A.; Meijer, E. W.
Angew. Chem., Int. Ed. 2015, 54, 8334.
Commun. 2013, 49, 1630. (f) García-Simon
Gomez, L.; Garcia-Bosch, I.; Osuna, S.; Swart, M.; Luis, J. M.; Rovira,
C.; Almeida, M.; Imaz, I.; Maspoch, D.; Costas, M.; Ribas, X. Chem. -
Eur. J. 2013, 19, 1445. (g) García-Simon, C.; Garcia-Borras, M.;
Gomez, L.; Parella, T.; Osuna, S.; Juanhuix, J.; Imaz, I.; Maspoch, D.;
Costas, M.; Ribas, X. Nat. Commun. 2014, 5, 5557. (h) Frischmann, P.
́ ̀
, C.; Garcia-Borras, M.;
́
(3) (a) de Ruiter, G.; van der Boom, M. E. Acc. Chem. Res. 2011, 44,
563. (b) Kloxin, C. J.; Bowman, C. N. Chem. Soc. Rev. 2013, 42, 7161.
(c) Hu, J.; Liu, S. Acc. Chem. Res. 2014, 47, 2084. (d) Isaacs, L. Acc.
Chem. Res. 2014, 47, 2052. (e) Qi, Z.; Schalley, C. A. Acc. Chem. Res.
2014, 47, 2222. (f) Su, X.; Aprahamian, I. Chem. Soc. Rev. 2014, 43,
1963. (g) Wojtecki, R. J.; Meador, M. A.; Rowan, S. J. Nat. Mater.
2011, 10, 14. (h) Brioche, J.; Pike, S. J.; Tshepelevitsh, S.; Leito, I.;
Morris, G. A.; Webb, S. J.; Clayden, J. J. Am. Chem. Soc. 2015, 137,
6680.
(4) (a) Gianneschi, N. C.; Nguyen, S. T.; Mirkin, C. A. J. Am. Chem.
Soc. 2005, 127, 1644. (b) Rodríguez-Llansola, F.; Meijer, E. W. J. Am.
Chem. Soc. 2013, 135, 6549. (c) Qian, H.; Aprahamian, I. Chem.
Commun. 2015, 51, 11158.
́
̀
́
D.; Kunz, V.; Wurthner, F. Angew. Chem. Int. Ed. 2015, 127, 7393.
̈
(i) Takezawa, H.; Akiba, S.; Murase, T.; Fujita, M. J. Am. Chem. Soc.
2015, 137, 7043.
(11) (a) Duriska, M. B.; Neville, S. M.; Lu, J.; Iremonger, S. S.; Boas,
J. F.; Kepert, C. J.; Batten, S. R. Angew. Chem., Int. Ed. 2009, 48, 8919.
(b) Riddell, I. A.; Smulders, M. M. J.; Clegg, J. K.; Nitschke, J. R.
Chem. Commun. 2011, 47, 457.
(12) (a) Horiuchi, S.; Murase, T.; Fujita, M. J. Am. Chem. Soc. 2011,
133, 12445. (b) Nakabayashi, K.; Kawano, M.; Fujita, M. Angew.
Chem., Int. Ed. 2005, 44, 5322. (c) Yoshizawa, M.; Kusukawa, T.;
Fujita, M.; Yamaguchi, K. J. Am. Chem. Soc. 2000, 122, 6311.
(13) (a) Hastings, C. J.; Pluth, M. D.; Bergman, R. G.; Raymond, K.
N. J. Am. Chem. Soc. 2010, 132, 6938. (b) Jiao, Y.; Wang, J.; Wu, P.;
Zhao, L.; He, C.; Zhang, J.; Duan, C. Chem. - Eur. J. 2014, 20, 2224.
(c) Murase, T.; Horiuchi, S.; Fujita, M. J. Am. Chem. Soc. 2010, 132,
(5) (a) Shaywitz, A.; Greenberg, M. Annu. Rev. Biochem. 1999, 68,
821. (b) Barrell, M. J.; Campana, A. G.; von Delius, M.; Geertsema, E.
