Allosteric regulation of supramolecular oligomerization and catalytic activity via coordination-based control of competitive hydrogen-bonding events

Article abstract of DOI:10.1021/ja508804n

Herein, we demonstrate that the activity of a hydrogen-bond-donating (HBD) catalyst embedded within a coordination framework can be allosterically regulated in situ by controlling oligomerization via simple changes in coordination chemistry at distal Pt(II) nodes. Using the halide-induced ligand rearrangement reaction (HILR), a heteroligated Pt(II) triple-decker complex, which contains a catalytically active diphenylene squaramide moiety and two hydrogen-bond-accepting (HBA) ester moieties, was synthesized. The HBD and HBA moieties were functionalized with hemilabile ligands of differing chelating strengths, allowing one to assemble them around Pt(II) nodes in a heteroligated fashion. Due to the hemilabile nature of the ligands, the resulting complex can be interconverted between a flexible, semiopen state and a rigid, fully closed state in situ and reversibly. FT-IR spectroscopy, ¹H DOSY, and ¹H NMR spectroscopy titration studies were used to demonstrate that, in the semiopen state, intermolecular hydrogen-bonding between the HBD and HBA moieties drives oligomerization of the complex and prevents substrate recognition by the catalyst. In the rigid, fully closed state, these interactions are prevented by steric and geometric constraints. Thus, the diphenylene squaramide moiety is able to catalyze a Friedel-Crafts reaction in the fully closed state, while the semiopen state shows no reactivity. This work demonstrates that controlling catalytic activity by regulating aggregation through supramolecular conformational changes, a common approach in Nature, can be applied to man-made catalytic frameworks that are relevant to materials synthesis, as well as the detection and amplification of small molecules.

Full text of DOI:10.1021/ja508804n

Article
pubs.acs.org/JACS
Allosteric Regulation of Supramolecular Oligomerization and
Catalytic Activity via Coordination-Based Control of Competitive
Hydrogen-Bonding Events
C. Michael McGuirk, Jose Mendez-Arroyo, Alejo M. Lifschitz, and Chad A. Mirkin*
Department of Chemistry and The International Institute for Nanotechnology, Northwestern University, 2145 Sheridan Road,
Evanston, Illinois 60208-3113, United States
S
* Supporting Information
ABSTRACT: Herein, we demonstrate that the activity of a hydrogen-
bond-donating (HBD) catalyst embedded within a coordination frame-
work can be allosterically regulated in situ by controlling oligomerization
via simple changes in coordination chemistry at distal Pt(II) nodes. Using
the halide-induced ligand rearrangement reaction (HILR), a heteroligated
Pt(II) triple-decker complex, which contains a catalytically active
diphenylene squaramide moiety and two hydrogen-bond-accepting
(HBA) ester moieties, was synthesized. The HBD and HBA moieties
were functionalized with hemilabile ligands of differing chelating strengths,
allowing one to assemble them around Pt(II) nodes in a heteroligated
fashion. Due to the hemilabile nature of the ligands, the resulting complex
can be interconverted between a flexible, semiopen state and a rigid, fully closed state in situ and reversibly. FT-IR spectroscopy,
1
1H DOSY, and H NMR spectroscopy titration studies were used to demonstrate that, in the semiopen state, intermolecular
hydrogen-bonding between the HBD and HBA moieties drives oligomerization of the complex and prevents substrate
recognition by the catalyst. In the rigid, fully closed state, these interactions are prevented by steric and geometric constraints.
Thus, the diphenylene squaramide moiety is able to catalyze a Friedel−Crafts reaction in the fully closed state, while the
semiopen state shows no reactivity. This work demonstrates that controlling catalytic activity by regulating aggregation through
supramolecular conformational changes, a common approach in Nature, can be applied to man-made catalytic frameworks that
are relevant to materials synthesis, as well as the detection and amplification of small molecules.
gen-bond-driven oligomerization, substrate recognition, and
catalytic activity. By developing a novel methodology for the
allosteric regulation of catalytically active coordination-based
complexes, we open the door to applications in small-molecule
detection and amplification, as well as multicomponent cascade
reactions. Using the weak-link approach (WLA) to the
synthesis of coordination-driven supramolecular structures, we
embedded both hydrogen-bond-donating (HBD) and hydro-
gen-bond-accepting (HBA) moieties within a coordination
framework.5,6 In these complexes, recognition of coordinating
small-molecule analytes at distal Pt(II) centers is followed by
conformational changes in the supramolecular structure,
thereby allosterically regulating the ability of the hydrogen-
bonding pairs to form intermolecular, catalytically deactivating
interactions (Scheme 1).
INTRODUCTION
■
In over 40 distinct metabolically active enzymes, allosteric up-
regulation of catalytic activity is characterized by the
dissociation of dormant higher-order enzyme oligomers into
catalytically active dimers or monomers.1 Previous to up-
regulation, the active site of such enzymes can participate in the
formation of higher order quaternary structures via hydrogen-
bonding interactions, forming tertiary peptide structures at the
active site that impede substrate recognition, and thus catalytic
activity.2,3 Like other allosteric mechanisms, the dissociation of
dormant oligomers is regulated by the binding of small-
molecule “effectors” to chemically orthogonal recognition sites
on the enzyme.4 Oligomer dissociation not only exposes the
enzymatic face containing the active site, but also causes
conformational changes in the tertiary structure of the active
site, alleviating catalytically detrimental hydrogen bonds
between key amino acids. The resulting hydrogen-bonding
network of the active site is then able to recognize and activate
an electrophilic substrate, initiating catalytic turnover. Inspired
by the role of hydrogen-bonding in the quaternary and tertiary
structures of such allosterically regulated enzymes, we set out to
design an analogous coordination-based system in which the
reversible binding of small-molecule effectors regulates hydro-
Among the various methodologies for synthesizing coordi-
nation-driven supramolecular constructs,7−15 the WLA is
unique for its modular and convergent synthesis of multi-
component assemblies that can be predictably and reversibly
toggled between multiple configurations in response to the
Received: August 26, 2014
Published: November 12, 2014
© 2014 American Chemical Society
16594
dx.doi.org/10.1021/ja508804n | J. Am. Chem. Soc. 2014, 136, 16594−16601

Journal of the American Chemical Society
Article
Scheme 1. Proposed Weak-Link Approach Triple-Decker Complex for the Allosteric Regulation of Squaramide Catalytic
a
Activity
a
The red circle represents catalytic feedstock, and the blue triangle represents catalytic product.
