Arene activation at iridium facilitates C-O bond cleavage of aryl ethers

Article abstract of DOI:10.1021/om5000166

An arene activation strategy for the selective disassembly of aryl ethers is reported. A variety of aryl ethers readily bind an electrophilic pentamethylcyclopentadienyl iridium center by η⁶-arene coordination, generating complexes that are activated toward hydrolysis and cleavage of the Ar-OR bond (R = Me, Et, Ph). Hydrolysis occurs rapidly at room temperature in aqueous pH 7 phosphate buffer (or upon modest heating under acidic conditions), releasing alcohol while forming cyclohexadienyl-one products. Under strongly acidic conditions, protonation of the dienyl-one followed by substitution with starting aryl ether completes a hydrolysis cycle. Mechanistic studies suggest that the key hydrolysis step proceeds via nucleophilic attack at the ipso position of the arene (S_NAr mechanism). The observed mechanism is considered in the context of lignocellulosic biomass conversion.

Full text of DOI:10.1021/om5000166

Article
pubs.acs.org/Organometallics
Arene Activation at Iridium Facilitates C−O Bond Cleavage of Aryl
Ethers
Alexander J. M. Miller,*,† Werner Kaminsky,‡ and Karen I. Goldberg*,‡
†Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290, United States
‡Department of Chemistry, University of Washington, Seattle, Washington 98195-1700, United States
S
* Supporting Information
ABSTRACT: An arene activation strategy for the selective
disassembly of aryl ethers is reported. A variety of aryl ethers
readily bind an electrophilic pentamethylcyclopentadienyl iridium
center by η6-arene coordination, generating complexes that are
activated toward hydrolysis and cleavage of the Ar−OR bond (R =
Me, Et, Ph). Hydrolysis occurs rapidly at room temperature in
aqueous pH 7 phosphate buffer (or upon modest heating under
acidic conditions), releasing alcohol while forming cyclohex-
adienyl-one products. Under strongly acidic conditions, protonation of the dienyl-one followed by substitution with starting aryl
ether completes a hydrolysis cycle. Mechanistic studies suggest that the key hydrolysis step proceeds via nucleophilic attack at the
ipso position of the arene (SNAr mechanism). The observed mechanism is considered in the context of lignocellulosic biomass
conversion.
We report herein the synthesis and aqueous reactivity of a
series of aryl ether complexes of iridium. Facile C−O bond
scission occurs for both alkyl aryl ethers and diaryl ethers upon
dissolution in buffered neutral water. Hydrolysis proceeds by a
nucleophilic aromatic substitution mechanism. The other steps
of the proposed redox-neutral C−O cleavage cycle shown in
Scheme 1 (protonation and arene exchange) are also observed.
INTRODUCTION
■
With the petrochemical landscape evolving, interest is growing
in alternative feedstocks for chemicals. Lignocellulosic biomass,
comprised of cellulose and lignin biopolymers, has received
attention as a feedstock that could potentially provide
petroleum-free routes to commodity chemicals.1 Cellulosic
biomass has been at the center of conversion studies, and the
chemistry and biology required to realize “biorefinery”
technologies have made great strides recently.2−5 Lignin, on
the other hand, has remained relatively underexplored.6 An
irregular biopolymer constructed primarily of aryl alkyl ether
linkages, lignin could in principle act as a starting material for
the production of aromatics. However, the insoluble polymer
must first be broken down into smaller pieces and then
selectively functionalized in order to realize efficient aromatic
synthesis. In nature, termites and fungi use peroxidase enzymes
to dissolve wood, but the radical mechanism breaks C−C and
C−O bonds indiscriminately.7
Scheme 1
Interest in utilizing discrete transition-metal catalysts to
achieve the valorization of lignin has been rapidly growing.8−15
Lignin conversion faces a variety of challenges, from insolubility
in common solvents to the need to handle a myriad of different
functional groups. Our strategy relies on homogeneous catalysis
in the hope that these species will be better able to intercalate
into mostly insoluble polymers in aqueous solvents. A single-
site, nonradical catalyst might also be anticipated to afford
higher selectivity in lignin conversion. With these strategies in
mind, we have explored an arene activation16−20 approach to
selective cleavage of the C−O bonds of aryl ethers. Aqueous
conditions were targeted for the benefits of low cost, low
toxicity, and pH control that could be amenable to lignin
processing.21
Received: January 12, 2014
Published: February 20, 2014
© 2014 American Chemical Society
1245
dx.doi.org/10.1021/om5000166 | Organometallics 2014, 33, 1245−1252

Organometallics
Article
substituted arene positions exhibiting Ir−C distances elongated
by ∼0.1 Å. The iridium arene complexes were soluble in
acetone but insoluble in THF and CH2Cl2. The arene
complexes were also soluble in MeCN, but arene displacement
by the solvent occurs to generate [Cp*Ir(MeCN)3]2+ and free
RESULTS AND DISCUSSION
■
Synthesis of Model Complexes. The desired arene
complexes were synthesized analogously to previously reported
Ir arene complexes.22 The one-pot procedure, outlined in
Scheme 2, involves initial halide abstraction from the chloride
1
arene (as observed by H NMR spectroscopy, over the course
of a few hours in the case of 1). Complexes displaying higher
Scheme 2
solubility in organic solvents were accessed by salt exchange.
Treatment of 1 with 2 equiv of NaBArF (ArF = 3,5-
4
bis(trifluoromethyl)phenyl) afforded [Cp*Ir(veratrole)]-
[BArF ] (1-BArF ), which is soluble in both THF and CH2Cl2.
4 2
4
1
The arene moieties of complexes 1−5 exhibit H NMR
resonances similar to those of the unbound arenes, confirming
that the η6 coordination mode is maintained in solution.
