Empirical and theoretical insights into the structural features and host-guest chemistry of M8L4 tube architectures

Article abstract of DOI:10.1021/ja412964r

We demonstrate a general method for the construction of M₈L ₄ tubular complexes via subcomponent self-assembly, starting from Cu^I or Ag^I precursors together with suitable elongated tetraamine and 2-formylpyridine subcomponents. The tubular architectures were often observed as equilibrium mixtures of diastereomers having two different point symmetries (D_2d or D₂ a? D₄) in solution. The equilibria between diastereomers were influenced through variation in ligand length, substituents, metal ion identity, counteranion, and temperature. In the presence of dicyanoaurate(I) and Au^I, the D ₄-symmetric hosts were able to bind linear Au(Au(CN) ₂)₂^- (with two different configurations) as the best-fitting guest. Substitution of dicyanoargentate(I) for dicyanoaurate(I) resulted in the formation of Ag(Au(CN)₂)₂^- as the optimal guest through transmetalation. Density functional theory was employed to elucidate the host-guest chemistries of the tubes.

Full text of DOI:10.1021/ja412964r

Article
pubs.acs.org/JACS
Empirical and Theoretical Insights into the Structural Features and
Host−Guest Chemistry of M8L4 Tube Architectures
Wenjing Meng,†,§ Aaron B. League,‡ Tanya K. Ronson,† Jack K. Clegg,†,∥ William C. Isley, III,‡
David Semrouni,‡ Laura Gagliardi,*,‡ Christopher J. Cramer,*,‡ and Jonathan R. Nitschke*,†
†Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, U.K.
‡Department of Chemistry, Chemical Theory Center, and Supercomputer Institute, University of Minnesota, 207 Pleasant Street SE,
Minneapolis, Minnesota 55455, United States
S
* Supporting Information
ABSTRACT: We demonstrate a general method for the
construction of M8L4 tubular complexes via subcomponent
self-assembly, starting from CuI or AgI precursors together
with suitable elongated tetraamine and 2-formylpyridine sub-
components. The tubular architectures were often observed
as equilibrium mixtures of diastereomers having two different
point symmetries (D2d or D2 ⇄ D4) in solution. The equilibria
between diastereomers were influenced through variation in
ligand length, substituents, metal ion identity, counteranion,
and temperature. In the presence of dicyanoaurate(I) and AuI,
the D4-symmetric hosts were able to bind linear Au(Au(CN)2)2−
(with two different configurations) as the best-fitting guest.
Substitution of dicyanoargentate(I) for dicyanoaurate(I) resulted in the formation of Ag(Au(CN)2)2− as the optimal guest through
transmetalation. Density functional theory was employed to elucidate the host−guest chemistries of the tubes.
Scheme 1. General Synthesis of M8L4 Tubes
INTRODUCTION
■
Metal−organic container molecules1−7 have attracted interest
due to their ability to isolate guest molecules in the micro-
environments provided by their internal cavities. Encapsulation
may alter the chemical behavior of a guest,8 leading to applica-
tions in catalysis,9−14 sensing,15−20 stabilization,21−25 and
transport.26−29 Subcomponent self-assembly, wherein dynamic-
covalent CN30,31 and coordinative M → L bonds are formed
during the same overall process,32−35 has proven particularly
useful for the synthesis of metal−organic hosts.36 The first such
hosts had tetrahedral37−41 or cubic42−46 structures, with
approximately spherical cavities suitable for binding compact
anions and small molecules.
Newer subcomponent-self-assembled hosts have been
prepared that have yet more complex structures, including
pseudoicosahedra,47 hexagonal48,49 and pentagonal50 prisms,
twisted cubes,51 asymmetric structures,52 and tubular archi-
tectures.53 Tubes represent interesting research targets due
to their potential biomimetic function as molecular channels
for selective transportation of ions and molecules, and as hosts
for linear guests. Although many tubular organic systems have
been reported,54−59 the structural properties and host−guest
chemistries of discrete metal−organic tubes have been less well-
studied.60,61
NC-Au-CN-Cu-NC-Au-CN−, which was not independently
observed, as the optimal guest for encapsulation. Building upon
Recently, we have reported the assembly of M8L4 tubular
capsule 1a from the reaction of tetraamine A, 6-methyl-2-
formylpyridine 1 and [Cu(MeCN)4]BF4 (Scheme 1).53 This
tube is able to transform Au(CN)2− into a linear complex anion
Received: December 20, 2013
Published: January 21, 2014
© 2014 American Chemical Society
3972
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980

Journal of the American Chemical Society
Article
our previous work on tube 1a, we demonstrate how the length,
shape and substituents of the ligands, the counteranions and
metals can influence the stereochemistry and host−guest
chemistry of the tubular complexes. Insights into the nature
and origin of some of these processes are provided by density
functional theory (DFT) analysis.
RESULTS AND DISCUSSION
■
Synthesis and Stereochemistry. A M8L4 tubular complex
can be constructed as the uniquely observed product using
elongated tetraamine A, B, or C (4 equiv), 2-pyridine-
carboxaldehyde derivatives (16 equiv), and a suitable salt of
CuI or AgI (8 equiv) in acetonitrile, as depicted in Scheme 1.
Depending on the orientation of the bidentate iminopyridine
binding sites, the M8L4 tube can adopt approximate D2d(/D2)
or D4 point symmetries where the metal ions define the vertices
of a cuboid. As we observed earlier,53 in the crystal structures of
tube 1a·BF4, the D2d isomer has isosceles trapezoids as the long
faces of the cuboid, with the shorter faces forming rectangles,
whereas in the D4 isomer the cuboid approximates a right
square prism in which one of the square faces is twisted with
respect to the other. The D4 isomer possesses a narrow linear
channel that is capable of trapping two acetonitrile molecules
inside. The difference in the symmetry of the two diastereomers
led to characteristic NMR peak multiplicities, allowing them
Figure 1. Crystal structure of 2a-D4·PF6. (a) Representation of the
complex with one ligand highlighted in yellow (hydrogen atoms not
−
shown). (b) CPK representation showing the proximity between PF6
anions and ligand hydrogens.
atoms and protons of the complex, which may account for the
extra stabilization effect brought by the PF6 anion.62,63
−
Host 2a·PF6 has an approximately equal distribution of both
isomers in solution. The interconversion between 2a-D4·PF6
and 2a-D2d·PF6 could be followed by 1H NMR spectroscopy as
the temperature was varied. Kinetic studies (described in the
Supporting Information) revealed ΔH‡ = 108 7 kJ mol−1 and
ΔS‡ = 71 24 J K−1 mol−1 for the isomerization from 2a-D4·
PF6 to 2a-D2d·PF6, and ΔH‡ = 58 8 kJ mol−1, and ΔS‡ =
−104 24 J K−1 mol−1 for the reverse transformation (from
2a-D2d·PF6 to 2a-D4·PF6), which appears more entropically
disfavored compared to the same process for the terphenyl
congener 1a·BF4 (ΔS = −62 21 J K−1 mol−1).53 The rate
constants for both transformations were identical at 283 K,
marking 2a-D4 as the dominant species in solution below this
temperature, and 2a-D2d above.
1
to be distinguished by H NMR. The population of the two
isomers in solution reflects their relative thermodynamic
stability, which can be tuned in several ways, as summarized
in Table 1.
Table 1. Summary of Isomers Formed in Acetonitrile upon
the Variation of Tetraamine, Aldehyde, and Counter Ion for
Cu8L4 Tubes
counter ion
−
−
complex
tetraamine
aldehyde
BF4
PF6
1a
A
1
D4:D2d
D4 only
Since the choice of counterions has been shown to have a
measurable but small impact on the stereochemistry of almost
all of the complexes listed in Table 1, a computational study
was undertaken to determine the differential effect of including
90:10%
D2d only
2a
2
D4:D2d
52:48%
unstable
D4:D2d
24:76%
D4:D2
−
3a
1b
3
1
D2d only
D4:D2d
1:99%
two PF6 counterions at the ends of the empty D4 versus the
B
D2d isomers of 1a. A relative stabilization of the D2d isomer by
only 4.1 kJ mol−1 was computed (see Computational Methods
section for theory details), a value commensurate with the small
energy changes associated with the variations in isomeric ratios
discussed above.
1c
C
1
D2 only
6:94%
Longer ligands also disfavored the D4 isomer: the reaction
between tetraamine B or C, 6-methyl-2-pyridine-carboxalde-
hyde 1 and [Cu(MeCN)4]BF4 in acetonitrile produced 1b-D2d
and 1c-D2 as the predominant isomers, respectively (Figures S27
and S33, Supporting Information). The hexafluorophosphate
anion was again found to slightly stabilize the D4 isomer; when
copper(I) hexafluorophosphate was used in place of the
tetrafluoroborate, the equilibrium ratios were found to be
24:76% and 6:94% for complexes 1b-D4:1b-D2d and 1c-D4:1c-
The substituent on the aldehyde subcomponent was observed
to influence the stability of the tube isomers. Replacing
a methyl group with a proton (aldehyde 2) or a bromine
(aldehyde 3) at the 6 position of pyridine-2-carboxaldehyde
resulted in the relative destabilization of the D4-symmetric
isomer, so that in the cases of 2a·BF4, 3a, and 1c·BF4, the D4
isomer did not form in solution.