̃
M.; Leigh, D. A. Angew. Chem., Int. Ed. 2011, 50, 285. (c) Wang, C.; Li,
Z.; Cao, D.; Zhao, Y.-L.; Gaines, J. W.; Bozdemir, O. A.; Ambrogio, M.
W.; Frasconi, M.; Botros, Y. Y.; Zink, J. I.; Stoddart, J. F. Angew. Chem.,
Int. Ed. 2012, 51, 5460. (d) Tian, F.; Jiao, D.; Biedermann, F.;
Scherman, O. A. Nat. Commun. 2012, 3, 1207. (e) Xiao, W.; Zeng, X.;
Lin, H.; Han, K.; Jia, H.-Z.; Zhang, X.-Z. Chem. Commun. 2015, 51,
1475.
2866. (d) Otte, M.; Kuijpers, P. F.; Troeppner, O.; Ivanovic-
Burmazovic, I.; Reek, J. N. H.; de Bruin, B. Chem. - Eur. J. 2013, 19,
10170. (e) Samanta, D.; Mukherjee, S.; Patil, Y. P.; Mukherjee, P. S.
Chem. - Eur. J. 2012, 18, 12322. (f) García-Simon, C.; Gramage-Doria,
́
́
(6) (a) Chen, S.; Itoh, Y.; Masuda, T.; Shimizu, S.; Zhao, J.; Ma, J.;
Nakamura, S.; Okuro, K.; Noguchi, H.; Uosaki, K.; Aida, T. Science
2015, 348, 555. (b) Fukino, T.; Joo, H.; Hisada, Y.; Obana, M.;
Yamagishi, H.; Hikima, T.; Takata, M.; Fujita, N.; Aida, T. Science
2014, 344, 499. (c) Matson, J. B.; Navon, Y.; Bitton, R.; Stupp, S. I.
ACS Macro Lett. 2015, 4, 43.
́
R.; Raoufmoghaddam, S.; Parella, T.; Costas, M.; Ribas, X.; Reek, J. N.
H. J. Am. Chem. Soc. 2015, 137, 2680.
(14) (a) Zarra, S.; Wood, D. M.; Roberts, D. A.; Nitschke, J. R. Chem.
Soc. Rev. 2015, 44, 419. (b) Northrop, B. H.; Zheng, Y.-R.; Chi, K.-W.;
Stang, P. J. Acc. Chem. Res. 2009, 42, 1554. (c) Meyer-Eppler, G.;
(7) For selected reviews, see: (a) Chakrabarty, R.; Mukherjee, P. S.;
Stang, P. J. Chem. Rev. 2011, 111, 6810. (b) Ward, M. D. Chem.
Commun. 2009, 4487. (c) Jin, P.; Dalgarno, S. J.; Atwood, J. L. Coord.
Chem. Rev. 2010, 254, 1760. (d) Harris, K.; Fujita, D.; Fujita, M. Chem.
Commun. 2013, 49, 6703. (e) Ronson, T. K.; Zarra, S.; Black, S. P.;
Nitschke, J. R. Chem. Commun. 2013, 49, 2476. (f) Han, M.;
Engelhard, D. M.; Clever, G. H. Chem. Soc. Rev. 2014, 43, 1848.
(g) Mukherjee, S.; Mukherjee, P. S. Chem. Commun. 2014, 50, 2239.
(8) For selected recent examples, see: (a) Custelcean, R.; Bonnesen,
P. V.; Duncan, N. C.; Zhang, X.; Watson, L. A.; Van Berkel, G.;
Parson, W. B.; Hay, B. P. J. Am. Chem. Soc. 2012, 134, 8525.
Topic, F.; Schnakenburg, G.; Rissanen, K.; Lutzen, A. Eur. J. Inorg.
́
̈
Chem. 2014, 2014, 2495. (d) Niess, F.; Duplan, V.; Sauvage, J.-P. J.
Am. Chem. Soc. 2014, 136, 5876. (e) Johnson, A. M.; Wiley, C. A.;
Young, M. C.; Zhang, X.; Lyon, Y.; Julian, R. R.; Hooley, R. J. Angew.