Despite the subtle difference in chemical shift upon complex-
ation, previous studies suggest that arenes bound to Ir(III) are
activated toward nucleophiles.23,24 Of interest is a related arene-
activation approach to lignin degradation, where Rauchfuss
found that Ru π complexes of lignin precursors such as
coniferyl alcohol showed significant upfield shifts upon
complexation; in that case, nucleophile addition did not result
in attack at the aromatic ring but rather deprotonation of the
phenol.25
dimer [Cp*Ir(Cl)2]2 by AgBF4 to generate the solvento species
[Cp*Ir(acetone)3][BF4]2 in situ. Treatment of the acetone
complex with various arenes produced η6-coordinated arene
complexes in high purity.
Veratrole (1,2-dimethoxybenzene) was chosen for initial
studies because o-alkoxyarene substitution is a prevalent
structural element of lignin.6 Coordination of veratrole, along
with anisole (methoxybenzene), phenetole (ethoxybenzene),
diphenyl ether, and 3-phenoxyanisole, afforded a series of
complexes [Cp*Ir(arene)]2+ (1−5; Chart 1). The iridium arene
Aqueous Reactivity of Arene Complexes. Complexes
1−5 all displayed good solubility in water. Upon dissolution of
complex 1 in neutral D2O (∼30 mM 1), the pD dropped to 3.2,
and formation of a small amount (<5%) of a single new
Chart 1
1
decomposition species was evident by H NMR spectroscopy.
After this initial reaction, slow partial displacement of veratrole
from complex 1 occurred, establishing an equilibrium between
[Cp*Ir(arene)]2+ and [Cp*Ir(H2O)3]2+ (6; Scheme 3). The
Scheme 3
complexes were fully characterized by 1H and 13C NMR
spectroscopy and elemental analysis. Single crystals of the
veratrole complex grew upon vapor diffusion of diethyl ether
into a nitromethane solution of 1; an X-ray diffraction (XRD)
study confirmed the expected η6 structure (Figure 1). The
arene ligand binds in a slightly asymmetric fashion, with the
equilibrium constant for the substitution reaction of Scheme 3
heavily favors the veratrole complex: Keq = 0.0049(1) M (eq
1).26 Even in aqueous solutions, complex 1 is the major species
Keq = [6][veratrole]/[1]
(1)
at equilibrium. A similar equilibrium constant was obtained
upon treatment of an aqueous solution of aquo complex 6 with
1 equiv of veratrole, where slow partial conversion to complex 1
was observed. Equilibrium arene binding suggests that arene
exchange should be viable under mild reaction conditions, with
the binding constant dictating that a significant portion of the Ir
will bind the arene, as needed for subsequent C−O bond
cleavage reactions.
Reactivity in Buffered Water. Given the rapid fall in pD
when 1 was dissolved in water and the slower equilibrium
formation of [Cp*Ir(H2O)3]2+, it seemed likely that the pD
change upon dissolution of 1 in D2O was related to the small
amount of new species that was generated. To investigate the
observed pD change, veratrole complex 1 was dissolved in D2O
buffered to pD 7.1 with 0.1 M phosphate. After only 15 min the
Figure 1. Structural representation of 1 with ellipsoids at 50%
probability. Hydrogen atoms, solvent molecules, and counteranions
are omitted for clarity. Bond lengths (Å): C11−O1 1.33(2), C12−O2
1.34(2), Ir−C11 2.34(2), Ir−C12 2.35(2), Ir−C13 2.23(2), Ir−C14
2.25(1), Ir−C15 2.22(1), Ir−C16 2.24(1), Ir−CCp* all between
2.17(1) and 2.19(1).
1246
dx.doi.org/10.1021/om5000166 | Organometallics 2014, 33, 1245−1252

Organometallics
Article
characteristic resonances of 1 were absent from the NMR
spectrum, replaced by resonances corresponding to the minor
species observed in unbuffered D2O. A single Cp* resonance
was observed, along with one methoxy group integrating to
three protons. The aromatic resonances had shifted significantly
upfield. Spectroscopic identification of the product as cyclo-
hexadienyl-one complex 7 (Scheme 4) was supported by
The methanol-releasing C−O hydrolysis reaction is notable
for the exceedingly mild conditions under which it occurs:
within 15 min in neutral, phosphate-buffered water at ambient
temperature in air. The effect of metal complexation is striking
in comparison to the commonly employed forcing conditions
for metal-free C−O cleavage of aryl ethers: for example, in
organic deprotections or lignin degradation.6
Two mechanisms were envisioned for C−O cleavage
(Scheme 4): SN2 (dashed red) and SNAr (solid blue). Base-
promoted C−O bond cleavage of a related aryl methyl ether
complex of Ir was reported to proceed through a SN2
mechanism.28 An SN2 mechanism would be ill-suited for lignin
degradation, however, due to the presence of branched,
substituted alkyl ether groups. To our delight, however, we
observed that all of the compounds in Chart 1 reacted
completely within 15 minincluding the diphenyl ether
complex, a substrate unable to participate in SN2 chemistry.
The reaction mechanism was further probed by monitoring
Scheme 4
1
comparison to similar complexes.27 Methanol was also present
in the mixture, confirming hydrolysis. Hydrolysis of complex 1
in pH 7 phosphate buffer, followed by anion exchange using
the hydrolysis of complexes 2−4 by H NMR spectroscopy in
0.1 M KH2PO4/K2HPO4 pD 7.7 at 10 °C.29 A small
1
concentration of an intermediate species was observed by H
NaBArF in a H2O/CH2Cl2 mixture, resulted in phase transfer
NMR spectroscopy over the course of each hydrolysis reaction.
It is notable that the spectral features of the intermediates in the
hydrolysis of the methoxybenzene complex 2 and the
ethoxybenzene complex 3 were distincta different inter-
mediate species was present in each hydrolysis reactionbut
4
of 7 into CH2Cl2 and allowed isolation of the analytically pure
cyclohexadienyl-one product 7-BArF . An XRD study on 7-
4
BArF confirmed the expected structure (Figure 2). The bond
4
1
featured similar H chemical shifts. Each intermediate had the
same number of protons as the starting material and displayed
the diagnostic resonances for the key functional groups (Cp*
and either MeO or EtO), but notably the aromatic resonances
had shifted significantly upfield, consistent with arene
dearomatization. The salient features of the spectra, and the
similarity with the dearomatized product 8, lead us to propose
the structure of the intermediate as a metal-bound Jackson−
Meisenheimer intermediate (Scheme 5).
The hydrolysis reactions of 2−4 each proceeded to
completion in less than 1 h at 10 °C. The pD dropped slightly
(0.1−0.2 pD unit), consistent with hydroxide consumption
over the course of the reaction. Figure 3 shows the reaction
profile of the hydrolysis of complex 2 over time. The
intermediate is apparent from the first data point (30−40 s
after mixing) and decays at roughly the same rate as anisole
complex 2 (see also Figure S11 in the Supporting Information).
The equilibrium constant (k1/k−1 in Scheme 5) was calculated
using the concentration of the starting material and
Figure 2. Structural representation of 7-BArF with ellipsoids at 50%
4
probability. Hydrogen atoms and counteranion are omitted for clarity.
Bond lengths (Å): C11−O1 1.34(1), C16−O2 1.23(1), Ir−C11
2.34(1), Ir−C12 2.22(1), Ir−C13 2.21(1), Ir−C14 2.20(1), Ir−C15
2.22(1), Ir−C16 2.55(1), Ir−CCp* between 2.17(1) and 2.20(1).
1
intermediate (measured in situ by H NMR integration) and
the concentration of hydroxide (measured by pH meter before
the reaction and assumed to be constant throughout). Values of
Keq were taken from early reaction times in order to avoid error
deriving from slight changes in pD observed at the end of the
reaction (see the Supporting Information for full details). The
reaction profile led us to propose the SNAr mechanism shown
lengths are consistent with a cyclohexadienyl-one formulation:
the organic fragment features a short CO double bond
(C16−O2 = 1.23(1) Å) that is bent slightly out of the arene
plane and is asymmetrically bound to Ir (Ir−C16 is much
longer than the other Ir−C bond distances).
Scheme 5
1247
dx.doi.org/10.1021/om5000166 | Organometallics 2014, 33, 1245−1252

Organometallics
Article
constant for Me vs Et substitution was roughly a factor of 4 for
k2 (∼5 for kobs). Notably, SN2 reactions tend to slow by a factor
of ∼30 upon addition of a substituent;31 the more subtle drop
in rateand the observation of an intermediateare strong
evidence against an SN2 mechanism. Diphenyl ether complex 4
reacted too rapidly to be accurately modeled, but the rapid
reaction of 4 is also consistent with a SNAr mechanism, as
electron-withdrawing groups would be expected to accelerate
the reaction. These results are in stark contrast with those of
Amouri and co-workers, who reported evidence of an SN2
mechanism for C−O bond cleavage in another Cp*Ir-
supported aryl alkyl ether.28 The substrate in that system was
a benzo-crown ether, which may lead to alternative selectivity as
observed in transition-metal-free systems.32 It is also noted that
the other study was conducted in acetone solvent, which could
influence the mechanism.28
With an SN2 mechanism having been ruled out by the
reactivity of diphenyl ether complex 4 and the minimal steric
influence on the rate constant, we sought to strengthen our
mechanistic proposal of SNAr hydrolysis by monitoring the
reaction at pD 8.1 and 9.0. In our proposed mechanism, kobs
should be proportional to hydroxide concentration, with added
base accelerating the reaction. Indeed, the observed rate
constants increase markedly with increasing hydroxide
concentration in the hydrolysis of 3 (Table 2, Figure S23 in
Figure 3. Reaction profile of the hydrolysis of anisole complex 2 from
in situ 1H NMR spectroscopic monitoring. Concentrations of 2 (filled
circles), the intermediate (empty circles), and the product 8 (filled
triangles) are plotted over time. These data are fit (solid red lines)
using the Copasi software,30 with Scheme 5 as a model and [OD−]
fixed on the basis of the buffer pD. The sum of all species (2,
intermediate, and 8) are shown with ×. Conditions: pD 7.7 D2O (0.1
M KH2PO4/K2HPO4), 6 mM 2, 10 °C.
in Scheme 5, in which pre-equilibrium is rapid (k−1 ≫ k2). The
observed rate constants for hydrolysis of complexes 2−4 were
obtained from a plot of ln([SM] + [I]) vs time (SM is starting
material 2−4 and I is the intermediate in the reaction) and are
collected in Table 1.
a
Table 2. pH Dependence of Hydrolysis of 3
compound 3 (R = Et)
param
pD 7.7
pD 8.1
pD 9.0
b
kobs (s−1
)
0.0013
0.011
3.3 × 105
0.0040
0.016
2.9 × 105
0.0071
0.015
9.2 × 104
Table 1. Rate Constants and Equilibrium Constants for
c
k2 (s−1
Keq (M−1
)
a
Scheme 5
c
)
compound
a
Conditions: 10 °C, D2O, 0.1 M KH2PO4/K2HPO4, pD 7.7, [Ir] = 4−
b
param
2, R = Me
3, R = Et
4, R = Ph
10 mM. kobs calculated from a linear fit to a plot of ln([SM] + [I]) vs
c
b
d
kobs (s−1
)
0.0062(2)
time, where SM is 2−4 and I is the intermediate. k2 and Keq were
0.0013(4)
0.011(2)
3.3(3)x105
0.21
∼0.01
obtained by fitting the concentration profiles to the model in Scheme
c
k2 (s−1
Keq (M−1
)
0.038(7)
4.6(2)x105
d
5, using the Copasi software suite.30
c
)
d
krel (kobs
krel (k2)
)
1
1
1.6
d
0.29
a
the Supporting Information). With only one kinetic measure-
ment at each value of pD, there is more uncertainty in the data;
however, the modeled constants Keq and k2 remained roughly
unchanged as a function of pD, while kobs increased
dramatically. The collective data provide convincing support
for a nucleophilic aromatic substitution mechanism for C−O
hydrolysis.
Conditions: 10 °C, D2O, 0.1 M KH2PO4/K2HPO4, pD 7.7, [Ir] = 5−
b
10 mM (except R = Ph, 30 mM). kobs calculated from linear fit to a
plot of ln([SM] + [I]) vs time, where SM is 2−4 and I is the
c
intermediate. k2 and Keq were obtained by fitting the concentration
profiles to the model in Scheme 5, using the Copasi software suite.30
d
The reaction proceeded too quickly for accurate kinetic data to be
obtained (more than 50% complete at first data point); values are
estimated.
A competition experiment probing the relative reactivities of
methyl and phenyl ethers was carried out using the doubly
substituted arene complex [Cp*Ir(3-phenoxyanisole)][BF4]2
(5). The synthesis of complex 5 proceeded less cleanly than
the syntheses of complexes 1−4, precluding complete
characterization. We hypothesize that the two arenes in 3-
phenoxyanisole bind Ir competitively, but with a strong
preference for the more electron rich dialkoxy-substituted
arene: integration of the Cp* region is consistent with 95%
selectivity. The aromatic region of complex 5 showed the
expected triplet/triplet/doublet pattern of a monosubstituted
arene further downfield, whereas the resonances for the doubly
substituted arene were shifted slightly upfield, consistent with
metal binding. The monosubstituted arene resonances of 5 also
resonated ∼0.5 ppm downfield of the bound anisole multiplets
in 2.
The data were further modeled according to Scheme 5 (see
the Supporting Information for differential equations).30 The
concentration profile of each Ir species was fitted by varying
only k1, k−1, and k2 (assuming no change in pD). The
hydrolysis of 2 and 3 showed very good agreement with the
model (Figure 3; see the Supporting Information for full
details). The best fits showed that equilibration is rapid relative
to release of alcohol. The equilibrium constants (Keq) were
obtained from the ratio k1/k−1 (and matched the values
obtained by NMR integration); the Keq values indicate that
hydroxide attack on the arene is slightly favored for complex 2
over 3 (Table 1). The rate constants for the rate-determining
step, k2, are also shown in Table 1, with alcohol release being
significantly favored for complex 2. The difference in rate
1248
dx.doi.org/10.1021/om5000166 | Organometallics 2014, 33, 1245−1252

Organometallics
Article
The mixture isolated upon metalation of 3-phenoxyanisole
exhibited hydrolysis reactivity consistent with the structural
formulation of complex 5 in Scheme 6. Dissolution of 5 (95:5
this product as the conjugate acid of the cyclohexadienyl-one,
[Cp*Ir(guaiacol)]2+ (9).
In an attempt to synthesize the guaiacol complex by an
alternative route, guaiacol was added to [Cp*Ir(H2O)3]2+ in 1
M HBF4/D2O (pD ∼0). The resulting NMR spectrum was
consistent with proton loss and formation of complex 7.
Addition of aliquots of acid resulted in a gradual downfield shift
of the aromatic resonances, eventually coalescing into the peak
shape assigned to 9 in 2 M HBF4/D2O. The pKa of 9 is
therefore very low, close to 0. Addition of guaiacol to
[Cp*Ir(H2O)3]2+ in 2 M HBF4/D2O resulted in resonances
similar to those obtained by protonation of complex 7, lending
further evidence to the notion that the product is [Cp*Ir-
(guaiacol)]2+.
Scheme 6
That quite acidic conditions are required for protonation is
not surprising, on the basis of literature examples (in organic
solvents) showing that treatment of [Cp*Ir(acetone)3]2+ with
phenols produced exclusively the cyclohexadienyl-one prod-
ucts.27 Finding a more easily protonated analogue of complexes
7 and 8 is likely to be required for efficient catalysis under mild
aqueous conditions.
Arene Exchange Reactions. After formation of guaiacol
complex 9, we attempted arene exchange by addition of excess
veratrole. After 15 h of heating at 60 °C, ∼30% conversion to
complex 1 was observed, along with resonances consistent with
free guaiacol (Scheme 8). Although prolonged heating (∼2
mixture of binding isomers) in pD 7.7 phosphate buffer led to
complete consumption of both isomers within 15 min. The
1
products were identified by H NMR spectroscopy. Methanol
and phenol were formed in a 2:1 ratio, leading to our
assignments of the two major Ir-containing products (also in a
2:1 ratio, see Scheme 6). The observation of small amounts of
the unsubstituted cyclohexadienyl-one complex 8 is consistent
with the impurity present in complex 5 being the
monosubstituted arene complex, which would hydrolyze to
produce the previously characterized product 8. Interestingly,
hydrolysis of diphenyl ether occurs more quickly than that of
anisole, but hydrolysis of an aromatic containing both phenyl
ether and methyl ether substituents released methanol
preferentially.
Scheme 8
The mechanistic studies described above lead to several
predictions on the eventual reactivity of lignin itself. First, the
hydrolysis reactivity described in this report proceeds by a SNAr
mechanism which should be amenable to very bulky
substituents, such as the substituted propyl groups found in
lignin. Second, regioselective C−O cleavage does seem
possible, although the origin of selectivity for cleavage of
alkyl vs aryl ethers remains unclear. Finally, the cleavage
reaction is rapid even at neutral pH and low temperatures,
conditions far milder than presently required for lignin
degradation.
Protonation of Dienyl-one Complex 7. After hydrolysis,
the next step in a redox-neutral cleavage cycle would be
protonation (Scheme 1). Upon addition of 200 μL of 8 M
HBF4(aq) to dienyl-one complex 7, NMR spectroscopy
showed quantitative conversion to a new species, as shown in
Scheme 7. The product contained a single Cp* resonance, a
MeO resonance (integrating to three protons), and aromatic
resonances corresponding to four protons. The aromatic
protons shifted downfield relative to those of 7, consistent
with rearomatization of the organic species. We therefore assign
weeks) did not lead to full conversion, this reaction suggested
that arene exchange was indeed possible. Furthermore, by
regenerating complex 1, the viability of each of the steps in
Scheme 1 was demonstrated.
To probe arene exchange under less harsh conditions, a
solution of anisole complex 2 was heated in the presence of
excess veratrole in unbuffered water at 80 °C. Three hours of
heating led to the formation of [Cp*Ir(veratrole)]2+ (1), free
anisole, and the hydrolysis products of both the veratrole and
anisole complexes (7 and 8, respectively), as shown in Scheme
9 (and Figure S13 in the Supporting Information). Complex 2
was completely consumed, with the 1:7:8 complex ratio being
roughly 2.8:1.0:1.3. This study confirmed that arene exchange is
possible and is competitive with hydrolysis under certain
conditions. Nevertheless, arene dissociation appears to be the
most difficult step under the present conditions.
Scheme 7
CONCLUSIONS
■
We have shown that the [Cp*Ir]2+ fragment can support all of
the key steps needed for a redox-neutral catalytic lignin
depolymerization scheme. Each stephydrolysis, protonation,
and arene exchangeis carried out under aqueous conditions
below 100 °C. The mechanism of hydrolysis was found to
1249
dx.doi.org/10.1021/om5000166 | Organometallics 2014, 33, 1245−1252

Organometallics
Article
Scheme 9
and CH2Cl2, sparingly soluble in acetone, and soluble in CH3CN,
CH3NO2, and H2O. The veratrole ligand was slowly displaced by
MeCN over time, presumably forming [Cp*Ir(CH3CN)3][BF4]2.
Single crystals suitable for X-ray diffraction studies were grown by
proceed through an SNAr mechanism; the metal-bound
Jackson−Meisenheimer complex was observed as an inter-
mediate, and the reactivity was only mildly retarded by
increased steric bulk at the alkyl ether group. This mechanistic
scheme is well suited for lignin degradation chemistry, as the
aromatic groups in lignin are sterically encumbered. Binding the
electron-rich arenes to the electrophilic Ir center rendered them
highly reactive toward nucleophilic attack, even in neutral
water. Currently strong acid is required to release the product
of hydrolysis from the metal; therefore, further work including
tuning the electronic properties of the metal center to facilitate
protonation is required in order to achieve catalysis. Work to
develop this promising concept of arene coordination for a mild
catalytic disassembly of lignin is continuing in our laboratories.
1
vapor diffusion of diethyl ether into a nitromethane solution of 1. H
NMR (CD3CN, 500 MHz): δ 2.23 (s, 15H, C5(CH3)5), 4.07 (s, 6H,
C6H4(OCH3)2), 6.8 (m, 2H, C6H4(OCH3)2), 7.2 (m, 2H,
C6H4(OCH3)2). 13C{1H} NMR (CD3CN, 126 MHz; over the course
of the experiment, a small amount of [Cp*Ir(CH3CN)3]2+ formed and
was visible by 13C NMR, δ 9.21, 94.24): δ 10.25 (C5(CH3)5), 60.68
(C6H4(OCH3)2), 82.92 (C6H4(OCH3)2), 91.38 (C5(CH3)5), 104.91
(C6H4(OCH3)2), 133.75 (C6H4(OCH3)2). 13C{1H} NMR (CD3NO2,
126 MHz): δ 9.62 (C5(CH3)5), 60.18 (C6H4(OCH3)2), 82.55
(C6H4(OCH3)2), 91.24 (C5(CH3)5), 105.14 (C6H4(OCH3)2), 133.71
(C6H4(OCH3)2). Anal. Calcd for C18H25B2F8IrO2: C, 33.82; H, 3.94.
Found: C, 33.78; H, 3.89.
Synthesis of [Cp*Ir(anisole)][BF4]2 (2). This compound was
synthesized analogously to complex 1 in 84% yield as a white powder.
Complex 2 has been previously reported by a similar preparation;
spectral data in acetone matched the literature data.24,36 1H NMR
(CD3CN, 500 MHz): δ 2.29 (s, 15H, C5(CH3)5), 4.06 (s, 3H,
C6H5(OCH3)), 6.99−7.03 (m, 1H, C6H5(OCH3)), 7.10−7.15 (m, 4H,
C6H5(OCH3)). Anal. Calcd for C17H23B2F8IrO: C, 33.52; H, 3.81.
Found: C, 33.70; H, 3.85.
EXPERIMENTAL SECTION
■
General Considerations. Most procedures were carried out in air,
without any special handling. A note is made below when procedures
were carried out under an inert atmosphere. HPLC grade water was
used without additional purification, and acetone was distilled from
calcium sulfate prior to use. All deuterated solvents were purchased
from Cambridge Isotopes Laboratories, Inc. When needed, solvents
were degassed by thorough sparging with N2 or by three freeze−
Synthesis of [Cp*Ir(ethoxybenzene)][BF4]2 (3). This com-
pump−thaw cycles and stored under nitrogen. [Cp*Ir(Cl)2]2,22
pound was synthesized analogously to complex 1 in 73% yield as a
1
[Cp*Ir(H2O)3][SO4] (6),33 and NaBArF
were synthesized by
34
white powder. H NMR (CD3CN, 500 MHz): δ 1.46 (t, J = 7.0 Hz,
4
3H, C6H5(OCH2CH3)), 2.28 (s, 15H, C5(CH3)5), 4.30 (q, J = 7.0 Hz,
literature methods. All other materials were readily commercially
available and used as received. 1H and 13C NMR spectra were
recorded on 300 and 500 MHz spectrometers at 25 °C, unless
otherwise noted. Chemical shifts in D2O are reported with respect to
NaTSP (3-(trimethylsilyl)propionic-2,2,3,3-d4 acid sodium salt, δ 0.0)
1
2H, C6H5(OCH2CH3)), 6.94−7.14 (m, 5H, C6H5(OCH2CH3)). H
NMR (acetone-d6, 500 MHz): δ 1.52 (t, J = 7.0 Hz, 3H,
C6H5OCH2CH3), 2.48 (s, 15H, C5(CH3)5), 4.56 (q, J = 7.0 Hz, 2H,
C6H5OCH2CH3), 7.40 (t, J = 5.8 Hz, 1H, C6H5OCH2CH3), 7.51 (t, J
= 6.5 Hz, 1H, C6H5OCH2CH3), 7.61 (d, J = 6.9 Hz, 1H,
C6H5OCH2CH3). 13C{1H} NMR (acetone-d6, 126 MHz): δ 10.09
(C5(CH3)5), 14.35 (C6H5OCH2CH3), 70.22 (C6H5OCH2CH3), 85.21
(C6H5OCH2CH3), 94.62 (C5(CH3)5), 98.25 (C6H5OCH2CH3),
106.04 (C6H5OCH2CH3), 144.65 (C6H5OCH2CH3). Anal. Calcd for
C18H25B2F8IrO: C, 34.69; H, 4.04. Found: C, 34.30, H, 3.81.
1
for H NMR spectra and with respect to methanol at δ 49.50 for
13C{1H} NMR spectra. Chemical shifts in organic solvents are
reported relative to the residual proteo solvent impurity. The solution
pH was measured with a pH meter. In D2O samples, pD is used to
describe the acidity. The value of pD is calculated by addition of +0.4
to the reading of a pH electrode that was calibrated using H2O
standards.35 Elemental analyses were performed by Atlantic Microlab
Inc. of Norcross, GA.
Synthesis of [Cp*Ir(diphenyl ether)][BF4]2 (4). This compound
was synthesized analogously to complex 1 in 71% yield as a white
1
powder. H NMR (CD3CN, 500 MHz): δ 2.30 (s, 15H, C5(CH3)5),
Synthesis of [Cp*Ir(veratrole)][BF4]2 (1). In a N2-filled glovebox,
a 30 mL scintillation vial was charged with 354.0 mg (0.444 mmol) of
[Cp*Ir(Cl)2]2, a magnetic stir bar, and 12 mL of acetone. To the
stirred orange slurry was added 346 mg (1.777 mmol, 4 equiv) AgBF4
as a solid. The mixture immediately turned yellow. After it was stirred
for 15 min at ambient temperature, the reaction mixture was filtered
through a bed of Celite on a medium-fritted filter, and the solids were
washed three times with 2 mL of acetone. The yellow filtrate was
transferred to another 30 mL scintillation vial, and 340 μL (2.664
mmol, 6 equiv) of veratrole was added by syringe. Some precipitate
formed over a few minutes, and the reaction mixture was stirred for 15
h. In air, the reaction mixture was filtered through sintered glass into a
125 mL Erlenmeyer flask, and 60 mL of Et2O was added to the filtrate,
prompting the precipitation of white solid (yellow supernatant). The
white solid was collected on a fritted-glass filter and washed with three
10 mL portions of Et2O. Collection of the solid and drying under
vacuum for 3 h afforded 457.4 mg (81% yield) of white, analytically
pure [Cp*Ir(veratrole)][BF4]2. The product was insoluble in THF
7.05 (m, 1H, PhOPh), 7.15 (m, 4H, PhOPh), 7.29 (m, 2H, PhOPh),
7.50 (m, 1H, PhOPh), 7.64 (m, 2H, PhOPh). H NMR (acetone-d6,
1
500 MHz): δ 2.50 (s, 15H), 7.44 (t, J = 5.9 Hz), 7.48−7.53 (m, 3H),
7.58 (m, 2H), 7.64−7.69 (m, 4H). 13C{1H} NMR (acetone-d6, 126
MHz): δ 10.22 (C5(CH3)5), 86.89 (Ir−o-C6H5OC6H5), 95.76 (Ir−p-
C6H5OC6H5), 98.47 (Ir−m-C6H5OC6H5), 106.73 (C5(CH3)5), 121.23
(Ir−C6H5O-o-C6H5), 128.94 (Ir−C6H5O-p-C6H5), 132.25 (Ir−
C6H5O-m-C6H5), 143.64 (Ir−ipso-C6H5OC6H5), 152.77 (Ir−C6H5O-
ipso-C6H5). Anal. Calcd for C22H25B2F8IrO: C, 39.36; H, 3.75. Found:
C, 39.69; H, 3.71.
Synthesis of [Cp*Ir(3-phenoxyanisole)][BF4]2 (5). This com-
pound was synthesized analogously to complex 1. Compound 5 was
obtained in 78% yield, but only in ∼95% purity (1H NMR). The other
species could possibly be the isomer bound to the phenoxy ring rather
than the anisole ring. This material was used as is for NMR scale test
reactions and was not further purified or sent for elemental analysis.
1
Data for the major product are as follows. H NMR (CD3CN, 300
1250
dx.doi.org/10.1021/om5000166 | Organometallics 2014, 33, 1245−1252

Organometallics
Article
MHz): δ 2.26 (s, 15H, C5(CH3)5), 4.05 (s, 3H, −OCH3), 6.9−7.03
(m, 3H), 7.27 (d, J = 7.7 Hz, 2H), 7.28 (s, 1H), 7.49 (d, J = 7.4 Hz,
1H), 7.62 (t, J = 8.0 Hz, 2H). Data for minor product are as follows.
1H NMR (CD3CN, 300 MHz): δ 2.30 (s), 3.84 (s), aromatic
resonances not clearly resolved.
pH was measured using an ISFET microelectrode. Changes in pH
were typically 0.1−0.2 pH unit, consistent with the expected change
based on the release of 4−10 mM H+ in the 0.1 M buffer system.
The organometallic product of hydrolysis was the same in each
reaction. The key spectroscopic features were similar to those for the
isolated, analytically pure methoxy-substituted species 7 (see above).
Characterization data for [Cp*Ir(cyclohexadienyl-one)][BF4] (8) are
as follows. 1H NMR (D2O, 283 K, 500 MHz): 2.23 (s, 15H, Cp*), 6.0
(m, 2H, C6H5O), 6.6 (m, 3H, C6H5O). 13C{1H} NMR (D2O, 238 K,
126 MHz): 9.60, 83.47, 86.20, 96.27, 101.65, 160.27. Complex 8 has
been previously reported.27
Synthesis of [Cp*Ir(veratrole)][BArF ] (1-BArF ). In air, a
4 2
4
scintillation vial was charged with 150.6 mg (0.1699 mmol) of
NaBArF , 54.3 mg (0.0849 mmol) of solid [Cp*Ir(veratrole)][BF4]2,
4
and a magnetic stir bar. A 10 mL portion of CH2Cl2 was added, and
the mixture was stirred for 2 h, during which time the larger clumps of
the white Ir complex were replaced with a fine white suspension. The
mixture was filtered through Celite, and the volatiles were removed
from the clear colorless filtrate under vacuum, affording 176 mg (95%
Arene Exchange Reactions. Arene exchange reactions were
performed in NMR tubes fitted with Teflon valves. The reactions were
initiated in a nitrogen-filled glovebox, sealed, and removed from the
glovebox for spectroscopic analysis and heating.
yield) of a white powder. The BArF salt was markedly more soluble
4
than the BF4 salt, with good solubility in THF and CH2Cl2. No arene
1
displacement was observed in these solvents. H NMR (THF-d8, 500
ASSOCIATED CONTENT
* Supporting Information
Figures, tables, text, and CIF files giving NMR spectra, kinetic
details, and crystallographic data for 1 and 7-BArF4. This
material is available free of charge via the Internet at http://
pubs.acs.org.
MHz): δ 2.36 (s, 15H, C5(CH3)5), 4.19 (s, 6H, C6H4(OCH3)2), 7.0
(m, 2H, C6H4(OCH3)2), 7.5 (m, 2H, C6H4(OCH3)2), 7.59 (s, 8H,
BArF ), 7.79 (s, 16H, BArF ). 1H NMR (CD2Cl2, 300 MHz): δ 2.24 (s,
■
S
4
4
15H, C5(CH3)5), 4.10 (s, 6H, C6H4(OCH3)2), 6.6 (m, 2H,
C6H4(OCH3)2), 6.8 (m, 2H, C6H4(OCH3)2), 7.57 (s, 8H, BArF ),
4
7.72 (s, 16H, BArF ). Anal. Calcd for C82H49B2F48IrO2: C, 44.93; H,
4
2.25. Found: C, 45.12; H, 1.97.
Synthesis of [Cp*Ir(methoxycyclohexadienyl-one)][BArF ]
4
(7-BArF ). In air, 39.2 mg (0.0613 mmol) of [Cp*Ir(veratrole)][BF4]2
AUTHOR INFORMATION
Corresponding Author
*E-mail: ajmm@email.unc.edu (A.J.M.M.); goldberg@chem.
washington.edu (K.I.G.).
Notes
4
■
was dissolved in 2 mL of 0.1 M phosphate buffer (pH 7.0). After the
reaction mixture was stirred for 30 min, 54.3 mg (0.0613 mmol) of
NaBArF4 was added as a solid. The white suspension was stirred for 1
h, and 2 mL of CH2Cl2 was added. Upon vigorous shaking of the
mixture, almost all of the white solid dissolved. In a small separator
funnel, the aqueous solution was extracted twice with 3 mL of CH2Cl2.
The organic fractions were combined, and the solvents were removed
under vacuum, providing 79.2 mg (98% yield) of a white solid of
analytically pure product. Single crystals suitable for an X-ray
diffraction study were grown from CH2Cl2/pentane. 1H NMR
(CD2Cl2, 500 MHz): δ 2.12 (s, 15H, C5(CH3)5), 3.88 (s, 3H,
−OCH3), 5.57 (dd, JHH = 0.9, 6.9 Hz, 1H, C6H4O(OCH3)), 5.85
(ddd, JHH = 1.3, 5.2, 6.7 Hz, 1H, C6H4O(OCH3)), 5.92 (m, 1H,
C6H4O(OCH3)), 6.04 (dd, JHH = 1.3, 5.7 Hz, 1H, C6H4O(OCH3)),
7.57 (s, 4H, BArF ), 7.72 (s, 8H, BArF ). 13C{1H} NMR (CD2Cl2, 126
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful for funding support from the Camille and
Henry Dreyfus Postdoctoral Program in Environmental
Chemistry, the NSF Center for Enabling New Technologies
Through Catalysis (CENTC, CHE-0650456 and CHE-
1205189), and the University of North Carolina at Chapel
Hill (A.J.M.M.). Jessica Young performed preliminary studies
on the metal binding of alkoxyarenes. Prof. James Mayer and
Prof. Michael Heinekey provided invaluable feedback and
experimental suggestions. K.I.G. thanks the Weizmann Institute
of Science for a Joseph Meyerhoff Visiting Professorship.
4
4
MHz): δ 10.02 (s, C5(CH3)5), 58.43, (s, −OCH3), 77.72 (s, O
C6H4OCH3), 80.56 (s, OC6H4OCH3), 80.71 (s, OC6H4OCH3),
89.33 (s, OC6H4OCH3), 99.25 (s, C5(CH3)5), 117.91 (m, B(3,5-
(CF3)2C6H3)4), 125.01 (q, JCF = 272 Hz, B(3,5-(CF3)2C6H3)4), 129.28
(q of 1:1:1:1 q, JCF = 31.8 Hz, JCB = 2.7 Hz, B(3,5-(CF3)2C6H3)4),
130.84 (s, OC6H4OCH3), 135.21 (br s, B(3,5-(CF3)2C6H3)4),
157.17 (s, OC6H4OCH3), 162.16 (1:1:1:1 q, 50 Hz, B(3,5-
(CF3)2C6H3)4). Anal. Calcd for C49H34BF24IrO2: C, 44.80; H, 2.61.
Found: C, 44.64; H, 2.46.
Equilibrium Measurements. The equilibrium constant for eq 1
was obtained by dissolving ∼30 mmol of complex 1 in D2O in air. The
reaction was monitored over 60 h, with equilibrium reached after
about 40 h. Three separate experiments were carried out, giving a Keq
value of 0.0049(1) M. Treatment of [Cp*Ir(H2O)3][SO4] with
veratrole also proceeded to a mixture of all the constituents of eq 1,
with a similar equilibrium constant. Side products (∼5%) included
hydrolysis product 7 and were ignored in the equilibrium measure-
ment.
Hydrolysis Kinetics. Kinetic experiments were carried out at 10
°C under air. Complexes 2−4 were weighed in NMR tubes and
transported to the NMR spectrometer. The probe was precooled to 10
°C, and a D2O solution (pD 7.7, 0.1 M KH2PO4/K2HPO4 containing
5 mM dioxane internal standard) was cooled in an ice bath. A sample
containing only the D2O buffer solution was used to lock and shim.
The reaction was initiated by addition of 0.5 mL of the D2O solution
in the NMR tube by syringe ([Ir] = 4−10 mM). The tube was shaken
and quickly injected into the NMR probe. Acquisition was started
immediately (1 s delay time, two scans per acquisition), with a 20 s
fixed delay between acquisitions. At the conclusion of the experiment,
the contents of the tube were transferred to a small glass vial and the
REFERENCES
■
(1) Bozell, J. J., Patel, M. K., eds. Feedstocks for the Future: Renewables
for the Production of Chemicals and Materials; American Chemical
Society: Washington, DC, 2006; ACS Symposium Series 921.
(2) Chheda, J. N.; Huber, G. W.; Dumesic, J. A. Angew. Chem., Int.
Ed. 2007, 46, 7164−7183.
(3) Carroll, A.; Somerville, C. Annu. Rev. Plant Biol. 2009, 60, 165−
182.
(4) Ruppert, A. M.; Weinberg, K.; Palkovits, R. Angew. Chem., Int. Ed.
2012, 51, 2564−2601.
(5) Gallezot, P. Chem. Soc. Rev. 2012, 41, 1538−1558.
(6) Zakzeski, J.; Bruijnincx, P. C. A.; Jongerius, A. L.; Weckhuysen, B.
M. Chem. Rev. 2010, 110, 3552−3599.
(7) Geib, S. M.; Filley, T. R.; Hatcher, P. G.; Hoover, K.; Carlson, J.
E.; Jimenez-Gasco, M. M.; Nakagawa-Izumi, A.; Sleighter, R. L.; Tien,
M. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 12932−12937.
(8) Hicks, J. C. J. Phys. Chem. Lett. 2011, 2, 2280−2287.
(9) Sergeev, A. G.; Hartwig, J. F. Science 2011, 332, 439−443.
(10) Kelley, P.; Lin, S.; Edouard, G.; Day, M. W.; Agapie, T. J. Am.
Chem. Soc. 2013, 134, 5480−5483.
(11) Nichols, J. M.; Bishop, L. M.; Bergman, R. G.; Ellman, J. A. J.
Am. Chem. Soc. 2010, 132, 12554−12555.
(12) Son, S.; Toste, F. D. Angew. Chem., Int. Ed. 2010, 49, 3791−
3794.
1251
dx.doi.org/10.1021/om5000166 | Organometallics 2014, 33, 1245−1252

Organometallics
Article
(13) Hanson, S. K.; Wu, R.; Silks, L. A. P. Angew. Chem., Int. Ed.
2012, 51, 3410−3413.
(14) Hanson, S. K.; Baker, R. T.; Gordon, J. C.; Scott, B. L.; Silks, L.
A. P.; Thorn, D. L. J. Am. Chem. Soc. 2010, 132, 17804−17816.
(15) Rahimi, A.; Azarpira, A.; Kim, H.; Ralph, J.; Stahl, S. S. J. Am.
Chem. Soc. 2013, 135, 6415−6418.
(16) Pape, A. R.; Kaliappan, K. P.; Kundig, E. P. Chem. Rev. 2000,
̈
100, 2917−2940.
(17) Pike, R.; Sweigart, D. Coord. Chem. Rev. 1999, 187, 183−222.
(18) Harman, W. Chem. Rev. 1997, 97, 1953−1978.
(19) Kane-Maguire, L. A. P.; Honig, E. D.; Sweigart, D. A. Chem. Rev.
1984, 84, 525−543.
(20) Davies, S. G.; Green, M. L. H.; Mingos, D. M. P. Tetrahedron
1978, 34, 3047−3077.
(21) Aqueous-Phase Organometallic Catalysis, 2nd ed.; Cornils, B.,
Herrmann, W. A., Eds.; Wiley-VCH: Weinheim, Germany, 2004.
(22) White, C.; Yates, A.; Maitlis, P. M. Inorg. Synth. 1992, 29, 228−
234.
(23) Bush, R. C.; Angelici, R. J. J. Am. Chem. Soc. 1986, 108, 2735−
2742.
(24) Grundy, S.; Maitlis, P. J. Organomet. Chem. 1984, 272, 265−282.
(25) Morvan, D.; Rauchfuss, T. B.; Wilson, S. R. Organometallics
2009, 28, 3161−3166.
(26) The activity of the water solvent is taken as unity.
(27) Le Bras, J.; El Amouri, H.; Vaissermann, J. Organometallics 1996,
15, 5706−5712.
(28) Amouri, H.; Vaissermann, J.; Rager, M. N.; Besace, Y. Inorg.
Chem. 1999, 38, 1211−1215.
(29) The solution acidity was measured by pH electrode, and the pD
was calculated by adding 0.4 decade to the meter reading.
(30) Hoops, S.; Sahle, S.; Gauges, R.; Lee, C.; Pahle, J.; Simus, N.;
Singhal, M.; Xu, L.; Mendes, P.; Kummer, U. Bioinformatics 2006, 22,
3067−3074.
(31) Smith, M. B.; March, J. March’s Advanced Organic Chemistry:
Reactions, Mechanisms, and Structure, 5th ed.; Wiley: New York, 2001.
(32) Gruter, G. J. M.; van Klink, G. P. M.; Akkerman, O. S.;
Bickelhaupt, F. Organometallics 1993, 12, 1180−1188.
(33) Ogo, S.; Makihara, N.; Watanabe, Y. Organometallics 1999, 18,
5470−5474.
(34) Yakelis, N. A.; Bergman, R. G. Organometallics 2005, 24, 3579−
3581.
(35) Glasoe, P. K.; Long, F. A. J. Phys. Chem. 1960, 64, 188−190.
(36) Espinet, P.; Bailey, P. M.; Downey, R. F.; Maitlis, P. M. J. Chem.
Soc., Dalton Trans. 1980, 1048−1054.
1252
dx.doi.org/10.1021/om5000166 | Organometallics 2014, 33, 1245−1252