−
−
In most cases, both BF4 and PF6 counterions allow the
formation of M8L4 tubes, and the formation of the D4 isomer is
preferred when PF6− is present. The crystal structure of 2a-D4·
1
D2, respectively, as revealed by their H NMR spectra (Figures
S21 and S30, Supporting Information). Models suggested that
the D4-symmetric tubes constructed from tetraamine B or C are
not long enough to accommodate a third acetonitrile molecule
inside the channel, leaving instead additional empty space, and
incurring an energetic penalty for doing so. Single crystals of 1b·
BF4 and 1c·PF6 were isolated by vapor diffusion of diethyl
ether (or diisopropyl ether) into an acetonitrile solution of the
−
PF6 (Figure 1) reveals that one PF6 anion is located at each
end of the tube with one fluorine atom pointing directly into
the channel, and four such anions associate at the junctions
between two neighboring terphenyl ligands, which are also
sandwiched between two pyridine residues. For all these anions,
short contacts (2.3−2.8 Å) are observed between fluorine
3973
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980

Journal of the American Chemical Society
Article
Figure 3. Crystal structure of AgI complex 4-D2d·BF4. (a) Side view
highlighting one ligand in thicker stick presentation. (b) Top view
showing the distortion at the AgI centers.
parallelogram (Figure 3b), whereas those in the CuI tubes
approximate a rectangle. Furthermore the AgI centers in 4-D2d·
BF4 show a greater degree of distortion from idealized
tetrahedral geometry compared to the CuI centers in 1a-D2d·
BF4 with N−Ag−N angles in the range 72−154° compared
N−Cu−N angles of 79−138° in its CuI analogue.
Host−Guest Chemistry. In previous work, we demon-
strated that tube 1a-D4·BF4 is capable of binding the complex
anion Cu(Au(CN)2)2 . The CuI ion bridges the two NC-
−
Au-CN−, and it could be substituted by AgI to give the
−
Ag(Au(CN)2)2 adduct of 1a-D4·BF4.
We have since determined DFT binding energies for these
guests, and for every analogous guest with a different combina-
tion of central group-11 metal and dicyano group-11-metalate,
inside of 1a in acetonitrile continuum solvent. The results are
shown in Table 2. Counterions were not included. Because of
Figure 2. Crystal structures of 1b-D2d·BF4 (a,b) and 1c-D2·PF6 (c,d).64
(a,c) Side view; (b,d) top view. Hydrogen atoms, solvent molecules,
and counterions are omitted for clarity.
Table 2. Computed Energies of Incorporation of Group-11
Metal Centers (Rows) And Dicyano Ends (Columns)
In kJ mol−1
respective complexes. X-ray analyses revealed the presence of
1b-D2d (Figure 2a,b) and 1c-D2 (Figure 2c,d), whose structures
resemble that of 1a-D2d. For 1b-D2d the elongation of the ligand
backbone from terphenylene to quaterphenylene did not result
in an increase of the width of the tube channel, but rather
narrows it. The shorter faces (Figure 2b) are slightly distorted
from a rectangular geometry. The average Cu−Cu distance of
the shorter edge of the top and bottom faces was 5.3 Å, 0.1 Å
shorter than in 1a-D2d. For 1c-D2 the presence of a naphthalene
spacer reduces the symmetry of the complex by removing the
mirror plane that bisects the ligand. The naphthalene spacer
also introduces an offset between the two terminal phenyl rings,
which slightly widens the tube channel. The shorter edge of
the rectangular face in 1c-D2 (5.6 Å) is 0.2 Å longer than that in
1a-D2d.
peripheral anions in NC−M′−CN−M−NC−
central cation
M′−CN
−
−
−
Cu(CN)2
Ag(CN)2
Au(CN)2
D4-host
CuI
−36.8
−41.8
−116.3
−52.7
−53.1
−129.3
−69.0
−72.4
−143.9
AgI
AuI
D2d-host
CuI
a
a
a
a
a
a
2.9
−15.1
−97.9
AgI
AuI
a
These values were not determined; no such binding is observed
AgI can also be used in place of CuI to form an M8L4 tube.
The reaction between tetraamine A, 6-methyl-2-pyridine-
carboxaldehyde 1 and AgBF4 in acetonitrile produced 4-D2d
experimentally.
1
as the only observed product in solution, as verified by H
the high computational cost of optimizing the geometry of the
large host−guest complexes, energies were not computed for
the experimentally unobserved binding of the dicyanoargentate
and dicyanocuprate guests in the D2d host isomer.
Binding energies were calculated by determining the dif-
ference in energy between each host−guest complex and its
corresponding separated starting compounds and acetonitrile-filled
D4 host isomer at 5 μM concentrations of the host−guest com-
plexes. For consistency with the experimental conditions employed
(vide infra), free AuI was modeled as the cationic moiety of the salt
Au(tmbn)2SbF6 (tmbn = 2,4,6-trimethoxybenzonitrile), whereas
NMR and MALDI-MS. Doublets were observed for the two
symmetry-independent imine protons, with J = 5.9 and 7.8 Hz
due to the coupling between 107/109Ag and the imine protons.
Vapor diffusion of diethyl ether into an acetonitrile solution
of 4-D2d·BF4 allowed the isolation of single crystals suitable for
X-ray analysis. The solid state structure reveals an approximate
D2d-symmetric M8L4 topology, consistent with solution observa-
tions (Figure 3). Compared to analogous CuI tubes (1a-D2d,
1b-D2d and 1c-D2) 4-D2d·BF4 is more distorted: the top view
of 4-D2d·BF4 shows that the shorter faces of the complex form a
3974
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980

Journal of the American Chemical Society
Article
free CuI and AgI were modeled as the tetrakis(acetonitrile)
complexes.
Scheme 2. Formation of Trigold Host−Guest Complex
a
[Au(Au(CN)2)2 ⊂ 1a-D4]·BF4 via Transmetalation
The computed energies of binding matched the exper-
imentally observed trend, where guests bound more strongly in
the D4 isomer and larger group-11 metals bound more strongly
than smaller ones. This trend is consistent, allowing for
−
reasonable extrapolation to the binding of the Ag(CN)2 and
−
Cu(CN)2 guests in the D2d host isomer. It is important to
note that these energies of the host−guest complexes are
relative to those of the solvent-filled cage and guest precursors,
not the polymeric precipitate actually observed when no host
is present. This distinction is likely to explain why we still
a
The two representations shown are X-ray crystal structures. One
−
obtain negative binding energies for the Cu(Ag(CN)2)2 and
configuration of the trigold guest is shown.
Ag(Ag(CN)2)2− guests, which are not observed to bind in situ,
as these energies were not calculated relative to the global
energy minimum.
gave rise to three doublets and four singlets with different
intensity, indicating the presence of multiple carbon environments.
The 13C NMR spectra of the labeled host−guest complexes
[Cu(Au(13CN)2)2 ⊂ 1a-D4] and [Ag(Au(13CN)2)2 ⊂ 1a-D4]
exhibited a pair of characteristic doublets with JC−C = 47 Hz for
the guest signals, consistent with conservation of the NC-Au-CN
aurocyanide configurations within the complex anion guests.
Similar signals were not observed in the 13C NMR spectrum for
[Au(Au(13CN)2)2 ⊂ 1a-D4]. This observation indicates that in
[Au(Au(CN)2)2 ⊂ 1a-D4], the guest configuration is different
from that in [Cu(Au(13CN)2)2 ⊂ 1a-D4] and [Ag(Au(13CN)2)2
⊂ 1a-D4]. We thus infer that the conformation NC-Au-CN-Au-
NC-Au-CN− is not adopted by the guest in [Au(Au(CN)2)2 ⊂
1a-D4]. Our data were consistent with the guest adopting the
conformations NC-Au-CN-Au-CN-Au-CN− and NC-Au-NC-
Au-CN-Au-CN−, in which each gold(I) center is bonded to at
least one carbon atom.
Contrary to our previous inference,53 it seems that the trend
of favoring heavier group-11 metals at the center of the
complex anion is predicated not upon increased cation-π
interaction with the organic linkers of the host cage, but upon
stronger intraguest binding. Figure 4 shows the DFT energetics
DFT calculations of the relative energies of the free complex
anions in continuum acetonitrile solvent predict NC-Au-CN-
Au-CN-Au-CN− and NC-Au-NC-Au-CN-Au-CN− to be more
stable than NC-Au-CN-Au-NC-Au-CN− by 15.5 and 14.6 kJ mol−1,
respectively. Thus, to have one gold atom not coordinated by
at least one cyanide carbon atom is disfavored energetically.
In so far as only two complex anion isomers are predicted to
dominate in the absence of encapsulation, and assuming that
binding energies are similar for the different complex anion
isomers, upon guest binding we expect to observe close to a
2:1 statistical distribution of NC-Au-CN-Au-CN-Au-CN−, and
NC-Au-NC-Au-CN-Au-CN−. We thus infer the tripling of host
signals in the NMR to result from one set of signals associated
with binding of NC-Au-NC-Au-CN-Au-CN− and two sets of
signals associated with binding of the asymmetric complex anion
NC-Au-CN-Au-CN-Au-CN−, which results in desymmetrization
of the two ends of the tube. The presence of multiple conforma-
tions is mirrored in the solid-state behavior of group-11
cyanides.65
Figure 4. Calculated energetics for stepwise formation and in-
corporation of group-11 metal-centered bis-dicyanoaurates into 1a-D4
(“Guest ⊂ Host” data from last column of Table 2, “Empty Host” for
guests as their dissociated precursors).
of the stepwise formation and insertion of the bis-dicyanoaurate
guests into both 1a-D4 and 1a-D2d. For this hypothetical pathway,
the global minimum energy of complex 1a was assumed to be
the D4 isomer with two incorporated acetonitrile guests, and
consequently this structure was chosen as the starting material for
the host cage in the second step of the pathway. By comparing
the energies of guest formation to those of host−guest
complexation, the role of the comparatively strong gold−nitrogen
bonds in stabilizing the [Au(Au(CN)2)2 ⊂ 1a-D4] complex
becomes apparent.
