Combination of redox-active ligand and Lewis acid for dioxygen reduction with π-bound molybdenum-quinonoid complexes

Article abstract of DOI:10.1021/ja5100405

A series of π-bound Mo-quinonoid complexes supported by pendant phosphines have been synthesized. Structural characterization revealed strong metal-arene interactions between Mo and the π system of the quinonoid fragment. The Mo-catechol complex (2a) was

Full text of DOI:10.1021/ja5100405

Article
pubs.acs.org/JACS
Combination of Redox-Active Ligand and Lewis Acid for Dioxygen
Reduction with π‑Bound Molybdenum−Quinonoid Complexes
Justin T. Henthorn, Sibo Lin, and Theodor Agapie*
Division of Chemistry and Chemical Engineering, California Institute of Technology, 1200 East California Boulevard, MC 127-72,
Pasadena, California 91125, United States
S
* Supporting Information
ABSTRACT: A series of π-bound Mo−quinonoid complexes
supported by pendant phosphines have been synthesized.
Structural characterization revealed strong metal−arene
interactions between Mo and the π system of the quinonoid
fragment. The Mo−catechol complex (2a) was found to react
within minutes with 0.5 equiv of O2 to yield a Mo−quinone
complex (3), H2O, and CO. Si- and B-protected Mo−
catecholate complexes also react with O2 to yield 3 along with
(R2SiO)n and (ArBO)3 byproducts, respectively. Formally, the
Mo−catecholate fragment provides two electrons, while the
elements bound to the catecholate moiety act as acceptors for
the O2 oxygens. Unreactive by itself, the Mo−dimethyl catecholate analogue reduces O2 in the presence of added Lewis acid,
B(C6F5)3, to generate a MoI species and a bis(borane)-supported peroxide dianion, [[(F5C6)3B]2O22−], demonstrating single-
electron-transfer chemistry from Mo to the O2 moiety. The intramolecular combination of a molybdenum center, redox-active
ligand, and Lewis acid reduces O2 with pendant acids weaker than B(C6F5)3. Overall, the π-bound catecholate moiety acts as a
two-electron donor. A mechanism is proposed in which O2 is reduced through an initial one-electron transfer, coupled with
transfer of the Lewis acidic moiety bound to the quinonoid oxygen atoms to the reduced O2 species.
a series of Mo−quinonoid complexes and their reactivity with
dioxygen.
1. INTRODUCTION
Biological reduction of dioxygen is performed by active sites that
employ redox-noninnocent ligands and proton relays to control
the transfer of electrons and protons to the substrate.1 In
synthetic transition-metal chemistry, redox-noninnocent li-
gands2 and second-coordination-sphere acid/base moieties that
facilitate proton transfer3 engender novel reactivity at the metal
center. However, ligand systems that engage in both electron and
proton transfers to substrates are less common.4 Metal−
quinonoid complexes in which the metal is π-bound to the
quinonoid fragment have the potential to access the two
electrons and two protons of the hydroquinone/quinone couple
in addition to any accessible metal-based redox couples. The
study of π-bound metal−quinonoid complexes5 has focused on
polymeric metal−organometallic coordination networks,5 with
only rare examples of substrate-based reactivity.4b,6 Although not
directly coordinated to a metal, hydroquinone has been
employed as distal redox mediator.4b
We have previously reported ligand designs that employ the π
system of an arene to support metals in various coordination
environments.7 A Ni−H complex underwent reversible migra-
tion of H between the metal and pendant arene, demonstrating
the reversible transfer of (formally) protons and electrons
between the Ni center and the ligand.8 Extending this chemistry
to multiproton, multielectron processes at a single metal site with
a pendant catechol moiety, we report herein the first synthesis of
2. RESULTS AND DISCUSSION
2.1. Synthesis and Characterization of π-Bound Mo−
Quinonoid Complexes. Access to the desired Mo−catechol
complex 2a was pursued through the use of the Si-protected
catechol diphosphine 1b, designed to avoid the formation of
oxygen-bound Mo−catecholate species. Heating diphosphine 1b
with (PhMe)Mo(CO)3 in tetrahydrofuran (THF) (Scheme 1)
generates a new species 2b as determined by NMR spectroscopy.
A singlet (50.6 ppm) is observed by 31P{1H} NMR spectroscopy,
1
while the protons assigned to the central arene ring in the H
NMR spectrum of 2b resonate upfield (5.7 ppm, CDCl3)
compared to the signal for the free phosphine 1b (6.8 ppm,
CDCl3) and have split into an apparent triplet (JHP = 4 Hz).
Additionally, two singlets account for the methyl groups bound
to silicon, indicating desymmetrization of the two faces of the
central ring. These data are consistent with a Cs-symmetric
molecule in which there is a strong metal−arene interaction with
the central ring of the terphenyl moiety. The solution IR
spectrum of 2b reveals three bands in the region corresponding
Received: September 29, 2014
Revised: December 4, 2014
Published: January 10, 2015
© 2015 American Chemical Society
1458
DOI: 10.1021/ja5100405
J. Am. Chem. Soc. 2015, 137, 1458−1464

Journal of the American Chemical Society
Article
accessed directly from diphosphine 1a through reaction with
Mo(CO)3(MeCN)3 at room temperature. The resulting product
exhibits spectroscopic features similar to those of 2b. A new
broad resonance at 5.6 ppm (1H NMR) is assigned to the
catechol OH protons. This assignment was confirmed by the loss
of this resonance upon the addition of D2O. An XRD study of 2a
(Figure 1) revealed structural parameters nearly identical to
those of 2b.
Scheme 1. Synthesis of π-Bound Mo−Quinonoid Complexes
2.2. Reduction of O2 by Mo−Catechol with Formation
of Mo−Quinone. While transition-metal σ-bound catecholate
complexes in general,1a,10 as well as Mo−catecholate complexes
specifically,10b,11 have been reported to react with dioxygen to
afford intra- and extra-diol cleavage products and oxidation to
quinones, we are unaware of any reports on π-bound transition-
metal−quinonoid complexes facilitating dioxygen reduction. To
test the propensity of the metal−catechol moiety of 2a to
perform the transfer of multiple electrons and protons, its
chemistry with O2 was studied. Exposure of a solution of 2a in
dichloromethane (DCM) to an atmosphere of O2 resulted in
quantitative conversion to a new diamagnetic species 3 (eq 3)
upon addition. The product displays a singlet at 72 ppm in the
31P{1H} NMR spectrum. In the 1H NMR spectrum, the resulting
species exhibits a new apparent triplet at 4.9 ppm, assigned to
olefinic protons coupled to the phosphines. The solution IR
spectrum of 3 reveals bands at 1875 and 1605 cm−1, consistent
with the stretching frequencies of a metal-bound carbonyl and a
quinone carbonyl, respectively.12 The solution 13C{1H} NMR
spectrum reveals two resonances at ca. 240 ppm, consistent with
two Mo-bound carbonyls. These spectroscopic features suggest
conversion to a quinone−Mo(CO)2 species, and this assignment
was confirmed by an XRD study (Figure 1). The Mo center
exhibits a pseudo-trigonal-prismatic geometry, with the vertices
defined by two phosphine donors, two CO ligands, and two
olefin moieties of the diene bound to the metal center in an η4
fashion. The C−O bonds of the organic fragment have
to CO stretches (νCO = 1959, 1843, and 1835 cm−1), consistent
with a Mo(CO)3 fragment.
The single-crystal X-ray diffraction (XRD) study of 2b (Figure
1) confirmed the spectroscopic findings, which are also
consistent with those for the previously reported analogue.9 In
the solid state, the metal center exhibits a pseudo-octahedral
geometry with the coordination sphere comprised of two trans
phosphines, three meridional carbonyls, and an η2 interaction
with the π system of the catechol fragment. Localization of
double-bond character in the central arene ring suggests
significant π back-bonding between Mo and the ring (the C7−
C
12 and C10−C11 bonds at 1.37 Å are considerably shorter than
the C7−C8, C9−C10, and C11−C12 bonds at 1.43 Å). The aryl C−
O bond distances at 1.37 Å are consistent with C−O single
bonds.
The catechol complex 2a can be accessed from 2b by removal
of the SiMe2 group upon treatment with NaOMe in MeOH,
followed by aqueous NH4Cl workup. Alternatively, 2a can be
Figure 1. Solid-state structures of 2a, 2b, 3, and [4+]2[[(F5C6)3B]2O22−]. Solvent molecules, hydrogen atoms, and the second 4+ cation have been
omitted for clarity. Carbon and fluorine atoms are depicted in black and green, respectively. Selected bond distances (average values for the two
molecules in the asymmetric unit for 2a) are given in Å.
1459
DOI: 10.1021/ja5100405
J. Am. Chem. Soc. 2015, 137, 1458−1464

Journal of the American Chemical Society
Article
GC−MS as cyclo-oligomers of dimethylsiloxane. It has been
reported that electrochemical reduction of O2 in the presence of
R2SiX2 (R = Me, Et, Ph; X = Cl, OMe) transiently generates
silanones (R2SiO species) that oligomerize to yield cyclo-
polysiloxanes.14 As silanones are highly reactive and rapidly
oligomerize, the presence of silanones in solution is typically
deduced via trapping experiments with linear siloxanes such as
hexamethyldisiloxane.14,15 Silanones insert into the Si−O bond
of Me3SiOSiMe3 to yield longer-linear-chain siloxanes of the
form Me3Si(OSiMe2)nOSiMe3 (n = 1, 2), which can be observed
by by GC−MS. Compound 2b displays a Si−O linkage for
potential silanone insertion. Indeed, the intermediate species 2h
(Scheme 2) is observable by 31P{1H} and 1H NMR spectroscopy
Scheme 2. Reactivity of 2b and 2c with O2
contracted to 1.23 Å, consistent with carbon−oxygen double
bonds. Formally, complex 2a was oxidized by two electrons,
coupled with the transfer of two protons, to generate 3. In this
net two-electron, two-proton transformation, the oxidation state
of the metal center remains unchanged (Mo0), with only the
organic fragment undergoing the redox transformation.
To deconvolute the effect of the catechol moiety versus the
metal moiety in the reaction with O2, control experiments were
performed with several species. 3,6-Bis(2-bromophenyl)catechol
(1Br), a phosphine-free alternative to 1a, exhibits no reaction
under an atmosphere of O2 in CD2Cl2 at room temperature, over
1
24 h as determined by H NMR spectroscopy (eq 1). Under
similar conditions, the dimethyl catecholate−Mo complex 2f and
the parent complex 2g also showed no conversion with O2 over
24 h (eq 2). These experiments indicate that the Mo−catechol
combination is required for the observed reactivity.
Toepler pump experiments (see pp S25−S26 in the
Supporting Information) were performed for the reaction of
2a with 5 equiv of O2. A net increase in gas content of the sample
during the reaction of 2b with O2, with a relative integration of
the Si−CH3 singlets to the central arene triplet of 6:2 (rather
than 3:2 as for 2b). These data are consistent with insertion of
generated Me2SiO into the Si−O bond of 2b (Scheme 2),
which was confirmed through independent synthesis. The
observation of 2h indicates that the Me2Si moiety acts as an
oxygen acceptor to generate Me2SiO, as protons do when
starting from 2a to generate H2O2/H2O.
was observed (0.48
0.02 equiv per Mo) after quantitative
conversion of 2a to 3. After removal of excess O2 by reaction with
a basic pyrogallol solution, the remaining gas was quantified
(1.05 0.05 equiv per Mo) and was found to be combustible
with CuO at 350 °C, consistent with CO. The identity of the
released gas was further confirmed by its reaction with a CuI
precursor to generate a previously reported CuI(CO) species.13
Overall, the Toepler pump experiments reveal that 0.5 equiv of
dioxygen is consumed and 1 equiv of CO is released per mole of
2a. This stoichiometry is consistent with four-electron reduction
of O2 to water involving 2 equiv of metal complex. This process
could occur via partial reduction of O2 to H2O2 by 1 equiv of 2a
followed by reduction of H2O2 with a second equivalent of 2a.
Indeed, 2a is cleanly converted to 3 upon treatment with H2O2,
while 3 exhibits only minor conversion to unidentified species
(<20%) with H2O2 (1 equiv) within 1 h. Thus, it is plausible that
H2O2 could be the initial O2 reduction product, which is then
rapidly consumed by a second equivalent of 2a.
The effect of increasing the steric bulk at Si on the overall
reaction with O2 was investigated with the Et2Si and Pr2Si
i
analogues 2c and 2d (Scheme 1). Electrochemical measurements
indicate that the nature of the alkyl group does not have a
significant effect on the reduction potentials and hence the
electronic properties of the complexes (see Figure 2 and Figure
S73 in the Supporting Information). Exposure of 2c to an
atmosphere of O2 leads to consumption of the starting material
within hours, similar to 2b, with the formation of a 1:1 mixture of
3 and the diethylsilanone insertion product 2i; however,
conversion of 2i to 3 is slower, requiring 48 h for full conversion.
The reaction of 2d with O2 is even slower than the conversion of
2i, with a <20% yield of 3 generated over the course of 5 days at
room temperature. The observed effect of steric bulk on reaction
rate indicates that the silicon center is accessed during a rate-
determining process; since the reduction potentials of 2b, 2c, and
2d are very similar this process likely involves Si−O bond
formation from O2.
To understand the O2 reduction process in more detail, 2b was
investigated as a metal complex with an electron-rich central ring
yet without easily transferable protons. Compound 2b also reacts
with O2 to generate 3, albeit slower than 2a (over the course of
several hours), with silyl-containing byproducts identified by
Reactivity with O2 was found to extend to boron-substituted
2e as well, though compared with 2b the reaction is slower. After
36 h at room temperature, 2e had been consumed and 3 formed
in ca. 80% yield. New unidentified species were observed by
1460
DOI: 10.1021/ja5100405
J. Am. Chem. Soc. 2015, 137, 1458−1464

Journal of the American Chemical Society
Article
diamagnetic species, 6+, resulting from ether demethylation.
Independent synthesis by addition of MeOTf to 3 supports the
structural assignment of 6+. After complete conversion of
intermediate 4+ to a mixture of 52+ and 6+, vacuum transfer of
the volatiles to a J. Young NMR tube revealed the formation of
Me2O and MeOH. To determine the origin of the O atom in
Me2O, the oxidation reaction was performed with 18O2 instead of
natural-abundance O2. Me2O generated in the reaction was
detected by GC−MS, and when the experiment was performed
with 18O2, the major isotopologue observed was Me218O. While
not quantitative, the observation of Me2O formation suggests
that the peroxide moiety reacts with 52+ via abstraction of Me+ to
yield 6+.18 Conversion of 52+ to 6+ demonstrates the ability of a
reduced oxygen species to cleave the O−element bond of the
resulting oxidized Mo complex, although cleavage of the aryl−O
bond has not been ruled out.18a
31P{1H} NMR spectroscopy in addition to multiple broad
resonances by 19F NMR spectroscopy. A qualitatively similar set
of spectra was obtained when 3 was combined with the boroxin
(ArBO)3 in CD2Cl2, likely as a result of the formation of Lewis
acid−base adducts and 3-mediated oligomerization of an “ArBO”
moiety similar to the polysiloxanes observed for 2b.
2.3. Investigation of the Role of Lewis Acids in O2
Reduction by Mo−Quinonoid Compounds. Considering
that pendant H+, [R2Si]2+, and [RB]2+ moieties can act as Lewis
acids and that all engage in the O2 activation process, more
mechanistic insight was sought by addition of an external Lewis
acid to target intermolecular reactivity. Compound 2f does not
exhibit O2 reactivity on its own. Addition of 2 equiv of B(C6F5)3
to 2f under N2 resulted in a broadening of the NMR
spectroscopic features of 2f, similar to what has been reported
for a zirconocene complex.16 This may be caused by a
combination of effects, including electron transfer and
coordination of borane to ether or carbonyl moieties. Exposure
The crystal structure of 4+ allows for an evaluation of the effect
of the metal oxidation state on the interaction with the arene.
While partial localization of double-bond character in the
catechol moiety was observed for 2a and 2b, with C−C distances
varying between 1.37 and 1.43 Å, the dimethyl catecholate
moiety of 4+ displays C−C distances in a narrower range (1.40−
1.42 Å). These structural parameters suggest that the Mo−arene
interaction shifts from predominantly Mo-to-arene π back-
bonding to arene-to-Mo σ donation upon oxidation of the metal
center.19 The Mo−C distances are ca. 0.05 Å shorter in 2a (ca.
2.55 Å) versus 4+ (ca. 2.50 Å), indicating a strong interaction
between the metal center and the ring. This interaction increases
the electrophilicity of the arene and of the E group bonded to the
catecholate oxygens. It is proposed that this activation of E for
nucleophilic attack facilitates the reaction with the O2 fragment.
Further insight into the redox chemistry of the reported Mo
complexes was provided by cyclic voltammetry (CV) studies
(Figure 2). Compound 2f shows a symmetric and fairly reversible
1
of this mixture to O2 affords a mixture that is silent by H and
31P{1H} NMR spectroscopy, and a gradual color change from
red-orange to brown to dark-purple (λmax = 575 nm) was
observed over the course of 30 min at room temperature. Purple
crystals of compound [4+]2[[(F5C6)3B]2O22−] were isolated
from the reaction mixture (Scheme 3), and an XRD study
Scheme 3. Reactivity of 2f with O2 in the Presence of B(C6F5)3
revealed a six-coordinate Mo(CO)3 unit bound by the terphenyl
diphosphine with methoxy moieties intact, analogous to 2a and
2b (Figure 1). The unit cell contains a bis(borane)-supported
peroxide dianion17 in a peroxide:Mo ratio of 1:2, indicating that
the metal complex is a monocation (formally MoI), consistent
with the lack of diamagnetic resonances by NMR spectroscopy.
The formation of the [[(F5C6)3B]2O22−] dianion upon treat-
ment of mixtures of ferrocenes and B(C6F5)3 with O2 was
recently reported.17 The observation of [4+]2[[(F5C6)3B]2O2
]
2−
in the reaction of 2f and B(C6F5)3 with O2 demonstrates the
ability of the Mo center to reduce O2 via outer-sphere one-
electron transfer.
Figure 2. Cyclic voltammograms of compounds 2b (blue), 2d (red), 2e
−
(green), and 2f (purple) in 0.1 M [nBu4N+][PF6 ] in THF recorded
with a glassy carbon electrode at a scan rate of 100 mV/s. Potentials are
Over the course of several hours, the purple solution of 4+
generated a new diamagnetic ion 52+, which was independently
synthesized by oxidation of 2f with 2 equiv of silver
trifluoromethanesulfonate (AgOTf). Upon treatment of 2f
with AgOTf, the solution initially turned purple, consistent
with the formation of the one-electron-oxidized product 4+, and
then became pale-yellow-orange as 52+ was produced via further
oxidation and loss of CO. In the presence of the bis(borane)-
supported peroxide dianion, 52+ is partially converted to a new
referenced to Cp2Fe/Cp2Fe+.
couple, consistent with the isolation of both neutral and oxidized
species and relatively small structural reorganization. Compound
2b exhibits a more asymmetric couple with lower cathodic
current compared with anodic current. Increasing the steric bulk
at Si in 2d shows a return to a symmetric couple, while the more
sterically accessible 2e shows a fully irreversible couple at scan
1461
DOI: 10.1021/ja5100405
J. Am. Chem. Soc. 2015, 137, 1458−1464

Journal of the American Chemical Society
Article
rates between 50 and 2000 mV/s. This electrochemical behavior
is reminiscent of the behavior of (η6-arene)Cr(CO)3 com-
pounds, wherein the reversibility of the one-electron redox event
is highly dependent on the presence of a nucleophile (an additive
such as MeCN, MeOH, H2O, or THF or the counteranion
20,21
−
−
ClO4 or PF6 )
because of chemical decomposition of the
generated radical cation by external nucleophiles. Consequently,
reversibility can be achieved in noncoordinating solvents such as
CH2Cl2 by either lowering the temperature or employing less-
21
−
nucleophilic electrolyte anions such as [B(C6F5)4 ].
For compounds 2b−e, we speculate that upon one-electron
oxidation, the Si or B bound to the catecholate oxygens develops
more electrophilic character and becomes susceptible to attack
by an external nucleophile, either from the supporting electrolyte
(tetrabutylammonium hexafluorophosphate) or even the solvent
itself (THF), resulting in the observed electrochemical
irreversibility. Indeed, chemical oxidation of 2b with either 1 or
2 equiv of Ag(OTf) or [Cp2Fe+][PF6 ] in THF resulted in a
+
1
mixture of species (as determined by H and 31P{1H} NMR
spectroscopy) that is capable of polymerizing THF over the
course of several hours, suggesting the generation of a very
electrophilic species. Conversely, the oxidation of 2d with 2 equiv
of Ag(OTf) in THF yields a single major diamagnetic species as
1
determined by H and 31P{1H} NMR spectroscopy, and the
resulting solution was not observed to polymerize THF. As the
alkyl substituents bound to Si are oriented away from both the
metal center and the catechol carbocyclic ring (Figure 1), the
recovery of electrochemical reversibility in going from 2b to 2d
and the lack of solvent polymerization upon chemical oxidation
of 2d suggest that Si is the site of nucleophilic attack. The bulkier
isopropyl groups of 2d better impede the approach of
nucleophiles to the Si center compared with the methyl groups
of 2b. These results are consistent with the development of
increased electrophilic character on the Lewis-acidic E upon
oxidation, indicative of ligand−metal cooperation by activation
of the catechol moiety upon Mo-based electron transfer. The
reactivity of these species with O2 follows the trends observed
electrochemically, with the bulky species displaying more-
reversible CV and reacting slower.
To further probe the role of metal−ligand (Mo−quinonoid)
cooperativity for dioxygen reactivity, the reaction of 2g with O2 in
the presence of external catechol was investigated. Compound 2g
was selected for this experiment to limit potential complications
due to the loss of Me+, as were observed during the reaction of 2f
and B(C6F5)3 with O2. Compound 2g is competent for O2
reduction in the presence of B(C6F5)3, generating the same
bis(borane)-supported peroxide, [[(F5C6)3B]2O22−], as 2f
according to 19F NMR spectroscopy (eq 4).
Accordingly, exposure of an orange mixture of 2g and catechol
to an atmosphere of O2 results in a slight darkening of the
solution over the course of 3 h (the time required for quantitative
conversion of 2b to 3; eq 5). 31P{1H} NMR spectroscopy shows
a complete loss of signal, while 1H NMR spectroscopy reveals a
broadening of the signals corresponding to 2g; however, the
majority of the catechol remains unaffected over the course of the
reaction (eq 6) with less than 15% conversion. After 36 h at room
temperature, a new signal is observed by 31P{1H} NMR
spectroscopy at 60 ppm, consistent with oxidation of the
phosphine to free phosphine oxide, indicative of decomposition
of the metal complex. These results suggest that the presence of
external catechol is sufficient to facilitate O2 reactivity at Mo.
However, this is a slow process, and the low conversion of
catechol and decomposition of the metal complex indicate that
the resulting reduced oxygen species preferentially reacts with
the Mo complex over the external catechol. Additionally, the
reaction of catechol with O2 in the presence of 1,1′,3,3′-
tetramethylferrocene (Me4Fc) as a surrogate outer-sphere
reductant of similar potential as the reported Mo complexes
was investigated (eq 7). Over the course of 3 h at room
temperature, slight oxidation of Me4Fc (<5%) was observed by
UV−vis spectroscopy; however, no conversion of catechol was
1
detected by H NMR spectroscopy. The low conversion of
catechol oxidation chemistry observed in these intermolecular
reactions emphasizes the cooperative nature of the reactivity
observed for the Mo−quinonoid complexes.
2.4. Proposed Mechanisms for O2 Reduction by Mo−
Quinonoid Complexes. The intermolecular reactivity of 2f
and O2 in the presence of B(C6F5)3 (Figure 3) offers insight
applicable to the intramolecular systems. The proposed
mechanism for O2 activation by 2f and B(C6F5)3 initiates via
outer-sphere electron transfer from Mo to O2 facilitated by the
strongly Lewis acidic borane. While the reduction potentials of
22
−
the Mo0/MoI couple (−0.176 V in THF) and the ·O2 /O2
couple (−1.18 V in dimethyl sulfoxide)23 are mismatched, it has
been demonstrated that electron transfer rates can be greatly
increased by coupling to Lewis acid binding.24 This pre-
equilibrium step is driven forward by the rapid disproportiona-
tion of the proposed borane-supported superoxide into bis-
(borane)-supported peroxide and dioxygen, as has been
previously reported.25 Disproportionation of the formally 17-
electron 4+ yields the 18-electron complex 52+ and regenerates
the starting material 2f. While it is anticipated that the MoI of 4+
should be an even weaker reductant than 2f, further oxidation of
4+ via O2 and B(C6F5)3 to yield 52+ cannot be ruled out. η6
coordination of phenol to a Cr(CO)3 unit resulted in a 4 pKa unit
increase in acidity,26 and thus it is presumed that similar
activation of the catecholate unit in 52+ results in increased
susceptibility of the methyl groups toward nucleophilic attack.
Both nucleophilic attack at the methyl carbon and at the aryl
carbon have been proposed in Cp*Ir(η6-anisole)2+ complexes.18
On the basis of our isotopic-labeling studies, 52+ is demethylated
by the bis(borane)-supported peroxide via nucleophilic sub-
1462
DOI: 10.1021/ja5100405
J. Am. Chem. Soc. 2015, 137, 1458−1464

Journal of the American Chemical Society
Article
Figure 3. Proposed mechanism for the intermolecular reactivity of 2f
and B(C6F5)3 with O2.
stitution at methyl to yield the methyl peroxide and 6+, with the
former reacting further to yield the observed Me2O or Me218O.
For the catechol and Si- and B-protected catecholate π
complexes, a related mechanism for O2 activation is proposed
(Figure 4A). Initial single-electron transfer is facilitated by
binding of ·O2− to the intramolecular Lewis acidic moiety E (E =
H, R2Si, ArB) as B(C6F5)3 does with 2f in intermolecular
fashion.24a The resulting intermediate 9 is an intramolecular
analogue of putative [4+][(F5C6)3BO2 ]. Oxidation of the Mo
−
complex through electron transfer to O2 results in increased
electrophilicity at E, as supported by structural analysis of 4+,
electrochemical analysis, and chemical oxidation of 2b. Loss of
carbonyl coupled with attack by ·O2− on the Lewis acid E results
in scission of a catecholate−E bond, with subsequent steps
leading to O−O bond cleavage (analogous to the formation of 6+
and Me2O). The intermediacy of MoII compound 10 accessed via
further oxidation of 9 cannot be ruled out, with attack by a
reduced oxygen species again resulting in formation of 11. While
it is unclear at which step the O−O bond is cleaved, one
demonstrated possibility is that 2a reduces H2O2 to yield H2O
and 3. Analogous EO2 species (E = R2Si, ArB) are anticipated to
be even more reactive than H2O2, and it is therefore presumed
that, if generated, they will be consumed via further reaction with
starting material 2. A recently computed mechanism for O2
reduction with hydroanthraquinones invokes H atom abstrac-
tion,27 and although the organoquinonoid 1Br does not react with
O2, it cannot be ruled out that 2a follows a similar mechanism
(Figure 4B) wherein the Mo center does not directly participate
in the reactivity but activates the catechol moiety.
Figure 4. (A) Proposed mechanism for intramolecular reactivity of 2a−
e,h,i with O2. (B) Alternative mechanism for the intramolecular reaction
of 2a with O2.
absence of Mo showed no reaction with O2. Additionally,
catechol added to solution but not covalently connected to Mo
(or tetramethylferrocene as an alternate single-electron reduc-
tant) is oxidized only partially. Altering the substitution on the
catechol oxygen centers from H to Si or B maintains the reactivity
with O2, but at lower rates. The dimethyl Mo−quinonoid
complex does not react with O2 independently, but in the
presence of B(C6F5)3 affords a bis(borane)-supported peroxide.
Mechanistically, the O2 activation is proposed to occur via
intramolecular Lewis acid-assisted electron transfer. The present
studies demonstrate the ability of a π-bound metal−quinonoid
complex to facilitate multielectron and proton transfer (as well as
silicon, boron, and carbon transfer) from the quinonoid moiety
to a small-molecule substrate and that coupling of the Mo and
quinonoid fragments is integral to the observed reactivity.
Current studies are focused on further exploiting this metal−
3. CONCLUSIONS
In summary, π-bound Mo−catechol complexes were synthesized
and their reactivity with dioxygen to yield a Mo−quinone
product was investigated. Control experiments of Mo complexes
in the absence of the catechol moiety or the catechol in the
1463
DOI: 10.1021/ja5100405
J. Am. Chem. Soc. 2015, 137, 1458−1464

Journal of the American Chemical Society
Article
metallics 2013, 32, 3161. (d) Kelley, P.; Lin, S.; Edouard, G.; Day, M. W.;
Agapie, T. J. Am. Chem. Soc. 2012, 134, 5480. (e) Velian, A.; Lin, S.;
Miller, A. J. M.; Day, M. W.; Agapie, T. J. Am. Chem. Soc. 2010, 132,
6296.
(8) Lin, S.; Day, M. W.; Agapie, T. J. Am. Chem. Soc. 2011, 133, 3828.
(9) Buss, J. A.; Edouard, G. A.; Cheng, C.; Shi, J.; Agapie, T. J. Am.
Chem. Soc. 2014, 136, 11272.
(10) (a) Lippert, C. A.; Arnstein, S. A.; Sherrill, C. D.; Soper, J. D. J. Am.
Chem. Soc. 2010, 132, 3879. (b) Morris, A. M.; Pierpont, C. G.; Finke, R.
G. J. Mol. Catal. A: Chem. 2009, 309, 137. (c) Vaillancourt, F. H.; Bolin,
J. T.; Eltis, L. D. Crit. Rev. Biochem. Mol. Biol. 2006, 41, 241. (d) Yin, C.-
X.; Finke, R. G. J. Am. Chem. Soc. 2005, 127, 9003. (e) Barbaro, P.;
Bianchini, C.; Mealli, C.; Meli, A. J. Am. Chem. Soc. 1991, 113, 3181.
(11) (a) Randolph, A. H.; Seewald, N. J.; Rickert, K.; Brown, S. N.
Inorg. Chem. 2013, 52, 12587. (b) Pierpoint, C. G.; Nordlander, E. In
Molybdenum: Its Biological and Coordination Chemistry and Industrial
Applications; Holder, A. A., Ed.; Nova Science Publishers: New York,
2012; p 161.
ligand cooperativity for O2 reduction and extending it to other
small molecules.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, characterization data, and crystallo-
graphic data (CIF). This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*agapie@caltech.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(12) Campos, J. L.; De Giovani, W. F.; Romero, J. R. Synthesis 1990,
597.
We thank Lawrence M. Henling and Dr. Michael Takase for
crystallographic assistance, Prof. John E. Bercaw for providing
access to a Toepler pump in his laboratory, and Christine Cheng
for assistance in organic synthesis. We are grateful to Caltech and
the NSF (CHE-1151918 to T.A.; GRFP to J.T.H.) for funding.
T.A. is a Sloan, Cottrell, and Dreyfus Fellow. The APEX II X-ray
diffractometer was purchased via an NSF CRIF:MU Award to
Caltech (CHE-0639094).
(13) Kretzer, R. M.; Ghiladi, R. A.; Lebeau, E. L.; Liang, H.-C.; Karlin,
K. D. Inorg. Chem. 2003, 42, 3016.
(14) (a) Keyrouz, R.; Jouikov, V. New J. Chem. 2003, 27, 902.
(b) Fattakhova, D. S.; Jouikov, V. V.; Voronkov, M. G. J. Organomet.
Chem. 2000, 613, 170.
(15) (a) Hussman, G.; Wulff, W. D.; Barton, T. J. J. Am. Chem. Soc.
1983, 105, 1263. (b) Voronkov, M. G.; Basenko, S. V. J. Organomet.
Chem. 1995, 500, 325.
(16) Harlan, C. J.; Hascall, T.; Fujita, E.; Norton, J. R. J. Am. Chem. Soc.
1999, 121, 7274.
(17) Henthorn, J. T.; Agapie, T. Angew. Chem., Int. Ed. 2014, 53, 12893.
(18) (a) Miller, A. J. M.; Kaminsky, W.; Goldberg, K. I. Organometallics
2014, 33, 1245. (b) Amouri, H.; Vaissermann, J.; Rager, M. N.; Besace,
Y. Inorg. Chem. 1999, 38, 1211.
REFERENCES
■
(1) (a) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr. Chem. Rev.
2004, 104, 939. (b) Whittaker, J. W. Chem. Rev. 2003, 103, 2347.
(c) Sono, M.; Roach, M. P.; Coulter, E. D.; Dawson, J. H. Chem. Rev.
1996, 96, 2841.
(2) (a) Luca, O. R.; Crabtree, R. H. Chem. Soc. Rev. 2013, 42, 1440.
(b) Lyaskovskyy, V.; de Bruin, B. ACS Catal. 2012, 2, 270. (c) Praneeth,
V. K. K.; Ringenberg, M. R.; Ward, T. R. Angew. Chem., Int. Ed. 2012, 51,
10228. (d) Eisenberg, R.; Gray, H. B. Inorg. Chem. 2011, 50, 9741.
(e) Chirik, P. J.; Wieghardt, K. Science 2010, 327, 794. (f) Haneline, M.
R.; Heyduk, A. F. J. Am. Chem. Soc. 2006, 128, 8410. (g) Pierpont, C. G.
Coord. Chem. Rev. 2001, 216−217, 99. (h) Stanciu, C.; Jones, M. E.;
Fanwick, P. E.; Abu-Omar, M. M. J. Am. Chem. Soc. 2007, 129, 12400.
(3) (a) Umehara, K.; Kuwata, S.; Ikariya, T. J. Am. Chem. Soc. 2013,
135, 6754. (b) Carver, C. T.; Matson, B. D.; Mayer, J. M. J. Am. Chem.
Soc. 2012, 134, 5444. (c) Hull, J. F.; Himeda, Y.; Wang, W.-H.;
Hashiguchi, B.; Periana, R.; Szalda, D. J.; Muckerman, J. T.; Fujita, E.
Nat. Chem. 2012, 4, 383. (d) Costentin, C.; Drouet, S.; Robert, M.;
(19) Falceto, A.; Carmona, E.; Alvarez, S. Organometallics 2014, 33,
6660.
(20) Geiger, W. E. Coord. Chem. Rev. 2013, 257, 1459.
(21) (a) Stone, N. J.; Sweigart, D. A.; Bond, A. M. Organometallics
1986, 5, 2553. (b) Zoski, C. G.; Sweigart, D. A.; Stone, N. J.; Rieger, P.
H.; Mocellin, E.; Mann, T. F.; Mann, D. R.; Gosser, D. K.; Doeff, M. M.;
Bond, A. M. J. Am. Chem. Soc. 1988, 110, 2109.
(22) All potentials are referenced to Cp2Fe/Cp2Fe+.
(23) Maricle, D. L.; Hodgson, W. G. Anal. Chem. 1965, 37, 1562.
(24) (a) Fukuzumi, S.; Ohkubo, K. Coord. Chem. Rev. 2010, 254, 372.
(b) Fukuzumi, S.; Morimoto, Y.; Kotani, H.; Naumov, P.; Lee, Y.-M.;
Nam, W. Nat. Chem. 2010, 2, 756. (c) Morimoto, Y.; Kotani, H.; Park, J.;
Lee, Y.-M.; Nam, W.; Fukuzumi, S. J. Am. Chem. Soc. 2010, 133, 403.
(d) Tarantino, K. T.; Liu, P.; Knowles, R. R. J. Am. Chem. Soc. 2013, 135,
10022.
́
Saveant, J.-M. Science 2012, 338, 90. (e) Gunanathan, C.; Milstein, D.
Acc. Chem. Res. 2011, 44, 588. (f) Rakowski DuBois, M.; DuBois, D. L.
Chem. Soc. Rev. 2009, 38, 62. (g) Kohl, S. W.; Weiner, L.; Schwartsburd,
L.; Konstantinovski, L.; Shimon, L. J. W.; Ben-David, Y.; Iron, M. A.;
Milstein, D. Science 2009, 324, 74.
(25) Zheng, D.; Wang, Q.; Lee, H.-S.; Yang, X.-Q.; Qu, D. Chem.Eur.
J. 2013, 19, 8679.
(26) Wu, A.; Biehl, E. R.; Reeves, P. C. J. Chem. Soc., Perkin Trans. 2
1972, 449.
(27) Nishimi, T.; Kamachi, T.; Kato, K.; Kato, T.; Yoshizawa, K. Eur. J.
Org. Chem. 2011, 4113.
(4) (a) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010, 110,
6961. (b) Purse, B. W.; Tran, L.-H.; Piera, J.; Åkermark, B.; Backvall, J.-E.
̈
Chem.Eur. J. 2008, 14, 7500. (c) Chang, C. J.; Chng, L. L.; Nocera, D.
G. J. Am. Chem. Soc. 2003, 125, 1866. (d) Chaudhuri, P.; Hess, M.;
Muller, J.; Hildenbrand, K.; Bill, E.; Weyhermuller, T.; Wieghardt, K. J.
̈
̈
Am. Chem. Soc. 1999, 121, 9599. (e) Collman, J. P.; Devaraj, N. K.;
Decreau, R. A.; Yang, Y.; Yan, Y.-L.; Ebina, W.; Eberspacher, T. A.;
́
Chidsey, C. E. D. Science 2007, 315, 1565. (f) Lu, F.; Zarkesh, R. A.;
Heyduk, A. F. Eur. J. Inorg. Chem. 2012, 467.
(5) (a) Kim, S. B.; Pike, R. D.; Sweigart, D. A. Acc. Chem. Res. 2013, 46,
2485. (b) Reingold, J. A.; Son, S. U.; Kim, S. B.; Dullaghan, C. A.; Oh,
M.; Frake, P. C.; Carpenter, G. B.; Sweigart, D. A. Dalton Trans. 2006,
2385.
(6) Son, S. U.; Kim, S. B.; Reingold, J. A.; Carpenter, G. B.; Sweigart, D.
A. J. Am. Chem. Soc. 2005, 127, 12238.
(7) (a) Horak, K. T.; Velian, A.; Day, M. W.; Agapie, T. Chem.
Commun. 2014, 50, 4427. (b) Herbert, D. E.; Lara, N. C.; Agapie, T.
Chem.Eur. J. 2013, 19, 16453. (c) Suseno, S.; Agapie, T. Organo-
1464
DOI: 10.1021/ja5100405
J. Am. Chem. Soc. 2015, 137, 1458−1464