Scheme 2. (A) Synthesis of and Reversible Toggling between Flexible, Semiopen (3) and Rigid, Fully Closed (4)
Configurations, and (B) Synthesis of and Reversible Toggling between Model Semiopen (6) and Model Fully Closed (7)
Configurations
recognition of small-molecule effectors at chemically orthogo-
nal regulatory sites.5,16−19 This is achieved via reversible
coordination of small-molecule effectors (e.g., Cl−, CO) to d8
metal-based structural nodes (e.g., Rh(I), Pt(II)), coordinated
by phosphino−heteroatom (P,X; X = S, O, Se, N) hemilabile
ligands. Therefore, we designed a heteroligated Pt(II) triple-
decker complex featuring both a symmetric, homoditopic P,S-
functionalized biphenylene squaramide HBD ligand (1), and
two monotopic P,S-functionalized methyl tetrafluorobenzoate
HBA ligands (2) (Scheme 2A). As we have previously
demonstrated, utilizing a 2:2:1 mixture of PtCl2(cod) and
two model P,S hemilabile ligands of significantly different
chelating strength (e.g., P,S-tetrafluorophenyl and P,S-phenyl)
affords the near-quantitative synthesis of a heteroligated Pt(II)
triple-decker complex via the halide-induced ligand rearrange-
ment reaction (HILR).20 The hemilabile nature of the P,S-
fluorophenyl motifs allows for the quantitative in situ
conversion of the complex between a flexible, semiopen
configuration and a rigid, geometrically constrained fully closed
configuration. We predicted that in the semiopen state (3)
intermolecular hydrogen-bonding interactions between the
HBA ester moieties and the HBD squaramide moiety would
16595
dx.doi.org/10.1021/ja508804n | J. Am. Chem. Soc. 2014, 136, 16594−16601

Journal of the American Chemical Society
Article
drive complex oligomerization analogous to enzymatic
quaternary structure (Scheme 1). This extended hydrogen-
bonding network would effectively outcompete substrate
recognition, thus hampering catalytic activity. Upon in situ
switching to the rigid, fully closed configuration (4),
intermolecular ester−squaramide hydrogen-bonding would be
disrupted by geometric and steric restraints, resulting in the
stabilization of the monomeric state. Additionally, the tertiary
structure would be such that the squaramide moiety is free to
hydrogen-bonding to potential HBA substrates and activate
them toward nucleophilic attack. To understand the differences
in shape and structure between semiopen 3 and fully closed 4,
as well as develop structure−function relationships, computa-
tional models were generated and optimized using DFT.
Characterization of intermolecular hydrogen-bonding and
complex oligomerization was performed using FT-IR spectros-
ppm (JP−P = 14 Hz, JP−Pt = 3466 Hz), arising from the chelated
P,S-phenylene squaramide-containing ligand. These values are
in strong agreement with previously reported values for
semiopen heteroligated Pt(II) structures.20 Fully closed 4 can
be isolated via two routes: (1) abstraction of inner-sphere
chlorides from semiopen 3(BF4)2 with 2 equiv of AgBF4,
followed by a counteranion exchange with NaBArF, or (2)
abstraction of inner-sphere chlorides from semiopen 3(BArF)2
via the addition of 2 equiv of a 0.1 M ethereal solution of
AgBArF. The 31P{1H} NMR spectrum of fully closed 4 shows
two downfield resonances at 44.57 ppm (JP−Pt = 3087 Hz) and
45.79 ppm (JP−Pt = 3244 Hz), corresponding to the fully
chelated heteroligated configuration. Based on their corre-
sponding P−Pt coupling constants, we can assign the resonance
at 44.57 ppm to the phosphorus atom of the P,S-
fluorophenylene ester-functionalized ligands (2), and the
resonance at 45.79 ppm to the phosphorus atom of the
symmetric P,S-phenylene squaramide bridging ligand (1). Both
semiopen 3 and fully closed 4 are stable under ambient
benchtop conditions.
1
copy concentration gradient studies and H diffusion-ordered
NMR spectroscopy (DOSY), respectively. The HBA substrate
recognition capabilities of the two distinct states were studied
1
using H NMR spectroscopy substrate binding studies. Finally,
to test the validity of our hypothesis, the ability of the WLA
construct to allosterically regulate the Friedel−Crafts (F-C)
reaction of indole and nitrostyrene in situ was determined. In
addition to the functional complex, a model system was
synthesized, possessing a simple proton in place of the HBA
methyl ester moiety (Scheme 2B). These models were used as a
control throughout this study to directly determine the role of
intermolecular hydrogen-bonding in the regulatory mechanism.
DFT Calculations. To gain further insight into the
structural organization of the synthesized structures, Density
Functional Theory (DFT) calculations were used to generate
models of the semiopen (3) and fully closed (4) states. Using
the same parameters (GGA:PBE TZ2P) we have previously
used to calculate the energy-minimized structures of hetero-
ligated WLA complexes,23,24 we calculated the structures of 3
and 4. As can be seen in Figure 1A,B, both the ball-and-stick
RESULTS AND DISCUSSION
■
Synthesis. In order to create a supramolecular complex in
which ligand−ligand hydrogen-bonding regulates the catalytic
activity of a HBD catalyst (e.g., biphenylene squaramide),21,22
we designed a structure in which the two conformational states
of the complex promote significantly disparate degrees of
intermolecular interaction. Whereas the flexibility of mono-
dentate ligands in the semiopen state would readily allow for
intermolecular hydrogen-bonding, the fully closed configura-
tion had to be properly designed to utilize geometric, steric,
and, potentially, electrostatic restraints to limit ligand−ligand
interactions. Previously, we reported on the single-crystal X-ray
diffraction structures of both the semiopen and fully closed
Pt(II) triple-decker model complexes.20 Using these structures
as an operational reference, we designed the aforementioned
triple-decker complex, for which we expected the steric bulk
associated with the “dumbbell-like” architecture of the fully
closed state, combined with the rigid nature of the bidentate
ligands and the tetravalent nature of the complex, to hinder
hydrogen-bonding between the HBD and HBA ligands.
To afford the designed triple-decker complex, we utilized an
adaptation of our previously reported syntheses for hetero-
ligated Pt(II) complexes.20 Specifically, 1 equiv of the bridging
homoditopic P,S-functionalized biphenylene squaramide ligand
(1) and 2 equiv of the monotopic P,S-functionalized methyl
tetrafluorobenzoate ligand (2) were added to 2 equiv of Pt(II)
precursor PtCl2(cod). Upon isolation of semiopen 3(BF4)2, a
counteranion exchange with NaBArF was performed, yielding
divalent semiopen 3(BArF)2. Utilizing BArF− as the counter-
nanion is critical for solubility in nonpolar solvents, which
promote the catalytic activity of HBD catalysts. 31P{1H} NMR
spectroscopy of semiopen 3 displays two resonances: one at
8.39 ppm (JP−P = 14 Hz, JP−Pt = 3231 Hz) corresponding to the
P-bound P,S-fluorophenylene ligands, and the other at 43.44
Figure 1. Energy-minimized structures: (A) space-filling front-on view
of semiopen 3, (B) ball-and-stick front-on view of semiopen 3, (C)
space-filling front-on view of fully closed 4, and (D) ball-and-stick top-
down view of fully closed 4. Colors: C, gray; P, magenta; S, orange; O,
red; N, blue; Pt, black; F, lime green; Cl, forest green; H, white.
Hydrogens omitted from ball-and-stick structures for clarity.
and space-filling structures of semiopen 3 suggest that there are
no geometric or steric constraints that would prevent the
intermolecular hydrogen-bonding association of multiple semi-
open complexes in solution. On the other hand, the geometry
optimization of fully closed 4 (Figure 1C,D) shows a rigid,
anisotropic “dumbbell-like” structure. This structure displays
geometric rigidity of the ester ligands imposed by chelation, as
well as the significant steric bulk of the end groups, which may
cooperatively deter intermolecular association of the fully
closed complexes.
Intermolecular Hydrogen-Bonding. In order to correlate
the structural modeling to function, we needed to directly
measure the self-associative behavior of the semiopen (3) and
fully closed (4) configurations in solution. Previously, we have
16596
dx.doi.org/10.1021/ja508804n | J. Am. Chem. Soc. 2014, 136, 16594−16601

Journal of the American Chemical Society
Article
demonstrated the ability to characterize hydrogen-bonding of
urea derivatives in a WLA architecture using FT-IR spectros-
copy by examining the N−H bond vibration that occurs from
3400 to 3200 cm−1.25 Unfortunately, the N−H bond vibration
of diphenyl squaramide derivatives occurs between 3200 and
3000 cm−1, thus overlapping with aliphatic and aromatic C−H
bond stretches. Therefore, we turned to the less prominent
CO bond vibration of the squaramide backbone to
characterize the degree of hydrogen-bonding in our structures.
Through extensive efforts by de Oliveira et al., it has been
shown that the CO bond vibration of a diphenyl squaramide
derivative occurs at approximately 1795 cm−1.26,27 It was also
compuationally demonstrated that upon hydrogen-bond
donation by the N−H moiety, the CO bond vibration shifts
to lower wavenumbers. Several computational investigations of
diphenyl squaramide derivatives have concluded that the
squaramide moiety is partially aromatic.28 Therefore, the
donation of electron density into the squaramide moiety via
hydrogen-bonding to some electron-rich HBA moiety will
increase the aromatic nature of the squaramide moiety. Such an
increase in the conjugation of a CO bond is well known to
cause carbonyl bond vibrations to shift to lower wave-
numbers.29 Thus, to study the propensity of the complexes
to form intermolecular hydrogen bonds between the ester and
squaramide ligands we performed a FT-IR spectroscopy
concentration gradient study in which the position of the
CO band was measured across a concentration range. As can
be seen in Figure 2, increasing the concentration of semiopen 3
we can conclude that whereas the flexible, semiopen state (3)
can form intermolecular hydrogen bonds, there is minimal
interaction in the rigid, fully closed configuration (4). To our
knowledge, this is the first example of direct characterization of
hydrogen-bonding of a diphenyl squaramide HBD catalyst. It
should be noted that no study of a mixture of the two free
ligands in solution could be performed due to the complete
insolubility of the free diphenylene squaramide ligand in any
solvent in which there is negligible hydrogen-bonding
interaction between solvent and squaramide.
Complex Oligomerization. Having characterized the
significant difference in intermolecular hydrogen-bonding
interactions, we sought to determine whether the coordination
mode at the Pt(II) structural node controlled the quaternary
1
structure of the construct. H diffusion-ordered NMR spec-
troscopy (DOSY) was employed to determine the relative state
of aggregation of 3 and 4. Recently, DOSY has become a
powerful tool for characterizing noncovalent aggregates of
supramolecular architectures and determining the degree of
oligomerization.30−32 Therefore, H DOSY experiments were
1
run on both semiopen 3 (Figure 3) and fully closed complex 4,
Figure 3. 1H DOSY spectrum of semiopen 3 with CH2Cl2 as internal
standard. Methyl protons of 3 are highlighted in green; ethylene
protons of 3 are highlighted in blue.
as well as on their respective models (6 and 7), at 0.01 M
concentration in dichloromethane. In order to determine the
degree of oligomerization of the two configurations, the
hydrodynamic volume (VH) of the functional complexes, as
1
determined by H DOSY, were compared to that of their
respective model complex.30 Using this method, the VH of
semiopen 3 was calculated to be approximately four times
greater than the VH of model semiopen 6. Thus, assuming the
model 6 is monomeric in solution, our study suggests that
semiopen 3 exists as a tetramer in solution. In contrast, the VH
of fully closed 4 is approximately equal to that of its respective
model complex, suggesting that fully closed 4 is monomeric.
Alternatively, we can make a direct comparison of semiopen 3
and fully closed 4 by accounting for the apparent rigid
structural anisotropy of 4, determined from the energy-
minimized structure, in our calculation of VH (see SI).30 We
thus derive a VH ratio of approximately 5:1 for semiopen 3:fully
closed 4. We attribute the deviation in the calculated state of
aggregation between four and five to our inability to
mathematically account for the difference in valency between
the semiopen and fully closed states. Importantly, we can
conclude that, at the given concentration in nonpolar solvent,
the prevalence of intermolecular hydrogen-bonding in the
flexible, semiopen state (3) drives oligomerization. Compara-
tively, fully closed 4 exists in a monomeric state.
Figure 2. Wavenumber (cm−1) of squaramide CO bond vibration
vs complex concentration in CH2Cl2 for semiopen 3 (red circle), fully
closed 4 (black square), and ester-free semiopen model 6 (blue
triangle).
from 0.0025 to 0.01 M in dichloromethane results in an
approximately linear shift to lower wavenumbers. This trend
suggests the presence of intermolecular hydrogen-bonding at
the squaramide moiety. In contrast, there is no such shift in the
CO band of the fully closed complex (4) (Figure 2),
suggesting negligible hydrogen-bonding interaction between
the ester and squaramide moieties. In order to confirm that the
observed trend for semiopen 3 is in fact due to hydrogen-
bonding between the ester and squaramide moieties, the same
experiment was performed with model semiopen 6. Indeed, the
CO bond vibration of model semiopen 6 does not shift with
increasing concentration (Figure 2), confirming the intermo-
lecular hydrogen-bonding interaction between the ester and
squaramide moiety for semiopen 3. From these observations,
16597
dx.doi.org/10.1021/ja508804n | J. Am. Chem. Soc. 2014, 136, 16594−16601

Journal of the American Chemical Society
Article
Figure 4. δ(ppm) of ethyl acetate methyl protons vs equivalents of ethyl acetate added to (A) semiopen 3, (B) fully closed 4, (C) ester-free model
semiopen 6, and (D) ester-free model fully closed 7. Titrations performed with 1.2 mM solution of complex in CD2Cl2.
HBA Substrate Association Studies. To evaluate the
effects that toggling intermolecular hydrogen-bonding has on
the HBA substrate recognition properties of semiopen 3 and
fully closed 4, the respective complexes were titrated with ethyl
1
acetate. Recognition was measured from the changes in H
NMR chemical shifts of the ethyl acetate methyl protons as a
function of ethyl acetate equivalents added. Figure 4A shows
negligible recognition of ethyl acetate by semiopen 3, even at
high equivalency of the substrate. This is remarkable, as ethyl
acetate is a more electron rich HBA substrate than the methyl
fluorobenzoate ligand. We attribute this effect to the high local
concentration of the HBA ligand around the HBD squaramide.
In contrast, fully closed 4 displays significant substrate
recognition (Figure 4B). Substrate recognition studies were
also performed with glutaric anyhydride and nitromethane.
These substrates show the same “on−off” contrast in binding
between the two complexes (see SI). Additional ethyl acetate
titration studies were performed with the model semiopen (6)
and fully closed (7) structures (Figure 4C,D). Both models
show very similar binding curves to fully closed 4. From these
studies we can infer a few points. First, intermolecular
hydrogen-bonding interactions between the diphenylene
squaramide and ester ligands in semiopen 3 prevent HBA
substrate recognition. Second, in fully closed 4 the tertiary
structure is such that the squaramide moiety is unimpeded by
intermolecular hydrogen-bonding, thus available to bind HBA
substrate. While geometric and steric restraints prevent ligand−
ligand association, the adopted structure also exposes the
squaramide active site. Third, there is no appreciable interaction
between the HBD and HBA ligands in fully closed 4, as the
substrate binding is very similar to the ester-free fully closed
model 7.
Allosteric Regulation of Catalysis. Upon elucidating the
significant differences in the quaternary and tertiary structures
of semiopen 3 and fully closed 4, we tested whether these
differences would translate into a significant difference in the
catalytic activity of the diphenylene squaramide active site.
Thus, we compared the ability of semiopen 3, fully closed 4,
and semiopen model 6 to catalyze the C−C bond-forming F-C
reaction between indole and nitrostyrene to form 3-(2-nitro-1-
phenylethyl)-1H-indole (Figure 5A).33−37 The Lewis acid-
catalyzed F-C addition of indole to nitroalkenes is the key initial
step in the synthesis of many pharmacologically active alkaloid
derivatives containing the tryptamine backbone.38 The test
reactions were carried out with a 1.5:1 indole:nitrostyrene
stoichiometric ratio (0.03 M: 0.02 M) in 1,2-dichlorobenzene-
d4 at 60 °C, in the presence of 20 mol % of complex. All
catalysis was performed under ambient, benchtop conditions.
As seen in Figure 5B, semiopen complex 3 produces a
negligible amount of product after 24 h, and in fact shows no
Figure 5. (A) Friedel−Crafts reaction between indole and nitro-
styrene. (B) Fully closed 4 and ester-free semiopen model 6 show rate
acceleration. Semiopen 3 shows no activity relative to catalyst free
control. Reaction progress was monitored by H NMR spectroscopy.
Conditions were the following: 1.5:1 indole:nitrostyrene (0.03 M:
1
0.02 M), 60 °C, 1,2-dichlorobenzene-d4, 20 mol % complex.
activity relative to the catalyst-free control (TOF = 1.74 × 10−6
s−1). Comparatively, fully closed 4 shows significant rate
acceleration and catalytic turnover, with a turnover frequency of
3.47 × 10−5 s−1, approximately 21.5 times that of semiopen 3.
Model semiopen complex 6 shows very similar rate acceleration
to fully closed 4. These results confirm that by regulating
oligomerization and substrate recognition via control of
competitive hydrogen-bonding events, we can control the
catalytic activity of the diphenylene squaramide moiety.
With the significant difference in activity between the two
configurations, we investigated the ability to allosterically
regulate catalytic activity in situ in the presence of the
substrates. Therefore, using the same conditions described
above, fully closed complex 4 was allowed to catalyze the F-C
reaction for 2 h, after which 2 equiv of N(Bu)4Cl were added to
the reaction solution. Upon the formation of the semiopen
complex 3, catalytic activity became negligible (Figure 6). After
2 h, during which no product was formed, fully closed 4 was re-
formed in situ via the addition of 2 equiv of AgBArF.39 After a
short dative period, the F-C reaction is re-initiated (Figure 6).
Therefore, by reversibly controlling the coordination mode of
16598
dx.doi.org/10.1021/ja508804n | J. Am. Chem. Soc. 2014, 136, 16594−16601

Journal of the American Chemical Society
Article
amplification. Moving forward we plan to incorporate
enantiomeric Lewis base cocataysts into the WLA structure
to expand the catalogue of applicable and controllable reactions
catalyzed by HBD catalysts.
EXPERIMENTAL SECTION
■
General Methods. Phosphino−thioether ligands 1, 2, and 5, and
precursors were prepared and stored using standard Schlenk line
techniques under an inert nitrogen atomophere, unless noted
otherwise. Ligand 5 was prepared as previously reported.40 Ligand
precursor 1-chloro-2-(diphenylphosphino)ethane was prepared as
previously reported.19 The synthesis of Pt(II) complexes, 3, 4, 6,
and 7, and their manipulations and characterization, were performed
under ambient conditions. All solvents used for ligand and complex
synthesis were anhydrous grade, purchased from Sigma-Aldrich.
Deuterated solvents were purchased from Cambridge Isotope
Laboratories and used as received. All other chemicals were purchased
1
from Aldrich Chemical Co. and used as received. H NMR spectra
were recorded on a Bruker Avance III 400 MHz spectrometer. H
1
Figure 6. In situ addition of 2 equiv of N(Bu)4Cl after 2 h to a
solution of fully closed 4, catalyzing the Friedel−Crafts reaction
between indole and nitrostyrene, shows immediate turn-off of catalysis,
due to formation of semiopen 3. In situ re-formation of 4 and re-
initiation of activity achieved by the addition of 2 equiv of AgBArF.
NMR spectra were referenced internally to residual protons in the
deuterated solvents (dichloromethane-d2 = δ 5.32 ppm, dimethylsulf-
oxide-d6 = δ 2.50 ppm). 31P{1H} NMR spectra were referenced to an
external 85% H3PO4 standard (δ 0 ppm). 19F{1H} NMR spectra were
referenced to an external CFCl3 standard (δ 0 ppm). Electrospray
ionization (ESI) mass spectra were recorded on a Bruker AmaZon SL
LC−MS instrument in positive ion mode.
the structural center, the ability of the diphenylene squaramide
site to catalyze the targeted reaction is allosterically regulated.
We hypothesize that the observed reduction in rate upon re-
initiating catalysis is due to the depletion of the catalytic
feedstock. This is the first example of allosteric regulation of
catalytic activity using a WLA construct under benchtop
conditions. Indeed, by controlling the effective quaternary and
tertiary structure the HBD squaramide-containing triple-decker
complex, we have developed a novel approach to reversible
regulation of catalytic activity in situ.
Synthesis. 4-(2-Diphenylphosphanylethylthio)phenylamine. 1-
Chloro-2-(diphenylphosphino)ethane (3.00 g, 12.06 mmol) and 4-
aminothiophenol (1.50 g, 12.06 mmol) were dissolved in degassed
anhydrous acetonitrile (15 mL) in a oven-dried Schlenk flask. With
stirring, cesium carbonate (3.93 g, 12.06 mmol) was added, and the
reaction mixture was heated to reflux for 12 h under nitrogen. After
cooling to room temperature, the mixture was filtered through a glass
frit, and the retentate was washed with acetonitrile. The solvent was
removed from the filtrate under reduced pressure. Silica-gel column
chromatography (dichloromethane) afforded the product as an off-
white solid (2.84 g, 70% yield). 1H NMR (400.16 MHz, 25 °C,
CD2Cl2): δ 7.41−7.35 (m, 10H), 7.22 (d, JH−H = 8 Hz, 2H), 6.72 (d,
JH−H = 8 Hz, 2H), 4.82 (br s, 2H), 2.85 (m, 2H), 2.31 (m, 2H).
31P{1H} NMR (161.94 MHz, 25 °C, CD2Cl2): δ −17.54 (s, 1P). ESI-
CONCLUSION
■
In using intermolecular hydrogen-bonding to regulate the
catalytic activity of a supramolecular construct, we have
developed a novel synthetic analogue of a key enzymatic
allosteric mechanism. Using the weak-link approach, we were
able to design and synthesize a supramolecular assembly with
predictable structural geometry, giving precise control of
hydrogen-bonding interactions. The hemilabile nature of the
phosphino−thioether ligands used to assemble the triple-decker
construct allows for the reversible in situ switching between a
flexible, oligomerized semiopen state and a rigid, monomeric
fully closed state. By doing so, we were able to mimic Nature’s
ability to reversible affect the quaternary structure of
metabolically important enzymes. Additionally, these changes
in quaternary structure coincide with changes in the tertiary
structure of the squaramide active site, manifested as a drastic
difference in catalytic activity. In contrast with previous
allosterically regulated WLA complexes, the complex herein is
catalytically dormant in the semiopen state, and active in the
fully closed state. This reversal of the structure−function
relationship opens doors to utilizing multiple WLA complexes
simultaneously in situ for small-molecule-regulated cascade-like
reaction pathways. The ability to precisely and simultaneously
control the activity of multiple orthogonal catalysts with simple
small-molecule effectors in situ would be unprecedented.
Additionally, this work sets the stage for the allosteric
regulation of functional WLA hydrogen-bonding-driven poly-
meric materials. Such stimuli responsive polymeric materials
could have significant applications in signal detection and
MS (m/z): 337 [M]+. Found 337.
3,4-((2-Diphenylphosphanylethylthio)phenylamino)cyclobut-3-
ene-1,2-dione (1). To an anhydrous mixture of toluene:dimethyl-
formamide (19:1, 7 mL) were added 3,4-dimethoxycyclobut-3-ene-1,2-
dione (dimethyl squarate) (200 mg, 1.41 mmol) and zinc trifluoro-
methanesulfonate (102.6 mg, 0.28 mmol). The suspension was stirred
vigorously for 10 min at room temperature. To the mixture was added
4-(2-diphenylphosphanylethylthio)phenylamine at once (1.00 g, 2.96
mmol). The mixture was heated to 100 °C and stirred overnight. The
reaction was cooled to room temperature and filtered. The retentate
was washed with toluene, methanol, and hexanes. The solid was dried
to give a yellow solid (775 mg, 73% yield). 1H NMR (400.16 MHz, 25
°C, DMSO-d6): δ 9.90 (br s, 2H), 7.42 (d, JH−H = 8 Hz, 4H), 7.38−
7.37 (m, 10H), 7.26 (d, JH−H = 8 Hz, 4H), 2.92 (m, 4H), 2.33 (m,
4H). 31P{1H} NMR (161.94 MHz, 25 °C, DMSO-d6): δ −17.51 (s,
2P).
(2-Diphenylphosphanylethylthio)triisopropylsilane (P,S-TIPS). To
a dry, degassed 500 mL Schlenck flask was added 150 mL of dry THF,
and then 100 mL of 0.5 M KPPh2 (50 mmol) was cannulated into the
flask. The solution was cooled in an ice bath. Ethylene sulfide (3.27
mL, 55 mmol) was added dropwise and stirred for 10 min at 0 °C.