In keeping with our theoretical predictions, the addition of
Au(tmbn)2SbF6 (1.2 equiv) to [Cu(Au(CN)2)2 ⊂ 1a-D4·BF4]
(1 equiv) led to the formation of a new host−guest complex
[Au(Au(CN)2)2 ⊂ 1a-D4·BF4] (Scheme 2), as verified by
ESI-MS. A low resolution crystal structure was obtained for the
product, showing that AuI replaced CuI as the bridging cation
within the guest.
The titration of Au(tmbn)2SbF6 into an acetonitrile solu-
tion of [Cu(Au(CN)2)2 ⊂ 1a-D4·BF4] allowed the stability
constant of 1.6 × 1011 M−3 for [Au(Au(CN)2)2 ⊂ 1a-D4·BF4] to
be determined, 129 times greater than that of [Cu(Au(CN)2)2
⊂ 1a-D4·BF4] and 3.7-fold higher than [Ag(Au(CN)2)2
⊂
1a-D4·BF4].
Tetraphenyl tube 1b·PF6 did not form any host−guest
NMR spectra of [Au(Au(CN)2)2 ⊂ 1a-D4·BF4] revealed
complex in the presence of KAu(CN)2, which suggests the
1
−
additional splitting: many H signals appeared as a set of
energy gained by trapping the guest Cu(Au(CN)2)2 is not
three closely spaced peaks of roughly equal intensity. Using
isotopically labeled KAu(13CN)2, in the 13C NMR spectrum
(Figure S37, Supporting Information) the 13C-labeled guest
enough to compensate energy lost during isomerization from
1b-D2d to 1b-D4. In contrast, for naphthalene-based tube 1c·
PF6 the addition of KAu(CN)2 resulted in a rapid and clean
3975
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980

Journal of the American Chemical Society
Article
a
Scheme 3. Formation of Host−Guest Complexes from 1c·PF6.
a
[Cu(Au(CN)2)2 ⊂ 1c-D4] Is Shown as the X-ray Crystal Structure.
−
+
transformation to [Cu(Au(CN)2)2 ⊂ 1c-D4] (Scheme 3),
despite the low abundance of 1c-D4 in solution. The crystal
structure of the product confirmed the encapsulation of
Ag(CN)2 + Au(tmbn)2 + 4CH3CN
→ Au(CN)2 + Ag(CH3CN)4+ + 2tmbn
is predicted to be exoergic by 146.4 kJ mol−1, making the
[Ag(Au(CN)2)2 ⊂ 1a-D4] complex more enthalpically favorable
than its gold bis-dicyanoargentate counterpart by 235.6 kJ mol−1.
−
−
Cu(Au(CN)2)2 within 1c-D4, consistent with NMR and
ESI-MS observations. The central CuI within the guest could
be replaced by AgI or AuI in a similar way to the analogous
terphenyl tube [Cu(Au(CN)2)2 ⊂ 1a-D4] (Scheme 3).
−
Linear dicyanoargentate, Ag(CN)2 , has very similar dimen-
CONCLUSION
−
■
sions to Au(CN)2 , yet no host−guest complex formation
−
In this study we have developed a general synthetic pro-
cedure for M8L4 tubular complexes using tetraamines with two
3,5-diaminophenylene moieties connected by a suitable spacer.
This technique allows facile investigations into the influences
of subtle changes in any of the subcomponents on the complex
structure. The M8L4 tubes are present in solution as either D4-
symmetric or D2d/D2-symmetric isomers, which are in dynamic
equilibrium. The D4 isomer, which is the only one observed to
was observed when 1a·PF6 was treated with Ag(CN)2 in the
presence of either CuI or AgI. In contrast, when AuI was added,
a new D4-symmetric complex was rapidly generated. This new
product was not the expected [Au(Ag(CN)2)2 ⊂ 1a-D4]·PF6,
but a transmetalated product [Ag(Au(CN)2)2 ⊂ 1a-D4]·PF6,
as verified by NMR and ESI-MS (Scheme 4). In the absence
Scheme 4. Formation of [Ag(Au(CN)2)2 ⊂ 1a-D4]·PF6 via
−
bind guests, is more stabilized when PF6 is present as the
−
Transmetalation between Ag(CN)2 and AuI
counteranion, whereas the D2d/D2 isomer is stabilized by the
elongation of the ligand or the introduction of an offset between
tube termini. Further systemic adaptation is revealed in the
host−guest chemistry of the tubes. Dicyanoaurate is a necessary
subcomponent of all guests that we observe to be bound by
any tube, and the system will undertake to transform guests
in order to achieve an optimal host−guest complex through
guest recombination or transmetalation. This work therefore
builds upon and contributes to fundamental studies of systems
chemistry,68 specifically the dynamic response of a system to
external stimuli, as is required in the design and creation of
increasingly complex molecular machines.69−73 The design of a
system that is specifically adapted to bind gold cyanides may
also be of relevance to the mining industry,74 and the ability to
specifically bind linear guests may allow for their catalytic
transformation, as has been observed in other systems.9−14
of the host 1a·PF6, mixing Ag(CN)2 with AuI resulted in
−
the formation of white precipitate, which we infer to be the
polymeric mixed-metal cyanide.66,67 Host 1a·PF6 therefore acts
as a solubilizing carrier, allowing the encapsulated guest to be
studied using routine spectroscopic methods.
The lack of observed binding of [Au(Ag(CN)2)2]− in 1a can
be explained computationally. DFT calculations show that while
EXPERIMENTAL SECTION
■
−
the fully formed guest Au(Ag(CN)2)2 is predicted to bind to
Computational Methods. All calculations employed the PBE-
D375,76 functional as implemented in the ADF 2013 software
package.77−79 TZP basis sets with large frozen cores were employed
for metal atoms, and DZP basis sets for the organic linkers.80
1a-D4 more strongly than Ag(Au(CN)2)2 (by 56.9 kJ mol−1,
−
see Table 2), the transmetalation of dicyanoargentate to
dicyanoaurate by the pathway
3976
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980

Journal of the American Chemical Society
Article
The zero-order regular approximation (ZORA) was employed to
account for scalar relativistic effects.81−83
[2a(PF6)5]3+ 1272.64. Found: C, 47.11; H, 7.03; N, 10.35%. Calc. for
C168H120Cu8F48N32P8·2H2O: C, 47.02; H, 2.91; N, 10.45%.
Empty cages and host−guest complexes were first optimized in the
gas phase, and final energies were computed from single-point calcula-
tions on these minima including acetonitrile solvation effects
computed from the COSMO continuum solvent model.84 When
representative host−guest complexes were subjected to reoptimization
including solvation effects, their energies were observed to fluctuate but
not to decrease (or converge, because of apparent numerical noise), on
which basis we concluded that for the large cage structures, solvated
single-point calculations on gas-phase geometries were sufficiently
accurate for our purposes. The geometries of small molecule guests and
guest precursors, however, were optimized including acetonitrile
solvation effects. Because of the size of the host structures, no frequency
calculations were performed, and consequently the theoretical energies
reported in this paper include no thermal corrections.
General Methods. Unless otherwise specified, all starting materials
were purchased from commercial sources and used as supplied.
Chromatographic separations were performed on silica gel 60 (particle
size: 0.040−0.063 mm) purchased from Aldrich. TLC was performed
on silica gel 60 F254 plates purchased from Merck and visualized
under ultraviolet light (254 nm). NMR spectra were recorded on
a Bruker DRX-400 and Bruker Avance 500 Cryo. Chemical shifts (δ)
are reported in parts per million (ppm) and are reported relative
to acetonitrile-d3 at 1.94 ppm at 298 K unless otherwise noted. Low
resolution electrospray ionization mass spectra (ESI-MS) were
obtained on a Micromass Quattro LC, infused from a Harvard Syringe
Pump at a rate of 10 μL per minute. MALDI was carried out by the
EPSRC National MS Service Centre at Swansea. Building blocks
[1,1′:4′,1″-terphenyl]-3,3″,5,5″-tetraamine A and Au(tmbn)2SbF6 were
synthesized following literature procedures.53,85
General Synthetic Procedure for 2a·BF4 and 3a·BF4. To a
Schlenk flask was added [1,1′:4′,1″-terphenyl]-3,3″,5,5″-tetraamine A
(4 equiv), suitable 2-pyridinecarboxaldehyde (16 equiv), Cu-
(CH3CN)4BF4 (8 equiv) and acetonitrile. The solution was degassed
by three evacuation/nitrogen fill cycles and stirred at room
temperature for 24 h. The product was purified by recrystallization:
diethyl ether was diffused into an acetonitrile solution of the complex.
The desired complex was isolated by filtration as a black solid.