Chem., Int. Ed. 2015, 54, 5641.
(15) (a) McConnell, A. J.; Wood, C. S.; Neelakandan, P. P.;
Nitschke, J. R. Chem. Rev. 2015, 115, 7729. (b) Clegg, J. K.; Cremers,
J.; Hogben, A. J.; Breiner, B.; Smulders, M. M. J.; Thoburn, J. D.;
Nitschke, J. R. Chem. Sci. 2013, 4, 68. (c) Han, M.; Michel, R.; He, B.;
Chen, Y.-S.; Stalke, D.; John, M.; Clever, G. H. Angew. Chem., Int. Ed.
2013, 52, 1319. (d) Meng, W.; Ronson, T. K.; Clegg, J. K.; Nitschke, J.
R. Angew. Chem., Int. Ed. 2013, 52, 1017. (e) Sørensen, A.; Castilla, A.
M.; Ronson, T. K.; Pittelkow, M.; Nitschke, J. R. Angew. Chem., Int. Ed.
2013, 52, 11273. (f) Bradberry, S. J.; Savyasachi, A. J.; Martinez-Calvo,
M.; Gunnlaugsson, T. Coord. Chem. Rev. 2014, 273−274, 226.
(g) Riddell, I. A.; Ronson, T. K.; Clegg, J. K.; Wood, C. S.; Bilbeisi, R.
A.; Nitschke, J. R. J. Am. Chem. Soc. 2014, 136, 9491.
(b) Pasquale, S.; Sattin, S.; Escudero-Adan
M.; de Mendoza, J. Nat. Commun. 2012, 3, 785. (c) Gutz, C.; Hovorka,
R.; Klein, C.; Jiang, Q.-Q.; Bannwarth, C.; Engeser, M.; Schmuck, C.;
́
, E. C.; Martínez-Belmonte,
̈
Assenmacher, W.; Mader, W.; Topic,
́
F.; Rissanen, K.; Grimme, S.;
Lutzen, A. Angew. Chem., Int. Ed. 2014, 53, 1693. (d) Metherell, A. J.;
̈
Ward, M. D. Chem. Commun. 2014, 50, 10979. (e) Young, M. C.;
Holloway, L. R.; Johnson, A. M.; Hooley, R. J. Angew. Chem., Int. Ed.
(16) (a) Smulders, M. M. J.; Zarra, S.; Nitschke, J. R. J. Am. Chem.
Soc. 2013, 135, 7039. (b) Ma, S.; Smulders, M. M. J.; Hristova, Y. R.;
Clegg, J. K.; Ronson, T. K.; Zarra, S.; Nitschke, J. R. J. Am. Chem. Soc.
́
2014, 53, 9832. (f) Ramsay, W. J.; Szczypinski, F. T.; Weissman, H.;
Ronson, T. K.; Smulders, M. M. J.; Rybtchinski, B.; Nitschke, J. R.
Angew. Chem., Int. Ed. 2015, 54, 5636. (g) Thorp-Greenwood, F. L.;
Kulak, A. N.; Hardie, M. J. Nat. Chem. 2015, 7, 526. (h) Wise, M. D.;
Holstein, J. J.; Pattison, P.; Besnard, C.; Solari, E.; Scopelliti, R.;
Bricogne, G.; Severin, K. Chem. Sci. 2015, 6, 1004.
́
2013, 135, 5678. (c) Jimenez, A.; Bilbeisi, R. A.; Ronson, T. K.; Zarra,
S.; Woodhead, C.; Nitschke, J. R. Angew. Chem., Int. Ed. 2014, 53,
4556. (d) Cullen, W.; Thomas, K. A.; Hunter, C. A.; Ward, M. D.
Chem. Sci. 2015, 6, 4025. (e) Cullen, W.; Turega, S.; Hunter, C. A.;
Ward, M. D. Chem. Sci. 2015, 6, 625.
(9) (a) Cook, T. R.; Vajpayee, V.; Lee, M. H.; Stang, P. J.; Chi, K.-W.
Acc. Chem. Res. 2013, 46, 2464. (b) Croue, V.; Goeb, S.; Salle, M.
Chem. Commun. 2015, 51, 7275. (c) Therrien, B. Chem. - Eur. J. 2013,
19, 8378. (d) Ward, M. D.; Raithby, P. R. Chem. Soc. Rev. 2013, 42,
1619. (e) Xu, L.; Wang, Y.-X.; Yang, H.-B. Dalton. Trans. 2015, 44,
867. (f) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int.
Ed. 2009, 48, 3418. (g) Brown, C. J.; Toste, F. D.; Bergman, R. G.;
Raymond, K. N. Chem. Rev. 2015, 115, 3012.
(17) (a) Bilbeisi, R. A.; Clegg, J. K.; Elgrishi, N.; de Hatten, X.;
Devillard, M.; Breiner, B.; Mal, P.; Nitschke, J. R. J. Am. Chem. Soc.
2012, 134, 5110. (b) Caulder, D. L.; Bruckner, C.; Powers, R. E.;
̈
Konig, S.; Parac, T. N.; Leary, J. A.; Raymond, K. N. J. Am. Chem. Soc.
̈
2001, 123, 8923. (c) Granzhan, A.; Riis-Johannessen, T.; Scopelliti, R.;
Severin, K. Angew. Chem., Int. Ed. 2010, 49, 5515. (d) Saalfrank, R. W.;
Glaser, H.; Demleitner, B.; Hampel, F.; Chowdhry, M. M.;
I
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Schunemann, V.; Trautwein, A. X.; Vaughan, G. B. M.; Yeh, R.; Davis,
(39) Compound B is insoluble in acetonitrile, whereas A is soluble
enough to be observed by 1H NMR at the concentrations used in this
work. In experiments not resulting in precipitation, NMR signals due
to free A were observed with integrals matching the concentration of 1
in the initial mixture.
̈
A. V.; Raymond, K. N. Chem. - Eur. J. 2002, 8, 493. (e) Albrecht, M.;
Shang, Y.; Rhyssen, T.; Stubenrauch, J.; Winkler, H. D. F.; Schalley, C.
A. Eur. J. Org. Chem. 2012, 2012, 2422.
(18) Castilla, A. M.; Ousaka, N.; Bilbeisi, R. A.; Valeri, E.; Ronson, T.
K.; Nitschke, J. R. J. Am. Chem. Soc. 2013, 135, 17999.
(19) Katayev, E. A.; Kolesnikov, G. V.; Sessler, J. L. Chem. Soc. Rev.
2009, 38, 1572 and references therein..
−
(40) Formation of the complex ReO4 ⊂2 was observed upon
addition of 2·[NTf2]8 to the final reaction mixture generated from
sequence I (after removal of the precipitate by filtration), thus
confirming that ReO4− remains in solution. See Figures S88 and S91 in
the Supporting Information.
(20) Busschaert, N.; Caltagirone, C.; Van Rossom, W.; Gale, P. A.
Chem. Rev. 2015, 115, 8038.
(21) (a) Kolesnikov, G. V.; German, K. E.; Kirakosyan, G.; Tananaev,
I. G.; Ustynyuk, Y. A.; Khrustalev, V. N.; Katayev, E. A. Org. Biomol.
Chem. 2011, 9, 7358. (b) Rambo, B. M.; Sessler, J. L. Chem. - Eur. J.
2011, 17, 4946. (c) Cao, R.; McCarthy, B. D.; Lippard, S. J. Inorg.
Chem. 2011, 50, 9499. (d) Alberto, R.; Bergamaschi, G.; Braband, H.;
Fox, T.; Amendola, V. Angew. Chem., Int. Ed. 2012, 51, 9772.
(e) Freye, S.; Michel, R.; Stalke, D.; Pawliczek, M.; Frauendorf, H.;
Clever, G. H. J. Am. Chem. Soc. 2013, 135, 8476. (f) Kataev, E. A.;
Pantos, P.; Karnas, E.; Kolesnikov, G. V.; Tananaev, I. G.; Lynch, V.
M.; Sessler, J. L. Supramol. Chem. 2015, 27, 346. (g) Lim, J. Y. C.;
Beer, P. D. Chem. Commun. 2015, 51, 3686.
(22) (a) Custelcean, R. Chem. Soc. Rev. 2014, 43, 1813. (b) Pluth, M.
D.; Raymond, K. N. Chem. Soc. Rev. 2007, 36, 161.
(23) Kleywegt, G. J.; Jones, T. A. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 1994, D50, 178.
(24) Mecozzi, S.; Rebek, J., Jr. Chem. - Eur. J. 1998, 4, 1016.
(25) Charged hosts have previously been observed not to obey
Rebek’s 55% optimal-occupancy “rule”. See refs 8f, 15b, and 22a.
(26) Chifotides, H. T.; Dunbar, K. R. Acc. Chem. Res. 2013, 46, 894.
(27) Anion Coordination Chemistry; Wiley-VCH: Weinheim,
Germany, 2011.
(28) Moyer, B. A.; Custelcean, R.; Hay, B. P.; Sessler, J. L.; Bowman-
James, K.; Day, V. W.; Kang, S.-O. Inorg. Chem. 2013, 52, 3473.
(29) See Figure S97 in the Supporting Information for an overlay of
the structures 2 and 3 in the solid state.
(30) Some association in solution between polycationic hosts and
anions has been previously described; for instance, see: Frantz, R.;
Grange, C. S.; Al-Rasbi, N. K.; Ward, M. D.; Lacour, J. Chem. Commun.
2007, 1459.
(31) Sanders, J. K. M.; Hunter, B. K. Modern NMR Spectroscopy: A
Guide for Chemists; Oxford University Press: Oxford, U.K., 1993.
(32) Pastor, A.; Martínez-Viviente, E. Coord. Chem. Rev. 2008, 252,
2314.
(33) Exchange spectroscopy (2D EXSY) has been applied for
studying kinetics of reversible systems that exchange slowly on the
chemical shift time scale but fast in the spin−lattice (T1) relaxation
time scale. In practice, exchange processes with rates rate constants
between 102 and 10−2 s−1 can be investigated. For selected examples
see references 10d, 16a, and b.
(34) Although ZnII is a labile metal ion, it has been shown that tightly
interlocked structures, like these ZnII L4 assemblies with rigid ligands,
4
̀
provide largely increased kinetic stabilities. See: (a) Charbonniere, L.
J.; Williams, A. F.; Frey, U.; Merbach, A. E.; Kamalaprija, P.; Schaad,
O. J. Am. Chem. Soc. 1997, 119, 2488. (b) Paul, R. L.; Bell, Z. R.;
Jeffery, J. C.; Harding, L. P.; McCleverty, J. A.; Ward, M. D. Polyhedron
2003, 22, 781.
(35) (a) Davis, A. V.; Raymond, K. N. J. Am. Chem. Soc. 2005, 127,
7912. (b) Davis, A. V.; Fiedler, D.; Seeber, G.; Zahl, A.; van Eldik, R.;
Raymond, K. N. J. Am. Chem. Soc. 2006, 128, 1324.
(36) Mal, P.; Schultz, D.; Beyeh, K.; Rissanen, K.; Nitschke, J. R.
Angew. Chem., Int. Ed. 2008, 47, 8297.
(37) Chepelin, O.; Ujma, J.; Barran, P. E.; Lusby, P. J. Angew. Chem.,
Int. Ed. 2012, 51, 4194.
(38) Formation of two different ZnII mononuclear complexes was
observed by 1H NMR: the octahedral complex zinc(II) tris(pyr-
idyliminoethyl)amine (Scheme 1) and a zinc(II) trigonal bipyramidal
complex where the ligand is formed from the condensation of tren
with only 1 equiv of 2-formylpyridine. See Figure S85 in the
Supporting Information for details.
J
DOI: 10.1021/jacs.5b13016
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX