C-H Activation by RuCo3O4Oxo Cubanes: Effects of Oxyl Radical Character and Metal-Metal Cooperativity

Article abstract of DOI:10.1021/jacs.1c04069

High-valent multimetallic-oxo/oxyl species have been implicated as intermediates in oxidative catalysis involving proton-coupled electron transfer (PCET) reactions, but the reactive nature of these oxo species has hindered the development of an in-depth understanding of their mechanisms and multimetallic character. The mechanism of C-H oxidation by previously reported RuCo3O4 cubane complexes bearing a terminal RuV-oxo ligand, with significant oxyl radical character, was investigated. The rate-determining step involves H atom abstraction (HAA) from an organic substrate to generate a Ru-OH species and a carbon-centered radical. Radical intermediates are subsequently trapped by another equivalent of the terminal oxo to afford isolable radical-trapped cubane complexes. Density functional theory (DFT) reveals a barrierless radical combination step that is more favorable than an oxygen-rebound mechanism by 12.3 kcal mol-1. This HAA reactivity to generate organic products is influenced by steric congestion and the C-H bond dissociation energy of the substrate. Tuning the electronic properties of the cubane (i.e., spin density localized on terminal oxo, basicity, and redox potential) by varying the donor ability of ligands at the Co sites modulates C-H activations by the RuV-oxo fragment and enables construction of structure-activity relationships. These results reveal a mechanistic pathway for C-H activation by high-valent metal-oxo species with oxyl radical character and provide insights into cooperative effects of multimetallic centers in tuning PCET reactivity.

Full text of DOI:10.1021/jacs.1c04069

pubs.acs.org/JACS
Article
C−H Activation by RuCo3O4 Oxo Cubanes: Effects of Oxyl Radical
Character and Metal−Metal Cooperativity
Jaruwan Amtawong, Bastian Bjerkem Skjelstad, Rex C. Handford, Benjamin A. Suslick, David Balcells,*
and T. Don Tilley*
Cite This: https://doi.org/10.1021/jacs.1c04069
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: High-valent multimetallic-oxo/oxyl species have
been implicated as intermediates in oxidative catalysis involving
proton-coupled electron transfer (PCET) reactions, but the
reactive nature of these oxo species has hindered the development
of an in-depth understanding of their mechanisms and multi-
metallic character. The mechanism of C−H oxidation by
previously reported RuCo3O4 cubane complexes bearing a terminal
RuV−oxo ligand, with significant oxyl radical character, was
investigated. The rate-determining step involves H atom
abstraction (HAA) from an organic substrate to generate a Ru−
OH species and a carbon-centered radical. Radical intermediates
are subsequently trapped by another equivalent of the terminal oxo
to afford isolable radical-trapped cubane complexes. Density functional theory (DFT) reveals a barrierless radical combination step
that is more favorable than an oxygen-rebound mechanism by 12.3 kcal mol−1. This HAA reactivity to generate organic products is
influenced by steric congestion and the C−H bond dissociation energy of the substrate. Tuning the electronic properties of the
cubane (i.e., spin density localized on terminal oxo, basicity, and redox potential) by varying the donor ability of ligands at the Co
sites modulates C−H activations by the RuV−oxo fragment and enables construction of structure−activity relationships. These
results reveal a mechanistic pathway for C−H activation by high-valent metal−oxo species with oxyl radical character and provide
insights into cooperative effects of multimetallic centers in tuning PCET reactivity.
the mechanisms and probing the essence of metal−metal
cooperativity.
INTRODUCTION
■
Biological and synthetic oxidation processes often involve high-
valent multimetallic-oxo species as reactive intermediates.1−7
Water oxidation catalyzed by the oxygen-evolving complex
(OEC) in photosystem II, for example, employs a Mn4CaO4
cubane cluster (Scheme 1a), which converts water to O2
through a stepwise oxidation mechanism.8 Current hypotheses
invoke metal ion cooperativity in the requisite accumulation of
redox equivalents in the OEC via a series of proton-coupled
electron transfer (PCET) steps; the resultant high-valent
metal−oxo/oxyl species directly catalyzes the formation of the
relevant O−O bond.9−11 Oxo−diiron cores in nonheme
hydroxylases highlight another key enzymatic example in
which multimetallic-oxo complexes engage in oxidative
reactions, specifically in the context of C−H hydroxylation
(Scheme 1a).12 The postulated mechanism involves an initial
H atom abstraction (HAA) from a C−H bond by the bis(μ-
oxo)diiron(IV) core; a subsequent rapid radical rebound step
occurs between the resultant (μ-oxo)(μ-hydroxo)diiron core
and the substrate-derived radical to produce an alcohol
(Scheme 1b).13 The importance of these biological oxidation
processes has inspired significant interest in fully characterizing
While significant efforts have elucidated the mechanisms of
metal−oxo mediated PCET reactions,14−16 a limited subset of
these studies employ high-valent multimetallic-oxo com-
plexes.17−20 Indeed, reactive multimetallic-oxo species bearing
well-characterized terminal metal−oxo/oxyl motifs are exceed-
ingly rare despite their importance as key intermediates in
oxidative catalysis.5,9,21−23 To date, few spectroscopically and
computationally characterized high-valent multimetallic-oxo/
oxyl species exist.24 Meyer and co-workers have provided
evidence for a dioxo−RuV intermediate, [(bpy)2(O)RuVO-
RuV(O)(bpy)2]4+ (bpy = 2,2′-bipyridine), derived from PCET
reactions of the binuclear oxo “blue dimer” [(bpy)2(H2O)-
RuIIIORuIII(H2O)(bpy)2]4+ (Scheme 1c).25−27 This dioxo−
RuV intermediate appears to possess considerable spin density
Received: April 18, 2021
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
imental and computational analyses revealed radical character
at the terminal oxo ligand, with significant electronic
communication between the Co and Ru ions within the
cluster. These interesting features and the reactive nature of
this class of complexes prompted further mechanistic
investigations into multimetallic effects and oxyl radical
character in PCET reactions.
Scheme 1
Herein, the mechanism of C−H bond oxidation reactions
involving a terminal Ru−oxo/oxyl fragment of oxo cubane
complexes is described. Mechanistic analyses, as well as DFT
computational studies, implicate a rate-determining HAA step
with a subsequent radical combination between the resulting
carbon-centered radical and the terminal RuV−oxo of a second
cubane complex. Steric congestion and the C−H bond energy
of the substrate strongly influence the HAA reactivity of the
oxo cubanes to form organic products. Additionally, the
electronic properties of the multimetallic cubane core are
shown to influence the reactivity toward C−H bond cleavage
via metal−metal cooperativity.
RESULTS AND DISCUSSION
■
Stoichiometric Reactions with 9,10-Dihydroanthra-
cene. This laboratory recently reported 2-O (Scheme 1c) as a
versatile oxidant toward a range of organic substrates.34 The
current study focuses on the C−H activation chemistry of 1-O,
a related derivative ligated by p-methoxypyridine. Significantly,
the distinct 1H NMR signature of the methoxy groups and the
higher degree of crystallinity relative to other cubane
derivatives make 1-O well suited for reactivity and mechanistic
studies. We first examined stoichiometric reactions between 1-
O and 9,10-dihydroanthracene (DHA) in chloroform-d (Table
S4). With 1 equiv of DHA, full conversion of 1-O was
at the terminal oxo ligands, and a radical coupling between the
two terminal RuV−oxo moieties has been proposed to mediate
O−O bond formation.5,28 Indeed, transient intermediate RuV−
oxo species are commonly invoked in water and organic
oxidation catalysis.29−31 Although metal−oxyl species are
known to exhibit unique reactivity,24 it is unclear how the
spin density localized on oxo ligands contributes to the PCET
reactivity of multimetallic-oxo clusters.32,33 Studies of high-
valent multimetallic-oxo clusters bearing terminal oxo/oxyl
moieties present an opportunity to probe the influence of
metal−metal cooperativity and provide insight into the role of
oxyl radical character in PCET reactions.