1
2a·BF4: H NMR (CD3CN, 500 MHz) 9.778 (8H, s, imine H),
9.483 (8H, s, imine H), 8.500 (4H, s, Ph-H), 8.319 (8H, d, J 4.8, py-
H), 8.225 (16H, py-H), 8.066 (24H, py-H), 7.946 (4H, s, Ph-H),
7.565 (16H, py-H), 7.175 (8H, s, Ph-H), 7.056 (8H, d, J 8.0, Ph-H),
6.701 (8H, d, J 8.0, Ph-H; 13C {1H} NMR (CD3CN, 125 MHz)
162.34, 158.96, 151.59, 151.54, 150.19, 149.82, 149.76, 147.72, 145.08,
143.94, 140.12, 139.93, 139.60, 139.58, 130.29, 130.05, 129.97, 129.91,
129.67, 127.99, 127.95, 124.26, 120.13, 106.75; MALDI-MS
[2a(BF4)7]+ 3701.4. Found: C, 49.70; H, 3.17; N, 10.84%. Calc. for
C168H120B8Cu8F32N32·14H2O: C, 49.92; H, 3.69; N, 11.09%.
1
3a·BF4: H NMR (CD3CN, 500 MHz) 9.783 (8H, s, imine H),
9.464 (8H, s, imine H), 8.599 (4H, s, Ph-H), 8.260 (8H, d, J 7.5,
py-H), 8.136 (8H, d, py-H), 8.113 (8H, s, Ph-H), 7.991 (8H, t, J 7.5,
py-H), 7.980 (8H, t, J 7.5, py-H), 7.822 (8H, d, J 7.5, py-H), 7.756
(8H, d, J 7.5, py-H), 7.322 (8H, s, Ph-H), 7.172 (8H, d, br, Ph-H),
6.784 (8H, d, J 7.0, Ph-H); 13C {1H} NMR (CD3CN, 125 MHz)
162.05, 157.83, 152.80, 152.75, 149.77, 146.69, 145.38, 144.11
142.84, 142.69, 142.52, 141.98, 139.90, 139.29, 134.15, 134.08,
129.64, 129.38, 129.07, 127.95, 123.72, 120.38, 106.47; MALDI-MS
[3a(BF4)7]+ 4964.1. Found: C, 39.15; H, 2.22; N, 8.58%. Calc. for
C168H104B8Br16Cu8F32N32·6H2O: C, 39.10; H, 2.27; N, 8.69%.
1b·PF6. To a Schlenk tube was added B, [1,1′:4′,1″:4″,1‴-
quaterphenyl]-3,3‴,5,5‴-tetraamine (30 mg, 0.08 mmol, 4 equiv),
6-methyl-2-pyridinecarboxaldehyde (39.7 mg, 0.32 mmol, 16 equiv),
Cu(CH3CN)4PF6 (61 mg, 0.16 mmol, 8 equiv) and acetonitrile
(10 mL). The solution was degassed by three evacuation/nitrogen fill
cycles and stirred at room temperature for 12 h. A dark pink solution
resulted. The product was purified by recrystallization: diisopropyl
ether was diffused into an acetonitrile solution of the complex. The
desired product 1b·PF6 was isolated by filtration as a black solid
(55 mg, 56%): 1H NMR (CD3CN, 500 MHz) the solution is a mixture
of two isomers, D4:D2d = 13:87%, as shown in Figure S21 (Supporting
Information), 10.30 (8H, s, br, imine H), 9.39 (8H, s, imine H), 9.17
(8H, s, imine H), 8.55 (4H, br, Ph-H), 8.58 (8H, t, J 7.75, py-H), 8.53
(4H, br, Ph-H), 8.10 (s, Ar-H), 8.06 (m, br, Ar-H), 7.99 (m, Ar-H),
7.92 (m, Ar-H), 7.83 (m, br, Ar-H), 7.77 (8H, d, J 7.40, Ph-H), 7.71
(8H, d, J 7.85, Ph-H), 7.67 (8H, s, Ar-H), 7.61 (8H, d, J 6.75, Ph-H),
7.57 (8H, d, J 6.75, Ph-H), 7.48 (8H, d, J 7.05, py-H), 7.38 (8H, d,
J 7.45, py-H), 7.27 (8H, d, J 7.35, Ph-H), 7.21 (8H, s, Ph-H), 7.13
(8H, d, J 7.00, Ph-H), 6.90 (8H, s, Ph-H), 6.66 (8H, d, J 7.05, Ph-H),
6.52 (8H, d, J 7.25, Ph-H), 2.55 (24H, s, CH3), 2.47 (24H, s, CH3),
2.08 (24H, s, CH3), 1.72 (24H, s, CH3); 13C {1H} NMR (CD3CN,
125 MHz) 163.19, 162.58, 162.36, 161.70, 160.96, 160.53, 159.53,
159.42, 158.68, 151.48, 150.91, 150.71, 150.29, 150.22, 150.03, 149.95,
147.85, 144.70, 144.57, 143.91, 140.49, 140.24, 139.69, 139.49, 139.46,
139.41, 139.35, 138.35, 131.59, 130.37, 130.09, 129.43, 129.37, 128.67,
128.51, 128.40, 128.27, 127.71, 127.60, 127.33, 127.26, 127.10, 124.36,
123.66, 119.76, 119.12, 114.88, 26.41, 25.99, 24.92, 23.88; ESI-MS
[1b]8+ 452.96, [1b(PF6)]7+ 538.53, [1b(PF4)2]6+ 652.22, [1b(PF6)3]5+
811.71, [1b(PF6)4]4+ 1051.12. Found: C, 52.91; H, 3.87; N, 9.50%.
Calc. for C208H168Cu8F48N32P8·2C6H4O (diisopropyl ether): C, 52.97;
H, 3.69; N, 8.99%.
Synthesis and Characterization of Metal Complexes. 1a·PF6.
To a Schlenk tube was added A (80 mg, 27.5 mmol, 4 equiv),
6-methyl-2-pyridinecarboxaldehyde (133.5 mg, 110.2 mmol, 16 equiv),
Cu(CH3CN)4PF6 (205 mg, 55.1 mmol, 8 equiv) and acetonitrile
(15 mL). The solution was degassed by three evacuation/nitrogen fill
cycles and stirred at room temperature for 12 h. A dark pink solution
resulted. The desired product 1a·PF6 was precipitated by adding
diethyl ether into the reaction mixture, and was isolated by filtration
1
as a black solid (200 mg, 65%): H NMR (CD3CN, 500 MHz) 9.31
(8H, s, imine H), 8.59 (8H, t, J 8.00, py-H), 8.07−8.05 (16H, d,
py-H), 7.91 (8H, d, J 8.00, py-H), 7.77 (8H, d, J 7.50, py-H), 7.67 (8H,
m, py-H), 7.65 (8H, s, Ph-H), 7.63 (8H, s, imine H), 6.89 (16H, s,
Ph-H), 6.82 (16H, s, Ph-H), 2.52 (24H, s, CH3), 2.44 (24H, s, CH3);
13C {1H} NMR (CD3CN, 125 MHz) 162.50, 161.38, 161.02, 159.35,
150.87, 150.31, 150.20, 149.99, 143.82, 140.44, 139.67, 139.64, 131.57,
130.23, 128.55, 127.80, 126.99, 123.55, 115.42, 26.42, 25.88; ESI-
MS [1a(PF6)2]6+ 601.76, [1a(PF6)3]5+ 751.08, [1a(PF6)4]4+ 975.15,
[1a(PF6)5]3+ 1348.58. Found: C, 48.60; H, 3.48; N, 9.80%. Calc. for
C184H152Cu8F48N32P8·3H2O: C, 48.75; H, 3.51; N, 9.89%.
2a·PF6. To a Schlenk tube was added A (10 mg, 34.4 μmol,
4 equiv), 2-pyridinecarboxaldehyde (13.1 μL, 0.13 mmol, 16 equiv),
Cu(CH3CN)4PF6 (25.6 mg, 68.8 mmol, 8 equiv) and acetonitrile
(5 mL). The solution was stirred at room temperature for 12 h to give a
dark pink solution. Diethyl ether was added into the reaction mixture;
the resulting mixture was centrifuged, and the solvent was decanted.
The solid was dried under a vacuum to give the desired product 2a·PF6
as dark pink solid (16.6 mg, 45%): 1H NMR (CD3CN, 500 MHz) 9.67
(8H, br, imine H), 9.33 (8H, s, imine H), 9.29 (8H, s, imine H), 8.87
(8H, d, J 4.70, Ar-H), 8.65 (8H, dt, J 7.90, 1.25, py-H), 8.62 (8H, d,
J 4.85, Ar-H), 8.30−7.89 (24H, Ar-H), 7.80 (8H, s, Ph-H), 7.75 (8H, t,
J 6.43, py-H), 7.65 (8H, s, Ph-H), 7.60 (8H, t, J 5.78, py-H), 7.56 (8H,
br, py-H), 7.65 (8H, s, Ph-H), 7.19 (8H, d, J 1.25, Ph-H), 7.00 (8H,
d, J 7.85, Ph-H), 6.97 (8H, s, Ph-H), 6.94 (8H, s, Ph-H), 6.84 (8H, s,
Ph-H), 6.71 (8H, d, J 7.80, Ph-H); 13C {1H} NMR (CD3CN, 125
MHz) 162.57, 162.20, 160.82, 159.05, 151.94, 151.64, 151.52, 151.41,
151.21, 150.76, 150.32, 150.19, 149.95, 149.75,147.81,144.97, 144.08,
143.91, 140.70, 140.20, 139.90, 139.76, 139.70, 139.65, 131.71, 130.67,
130.32, 130.24, 130.05, 129.80, 129.63, 129.47, 128.68, 128.15, 123.95,
123.78, 120.16, 118.92, 115.05; ESI-MS [2a(PF6)4]4+ 918.57,
1b·BF4. To a NMR tube was added B, [1,1′:4′,1″:4″,1‴-
quaterphenyl]-3,3‴,5,5‴-tetraamine (8 mg, 0.02 mmol, 4 equiv),
6-methyl-2-pyridinecarboxaldehyde (10.5 mg, 0.08 mmol, 16 equiv),
Cu(CH3CN)4BF4 (13.7 mg, 0.04 mmol, 8 equiv) and acetonitrile
(1 mL). The resulting dark pink solution was kept at 50 °C for 12 h.
Diethyl ether was added into the reaction mixture; the resulting
mixture was centrifuged, and the solvent was decanted. The solid
was dried under a high vacuum to give the desired product 1b·BF4 as
3977
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980

Journal of the American Chemical Society
Article
a dark pink solid (22.2 mg, 94%). 1b-D4:1b-D2d = 99:1%, cal-
culated from the integration of CH3 signals in the 1H NMR spectrum
(i.e., peaks at 2.55, 2.48 ppm). NMR data for D2d isomer reported
[Au(Au(CN)2)2 ⊂ 1a]·BF4: 1H NMR (CD3CN, 400 MHz) 9.44−9.38
(8H, imine H), 8.44−8.39 (8H, py-H), 8.05−7.79 (48H, Ar-H),
7.62 (8H, br, Ar-H), 7.30−7.13 (16H, Ar-H), 7.04−6.95 (16H, Ar-H),
2.50 (24H, s, CH3), 2.39 (24H, br, CH3); 13C {1H} NMR (CD3CN,
125 MHz) peaks split from one carbon signal are grouped in
parentheses (161.48, 161.41, 161.34), 160.53, 160.08, 159.28, 151.53,
(151.03, 150.98), 150.54, (149.59, 149.56, 149.52), (149.14, 149.10),
(143.48, 143.19, 143.02), 139.61, 139.45, 138.86, 138.60, 138.43,
131.06, 130.62, 129.95, (129.15, 129.06, 128.90), 128.05, 127.53,
(124.65, 124.42, 124.29), 117.74, 114.94, 26.19, 25.88; guest signals
152.0 (d, J 10.0), 152.5 (d, J 13.75), 152.8 (d, J 12.8), 149.24, 135.86,
135.44, 134.77; ESI-MS [Au(Au(CN)2)2 ⊂ 1a (BF4)]6+ 683.52,
[Au(Au(CN)2)2 ⊂ 1a (BF4)2]5+ 837.51, [Au(Au(CN)2)2 ⊂ 1a
(BF4)3]4+ 1068.79, [Au(Au(CN)2)2 ⊂ 1a (BF4)4]3+ 1453.69.
[Ag(Au(CN)2)2 ⊂ 1a]·PF6. 1a·PF6 (30 mg, 7 μmol, 1 equiv),
KAu(CN)2 (4.1 mg, 14 μmol, 2 equiv), Cu(NCMe)4PF6 (2.6 mg,
7 μmol, 1 equiv) and MeCN (5 mL) were mixed in a vial. The reaction
mixture was stirred at room temperature for 12 h. Diethyl ether was
then added, and the desired complex [Ag(Au(CN)2)2 ⊂ 1a]·PF6 was
collected by filtration as a plum colored solid (38 mg, 70%): 1H NMR
(CD3CN, 400 MHz) 9.32 (8H, s, imine H), 8.39 (8H, t, J 7.76, Ph-H),
8.06−8.00 (24H, Ar-H), 7.83 (8H, d, J 7.96, py-H), 7.80 (8H, d, J 7.60,
py-H), 7.74 (8H, s, Ph-H), 7.65 (8H, dd, J 6.72, 2.16, py-H), 7.11
(16H, s, Ph-H), 6.98 (16H, d, J 1.12, Ph-H), 2.49 (24H, s, CH3), 2.39
(24H, s, CH3); 13C {1H} NMR (CD3CN, 125 MHz) 161.40, 160.93,
160.09, 159.30, 150.96, 150.53, 149.66, 149.37, 143.48, 139.56, 130.60,
130.04, 128.79, 128.15, 127.51, 124.83, 115.15, 26.21, 25.77; ESI-MS
[Ag(Au(CN)2)2 ⊂ 1a]7+ 560.72, [Ag(Au(CN)2)2 ⊂ 1a (PF6)]6+
678.22, [Ag(Au(CN)2)2 ⊂ 1a (PF6)2]5+ 842.87, [Ag(Au(CN)2)2 ⊂
1a (PF6)3]4+ 1090.03, [Ag(Au(CN)2)2 ⊂ 1a (PF6)4]3+ 1501.61;
Found: C, 43.56; H, 3.08; N, 9.66%. Calc. for C188H152AgAu2Cu8-
F42N36P7·13H2O: C, 43.64; H, 3.47; N, 9.74%.
1
here: H NMR (CD3CN, 400 MHz) 9.76 (8H, s, imine H), 9.39
(8H, s, imine H), 8.46 (4H, s, Ph-H), 8.09−8.03 (20H, Ar-H), 7.99−
7.92 (24H, Ar-H), 7.61−7.54 (16H, Ar-H), 7.46 (8H, s, Ph-H), 7.44
(8H, s, Ph-H), 7.23 (8H, d, J 8.08, Ph-H), 7.15 (8H, s, Ph-H), 6.49
(8H, d, J 8.08, Ph-H), 2.13 (24H, s, CH3), 1.70 (24H, s, CH3); 13C
{1H} NMR (CD3CN, 125 MHz) 162.54, 159.47, 159.26, 158.98,
150.85, 150.85, 149.81, 147.61, 144.65, 144.24, 140.22, 139.65, 139.36,
139.21, 139.09, 138.44, 130.06, 129.72, 129.34, 128.67, 128.31, 128.11,
127.81, 127.58, 127.31, 127.25, 124.23, 119.65, 106.89, 25.01, 23.74;
ESI-MS [1b]8+ 439.92, [1b(BF4)]7+ 515.24, [1b(BF4)2]6+ 615.43,
[1b(BF4)3]5+ 755.93, [1b(BF4)4]4+ 966.81, [1b(BF4)5]3+ 1317.78.
Found: C, 55.84; H, 3.96; N, 10.00%. Calc. for C208H168Cu8F32N32B8·
12H2O: C, 55.09; H, 4.27; N, 9.88%.
1c·PF6. To a Schlenk tube was added C, 5,5′-(naphthalene-2,6-
diyl)bis(benzene-1,3-diamine) (20 mg, 0.06 mmol, 4 equiv), 6-methyl-
2-pyridinecarboxaldehyde (28.5 mg, 0.23 mmol, 16 equiv), Cu-
(CH3CN)4PF6 (44 mg, 0.12 mmol, 8 equiv) and acetonitrile (5 mL).
The solution was degassed by three evacuation/nitrogen fill cycles and
stirred at room temperature for 12 h. A dark pink solution resulted.
The product was precipitated by adding diethyl ether into the reaction
mixture and was isolated by filtration as dark pink solid (20 mg, 29%).
1c-D2:1c-D4 = 96:4%, calculated from the integration of CH3 signals in
1
the H NMR spectrum (i.e., peaks at 2.54, 2.46 ppm). NMR data for
D2 isomer reported here: 1H NMR (CD3CN, 400 MHz) 9.84 (8H, br,
imine H), 9.40 (8H, s, imine H), 8.54 (4H, br, Ar-H), 8.22 (4H, br,
Ar-H), 8.12 (4H, s, Ph-H), 8.00−7.91 (36H, Ar-H), 7.86 (4H, d,
J 8.32, naph-H), 7.73 (4H, d, J 8.36, naph-H), 7.48−7.37 (28H, Ar-H),
7.00 (4H, s, naph-H), 6.68 (4H, d, J 8.68, naph-H), 6.47 (4H, d, J 8.28,
naph-H), 2.13 (24H, s, CH3, overlapping with H2O signals), 1.73
(24H, s, CH3); 13C {1H} NMR (CD3CN, 125 MHz) 162.46, 160.04,
159.55, 159.01, 151.15, 151.01, 149.93, 147.99, 145.10, 143.62, 140.07,
139.79, 137.73, 137.09, 133.61, 133.48, 130.26, 130.20, 129.95, 129.74,
128.23, 128.07, 127.65, 127.08, 126.33, 125.33, 124.53, 119.88, 107.81,
26.39, 25.93, 24.98, 23.79; ESI-MS [1c]8+ 439.95, [1c(PF6)]7+ 523.48,
[1c(PF6)2]6+ 634.87, [1c(PF6)3]5+ 790.91, [1c(PF6)4]4+ 1024.77,
[1c(PF6)5]3+ 1414.65. Found: C, 50.29; H, 3.45; N, 9.20%. Calc. for
C200H160Cu8F48N32P8·5H2O: C, 50.36; H, 3.59; N, 9.40%.
[Cu(Au(CN)2)2 ⊂ 1c]·PF6. 1c·PF6 (50 mg, 10.7 mmol, 1 equiv),
KAu(CN)2 (6.1 mg, 21.4 mmol, 2 equiv), Cu(NCMe)4PF6 (4.0 mg,
10,7 mmol, 1 equiv) and MeCN (5 mL) were mixed in a Schlenk flask.
The reaction mixture was stirred at room temperature for 5 h. Diethyl
ether was added into the reaction mixture; the resulting mixture was
centrifuged, and the solvent was decanted. The solid was dried under a
high vacuum to give the desired product [Cu(Au(CN)2)2 ⊂ 1c]·PF6 as
a dark pinkish-red solid (30 mg, 86%): 1H NMR (CD3CN, 400 MHz)
9.27 (8H, s, imine-H), 8.59 (8H, t, J 7.48, py-H), 8.06 (8H, t, J 7.68,
py-H), 8.04 (8H, s, imine-H), 7.96 (8H, d, J 6.40, py-H), 7.94 (8H, d,
J 7.20, py-H), 7.87 (8H, d, J 7.88, py-H), 7.69 (8H, d, J 8.04, py-H),
7.67 (8H, s, Ph-H), 7.45 (8H, s, Ph-H), 7.04 (8H, s, naph-H),
6.90 (16H, s, naph-H), 6.80 (8H, s, Ph-H), 2.50 (24H, s, CH3), 2.49
(24H, s, CH3); 13C {1H} NMR (CD3CN, 125 MHz) 161.97, 161.28,
160.47, 159.32, (Guest signals 153.29, 152.91, d, JC−Au−C 47.8, 151.42,
151.04, d, JC−Au−C 47.8) 150.95, 150.86, 150.85, 150.08, 149.53,
144.06, 140.27, 139.55, 137.39, 133.41, 130.94, 130.10, 129.98, 127.74,
127.53, 127.49, 126.85, 123.88, 118.25, 115.38, 26.48, 25.73; ESI-MS
[Cu(Au(CN)2)2 ⊂ L4Cu8]7+ 583.05, [Cu(Au(CN)2)2 ⊂ 1c(PF6)]6+
1c·BF4. To a NMR tube was added C, 5,5′-(naphthalene-2,6-
diyl)bis(benzene-1,3-diamine) (6 mg, 17.6 μmol, 4 equiv), 6-methyl-2-
pyridinecarboxaldehyde (8.5 mg, 70.5 μmol, 16 equiv), Cu(CH3CN)4BF4
(11 mg, 35.2 μmol, 8 equiv) and acetonitrile (1 mL). The resulting
dark pink solution was kept at 50 °C for 12 h. Diethyl ether was added
into the reaction mixture; the resulting mixture was centrifuged, and the
solvent was decanted. The solid was dried under a high vacuum to give
1
the desired product 1c·BF4 as a dark pink solid (14.4 mg, 77%): H
NMR (CD3CN, 400 MHz) 9.78 (8H, s, imine H), 9.43 (8H, s, imine
H), 8.51 (4H, s, Ph-H), 8.13−8.09 (12H, py-H), 8.00−7.93 (32H,
Ar-H), 7.87 (4H, d, J 8.36, naph-H), 7.75 (4H, d, J 8.28, naph-H), 7.47
(8H, s, Ph-H), 7.45 (8H, s, Ph-H), 7.41 (4H, s, naph-H), 7.35 (8H, s,
naph-H), 6.66 (4H, d, J 8.36, naph-H), 6.48 (4H, d, J 8.36, naph-H),
2.14 (24H, s, CH3), 1.72 (24H, s, CH3); 13C {1H} NMR (CD3CN,
125 MHz) 162.58, 159.58, 159.49, 159.08, 150.97, 150.83, 149.86, 147.88,
145.00, 143.68, 140.27, 139.74, 137.73, 136.90, 133.54 133.42, 130.25,
130.12, 129.84, 127.90, 127.80, 127.61, 127.02, 126.95, 126.35, 125.23,
124.41, 119.83, 107.25, 24.98, 23.70; ESI-MS [1c]8+ 452.91, [1c(BF4)]7+
530.29, [1c(BF4)2]6+ 632.81, [1c(BF4)3]5+ 776.73, [1c(BF4)4]4+ 992.65,
[1c(BF4)5]3+ 1352.19. Found: C, 53.84; H, 3.98; N, 10.09%. Calc. for
C200H160Cu8F32N32B8·14H2O: C, 53.78; H, 4.24; N, 10.03%.
704.37, [Cu(Au(CN)2)2 ⊂ 1c(PF6)2]5+ 874.26, [Cu(Au(CN)2)2
⊂
1c(PF6)3]4+ 1129.01, [Cu(Au(CN)2)2 ⊂ 1c(PF6)4]3+ 1553.50. Found:
C, 45.50; H, 3.15; N, 9.22%. Calc. for C204H160Cu9Au2F42N36P7·
15H2O: C, 45.66; H, 3.57; N, 9.40%.
[Ag(Au(CN)2)2 ⊂ 1c]·PF6. To [Cu(Au(CN)2)2 ⊂ 1c-D4]·PF6 (5
mg, 0.98 μmol, 1 equiv) was added a stock solution of AgPF6 (0.3 mg,
1.2 μmol, 1.2 equiv of stock solution prepared using 37 mg AgPF6 and
0.5 mL CD3CN). The resulting solution was heated at 40 °C for 4 h.
Diethyl ether was added into the reaction mixture, which was
centrifuged, and then the solvent was decanted. The solid was dried
under a high vacuum to give the desired product [Ag(Au(CN)2)2 ⊂
1c]·PF6 as a dark pinkish-red solid (5.3 mg, 96%): 1H NMR (CD3CN,
500 MHz) 9.26 (8H, s, imine H), 8.56 (8H, t, J 7.77, py-H), 8.06
(16H, imine-H and py-H), 7.96 (8H, d, J 7.60, py-H), 7.94 (8H, d,
J 7.60, py-H), 7.87 (8H, d, J 7.90, py-H), 7.69 (8H, d, J 7.55, py-H),
7.65 (8H, s, Ph-H), 7.46 (8H, s, naph-H), 7.04 (8H, s, Ph-H), 6.90−
6.87 (16H, m, naph-H), 6.82 (8H, s, Ph-H), 2.50 (24H, s, CH3), 2.48
(24H, s, CH3); 13C {1H} NMR (CD3CN, 125 MHz) 162.04, 161.32,
160.40, 159.36, 150.97, 150.83, 150.05, 149.57, 144.09, 140.19, 139.57,
[Au(Au(CN)2)2 ⊂ 1a]·BF4. [Cu(Au(CN)2)2 ⊂ 1a]·BF4 (6.6 mg,
1.5 μmol, 1 equiv), Au(tmbn)2SbF6 (1.4 mg, 1.7 μmol, 1.1 equiv) and
MeCN (0.35 mL) were mixed in a NMR tube. The tube was rotated on
a turner at room temperature for 12 h. Diethyl ether was then added,
and the product was collected by filtration as a plum-colored solid
1
(4 mg, 59%). H NMR revealed the presence of [Au(Au(CN)2)2 ⊂
1a]·BF4, [Cu(Au(CN)2)2 ⊂ 1a]·BF4 and 1a·BF4 in a ratio of 87:9:4%.
Further addition of Au(tmbn)2SbF6 did not increase the amount of
the desired product but produced more 1a. Characterization data for
3978
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980

Journal of the American Chemical Society
Article
137.59, 133.40 130.91, 130.13, 129.51, 127.96, 127.77, 127.54, 127.03,
124.06, 115.37, 26.47, 25.75 (Guest signals 153.1, dd, JC−Au−C, C−N−Ag
47.1, 26.0, and 152.3, d, JC−Au−C 46.8); ESI-MS [Ag(Au(CN)2)2 ⊂
1c]7+ 589.40, [Ag(Au(CN)2)2 ⊂ 1c (PF6)]6+ 711.55, [Ag(Au(CN)2)2
⊂ 1c(PF6)2]5+ 883.05, [Ag(Au(CN)2)2 ⊂ 1c(PF6)3]4+ 1140.01,
[Ag(Au(CN)2)2 ⊂ 1c(PF6)4]3+ 1568.57. Found: C, 46.35; H, 3.07;
N, 9.87%. Calc. for C204H160AgCu8Au2F42N36P7·7H2O: C, 46.52; H,
3.33; N, 9.57%.
Present Addresses
§Clarendon Laboratory, Department of Physics, University of
Oxford, Oxford, OX1 3PU, U.K.
∥School of Chemistry and Molecular Biosciences, The
University of Queensland, Brisbane St Lucia, QLD, Australia,
4072.
Notes
[Au(Au(CN)2)2 ⊂ 1c]·PF6. To [Cu(Au(CN)2)2 ⊂ 1c-D4]·PF6
(6.2 mg, 1.2 μmol, 1 equiv) in CD3CN (0.35 mL) in a j-young tube
was added Au(tmbn)2SbF6 (1.2 mg, 1.5 μmol, 1.2 equiv). The tube was
rotated on a turner at room temperature for 12 h. 1H NMR showed 50%
of the starting material was converted to [Au(Au(CN)2)2 ⊂ 1c]·PF6.
Further addition of Au(tmbn)2SbF6 did not increase the amount of
the desired product but converted left over [Cu(Au(CN)2)2 ⊂ 1c]·PF6
into 1c-D2; ESI-MS [Au(Au(CN)2)2 ⊂ 1c]7+ 602.06, [Au(Au(CN)2)2 ⊂
1c(PF6)]6+ 726.60, [Ag(Au(CN)2)2 ⊂ 1c(PF6)2]5+ 900.87, [Ag(Au-
(CN)2)2 ⊂ 1c(PF6)3]4+ 1162.46, [Ag(Au(CN)2)2 ⊂ 1c (PF6)4]3+ 1598.41.
4·BF4. To a NMR tube were added [1,1′:4′,1″-terphenyl]-3,3″,5,5″-
tetraamine A (1.5 mg, 5.2 μmol, 4 equiv), 6-methyl-2-pyridinecarbox-
aldehyde (2.5 mg, 20.7 μmol, 16 equiv), and acetonitrile (0.5 mL).
The resulting mixture was heated at 50 °C overnight before AgBF4
(2.0 mg, 10.3 μmol, 8 equiv) was added. The tube was turned at room
temperature overnight. A bright yellow solution resulted. The product
was purified by recrystallization: diethyl ether was diffused into an
acetonitrile solution of the complex. The desired product 4·BF4 was
isolated by filtration as yellow crystals (4.4 mg, 78%): 1H NMR
(CD3CN, 500 MHz) 9.45 (8H, d, JAg−H 5.9, imine H), 9.33 (8H, d,
JAg−H 7.8, imine H), 8.33 (4H, t, Ph-H), 8.06 (4H, t, J 1.75, Ph-H),
8.05−8.03 (16H, d, J 4.7, py-H), 7.96 (8H, s, Ph-H), 7.93 (8H, t, J 7.7,
py-H), 7.87 (8H, d, J 7.6, py-H), 7.54−7.51 (16H, py-H), 7.27−7.26
(16H, Ph-H), 6.66 (8H, d, J 7.9, Ph-H), 2.51 (24H, CH3), 2.07 (24H,
CH3); 13C {1H} NMR (CD3CN, 125 MHz) 163.71, 160.76, 160.30,
159.63, 150.10, 149.54, 149.21, 147.98, 144.43, 143.70, 141.22, 140.54,
140.27, 139.84, 130.01, 129.54, 129.17, 128.97, 128.90, 128.81, 128.67,
128.13, 125.59, 118.81, 27.18, 25.58; MALDI-MS using DCTB (2-
[(2E)-3-(4-tert-butylphenyl)-2-methylprop-2-enylidene]malononitrile)
matrix observed 4280.6, calc. for [L4Cu8(BF4)7]+ 4281.94. Found: C,
49.35; H, 3.43; N, 9.51%. Calc. for C184H152B8Ag8F32N32·5H2O: C,
49.56; H, 3.66; N, 10.05%.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the U.K. Engineering and Physical
Sciences Research Council (EPSRC), the US National Science
Foundation (NSF) and the Marie Curie IIF Scheme of the
7th EU Framework Program. We thank the EPSRC Mass
Spectrometry Service at Swansea for MALDI-TOF experiments,
Diamond Light Source (U.K.) for synchrotron beamtime on I19
(MT7114 and MT7569), and the EPSRC National Crystallog-
raphy Service at the University of Southampton for the collection
of crystallographic data. We thank Ana M. Castilla for the
preparation of Au(tmbn)2SbF6 and Emily V. Dry for the
preparation of Cu(CH3CN)4BF4.
REFERENCES
■
(1) Chakrabarty, R.; Mukherjee, P. S.; Stang, P. J. Chem. Rev. 2011,
111, 6810−6918.
(2) Ward, M. D. Chem. Commun. 2009, 4487−4499.
(3) Fujita, M.; Tominaga, M.; Hori, A.; Therrien, B. Acc. Chem. Res.
2005, 38, 369−378.
(4) Dalgarno, S. J.; Power, N. P.; Atwood, J. L. Coord. Chem. Rev.
2008, 252, 825−841.
(5) Tranchemontagne, D. J.; Ni, Z.; O’Keeffe, M.; Yaghi, O. M.
Angew. Chem., Int. Ed. 2008, 47, 5136−5147.
(6) Saalfrank, R. W.; Maid, H.; Scheurer, A. Angew. Chem., Int. Ed.
2008, 47, 8794−8824.
(7) Young, N. J.; Hay, B. P. Chem. Commun. 2013, 49, 1354−1379.
(8) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int.
Ed. 2009, 48, 3418−3438.
(9) Yoshizawa, M.; Tamura, M.; Fujita, M. Science 2006, 312, 251−
254.
Transmetalation. A stock solution was prepared using 18-crown-6
(6.2 mg, 23.4 μmol, 1.01 equiv), KAg(CN)2 (4.6 mg, 23.1 μmol,
1 equiv) and CD3CN (0.35 mL). To 1a·PF6 (4 mg, 0.9 μmol, 1 equiv)
in CD3CN (0.4 mL) in a NMR tube were added KAg(CN)2 (0.37 mg,
1.8 μmol, 2 equiv, 28 μL stock solution) and Au(tmbn)2SbF6 (0.77 mg,
0.9 μmol, 1 equiv). The tube was rotated on a turner at room
(10) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Science 2007,
316, 85−88.
(11) Hastings, C. J.; Fiedler, D.; Bergman, R. G.; Raymond, K. N. J.
Am. Chem. Soc. 2008, 130, 10977−10983.
1
temperature for 12 h. H NMR and ESI-MS showed the formation of
(12) Kuil, M.; Soltner, T.; Van Leeuwen, P. W. N. M.; Reek, J. N. H.
J. Am. Chem. Soc. 2006, 128, 11344−11345.
[Ag(Au(CN)2)2 ⊂ 1a-D4]·PF6.
́ ́
(13) Otte, M.; Kuijpers, P. F.; Troeppner, O.; Ivanovic-Burmazovic,
ASSOCIATED CONTENT
I.; Reek, J. N. H.; de Bruin, B. Chem.Eur. J. 2013, 19, 10170−10178.
(14) Wang, Z. J.; Clary, K. N.; Bergman, R. G.; Raymond, K. N.;
Toste, F. D. Nat. Chem. 2013, 5, 100−103.
■
S
* Supporting Information
Synthesis and characterization for tetraamine B and C, NMR
spectra for 1a·PF6, 2a·BF4, 2a·PF6, 3a·BF4, 1b·BF4, 1b·PF6, 1c·
BF4, 1c·PF6, 4·BF4, [Au(Au(CN)2)2 ⊂ 1a]·BF4, [Ag(Au(CN)2)2
⊂ 1a]·PF6, [Cu(Au(CN)2)2 ⊂ 1c]·PF6, [Ag(Au(CN)2)2 ⊂ 1c]·
PF6, and [Au(Au(CN)2)2 ⊂ 1c]·PF6, kinetic study of isomer-
ization for 1b·BF4, calculation of binding constants for Au(Au-
(CN)2)2, and crystallographic data (CIF). Crystallographic data
has been deposited with the CCDC (numbers 846046, 955699,
and 905144−905149). This material is available free of charge via
the Internet at http://pubs.acs.org.
(15) Gunnlaugsson, T.; Glynn, M.; Tocci, G. M.; Kruger, P. E.;
Pfeffer, F. M. Coord. Chem. Rev. 2006, 250, 3094−3117.
(16) Kondo, K.; Suzuki, A.; Akita, M.; Yoshizawa, M. Angew. Chem.,
Int. Ed. 2013, 52, 2308−2312.
(17) Li, S.-S.; Northrop, B. H.; Yuan, Q.-H.; Wan, L.-J.; Stang, P. J.
Acc. Chem. Res. 2008, 42, 249−259.
(18) Sessler, J. L.; Camiolo, S.; Gale, P. A. Coord. Chem. Rev. 2003,
240, 17−55.
(19) Smith, D. K.; Diederich, F. Top. Curr. Chem. 2000, 210, 183−
227.
(20) Watt, M. M.; Zakharov, L. N.; Haley, M. M.; Johnson, D. W.
Angew. Chem., Int. Ed. 2013, 52, 10275−10280.
(21) Dong, V. M.; Fiedler, D.; Carl, B.; Bergman, R. G.; Raymond, K.
N. J. Am. Chem. Soc. 2006, 128, 14464−14465.
(22) Mal, P.; Breiner, B.; Rissanen, K.; Nitschke, J. R. Science 2009,
324, 1697−1699.
AUTHOR INFORMATION
■
Corresponding Author
jrn34@cam.ac.uk; cramer@umn.edu; gagliard@umn.edu
3979
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980

Journal of the American Chemical Society
Article
(23) Sawada, T.; Yoshizawa, M.; Sato, S.; Fujita, M. Nat. Chem. 2009,
1, 53−56.
(55) Han, M.; Michel, R.; He, B.; Chen, Y.-S.; Stalke, D.; John, M.;
Clever, G. H. Angew. Chem., Int. Ed. 2013, 52, 1319−1323.
(56) Bong, D. T.; Clark, T. D.; Granja, J. R.; Ghadiri, M. R. Angew.
Chem., Int. Ed. 2001, 40, 988−1011.
(24) Yoshizawa, M.; Tamura, M.; Fujita, M. Angew. Chem., Int. Ed.
2007, 46, 3874−3876.
(57) Kimizuka, N.; Kawasaki, T.; Hirata, K.; Kunitake, T. J. Am.
Chem. Soc. 1995, 117, 6360−6361.
(25) Riddell, I. A.; Smulders, M. M. J.; Clegg, J. K.; Nitschke, J. R.
Chem. Commun. 2011, 47, 457−459.
(26) Kishi, N.; Akita, M.; Kamiya, M.; Hayashi, S.; Hsu, H.-F.;
Yoshizawa, M. J. Am. Chem. Soc. 2013, 135, 12976−12979.
(27) Berna, J.; Leigh, D. A.; Lubomska, M.; Mendoza, S. M.; Perez, E.
M.; Rudolf, P.; Teobaldi, G.; Zerbetto, F. Nat. Mater. 2005, 4, 704−
710.
(58) Matile, S.; Som, A.; Sorde, N. Tetrahedron 2004, 60, 6405−
6435.
(59) Ajami, D.; Rebek, J. Acc. Chem. Res. 2013, 46, 990−999.
(60) Aoyagi, M.; Tashiro, S.; Tominaga, M.; Biradha, K.; Fujita, M.
Chem. Commun. 2002, 2036−2037.
(61) Tominaga, M.; Tashiro, S.; Aoyagi, M.; Fujita, M. Chem.
Commun. 2002, 2038−2039.
(62) Chifotides, H. T.; Giles, I. D.; Dunbar, K. R. J. Am. Chem. Soc.
2013, 135, 3039−3055.
(28) Sakai, N.; Matile, S. Langmuir 2013, 29, 9031−9040.
(29) Custelcean, R.; Bonnesen, P. V.; Duncan, N. C.; Zhang, X. H.;
Watson, L. A.; Van Berkel, G.; Parson, W. B.; Hay, B. P. J. Am. Chem.
Soc. 2012, 134, 8525−8534.
(63) Glasson, C. R. K.; Meehan, G. V.; Motti, C. A.; Clegg, J. K.;
Davies, M. S.; Lindoy, L. F. Aust. J. Chem. 2012, 65, 1371−1376.
(64) Coles, S. J.; Gale, P. A. Chem. Sci. 2012, 3, 683−689.
(65) Hibble, S. J.; Cheyne, S. M.; Hannon, A. C.; Eversfield, S. G.
Inorg. Chem. 2002, 41, 1042−1044.
(30) Belowich, M. E.; Stoddart, J. F. Chem. Soc. Rev. 2012, 41, 2003−
2024.
̌
(31) Osowska, K.; Miljanic,
8345−8349.
́
O. S. Angew. Chem., Int. Ed. 2011, 50,
(32) Bunzen, H.; Nonappa; Kalenius, E.; Hietala, S.; Kolehmainen, E.
Chem.Eur. J. 2013, 19, 12978−12981.
(66) Veldkamp, A.; Frenking, G. Organometallics 1993, 12, 4613−
4622.
(33) Campbell, V. E.; de Hatten, X.; Delsuc, N.; Kauffmann, B.; Huc,
I.; Nitschke, J. R. Chem.Eur. J. 2009, 15, 6138−6142.
(67) Selig, W. S. Microchem. J. 1985, 32, 18−23.
(68) Ludlow, R. F.; Otto, S. Chem. Soc. Rev. 2008, 37, 101−108.
(69) Badjic, J. D.; Balzani, V.; Credi, A.; Silvi, S.; Stoddart, J. F. Science
2004, 303, 1845−1849.
(34) Domer, J.; Slootweg, J. C.; Hupka, F.; Lammertsma, K.; Hahn,
̈
F. E. Angew. Chem., Int. Ed. 2010, 49, 6430−6433.
(35) Wu, Y.; Zhou, X.-P.; Yang, J.-R.; Li, D. Chem. Commun. 2013,
49, 3413−3415.
(36) Nitschke, J.; Ronson, T.; Zarra, S.; Black, S. P. Chem. Commun.
2012, 49, 2476−2490.
(70) Hernandez, J. V.; Kay, E. R.; Leigh, D. A. Science 2004, 306,
1532−1537.
(71) Gianneschi, N. C.; Nguyen, S. T.; Mirkin, C. A. J. Am. Chem.
Soc. 2005, 127, 1644−1645.
(37) Bilbeisi, R. A.; Clegg, J. K.; Elgrishi, N.; Hatten, X. d.; Devillard,
M.; Breiner, B.; Mal, P.; Nitschke, J. R. J. Am. Chem. Soc. 2012, 134,
5110−5119.
(72) Thordarson, P.; Bijsterveld, E. J. A.; Rowan, A. E.; Nolte, R. J.
M. Nature 2003, 424, 915−918.
(73) Ray, D.; Foy, J. T.; Hughes, R. P.; Aprahamian, I. Nat. Chem.
2012, 4, 757−762.
(38) Ferguson, A.; Squire, M. A.; Siretanu, D.; Mitcov, D.;
Mathoniere, C.; Clerac, R.; Kruger, P. E. Chem. Commun. 2013, 49,
1597−1599.
(74) Hilson, G.; Monhemius, A. J. J. Cleaner Prod. 2006, 14, 1158−
1167.
(39) Mal, P.; Schultz, D.; Beyeh, K.; Rissanen, K.; Nitschke, J. R.
Angew. Chem., Int. Ed. 2008, 47, 8297−8301.
(75) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104−154119.
(40) Yi, S.; Brega, V.; Captain, B.; Kaifer, A. E. Chem. Commun. 2012,
48, 10295−10297.
(76) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,
3865−3868.
(41) Chepelin, O.; Ujma, J.; Wu, X. H.; Slawin, A. M. Z.; Pitak, M. B.;
Coles, S. J.; Michel, J.; Jones, A. C.; Barran, P. E.; Lusby, P. J. J. Am.
Chem. Soc. 2012, 134, 19334−19337.
(77) ADF; SCM, Theoretical Chemistry, Vrije Universiteit:
Amsterdam, The Netherlands, 2013.
(78) Fonseca Guerra, C.; Snijders, J. G.; te Velde, G.; Baerends, E. J.
Theor. Chem. Acc. 1998, 99, 391−403.
(42) Meng, W.; Breiner, B.; Rissanen, K.; Thoburn, J. D.; Clegg, J. K.;
Nitschke, J. R. Angew. Chem., Int. Ed. 2011, 50, 3479−3483.
(43) Liu, Y.; Kravtsov, V.; Walsh, R. D.; Poddar, P.; Srikanth, H.;
Eddaoudi, M. Chem. Commun. 2004, 2806−2807.
(44) Stephenson, A.; Ward, M. D. Dalton Trans. 2011, 10360−
10369.
(79) te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca
Guerra, C.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J.
Comput. Chem. 2001, 22, 931−967.
(80) van Lenthe, E.; Baerends, E. J. J. Comput. Chem. 2003, 24,
1142−1156.
(45) Zhou, X.-P.; Liu, J.; Zhan, S.-Z.; Yang, J.-R.; Li, D.; Ng, K.-M.;
Sun, R. W.-Y.; Che, C.-M. J. Am. Chem. Soc. 2012, 134, 8042−8045.
(46) Zhou, X. P.; Wu, Y.; Li, D. J. Am. Chem. Soc. 2013, 135, 16062−
16065.
(81) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.
1993, 99, 4597−4610.
(82) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.
1994, 101, 9783−9792.
(47) Bilbeisi, R. A.; Ronson, T. K.; Nitschke, J. R. Angew. Chem., Int.
(83) van Lenthe, E.; Ehlers, A.; Baerends, E.-J. J. Chem. Phys. 1999,
110, 8943−8953.
Ed. 2013, 52, 9027−9030.
(48) Meng, W.; Ronson, T. K.; Clegg, J. K.; Nitschke, J. R. Angew.
Chem., Int. Ed. 2013, 52, 1017−1021.
(84) Pye, C. C.; Ziegler, T. Theor. Chem. Acc. 1999, 101, 396−408.
(85) Raducan, M.; Rodriguez-Escrich, C.; Cambeiro, X. C.; Escudero-
Adan, E. C.; Pericas, M. A.; Echavarren, A. M. Chem. Commun. 2011,
47, 4893−4895.
(49) Sham, K.-C.; Yiu, S.-M.; Kwong, H.-L. Inorg. Chem. 2013, 52,
5648−5650.
(50) Riddell, I. A.; Smulders, M. M. J.; Clegg, J. K.; Hristova, Y. R.;
Breiner, B.; Thoburn, J. D.; Nitschke, J. R. Nat. Chem. 2012, 4, 751−
756.
(51) Riddell, I. A.; Hristova, Y. R.; Clegg, J. K.; Wood, C. S.; Breiner,
B.; Nitschke, J. R. J. Am. Chem. Soc. 2013, 135, 2723−2733.
(52) Meng, W.; Ronson, T. K.; Nitschke, J. R. Proc. Natl. Acad. Sci. U.
S. A. 2013, 110, 10531−10535.
(53) Meng, W.; Clegg, J. K.; Nitschke, J. R. Angew. Chem., Int. Ed.
2012, 51, 1881−1884.
(54) Ustinov, A.; Weissman, H.; Shirman, E.; Pinkas, I.; Zuo, X.;
Rybtchinski, B. J. Am. Chem. Soc. 2011, 133, 16201−16211.
3980
dx.doi.org/10.1021/ja412964r | J. Am. Chem. Soc. 2014, 136, 3972−3980