Triisopropylsilyl chloride (10.7 mL, 50 mmol) was added slowly
dropwise. The reaction was stirred for 2 h. (Additional dry THF was
added if reaction congealed.) The solvent was removed under reduced
pressure, giving a gray oil. The crude mixture was extracted into the
organic phase using a CH2Cl2:water mixture (3×). The organic phase
was dried with magnesium sulfate, and then the drying agent was
removed by filtering through a fritted funnel. CH2Cl2 was removed
under reduced pressure. The crude oil was purified using silica-gel
16599
dx.doi.org/10.1021/ja508804n | J. Am. Chem. Soc. 2014, 136, 16594−16601

Journal of the American Chemical Society
Article
chromatography (2:1 hexanes:CH2Cl2), affording a colorless, crude oil
under reduced pressure, giving a yellow oil. The oil was dried under
high vacuum to give yellow solid 4 (134 mg, 94% yield). H NMR
1
1
(7.02 g, 34% yield). H NMR (400.16 MHz, 25 °C, CD2Cl2): 7.47−
7.38 (m, 10H), 2.59 (m, 2H), 2.39 (m, 2H), 1.16 (m, 3H), 1.07 (d,
JH−H = 8 Hz, 18H). 31P{1H} NMR (161.94 MHz, 25 °C, CD2Cl2): δ
−16.08 (s, 1P). ESI-MS (m/z): 402.65 [M]+. Found 403.65.
(400.16 MHz, 25 °C, CD2Cl2): δ 7.75 (s, 32 H), 7.66 (br m, 10H),
7.57 (s, 16H), 7.50 (br m, 20H), 7.35 (br m, 20H), 3.95 (s, 6H),
3.15−2.90 (br m, 16H). 31P{1H} NMR (161.98 MHz, 25 °C, DMSO-
d6): 45.79 (br s, JP−Pt = 3244 Hz, 2P), 44.57 (br s, JP−Pt = 3087 Hz,
2P). ESI-MS (m/z): 682.61 [M − 4BArF−]3+. Found 681.98
Methyl (2-Diphenylphosphanylethylthio)-2,3,5,6-tetrafluoro-
benzoate (2). To minimal dry, degassed DMF in a Schlenck flask
were added P,S-TIPS (1.00 g, 2.48 mmol) and methyl pentafluoro-
benzoate (0.36 mL, 2.48 mmol). The solution was stirred for 10 min.
Cesium fluoride (376 mg, 2.48 mmol) was added, and the reaction was
stirred overnight at room temperature. The solvent was removed
under reduced pressure. The crude was extracted with CH2Cl2:water
(3 × 50 mL CH2Cl2). The organic phase was dried with magnesium
sulfate and then filtered through a fritted funnel. CH2Cl2 was removed
under reduced pressure giving a yellow oil, which upon drying on the
high-vacuum line yielded 2 as a pale yellow solid (1.07 g, 95% yield). If
need be, 2 can be purified by silica-gel chromatography (1:1 hexanes:
Semiopen Complex (BArF)2 (6). A solution of PtCl2(cod) (150 mg,
0.40 mmol) in CH2Cl2 (20 mL) was added dropwise to a CH2Cl2
slurry of 1 (150.9 mg, 0.20 mmol). To this mixture was added,
dropwise, a solution of 5 (157.7 mg, 0.40 mmol) in CH2Cl2 (5 mL).
The resulting solution was stirred for 15 min. MeOH (15 mL) was
added, and the CH2Cl2 was removed under reduced pressure. An
additional 150 mL of MeOH was added, and the solution was stirred
overnight. The MeOH was reduced under reduced pressure to 1/10th
the volume. To the MeOH solution was added a methanolic solution
(5 mL) of excess NaBF4 (65.8 mg, 0.60 mmol). The solution was
stirred overnight and then filtered over a fritted funnel. The retentate
was washed with minimal cold MeOH and then hexanes, yielding a
yellow solid (6(BF4)2). The yellow solid was dried under vacuum
overnight. The resulting yellow solid was added to a solution of 2
equiv of NaBArF in CH2Cl2 (10 mL) and then stirred overnight. The
resulting slurry was filtered over Celite on a fritted funnel and then
washed with dry CH2Cl2. The filtrate was collected, and the CH2Cl2
was removed under reduced pressure to yield 6 (564 mg, 76% yield).
1H NMR (400.16 MHz, 25 °C, CD2Cl2): δ 8.08 (m, 2H), 7.75 (s, 16
H), 7.64 (br m, 8H), 7.58 (s, 8H), 7.49 (br m, 20H), 7.28 (br m,
20H), 3.00−2.20 (br m, 16H). 31P{1H} NMR (161.98 MHz, 25 °C,
DMSO-d6): 43.72 (d, JP−P = 15 Hz, JP−Pt = 3540 Hz, 2P), 8.72 (d, JP−P
= 15 Hz, JP−Pt = 3236 Hz, 2P). ESI-MS (m/z): 1017.57 [M + Cl− −
2BArF−]1+. Found 1018.97
Fully Closed Complex (BArF)4 (7). 7 was synthesized via two
methods: (a) abstraction of 6(BF4)2 in nitromethane with AgBF4,
followed by a counteranion exchange with NaBArF in CH2Cl2, and (b)
abstraction of 6 in CH2Cl2 with an ethereal solution of AgBArF.
Method (a). To a solution of 6(BF4)2 (216 mg, 0.1 mmol) in
nitromethane (10 mL) was added 2 equiv of AgBF4 (38.8 mg, 0.2
mmol). The solution was stirred overnight occluded from light. The
resulting slurry was filtered through Celite on a fritted funnel. The
Celite was washed with dry CH2Cl2. The filtrate was collected, and
then the CH2Cl2 was minimized under reduced pressure. Hexanes (20
mL) was added, and the suspension was stored at −20 °C for 1 h. The
suspension was filtered over a fritted funnel and then washed with
hexanes. The yellow retentate was dried overnight. The resulting
yellow solid was added to a solution of 4 equiv of NaBArF in CH2Cl2
and then stirred overnight. The resulting suspension was filtered over
Celite on a fritted funnel, and the Celite was washed with dry CH2Cl2.
The filtrate was collected, and the CH2Cl2 was removed under reduced
pressure, giving a yellow oil. The oil was dried under high vacuum to
give yellow solid 7 (461 mg, 86% yield).
1
CH2Cl2). H NMR (400.16 MHz, 25 °C, CD2Cl2): δ 7.44−7.37 (m,
10H), 4.00 (s, 3H), 3.09 (m, 2H), 2.37 (m, 2H). 31P{1H} NMR
(161.94 MHz, 25 °C, CD2Cl2): δ −17.35 (s, 1P). 19F {1H} NMR
(376.49 MHz, 25 °C, CD2Cl2): δ −133.73 (m, 2F), −139.86 (m, 2F).
ESI-MS (m/z): 452.06 [M]+. Found 453.06.
Semiopen Complex (BArF)2 (3). A solution of PtCl2(cod) (150 mg,
0.40 mmol) in CH2Cl2 (20 mL) was added dropwise to a CH2Cl2
slurry of 1 (150.9 mg, 0.20 mmol). To this mixture was added,
dropwise, a solution of 2 (181.4 mg, 0.40 mmol) in CH2Cl2 (5 mL).
The resulting solution was stirred for 15 min. MeOH (15 mL) was
added, and the CH2Cl2 was removed under reduced pressure. An
additional 150 mL of MeOH was added, and the solution was stirred
overnight. The MeOH was reduced under reduced pressure to 1/10th
the volume. To the MeOH solution was added a methanolic solution
(5 mL) of excess NaBF4 (65.8 mg, 0.60 mmol). The solution was
stirred overnight and then filtered over a fritted funnel. The retentate
was washed with minimal cold MeOH and then hexanes, yielding a
yellow solid (3(BF4)2). The yellow solid was dried under vacuum
overnight. The resulting yellow solid was added to a solution of 2
equiv of NaBArF in CH2Cl2 (10 mL) and then stirred overnight. The
resulting slurry was filtered over Celite on a fritted funnel and then
washed with dry CH2Cl2. The filtrate was collected, and the CH2Cl2
was removed under reduced pressure to yield 3 (550 mg, 72% yield).
1H NMR (400.16 MHz, 25 °C, CD2Cl2): δ 7.71 (s, 16 H), 7.63 (br m,
10H), 7.54 (s, 8H), 7.46 (br m, 20H), 7.26 (br m, 20H), 3.97 (s, 6H),
3.22−2.70 (br m, 16H). 31P{1H} NMR (161.98 MHz, 25 °C, DMSO-
d6): δ 43.44 (d, JP−P = 14 Hz, JP−Pt = 3466 Hz, 2P), 8.39 (d, JP−P = 14
Hz, JP−Pt = 3231 Hz, 2P). ESI-MS (m/z): 1058.09 [M − 2BArF−]2+.
Found 1059.49
Fully Closed Complex (BArF)4 (4). 4 was synthesized via two
methods: (a) abstraction of 3(BF4)2 in nitromethane with AgBF4,
followed by a counteranion exchange with NaBArF in CH2Cl2, and (b)
abstraction of 3 in CH2Cl2 with an ethereal solution of AgBArF.
Method (a). To a solution of 3(BF4)2 (250 mg, 0.11 mmol) in
nitromethane (10 mL) was added 2 equiv of AgBF4 (42.8 mg, 0.22
mmol). The solution was stirred overnight occluded from light. The
resulting slurry was filtered through Celite on a fritted funnel. The
Celite was washed with dry CH2Cl2. The filtrate was collected, and
then the CH2Cl2 was minimized under reduced pressure. Hexanes (20
mL) was added, and the suspension was stored at −20 °C for 1 h. The
suspension was filtered over a fritted funnel , and then washed with
hexanes. The yellow retentate was dried overnight. The resulting
yellow solid was added to a solution of 4 equiv of NaBArF in CH2Cl2
and then stirred overnight. The resulting suspension was filtered over
Celite on a fritted funnel, and the Celite was washed with dry CH2Cl2.
The filtrate was collected, and the CH2Cl2 was removed under reduced
pressure, giving a yellow oil. The oil was dried under high vacuum to
give yellow solid 4 (531 mg, 88% yield).
Method (b). To a CH2Cl2 solution of 6 (100 mg, 0.027 mmol) was
added 2 equiv of a 0.1 M ethereal solution of AgBArF.39 The solution
was stirred overnight occluded from light. The resulting suspension
was filtered over Celite on a fritted funnel. The Celite was washed with
dry CH2Cl2. The filtrate was collected, and the CH2Cl2 was removed
under reduced pressure, giving a yellow oil. The oil was dried under
1
high vacuum to give yellow solid 7 (132 mg, 92% yield). H NMR
(400.16 MHz, 25 °C, CD2Cl2): δ 8.08 (m, 2H) 7.75 (s, 32 H), 7.66
(br m, 8H), 7.57 (s, 16H), 7.51 (br m, 20H), 7.32 (br m, 20H), 3.92−
2.13 (br m, 16H). 31P{1H} NMR (161.98 MHz, 25 °C, DMSO-d6):
45.04 (d, JP−P = 13 Hz, JP−Pt = 3250 Hz, 2P), 44.63 (d, JP−P = 13 Hz,
JP−Pt = 3113 Hz, 2P).
ASSOCIATED CONTENT
* Supporting Information
■
S
Method (b). To a CH2Cl2 solution of 3 (100 mg, 0.026 mmol) was
added 2 equiv of a 0.1 M ethereal solution of AgBArF.39 The solution
was stirred overnight occluded from light. The resulting suspension
was filtered over Celite on a fritted funnel. The Celite was washed with
dry CH2Cl2. The filtrate was collected, and the CH2Cl2 was removed
Details of computational geometry optimization, FT-IR study
of self-association, 1H DOSY study of oligomerization, 1H
NMR spectroscopy study of substrate binding and catalytic
1
study, 31P{1H} and H NMR spectra of 1, 3, 4, 6 and 7, and
16600
dx.doi.org/10.1021/ja508804n | J. Am. Chem. Soc. 2014, 136, 16594−16601

Journal of the American Chemical Society
Article
31P{1H},1H and 19F {1H} NMR spectra of 2 and 5. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(27) Lunelli, B.; Giorgini, M. G. Spectrochim. Acta A−M 1988, 44,
1153.
(28) Storer, R. I.; Aciro, C.; Jones, L. H. Chem. Soc. Rev. 2011, 40,
2330.
(29) Wade, L. G., Jr. Organic Chemistry, 7th ed.; Prentice Hall: Upper
Saddle River, NJ, 2010.
(30) Macchioni, A.; Ciancaleoni, G.; Zuccaccia, C.; Zuccaccia, D.
Chem. Soc. Rev. 2008, 37, 479.
(31) Cohen, Y.; Avram, L.; Frish, L. Angew. Chem., Int. Ed. 2005, 44,
520.
(32) Kapur, G. S.; Cabrita, E. J.; Berger, S. Tetrahedron Lett. 2000, 41,
AUTHOR INFORMATION
■
Corresponding Author
chadnano@northwestern.edu
Notes
The authors declare no competing financial interest.
7181.
(33) Herrera, R. P.; Sgarzani, V.; Bernardi, L.; Ricci, A. Angew. Chem.,
Int. Ed. 2005, 44, 6576.
(34) Ganesh, M.; Seidel, D. J. Am. Chem. Soc. 2008, 130, 16464.
(35) da Silva, R. C.; da Silva, G. P.; Sangi, D. P.; Pontes, J. G. d. M.;
ACKNOWLEDGMENTS
■
This material is based upon work supported by awards from the
National Science Foundation (CHE-1149314) and U.S. Army
(W911NF-11-1-0229). J.M.-A. acknowledges a fellowship from
Consejo Nacional de Ciencia y Tecnologia (CONACYT).
Ferreira, A. G.; Correa
9007.
̂
, A. G.; Paixao, M. W. Tetrahedron 2013, 69,
̃
(36) Dessole, G.; Herrera, R. P.; Ricci, A. Synlett 2004, 2004, 2374.
(37) Loh, C. C. J.; Atodiresei, I.; Enders, D. Chem.Eur. J. 2013, 19,
REFERENCES
■
10822.
(1) Traut, T. W. Crit. Rev. Biochem. Mol. Biol. 1994, 29, 125.
(2) Barford, D.; Johnson, L. Nature 1989, 340, 609.
(3) Ali, M. H.; Imperiali, B. Bioorg. Med. Chem. 2005, 13, 5013.
(4) Goodey, N. M.; Benkovic, S. J. Nat. Chem. Biol. 2008, 4, 474.
(5) Gianneschi, N. C.; Masar, M. S.; Mirkin, C. A. Acc. Chem. Res.
2005, 38, 825.
(38) Lancianesi, S.; Palmieri, A.; Petrini, M. Chem. Rev. 2014, 114,
7108.
(39) Hayashi, Y.; Rohde, J. J.; Corey, E. J. J. Am. Chem. Soc. 1996,
118, 5502.
(40) Brown, A. M.; Ovchinnikov, M. V.; Mirkin, C. A. Angew. Chem.,
Int. Ed. 2005, 44, 4207.
(6) Oliveri, C. G.; Ulmann, P. A.; Wiester, M. J.; Mirkin, C. A. Acc.
Chem. Res. 2008, 41, 1618.
(7) Molenveld, P.; Engbersen, J. F. J.; Reinhoudt, D. N. Chem. Soc.
Rev. 2000, 29, 75.
(8) Prins, L. J.; Huskens, J.; de Jong, F.; Timmerman, P.; Reinhoudt,
D. N. Nature 1999, 398, 498.
(9) Lin, W. Supramolecular Catalysis; Wiley-VCH Verlag GmbH &
Co. KGaA: Weinheim, 2008; p 93.
(10) Lee, S. J.; Lin, W. Acc. Chem. Res. 2008, 41, 521.
(11) Fujita, M.; Tominaga, M.; Hori, A.; Therrien, B. Acc. Chem. Res.
2005, 38, 369.
(12) Ikemoto, K.; Inokuma, Y.; Fujita, M. J. Am. Chem. Soc. 2011,
133, 16806.
(13) Caulder, D. L.; Raymond, K. N. Acc. Chem. Res. 1999, 32, 975.
(14) Das, S.; Incarvito, C. D.; Crabtree, R. H.; Brudvig, G. W. Science
2006, 312, 1941.
(15) Sauvage, J. P.; Collin, J. P.; Chambron, J. C.; Guillerez, S.;
Coudret, C.; Balzani, V.; Barigelletti, F.; De Cola, L.; Flamigni, L.
Chem. Rev. 1994, 94, 993.
(16) Farrell, J. R.; Mirkin, C. A.; Guzei, I. A.; Liable-Sands, L. M.;
Rheingold, A. L. Angew. Chem., Int. Ed. 1998, 37, 465.
(17) Gianneschi, N. C.; Bertin, P. A.; Nguyen, S. T.; Mirkin, C. A.;
Zakharov, L. N.; Rheingold, A. L. J. Am. Chem. Soc. 2003, 125, 10508.
(18) Holliday, B. J.; Mirkin, C. A. Angew. Chem., Int. Ed. 2001, 40,
2022.
(19) Yoon, H. J.; Kuwabara, J.; Kim, J.-H.; Mirkin, C. A. Science 2010,
330, 66.
(20) Kennedy, R. D.; Machan, C. W.; McGuirk, C. M.; Rosen, M. S.;
Stern, C. L.; Sarjeant, A. A.; Mirkin, C. A. Inorg. Chem. 2013, 52, 5876.
(21) Malerich, J. P.; Hagihara, K.; Rawal, V. H. J. Am. Chem. Soc.
2008, 130, 14416.
(22) Aleman
2011, 17, 6890.
́
, J.; Parra, A.; Jiang, H.; Jørgensen, K. A. Chem.Eur. J.
(23) Park, J. S.; Lifschitz, A. M.; Young, R. M.; Mendez-Arroyo, J.;
Wasielewski, M. R.; Stern, C. L.; Mirkin, C. A. J. Am. Chem. Soc. 2013,
135, 16988.
(24) Lifschitz, A. M.; Young, R. M.; Mendez-Arroyo, J.;
Roznyatovskiy, V. V.; McGuirk, C. M.; Wasielewski, M. R.; Mirkin,
C. A. Chem. Commun. 2014, 50, 6850.
(25) McGuirk, C. M.; Stern, C. L.; Mirkin, C. A. J. Am. Chem. Soc.
2014, 136, 4689.
(26) Silva, C. E.; Dos Santos, H. l. F.; Speziali, N. L.; Diniz, R.; de
Oliveira, L. F. C. J. Phys. Chem. A 2010, 114, 10097.
16601
dx.doi.org/10.1021/ja508804n | J. Am. Chem. Soc. 2014, 136, 16594−16601