This laboratory recently reported a series of mixed-metal
cubane clusters RuV(O)CoIII3(μ3-O)4(OAc)4(4-R-py)3 (1-O−
4-O, R = OMe, H, CF3, and Me, respectively; Scheme 1c)
featuring a rare terminal RuV−oxo and preliminary studies of
their oxidative reactivity toward organic substrates.34 Exper-
1
observed after 22 h at 23 °C as determined by H NMR
spectroscopy; in contrast, only 51% of DHA was consumed to
primarily produce a new paramagnetic species (54% yield with
respect to 1-O). Interestingly, only a trace amount of
anthracene was observed as an organic product. When an
excess of cubane (2 equiv) was exposed to DHA (1 equiv),
both reactants were fully consumed to afford the same
paramagnetic species in 50% yield with respect to 1-O
(Scheme 2).
Single crystals of the paramagnetic species were isolated and
characterized as the radical-trapped cubane complex Ru-
(OC14H11)Co3(μ3-O)4(OAc)4(4-OMe-py)3, 1-OHAn. This
1
species exhibits H NMR spectroscopic features consistent
a
Scheme 2. Synthesis of 1-OHAn
a
Inset: solid-state molecular structure of 1-OHAn as determined by XRD. Solvents and hydrogen atoms are omitted for clarity. Thermal ellipsoids
are shown at 50% probability.
B
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
with a Cs symmetric structure (Figure S43), in good agreement
with the solid-state molecular structure as determined by
single-crystal X-ray diffraction (XRD) (Scheme 2, inset). The
Ru1−O1 distance of 1.928(2) Å is consistent with
corresponding bond lengths for reported RuIV−OR alkoxide
complexes.34 An Evans method measurement of the magnetic
moment of 1-OHAn (3.2 μB) supports an S = 1 configuration
for this formally RuIVCoIII3O4 cubane. Consistently, geometric
optimization of various spin states of 1-OHAn by DFT
indicated that the triplet is the lowest energy spin state (see the
Supporting Information).
The observed stoichiometry in Scheme 2 implies that a likely
byproduct is the formally RuIV−OH species (1-OH), which
accounts for the consumption of a second equivalent of 1-O as
well as the loss of a hydrogen atom from DHA. Although this
species could not be isolated, 1H NMR spectroscopic
monitoring of the reaction between 1-O and DHA indicated
the production of 1-OH. Thus, in addition to proton
resonances corresponding to 1-OHAn, a new set of peaks
grew over 1 h and slowly decreased over the course of 4 h
(Figure S44). The number of peaks (8) and their relative
integration are consistent with the expected inequivalent
protons of acetate and p-methoxypyridine ligands of a Cs
symmetric cubane species which is tentatively assigned as 1-
OH (a chemical shift for the Ru−OH could not be identified).
Furthermore, analysis of an aliquot of the reaction mixture by
HR-ESI-MS in acetonitrile after 4 h revealed the presence of
the expected RuIV−OH species (m/z = 923.9091, [Ru(OH)-
Co3(μ3-O)4(OAc)4(4-OMe-py)3]·H+), 1-OHAn, and formally
RuIV species such as [Ru(L)Co3(μ3-O)4(OAc)4(4-OMe-py)3]+
(L = MeCN, 4-OMe-py) (Table S22, Figures S45 and S46).
The relatively weak signal of 1-OH disappeared within a few
minutes after sample injection, suggesting that this compound
is not persistent under the ionization conditions. Note that an
O−H stretching frequency could not be identified by solution-
state IR measurements, indicating the instability of 1-OH
under these conditions.
Figure 1. (a) Plot of kinit vs [DHA]0 ([1-O]0 = 9.0 mM, kobs = 7.8 ×
10−5 s−1). (b) Plot of kinit vs [1-O]0 ([DHA]0 = 9.0 mM). (c) Eyring
analysis of the reaction of 1-O and DHA. (d) Plot of log kobs vs
BDEC−H of organic substrates with linear regression (R2 = 0.9732).
Observed rate constants (kobs) were obtained from the slope of the
linear correlation between kinit and [substrate]0.
To gain further information about the HAA mechanism, 1-
O was treated with xanthene, fluorene, and 1,4-cyclohexadiene
(CHD). Notably, the radical-trapped cubane species and 1-
OH were observed in the reactions with each of these H atom
1
donors (by H NMR spectroscopy and HR-ESI-MS, Tables
S4−S7 and Figures S1−S8). Kinetic analyses revealed a linear
correlation between the reaction rates and the bond
dissociation energy (BDE) values of the substrates15 (Figure
1d). The slope of −0.35 is within the reported range for H
atom abstraction reactions mediated by several other
mononuclear Ru−oxo complexes, −0.21 to −0.61.36−39 The
absolute value of the slope (0.35) is reasonably close to the
theoretical slope of 0.5 predicted by the Bell−Evans−Polanyi
correlation between activation energy and enthalpic driving
force, which indicates an H atom transfer mechanism.40,41
However, no apparent correlation between the reaction rates
and the pKa and ionization energies of the C−H substrates
exists (Figures S47 and S48). Note that the reactivity of 1-O
exhibits an inverse relationship with the pKa values of fluorene,
DHA, and CHD; namely, 1-O reacts faster with substrates
having higher pKa values. There is, however, a large rate
difference of a half order of magnitude for DHA and xanthene
despite the similarity in their pKa values (30 and 30.1,
respectively, in DMSO).42 These observations support a
concerted HAA mechanism over a stepwise transfer of the
proton and electron.42
Mechanism of C−H Bond Activation by 1-O. Kinetic
1
experiments using H NMR spectroscopy were performed to
elucidate the operative mechanism of this C−H activation
process. Monitoring the reaction of 1-O and DHA at 25 °C
revealed that the initial decay rate of DHA (kinit) depends
linearly on the concentrations of both reactants (i.e., first order
with respect to each substrate; Figures 1a,b and Figures S16
and S17), and the overall second-order reaction proceeds with
an observed rate constant of (1.1 0.1) × 10−2 M−1 s−1. An
Eyring analysis elucidated the temperature dependency on the
reaction and provided activation parameters of ΔH‡ = 8
1
kcal mol−1 and ΔS‡ = −41
4 eu, which correspond to a
ΔG‡(298 K) of 20 3 kcal mol−1 (Figure 1c). The negative
entropy of activation is consistent with a bimolecular
mechanism.
Given that 1-OH is not isolable, we sought to determine its
bond dissociation free energy (BDFE) by evaluating reactions
of 1-O with substrates of various C−H bond strengths and by
carrying out DFT calculations. Compound 1-O reacted rapidly
in stoichiometric reactions with substrates exhibiting BDFEC−H
∼ 70 kcal mol−1 (e.g., xanthene and CHD)42 under ambient
conditions. However, with a 1000-fold excess of THF (DFT-
predicted BDFEC−H ∼ 78.5 kcal mol−1) as a substrate, 1-O
slowly activated the α-C−H position of THF to generate the
radical-trapped cubane species Ru(OC4H7O)Co3(μ3-
Kinetic isotope effects (KIEs) were determined to further
probe the nature of the C−H activation.35 The reaction of 1-O
with DHA-d4 provided a large primary KIE value of 20 2,
which implicates a rate-limiting C−H bond cleavage in the
operative mechanism.15 Indeed, KIE values greater than ca. 7
typically reflect a contribution from tunneling effects,14 which
was corroborated by DFT calculations for this reaction. By
treating only the electrons quantum mechanically (i.e., nuclei
are treated classically, without tunneling effects), a significantly
lower KIE value of 6.8 is predicted.
1
O)4(OAc)4(4-OMe-py)3 (1-OTHF) (by H NMR spectros-
copy and HR-ESI-MS; Table S8, Figures S9 and S10). This
reaction sets the BDFEC−H of THF as an approximate upper
C
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
bound for the BDFEO−H of 1-OH. Additionally, DFT
calculations predict an H atom abstraction barrier of 27.1
kcal mol−1 for the reaction between 1-O and THF, which is at
the upper limit under the conditions considered here.
Consistently, DFT calculations predict a BDFEO−H value for
1-OH of 65.5 kcal mol−1 in acetonitrile at 23 °C.
aforementioned reported complexes, 1-O is far less oxidizing,
likely a consequence of the lower overall charge of 1-O and the
strong π-electron donation from the Co3O4 fragment to the
RuV−oxo moiety.34 Nevertheless, the weak oxidizing ability of
1-O is compensated by the high pKa value of 1-OH to afford a
moderate BDFEO−H. In fact, the pKa of 1-OH is one of the
highest reported values among metal−oxo/hydroxo com-
plexes.43
Mechanism of O−C Bond Formation. The kinetic
analyses indicate that the first and rate-determining step of
the reaction between 1-O and DHA occurs via a concerted
HAA mechanism. However, there are two plausible pathways
after this rate-limiting step to account for formation of 1-
OHAn (Scheme 3). After the HAA step (step 1), the
corresponding 9-hydroanthracyl radical may be readily trapped
by a second molecule of 1-O to form 1-OHAn (step 2A).
Alternatively, the 9-hydroanthracyl radical may undergo
rebound with 1-OH to form an anthracenol-bound Ru(III)
intermediate that then acts as a hydrogen-atom donor to an
unreacted equivalent of 1-O to give 1-OHAn and 1-OH (step
2B). Both mechanisms are possible as they provide the same
overall net reaction described in Scheme 2. Note that the
conventional oxygen rebound mechanism12 (Scheme 1b) is
discounted since 9-anthracenol and the formally Ru(III)
cubane species were not observed.
DFT was employed to probe the degree of oxyl character in
1-O and 1-OH and to provide insight into which species
couples more readily with 9-hydroanthracyl radical. Optimiza-
tion of the doublet state of 1-O yielded an energy minimum in
which the spin density of the unpaired electron (ρ) is mostly
located on Ru (ρ = 0.56α) and the terminal oxo ligand (ρ =
0.36α) (Figure 2a). Conversely, the hydroxyl ligand of 1-OH
in the triplet state contains a smaller portion of the spin density
(ρ = 0.25α) (Figure 2b). The stronger localization of the spin
density on the terminal oxo of 1-O than on the hydroxo ligand
of 1-OH suggested that the 9-hydroanthracyl radical
preferentially couples with 1-O to form 1-OHAn in this
Using the calculated BDFEO−H and the RuV/RuIV cathodic
peak potential (Ep,c) of 1-O (−0.95 V vs Fc/Fc+ in
acetonitrile)34 as an approximation of E1/2, a thermodynamic
square scheme was constructed (Figure S40) to estimate a pKa
value of 25.4 for 1-OH (see the Supporting Information for
details). The BDFEO−H of 1-OH is comparatively lower than
those of other well-studied mononuclear Ru−oxo/hydroxo
complexes (Table 1). When compared to E0 values of the
Table 1. Thermodynamic Parameters for Selected Ru−Oxo
Complexes
E0 (vs
pKa
BDFE
a
b
O−H c
Fc/Fc+) (MeCN) (kcal mol−1
Run+−oxo complex
)
ref
this
RuV(O)CoIII3(μ3-
O)4(OAc)4(4-OMe-py)3
(1-O)
−0.95
25.4
65.5
work
d
[RuIV(O)(TPA)(H2O)]2+
1.67
1.74
9.1
9.4
104
44
[RuIV(O)(N4Py)
91.1
83.9
90.4
44
38
38
e
(H2O)]2+
trans-[RuVI(O)2(14-
1.19
1.57
2.8
1.2
f
TMC)]2+
trans-
g
[RuVI(O)2(Me2BAdC]2+
a
E0 for Run+−oxo/Ru(n−1)+−oxo adjusted vs Fc/Fc+ as necessary.45
b
pKa in H2O for Ru(n−1)+−OH/Ru(n−1)+−oxo converted to that in
MeCN by using the correlation between pKa values in H2O and
MeCN as necessary.46 BDFEO−H in MeCN calculated by using
c
Bordwell’s equation with adjusted thermodynamic parameters and CG
d
value of 52.6 kcal mol−1 as necessary.47 TPA = tris(2-pyridylmethyl)-
e
amine. N4Py = N,N-(dipyridin-2-ylmethyl)bis(pyri-din-2-ylmethyl)-
f
amine. 14-TMC = 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetrade-
g
cane. Me2BAdC = 1,12-dimethyl-3,4:9,10-dibenzo-1,12-diaza-5,8-
dioxacyclopentadecane.
Scheme 3. Possible Mechanisms for Formation of 1-OHAn
D
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
may take place in the triplet stateeither way, the radical
coupling is faster than the rebound, which has a higher energy
3
barrier of 12.3 kcal mol−1. In TSRC, the Ru−O distance is
slightly longer than that of 1-O (1.74 vs 1.70 Å), and the O−C
distance is rather long (2.56 Å). The reactant-like nature of
3TSRC indicates that it should be easily accessible without
significant changes to the structure of the cubane. However,
the barrier height (9.3 kcal mol−1) and the significant amount
of α spin density localized on both O and C (ρ = 0.33α and
0.38α, respectively) suggest electronic repulsion between these
two moieties and provide a plausible explanation for the
greater barrier height of the triplet state relative to the lack of
barrier in the singlet.
The lack of anthracene formation prompted an investigation
into the free energy barriers for possible pathways to generate
anthracene. DFT predicts an energy barrier of 13.5 kcal mol−1
for an HAA reaction between 1-OH and 9-hydroanthracyl
radical, which is slightly higher than the barrier height of the
radical rebound (12.3 kcal mol−1). Because the formation of
anthracene is not an experimentally observed pathway, the
small energy difference between the two processes (1.2 kcal
mol−1) implies that the radical rebound is also unfavorable,
further strengthening the proposed radical combination
mechanism.
Figure 2. (a) Spin density of 1-O (ρ(Ru:O) = 0.56:0.36α). (b) Spin
density of 1-OH (ρ(Ru:O) = 1.4:0.25α). Isovalue is 0.02 for both 1-
O and 1-OH. All hydrogen atoms except that of the hydroxo ligand of
1-OH were removed for clarity.
reaction (Scheme 3, step 2A). This hypothesis was
corroborated by computing the associated reaction pathways.
The DFT-calculated energy profiles of the plausible reaction
mechanisms for formation of 1-OHAn solvated by chloroform
(conductor-like polarizable continuum model, C-PCM)48,49 at
296.15 K are shown in Figure 3. The first step involves
formation of 1-OH and 9-hydroanthracyl radical (HAn•) via
HAA from DHA by 1-O (Figure 3, top). This reaction
traverses the doublet transition state 2TSHAA with a rate-
determining barrier of 18.9 kcal mol−1. Over the course of the
reaction, the spin density steadily accumulates on Ru (ρ =
0.56α, 1.06α, and 1.40α in 1-O, 2TSHAA, and 1-OH,
respectively). In contrast, the spin density on the oxo decreases
Additionally, the HAA reaction between 1-O and 9-
hydroanthracyl radical to form anthracene was found to have
a barrier of 7.2 kcal mol−1 in the triplet state. Because this
reaction is not observed experimentally, the fact that the
barrier is lower than that of the triplet radical combination (9.3
kcal mol−1) implies that the latter pathway can be discounted.
Note that the singlet transition for the HAA reaction between
1-O and 9-hydroanthracyl radical was found to be barrierless,
but this possibility is not preferred since the triplet state
transition is lower in energy (Figure S56). Taken together,
DFT calculations reveal a barrierless radical combination step
between 1-O and 9-hydroanthracyl radical in the singlet state
until the transition state, after which a slight increase of spin
2
density occurs (ρ = 0.36α, 0.12α, and 0.25α in 1-O, TSHAA
,
and 1-OH, respectively). β spin density originating from the
increasing radical character of DHA is shared between the
carbon undergoing HAA and the terminal oxo. The observed
polarization of spin density toward the Ru−O···H···C moiety is
consistent with a concerted HAA mechanism.
The oxygen rebound mechanism in the reaction of 1-OH
with the 9-hydroanthracyl radical to yield the hypothesized
anthracenol-bound Ru(III) intermediate (2POR) (step 2B in
Scheme 3) was found to have a feasible energy barrier of 12.3
kcal mol−1 (Figure 3, top). However, a competing reactiona
radical combination between starting material 1-O and the 9-
hydroanthracyl radical to yield the observed 1-OHAn species
(step 2A in Scheme 3)was found to be essentially barrierless
(vide infra) and should therefore outcompete the oxygen
rebound pathway.
3
and a subsequent singlet-to-triplet SCO yielding 1-OHAn.
The divergence in mechanism from a conventional oxygen
rebound is reminiscent of several reports on mononuclear
nonheme metal−oxo complexes.50−52 In these systems, the
oxygen rebound mechanism is disfavored in preference for a
pathway involving dissociation of the substrate radical from the
metal−hydroxo intermediate and further reaction with an
oxidizing metal center or a radical trap.53 Consistent with the
latter scenario, 1-O may be regarded as a radical trap that
intercepts the diffusing 9-hydroanthracyl radical to form 1-
OHAn. An analogous radical coupling reaction was observed
in the HAA reaction of methane with a ZnII−oxyl species
stabilized in an MFI-type zeolite.54 In this system, after a ZnII−
OH species and a methyl radical are generated in the HAA
step, another site of ZnII−oxyl readily traps the methyl radical
to form a ZnII−OCH3 species.
In the radical-combination pathway, two different electronic
states, singlet and triplet, were considered; the spin state
depended on whether the spins at the terminal oxo and the
carbon-centered radical were aligned antiparallel or parallel
(Figure 3, bottom). The energy barrier required for the triplet
To further support the dissociation of the free radical from
1-OH, a 2:1 reaction of 1-O with DHA was performed under
an atmosphere of O2, a commonly used radical trap.53 In
contrast to the identical reaction performed in an N2
atmosphere (vide supra), anthraquinone was observed as a
major product (66%) while only a small amount of 1-OHAn
3
transition state TSRC was found to be 9.3 kcal mol−1. The
singlet transition, however, was found to be barrierless due to
spin coupling toward bond formation.
Because of the small energy difference between the singlet
and triplet states in the reactant complexes of the radical-
combination pathway (ΔG(1RRC) − ΔG(3RRC) = 0.6 kcal
mol−1), the reaction may take place in the singlet state, without
an energy barrier, and yield the most stable triplet product, 31-
OHAn, after spin crossover (SCO). Alternatively, the reaction
1
(15%) was observed after 22 h as determined by H NMR
spectroscopy (Table S4). The interference of O2 with this
reaction resulted in the formation of anthraquinone and the
reduction in yield of 1-OHAn. These observations suggest that
the substrate radical likely diffuses away from 1-OH and is
E
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 3. Free energy profiles for the plausible reaction pathways, including the HAA and oxygen rebound steps in the doublet state (top), and the
radical combination step in the singlet and triplet states (bottom). Reaction coordinate, ξ, is given on the x-axis, and the relative free energy, ΔG, is
given in kcal mol−1 on the y-axis. The labels R, TS, and P refer to reactant complex, transition state, and product complex, respectively. Spin
multiplicities are given as superscripts preceding the labels, whereas subscripts following the labels describe the reaction mechanism, with HAA,
OR, and RC referring to H atom abstraction, oxygen rebound, and radical combination mechanisms, respectively.
subsequently intercepted by O2 and 1-O, which lend support
to the radical combination mechanism in this chemistry.
Given that 1-O is present in much higher concentrations
than the substrate-based radicals generated in these stoichio-
metric reactions, it is of interest to determine whether
substantially decreasing the concentration of 1-O could favor
the rebound of organic radicals at 1-OH or the dimerization of
substrate radicals. Notably, a reaction with a 20:1 ratio of DHA
to 1-O in CDCl3 afforded 1-OHAn in ∼50% yield with respect
trityl radical to quantitatively produce a radical-trapped cubane
complex, Ru(OC19H15)Co3(μ3-O)4(OAc)4(4-OMe-py)3 (1-
OT), after 1 h at 23 °C (Scheme 4). Single-crystal XRD
analysis, as well as other characterization techniques (i.e.,
1
elemental analysis, H NMR spectroscopy, and HR-ESI-MS),
confirmed the identity of the product as 1-OT (Figures S49
and S50). This type of radical coupling reaction where organic
radicals are readily intercepted with the terminal oxyl of 2-O
has been demonstrated previously by this laboratory.34
Separately, treatment of 1-O (2 equiv) with triphenylmethane
(1 equiv) afforded 1-OT in quantitative yield with respect to
triphenylmethane (50% yield with respect to 1-O) as indicated
1
to 1-O after 24 h as determined by H NMR spectroscopy.
Additionally, neither 9-anthracenol nor 9,9′-bianthracene was
observed. The exclusive formation of 1-OHAn in the highly
diluted solution of 1-O indicates a strong propensity of 1-O for
coupling with a substrate radical.
To further support the feasibility of the radical combination
step, 1-O was treated with Gomberg’s dimer as a source of
1
by H NMR spectroscopy (Scheme 4). The observed product
distribution is consistent with the stoichiometry established for
the reaction with DHA. Taken together, these results provide
strong evidence that coupling of the substrate radicals with the
F
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
a
Scheme 4. Synthesis of 1-OT and Subsequent Protonolysis
a
Inset: solid-state molecular structure of 1-OT as determined by single-crystal XRD. Solvents and hydrogen atoms are omitted for clarity. Thermal
ellipsoids are shown at 50% probability.
terminal oxo/oxyl of 1-O may be the operative mechanism in
this chemistry.
been consumed to produce benzene in 57% yield. Additional
proton resonances consistent with formation of 1-OH were
also observed (Figure S7). HR-ESI-MS analysis of the reaction
mixture in acetonitrile revealed the presence of a [Ru-
(OC6H7)Co3(μ3-O)4(OAc)4(4-OMe-py)3]·H+ ion (1-
OCHD, m/z = 1001.8933) though only a trace amount of
this radical-trapped species was observed by 1H NMR
spectroscopy (Figure S8).
The formation of 1-OH and 1-OCHD suggests that the first
step of this reaction is abstraction of a hydrogen atom from
CHD followed by radical combination involving the cyclo-
hexadienyl radical (•CHD) and a second molecule of 1-O (eqs
1 and 2), in a manner similar to that observed in the reaction
with DHA. However, it appears that a second HAA reaction at
1-OCHD occurs to produce benzene. The most likely H atom
abstractor for the latter step is 1-OH to form a RuIII−OH2
cubane species (1-OH2) (eq 3). This is consistent with the
RuIII−cubane species observed by HR-ESI-MS. Taken
together, these steps give the overall reaction depicted in eq
4, which is consistent with the observed stoichiometry. This
implies that 1-OCHD is a reactive intermediate in the C−H
activation reaction of CHD and functions as an H atom donor
to 1-OH. Nevertheless, a HAA reaction between 1-OH and
•CHD cannot be ruled out and may be a competing pathway
to produce benzene.
Interestingly, an overall C−H hydroxylation reaction was
achieved by protonolysis of 1-OT with bistriflimidic acid,
resulting in quantitative formation of triphenylmethanol and a
formally RuIV species, [Ru(L)Co3(μ3-O)4(OAc)4(py)3]+ (L =
4-OMe-py) as determined by 1H NMR spectroscopy and HR-
ESI-MS (Scheme 4 and Figure S51). Note that formation of
the p-methoxypyridine adduct observed by HR-ESI-MS is
likely a consequence of fragmentation of cubane species under
ionization conditions. This experiment demonstrates that the
radical-trapped cubane species may be viable intermediates for
conversion of hydrocarbons to alcohols.
Influence of Steric Congestion and C−H Bond Energy
of Radical Substrates on Formation of Organic
Products. Previous work from this laboratory demonstrated
the oxidation of 1,4-cyclohexadiene (CHD) by 2-O to afford
benzene and the reduced RuIIICo3O4 cubane with a 1:1
reaction stoichiometry.34 Similarly, compound 1-O reacts
cleanly with CHD to produce benzene in quantitative yield
(by 1H NMR spectroscopy) and a formal RuIII species,
detected by HR-ESI-MS after 24 h at 23 °C as [Ru(L)Co3(μ3-
O)4(OAc)4(4-OMe-py)3]·H+ (L = MeCN, 4-OMe-py). With 2
equiv of 1-O for each equivalent of CHD, 53% of 1-O was
consumed to fully convert CHD to benzene; thus, this result
stands in contrast to that observed under analogous conditions
with DHA. These observations raise two important mecha-
nistic questions. First, what is the origin of the discrepancy
between the stoichiometries used to fully convert reactants in
the oxidations of CHD (1:1; cubane:CHD) vs other substrates
(2:1; cubane:substrate)? Second, why is benzene formed
quantitatively in the reaction with CHD, but with other
substrates only trace amounts of organic products are
detected?
RDS
•
(1)
(2)
(3)
(4)
1‐O + CHD ⎯⎯⎯→ 1‐OH + CHD
•
1‐O + CHD → 1‐OCHD
1‐OH + 1‐OCHD → 1‐O + C6H6 + 1‐OH2
net: 1‐O + CHD → C6H6 + 1‐OH2
The above observations imply that 1-OH might also undergo a
HAA reaction with 1-OHAn or 9-hydroanthracyl radical to
generate anthracene. However, as mentioned above, only a
small amount of anthracene (<5%) was produced in the
reaction between 1-O and DHA after 22 h at 23 °C. At 50 °C,
To answer these questions, a stoichiometric reaction of 1-O
with CHD was monitored by 1H NMR spectroscopy. After 1 h
at 23 °C, 62% and 59% of 1-O and CHD, respectively, had
G
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
this reaction slowly converted to anthracene and anthraqui-
none in 16% and 30% yield, respectively, after 24 h (by H
NMR spectroscopy). The greater quantities of organic
products formed with increasing temperature suggest that the
pathways leading to anthracene and anthraquinone occur at
relatively slow rates.
reactivity. Relative bond energies of the radical-trapped cubane
intermediates vs those in 1-OH2 may also play a pivotal role in
determining the rates of subsequent HAA and formation of
organic products.
1
Influence of Oxyl Radical Character and Metal−Metal
Cooperativity on C−H Bond Activations. Significant
metal−metal interactions within the RuCo3O4 cubane clusters
give rise to redox properties that can be tuned through
modulation of the electron-donating properties of the Co-
bound pyridine ligands.34 These cooperative interactions also
influence the C−H activation reactivity of these cubane
clusters. Notably, in reactions between DHA and Ru(O)-
Co3(μ3-O)4(OAc)4(4-R-py)3 cubanes (1-O−4-O, R = OMe,
H, CF3, and Me), the HAA reaction rate decreases linearly
with the conjugate acid pKa values of the 4-R-pyridine ligand in
water (Figure 4a).56 To probe the origin of this variable
reactivity, the electronic properties of the cubanes as a function
of substituents were investigated by using DFT calculations.
The lower reactivity of 1-OHAn with respect to 1-OCHD in
HAA reactions indicates that interactions with 1-OH are
influenced by the steric bulk of the trapped substrate radical
(Ru−alkoxide cubane). For 1-OHAn, the steric congestion
introduced by the tricyclic 9-hydroanthracyl moiety may
hinder the approach of 1-OH as an H atom acceptor. The
persistence of the radical-trapped cubane species Ru-
(OC13H9O)Co3(μ3-O)4(OAc)4(4-OMe-py)3 (1-OX) and Ru-
(OC13H9)Co3(μ3-O)4(OAc)4(4-OMe-py)3 (1-OFl) in the
reactions with related tricyclic substrates, xanthene, and
fluorene further supports this hypothesis (Figures S1−S8).
However, in a 2:1 reaction of 1-O with xanthene, 1-OX slowly
converted to xanthone in 15% yield after 25 h at 23 °C (by 1H
NMR spectroscopy). Xanthene possesses a relatively weaker
C−H bond (ΔBDEC−H ∼ 0.8 kcal mol−1) but is more sterically
demanding than CHD. The slow conversion of 1-OX therefore
implies that steric hindrance and C−H bond energies of the
bound substrates are competing factors that contribute to the
formation rates of organic products.
To further characterize the steric and electronic effects, 1-O
was treated with an excess of cyclohexene (20 equiv), a
relatively small allylic C−H substrate with a relatively strong
C−H bond (BDEC−H = 87 3 kcal mol−1),55 in chloroform-d
(Scheme 5). After 23 h at 23 °C, no organic products were
a
Scheme 5. Synthesis of 1-OCH
Figure 4. (a) Plot of log(kobs) of reactions between Ru(O)Co3(μ3-
O)4(OAc)4(4-R-py)3 cubanes (1-O−4-O, R = OMe, H, CF3, and
Me) and DHA vs conjugate acid pKa value of 4-R-pyridine with R2 =
0.9993. (b) Plot of log(kobs) vs spin density at terminal oxo ligand
with R2 = 0.9939. (c) Plot of log(kobs) vs local natural charge of
terminal oxo ligand with R2 = 0.9916. (d) Plot of log(kobs) vs
reduction potential (in o-C6H4F2) with R2 = 0.9939. The observed
rate constants (kobs) were obtained from the slope of the linear
correlation between kinit and [substrate]0.
a
Inset: Solid-state molecular structure of 1-OCH as determined by
single-crystal XRD. Solvents and hydrogen atoms are omitted for
clarity. Thermal ellipsoids are shown at 50% probability.
The electronic tunability of the cubane clusters, and
hypotheses concerning the possible roles of oxyl radical
character in C−H activation of metal−oxo com-
plexes,32,33,57−61 prompted a study of the effect of spin density
on HAA reactivity. Metal−oxyl species are thought to be more
reactive than closed-shell systems in C−H bond activation
since less energy is required to reorganize spins prior to
accepting an H atom.32,59,62 An alternative perspective suggests
that spin density at the oxo ligand can influence the reactivity
but cannot be a dominant factor.33 Significantly, the influence
of spin density may be investigated with a metal−oxo system in
which the degree of ground-state oxyl radical character can be
fine-tuned and correlated to C−H activation reactivity.
Notably, the spin density localized on the terminal oxo moiety
(ρO) of the RuCo3O4 cubanes increases linearly with the
electron-withdrawing ability of the pyridine ligand set (Figure
S53) and with the rates of the reactions between 1-O−4-O and
observed; however, new resonances consistent with formation
of 1-OH and a new radical-trapped cubane species were
1
observed by H NMR spectroscopy (Figures S11 and S12).
The resulting cyclohexenyl complex, Ru(OC6H9)Co3(μ3-
O)4(OAc)4(4-OMe-py)3 (1-OCH), was isolated in moderate
yield (22%) compared to the theoretical yield (50%) with
respect to 1-O as the limiting reactant. Recrystallization from
dichloromethane/hexamethyldisiloxane afforded X-ray quality
single crystals of 1-OCH (Scheme 5, inset). The C−C single
bond lengths of the cyclohexenyl moiety span the range of
1.482(5)−1.523(5) Å, while the short C28−C29 bond length
(1.310(5) Å) is consistent with the presence of a cyclohexenyl
moiety.
The isolation of 1-OCH and its persistence further suggest
that steric effects are not solely responsible for the observed
H
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
DHA (1-O, R = OMe < 4-O, R = Me < 2-O, R = H < 3-O, R =
CF3) (Figure 4b). The total increase in spin density over this
series (ΔρO = 0.05α) is relatively small, which is consistent
with the small change in free energy of activation (ΔΔG‡ ∼ 1.1
kcal/mol) at 298 K. The increase in spin density may be
rationalized based on simplified π-bonding interactions of the
Ru−oxo moiety (Figure S54). The reduction of π-electron
donation from the [Co3O4] framework induced by electron-
withdrawing substituents34 lowers the energies of the π-
Ru(dxz/dyz) orbitals. This results in a net decrease in energy for
the antibonding π*-Ru−oxo(d−p) orbital (SOMO), in turn
increasing the contribution of O(px/py) orbitals to the SOMO.
This explanation is consistent with the lower SOMO energy34
and stronger oxyl radical character of the more electron-
deficient clusters. Notably, enhanced oxyl radical character may
provide a stronger driving force for H atom transfer from DHA
to the Ru−oxo center, consistent with the observed reactivity
trend and the influence for spin density contribution in
increasing C−H oxidation reactivity.
While the spin density descriptor applies to metal−oxyl
systems or organic radicals undergoing H atom transfer
reactions in which the proton and electron travel in concert,63
more general PCET reactivity determinants for metal−oxo
complexes are the charge on the proton-receiving oxo ligand
(basicity) and the reduction potential at the metal center
(oxidizing power).62,64 Because the cubanes belong to an
intermediate class where the spin is delocalized over the Ru−
oxo bond, their oxidizing power and/or the basicity should also
contribute to the observed reactivity. In fact, the rates of
reactions between 1-O−4-O and DHA are linearly dependent
on both the local natural charges of the terminal oxo (qO) and
the experimentally determined reduction potentials (Ep,c) in o-
difluorobenzene34 (Figures 4c,d). The local natural charges
predicted by DFT reveal a decrease (ΔqO ∼ 0.02e) as the
cluster becomes more electron deficient (i.e., less basic), which
should result in a lower driving force for proton transfer in a
HAA reaction. Therefore, the increase in reactivity is predicted
to be dominated by the decrease in reduction potential (ΔEp,c
∼ 0.13 V vs Fc/Fc+) as the reduction is rendered more facile
by electron-withdrawing ligands. Note that the locus of
reduction in these complexes is centered at the Ru−oxo as
evidenced by the LUMOs located upon the Ru−oxo moieties
(Figure S55).
radical-like) Ru−oxo moieties. DFT calculations predict a
barrierless radical combination between 9-hydroanthracyl
radical and the terminal Ru−oxo whereas the radical rebound
at the Ru−hydroxo species is less favorable, consistent with the
higher spin density localized upon oxygen of the terminal Ru−
oxo vs Ru−hydroxo. These results suggest that the degree of
oxyl character could be a critical factor in determining the
preferred pathway(s) for C−H activation by high-valent
metal−oxo/oxyl species, and such a radical combination
reaction may be a common pathway in other metal−oxyl
mediated PCET reactions.
The strong electronic communication within the cubane
core allows for establishing structure−function relationships.
The reactivity of the terminal Ru−oxo moiety is readily
manipulated by the donating properties of the pyridine ligand
set for the Co sites. Notably, it was found that electron-
deficient clusters exhibit faster reaction rates toward C−H
cleavage. The origins of the observed reactivity may be
attributed to the reduction potential and the spin density
localized upon the terminal oxo ligand as important properties
that can be tuned in these cubane clusters. Such relationships
may be important in other multimetallic oxo systems wherein
the PCET reactivity of the terminal oxo/oxyl can be fine-tuned
without changing the formal oxidation state of the supporting
metal center; however, subtle electronic changes of nearby
metals directly influence the electronic properties and thus
reactivity of the overall cluster. This cooperativity may also
play a role in enzymatic multimetallic catalysts in which the
highly dynamic local environments and ligand binding to the
active sites could greatly affect the reactivity of the reactive oxo
moiety.65 The fundamental understanding developed in this
study regarding multimetallic cooperative effects should prove
useful in designing and evaluating multimetallic oxo catalysts.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c04069.
Experimental and computational details (PDF)
Accession Codes
The strong dependence of the reaction rates on the
electronic properties of the cubanes demonstrates that simple
modifications to the distal ligands bound at Co directly
modulates the reactivity at the Ru−oxo moiety. Significantly,
electronic changes at the ligand set are transmitted through all
three Co centers to the Ru−oxo moiety, indicating that the
metal centers are working cooperatively as an assembly in
oxidizing C−H substrates. Similar multimetallic cooperative
effects have been predicted by DFT in the C−H activation
study of the related Co4O4 cubane.18 These studies indicate a
direct influence of the electronic communication within the
cubane clusters on C−H activation reactivity.
CCDC 2081075−2081077 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
■
Corresponding Authors
T. Don Tilley − Department of Chemistry, University of
California, Berkeley, Berkeley, California 94720-1460,
United States; Chemical Sciences Division, Lawrence Berkeley
National Laboratory, Berkeley, California 94720, United
States; orcid.org/0000-0002-6671-9099;
CONCLUDING REMARKS
■
Mechanistic insights described above highlight the importance
of two structural features that contribute to the reactivity of
RuCo3O4 cubanes in oxidative catalysis: high-valent metal−
oxyl moieties and multimetallic oxo frameworks. These clusters
abstract H atoms from a range of C−H substrates to generate
substrate radicals that are readily trapped by available (and
Email: tdtilley@berkeley.edu
David Balcells − Hylleraas Centre for Quantum Molecular
Sciences, Department of Chemistry, University of Oslo, 0315
Oslo, Norway; orcid.org/0000-0002-3389-0543;
Email: david.balcells@kjemi.uio.no
I
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(4) Okamura, M.; Kondo, M.; Kuga, R.; Kurashige, Y.; Yanai, T.;
Hayami, S.; Praneeth, V. K. K.; Yoshida, M.; Yoneda, K.; Kawata, S.;
Masaoka, S. A pentanuclear iron catalyst designed for water oxidation.
Nature 2016, 530 (7591), 465−468.
(5) Liu, F.; Concepcion, J. J.; Jurss, J. W.; Cardolaccia, T.;
Templeton, J. L.; Meyer, T. J. Mechanisms of water oxidation from
the blue dimer to photosystem II. Inorg. Chem. 2008, 47 (6), 1727−
1752.
(6) Khenkin, A. M.; Kumar, D.; Shaik, S.; Neumann, R.
Characterization of manganese(V)-oxo polyoxometalate intermedi-
ates and their properties in oxygen-transfer reactions. J. Am. Chem.
Soc. 2006, 128 (48), 15451−15460.
(7) Buchwalter, P.; Rosé, J.; Braunstein, P. Multimetallic catalysis
based on heterometallic complexes and clusters. Chem. Rev. 2015, 115
(1), 28−126.
(8) Yano, J.; Yachandra, V. Mn4Ca cluster in photosynthesis: Where
and how water is oxidized to dioxygen. Chem. Rev. 2014, 114 (8),
4175−4205.
(9) Suga, M.; Akita, F.; Yamashita, K.; Nakajima, Y.; Ueno, G.; Li,
H.; Yamane, T.; Hirata, K.; Umena, Y.; Yonekura, S.; Yu, L. J.;
Murakami, H.; Nomura, T.; Kimura, T.; Kubo, M.; Baba, S.;
Kumasaka, T.; Tono, K.; Yabashi, M.; Isobe, H.; Yamaguchi, K.;
Yamamoto, M.; Ago, H.; Shen, J. R. An oxyl/oxo mechanism for
oxygen-oxygen coupling in PSII revealed by an X-ray free-electron
laser. Science (Washington, DC, U. S.) 2019, 366 (6463), 334−338.
(10) Cox, N.; Pantazis, D. A.; Lubitz, W. Current Understanding of
the Mechanism of Water Oxidation in Photosystem II and Its
Relation to XFEL Data. Annu. Rev. Biochem. 2020, 89, 795−820.
(11) Pantazis, D. A. Missing Pieces in the Puzzle of Biological Water
Oxidation. ACS Catal. 2018, 8 (10), 9477−9507.
(12) Huang, X.; Groves, J. T. Beyond ferryl-mediated hydroxylation:
40 years of the rebound mechanism and C−H activation. JBIC, J. Biol.
Inorg. Chem. 2017, 22, 185−207.
(13) Baik, M. H.; Newcomb, M.; Friesner, R. A.; Lippard, S. J.
Mechanistic studies on the hydroxylation of methane by methane
monooxygenase. Chem. Rev. 2003, 103 (6), 2385−2419.
(14) Gunay, A.; Theopold, K. H. C-H bond activations by metal oxo
compounds. Chem. Rev. 2010, 110 (2), 1060−1081.
(15) Xue, X. S.; Ji, P.; Zhou, B.; Cheng, J. P. The Essential Role of
Bond Energetics in C-H Activation/Functionalization. Chem. Rev.
2017, 117 (13), 8622−8648.
(16) Borovik, A. S. Role of metal-oxo complexes in the cleavage of
C-H bonds. Chem. Soc. Rev. 2011, 40 (4), 1870−1874.
(17) Reed, C. J.; Agapie, T. A Terminal FeIII-Oxo in a Tetranuclear
Cluster: Effects of Distal Metal Centers on Structure and Reactivity. J.
Am. Chem. Soc. 2019, 141 (24), 9479−9484.
Authors
Jaruwan Amtawong − Department of Chemistry, University of
California, Berkeley, Berkeley, California 94720-1460,
United States; Chemical Sciences Division, Lawrence Berkeley
National Laboratory, Berkeley, California 94720, United
States; orcid.org/0000-0002-3850-3290
Bastian Bjerkem Skjelstad − Hylleraas Centre for Quantum
Molecular Sciences, Department of Chemistry, University of
Oslo, 0315 Oslo, Norway; orcid.org/0000-0002-9769-
6459
Rex C. Handford − Department of Chemistry, University of
California, Berkeley, Berkeley, California 94720-1460,
United States; orcid.org/0000-0002-3693-1697
Benjamin A. Suslick − Department of Chemistry, University of
California, Berkeley, Berkeley, California 94720-1460,
United States; Chemical Sciences Division, Lawrence Berkeley
National Laboratory, Berkeley, California 94720, United
States; orcid.org/0000-0002-6499-3625
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c04069
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was primarily funded by the U.S. Department of
Energy, Office of Science, Office of Basic Energy Sciences,
Chemical Sciences, Geosciences, and Biosciences Division,
under Contract DE-AC02-05CH11231. We thank the College
of Chemistry (CoC) NMR facility for resources provided.
Instruments in CoC NMR facility are supported in part by
NIH S10OD024998. We thank the Catalysis Center of the
Lawrence Berkeley National Laboratory and the QB3/
Chemistry Mass Spectrometry Facility at UC Berkeley. This
research used resources of the Advanced Light Source, which is
a DOE Office of Science User Facility under Contract DE-
AC02-05CH11231. J.A. is supported by a scholarship from the
Development and Promotion of Science and Technology
(DPST), Thailand. B.B.S. and D.B. acknowledge the support
from the Norwegian Research Council through the Hylleraas
Centre for Quantum Molecular Sciences (Project 262695) and
the Norwegian Metacenter for Computational Science
(NOTUR; Grant nn4654k). R.C.H. acknowledges NSERC
of Canada for a PGS-D award. We thank Simon J. Teat for his
crystallographic support and expertise. We thank Dr. Hasan
Celik for NMR spectroscopy advice and Dr. Miao Zhang for
IR spectroscopy advice. Finally, we thank Dr. Stefan Kuenzi,
Dr. Pablo Rios, T. Alex Wheeler, Josh Tsang, and Prof. Andy I.
Nguyen of the University of Illinois at Chicago for useful and
insightful discussions.
(18) Amtawong, J.; Skjelstad, B. B.; Balcells, D.; Tilley, T. D.
Concerted Proton−Electron Transfer Reactivity at a Multimetallic
Co4O4 Cubane Cluster. Inorg. Chem. 2020, 59 (20), 15553−15560.
(19) Symes, M. D.; Surendranath, Y.; Lutterman, D. A.; Nocera, D.
G. Bidirectional and unidirectional PCET in a molecular model of a
cobalt-based oxygen-evolving catalyst. J. Am. Chem. Soc. 2011, 133
(14), 5174−5177.
(20) Xue, G.; De Hont, R.; Munck, E.; Que, L. Million-fold
̈
activation of the [Fe2(μ-O)2] diamond core for C-H bond cleavage.
Nat. Chem. 2010, 2 (5), 400−405.
(21) Moysiadou, A.; Lee, S.; Hsu, C. S.; Chen, H. M.; Hu, X.
Mechanism of Oxygen Evolution Catalyzed by Cobalt Oxyhydroxide:
Cobalt Superoxide Species as a Key Intermediate and Dioxygen
Release as a Rate-Determining Step. J. Am. Chem. Soc. 2020, 142 (27),
11901−11914.
(22) Ullman, A. M.; Brodsky, C. N.; Li, N.; Zheng, S. L.; Nocera, D.
G. Probing Edge Site Reactivity of Oxidic Cobalt Water Oxidation
Catalysts. J. Am. Chem. Soc. 2016, 138 (12), 4229−4236.
(23) Koroidov, S.; Anderlund, M. F.; Styring, S.; Thapper, A.;
Messinger, J. First turnover analysis of water-oxidation catalyzed by
Co-oxide nanoparticles. Energy Environ. Sci. 2015, 8 (8), 2492−2503.
REFERENCES
■
(1) McEvoy, J. P.; Brudvig, G. W. Water-splitting chemistry of
photosystem II. Chem. Rev. 2006, 106 (11), 4455−4483.
(2) Larson, V. A.; Battistella, B.; Ray, K.; Lehnert, N.; Nam, W. Iron
and manganese oxo complexes, oxo wall and beyond. Nat. Rev. Chem.
2020, 4, 404−419.
(3) Nguyen, A. I.; Ziegler, M. S.; Ona-Burgos, P.; Sturzbecher-
̃
Hohne, M.; Kim, W.; Bellone, D. E.; Tilley, T. D. Mechanistic
Investigations of Water Oxidation by a Molecular Cobalt Oxide
Analogue: Evidence for a Highly Oxidized Intermediate and Exclusive
Terminal Oxo Participation. J. Am. Chem. Soc. 2015, 137 (40),
12865−12872.
J
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(24) Shimoyama, Y.; Kojima, T. Metal-Oxyl Species and Their
Possible Roles in Chemical Oxidations. Inorg. Chem. 2019, 58 (15),
9517−9542.
(25) Yamada, H.; Hurst, J. K. Resonance raman, optical
spectroscopic, and EPR characterization of the higher oxidation
states of the water oxidation catalyst, cis,cis-[(bpy)2Ru(OH2)]2O4+. J.
Am. Chem. Soc. 2000, 122 (22), 5303−5311.
(26) Binstead, R. A.; Chronister, C. W.; Ni, J.; Hartshorn, C. M.;
Meyer, T. J. Mechanism of water oxidation by the μ-Oxo dimer
[(bpy)2(H2O)Ru(III)ORu(III)(OH2)(bpy)2]4+. J. Am. Chem. Soc.
2000, 122 (35), 8464−8473.
(27) Gilbert, J. A.; Eggleston, D. S.; Murphy, W. R.; Geselowitz, D.
A.; Gersten, S. W.; Hodgson, D. J.; Meyer, T. J. Structure and Redox
Properties of the Water-Oxidation Catalyst [(bpy)2(OH2)RuORu-
(OH2)(bpy)2]. J. Am. Chem. Soc. 1985, 107 (13), 3855−3864.
(28) Moonshiram, D.; Alperovich, I.; Concepcion, J. J.; Meyer, T. J.;
Pushkar, Y. Experimental demonstration of radicaloid character in a
RuV=O intermediate in catalytic water oxidation. Proc. Natl. Acad. Sci.
U. S. A. 2013, 110 (10), 3765−3770.
(29) Sala, X.; Maji, S.; Bofill, R.; García-Antón, J.; Escriche, L.;
Llobet, A. Molecular Water Oxidation Mechanisms Followed by
Transition Metals : State of the Art. Acc. Chem. Res. 2014, 47 (2),
504−516.
(30) Shaffer, D. W.; Xie, Y.; Concepcion, J. J. O−O bond formation
in ruthenium-catalyzed water oxidation: Single-site nucleophilic
attack: vs. O−O radical coupling. Chem. Soc. Rev. 2017, 46 (20),
6170−6193.
(31) Shing, K. P.; Cao, B.; Liu, Y.; Lee, H. K.; Li, M. De; Phillips, D.
L.; Chang, X. Y.; Che, C. M. Arylruthenium(III) Porphyrin-Catalyzed
C−H Oxidation and Epoxidation at Room Temperature and
[RuV(Por)(O)(Ph)] Intermediate by Spectroscopic Analysis and
Density Functional Theory Calculations. J. Am. Chem. Soc. 2018, 140
(22), 7032−7042.
(32) Dietl, N.; Schlangen, M.; Schwarz, H. Thermal hydrogen-atom
transfer from methane: The role of radicals and spin states in oxo-
cluster chemistry. Angew. Chem., Int. Ed. 2012, 51 (23), 5544−5555.
(33) Saouma, C. T.; Mayer, J. M. Do spin state and spin density
affect hydrogen atom transfer reactivity? Chem. Sci. 2014, 5 (1), 21−
31.
(34) Amtawong, J.; Balcells, D.; Wilcoxen, J.; Handford, R. C.;
Biggins, N.; Nguyen, A. I.; Britt, R. D.; Tilley, T. D. Isolation and
Study of Ruthenium-Cobalt Oxo Cubanes Bearing a High-Valent,
Terminal RuV-Oxo with Significant Oxyl Radical Character. J. Am.
Chem. Soc. 2019, 141, 19859−19869.
(35) Simmons, E. M.; Hartwig, J. F. On the interpretation of
deuterium kinetic isotope effects in C-H bond functionalizations by
transition-metal complexes. Angew. Chem., Int. Ed. 2012, 51 (13),
3066−3072.
(36) Bryant, J. R.; Mayer, J. M. Oxidation of C-H bonds by
[(bpy)2(py)RuIVO]2+ occurs by hydrogen atom abstraction. J. Am.
Chem. Soc. 2003, 125 (34), 10351−10361.
(37) Matsuo, T.; Mayer, J. M. Oxidations of NADH Analogues by
cis-[RuIV(bpy)2(py)(O)]2+ Occur by Hydrogen-Atom Transfer
Rather than by Hydride Transfer. Inorg. Chem. 2005, 44 (7),
2150−2158.
(38) Lam, W. W. Y.; Man, W. L.; Lau, T. C. Mechanisms of
oxidation by trans-dioxoruthenium(VI) complexes containing macro-
cyclic tertiary amine ligands. Coord. Chem. Rev. 2007, 251 (17−20),
2238−2252.
(39) Ishizuka, T.; Kotani, H.; Kojima, T. Characteristics and
reactivity of ruthenium-oxo complexes. Dalt. Trans. 2016, 45 (42),
16727−16750.
(42) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Thermochemistry of
proton-coupled electron transfer reagents and its implications. Chem.
Rev. 2010, 110 (12), 6961−7001.
(43) Parsell, T. H.; Yang, M.-Y.; Borovik, A. S. C-H bond cleavage
with reductants: re-investigating the reactivity of monomeric MnIII/IV
-
Oxo complexes and the role of oxo ligand basicity. J. Am. Chem. Soc.
2009, 131 (8), 2762−2763.
(44) Ishizuka, T.; Ohzu, S.; Kotani, H.; Shiota, Y.; Yoshizawa, K.;
Kojima, T. Hydrogen atom abstraction reactions independent of C-H
bond dissociation energies of organic substrates in water: Significance
of oxidant-substrate adduct formation. Chem. Sci. 2014, 5 (4), 1429−
1436.
(45) Pavlishchuk, V. V.; Addison, A. W. Conversion constants for
redox potentials measured versus different reference electrodes in
acetonitrile solutions at 25°C. Inorg. Chim. Acta 2000, 298 (1), 97−
102.
(46) Augustin-Nowacka, D.; Chmurzynski, L. A potentiometric
̃
study of acid-base equilibria of substituted pyridines in acetonitrile.
Anal. Chim. Acta 1999, 381 (2−3), 215−220.
(47) Wise, C. F.; Agarwal, R. G.; Mayer, J. M. Determining Proton-
Coupled Standard Potentials and X−H Bond Dissociation Free
Energies in Nonaqueous Solvents Using Open-Circuit Potential
Measurements. J. Am. Chem. Soc. 2020, 142 (24), 10681−10691.
(48) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. Energies,
Structures, and Electronic Properties of Molecules in Solution with
the C-PCM Solvation Model. J. Comput. Chem. 2003, 24 (6), 669−
681.
(49) Barone, V.; Cossi, M. Quantum Calculation of Molecular
Energies and Energy Gradients in Solution by a Conductor Solvent
Model. J. Phys. Chem. A 1998, 102 (11), 1995−2001.
(50) Cho, K.-B.; Wu, X.; Lee, Y.-M.; Kwon, Y. H.; Shaik, S.; Nam,
W. Evidence for an alternative to the oxygen rebound mechanism in
C-H bond activation by non-heme FeIVO complexes. J. Am. Chem.
Soc. 2012, 134 (50), 20222−20225.
(51) Cho, K.-B.; Shaik, S.; Nam, W. Theoretical investigations into
C-H bond activation reaction by nonheme MnIVO complexes:
Multistate reactivity with no oxygen rebound. J. Phys. Chem. Lett.
2012, 3 (19), 2851−2856.
(52) Dhuri, S. N.; Cho, K.-B.; Lee, Y.-M.; Shin, S. Y.; Kim, J. H.;
Mandal, D.; Shaik, S.; Nam, W. Interplay of Experiment and Theory
in Elucidating Mechanisms of Oxidation Reactions by a Nonheme
RuIVO Complex. J. Am. Chem. Soc. 2015, 137 (26), 8623−8632.
(53) Cho, K.-B.; Hirao, H.; Shaik, S.; Nam, W. To rebound or
dissociate? This is the mechanistic question in C-H hydroxylation by
heme and nonheme metal-oxo complexes. Chem. Soc. Rev. 2016, 45
(5), 1197−1210.
(54) Oda, A.; Ohkubo, T.; Yumura, T.; Kobayashi, H.; Kuroda, Y.
Room-Temperature Activation of the C-H Bond in Methane over
Terminal ZnII-Oxyl Species in an MFI Zeolite: A Combined
Spectroscopic and Computational Study of the Reactive Frontier
Molecular Orbitals and Their Origins. Inorg. Chem. 2019, 58 (1),
327−338.
(55) Zhao, X.; Continetti, R. E.; Yokoyama, A.; Hintsa, E. J.; Lee, Y.
T. Dissociation of cyclohexene and 1,4-cyclohexadiene in a molecular
beam. J. Chem. Phys. 1989, 91 (7), 4118−4127.
(56) CRC Handbook of Chemistry and Physics, 87th ed.; Lide, D. R.,
Ed.; CRC Press: Boca Raton, FL, 2006.
(57) Kojima, T.; Hirai, Y.; Ishizuka, T.; Shiota, Y.; Yoshizawa, K.;
Ikemura, K.; Ogura, T.; Fukuzumi, S. A low-spin ruthenium(IV)-oxo
complex: Does the spin state have an impact on the reactivity? Angew.
Chem., Int. Ed. 2010, 49 (45), 8449−8453.
(58) Mayer, J. M. Understanding hydrogen atom transfer: From
bond strengths to marcus theory. Acc. Chem. Res. 2011, 44 (1), 36−
46.
(59) Lai, W.; Li, C.; Chen, H.; Shaik, S. Hydrogen-abstraction
reactivity patterns from A to Y: The valence bond way. Angew. Chem.,
Int. Ed. 2012, 51 (23), 5556−5578.
(40) Evans, M. G.; Polanyi, M. Inertia and Driving Force of
Chemical Reactions. Trans. Faraday Soc. 1938, 34, 11.
(41) Warren, J. J.; Mayer, J. M. Application of the Marcus Cross
Relation to Hydrogen Atom Transfer/Proton-Coupled Electron
Transfer Reactions. In Proton-Coupled Electron Transfer: A Carrefour
of Chemical Reactivity Traditions; The Royal Society of Chemistry:
2012; pp 1−31.
K
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(60) Ding, X. L.; Wu, X. N.; Zhao, Y. X.; He, S. G. C-H bond
activation by oxygen-centered radicals over atomic clusters. Acc. Chem.
Res. 2012, 45 (3), 382−390.
(61) Schneider, J. E.; Goetz, M. K.; Anderson, J. S. Statistical analysis
of C−H activation by oxo complexes supports diverse thermodynamic
control over reactivity. Chem. Sci. 2021, 12, 4173−4183.
(62) Gaggioli, C. A.; Sauer, J.; Gagliardi, L. Hydrogen Atom or
Proton Coupled Electron Transfer? C-H Bond Activation by
Transition-Metal Oxides. J. Am. Chem. Soc. 2019, 141 (37),
14603−14611.
(63) Huynh, M. H. V.; Meyer, T. J. Proton-coupled electron transfer.
Chem. Rev. 2007, 107 (11), 5004−5064.
(64) Prokop, K. A.; De Visser, S. P.; Goldberg, D. P. Unprecedented
rate enhancements of hydrogen-atom transfer to a manganese(V)-oxo
corrolazine complex. Angew. Chem., Int. Ed. 2010, 49 (30), 5091−
5095.
(65) Merkx, M.; Kopp, D. A.; Sazinsky, M. H.; Blazyk, J. L.; Muller,
̈
J.; Lippard, S. J. Dioxygen activation and methane hydroxylation by
soluble methane monooxygenase: A tale of two irons and three
proteins. Angew. Chem., Int. Ed. 2001, 40 (15), 2782−2807.
L
https://doi.org/10.1021/jacs.1c04069
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX