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

Article abstract of DOI:10.1021/jacs.8b04470

The development of highly active and selective metal catalysts for efficient oxidation of hydrocarbons and identification of the reactive intermediates in the oxidation catalysis are long-standing challenges. In the rapid hydrocarbon oxidation catalyzed by ruthenium(IV) and -(III) porphyrins, the putative Ru(V)-oxo intermediates remain elusive. Herein we report that arylruthenium(III) porphyrins are highly active catalysts for hydrocarbon oxidation. Using catalyst [Ru^III(TDCPP)(Ph)(OEt₂)] (H₂TDCPP = 5,10,15,20-tetrakis(2,6-dichlorophenyl)porphyrin), the oxidation of C-H bonds of various hydrocarbons with oxidant m-CPBA at room temperature gave alcohols/ketones in up to 99% yield within 1 h; use of [ⁿBu₄N]IO₄ as a mild alternative oxidant avoided formation of lactone from cyclic ketone in C-H oxidation, and the catalytic epoxidation with up to 99% yield and high selectivity (no aldehydes as side product) was accomplished within 5 min. UV-vis, electrospray ionization-mass spectrometry, resonance Raman, electron paramagnetic resonance, and kinetic measurements and density functional theory calculations lend evidence for the formation of Ru(V)-oxo intermediate [Ru^V(TDCPP)(O)(Ph)].

Full text of DOI:10.1021/jacs.8b04470

Article
pubs.acs.org/JACS
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
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
Ka-Pan Shing,† Bei Cao,† Yungen Liu,† Hung Kay Lee,‡ Ming-De Li,† David Lee Phillips,†
Xiao-Yong Chang,† and Chi-Ming Che*,†,§
†Department of Chemistry and State Key Laboratory of Synthetic Chemistry, The University of Hong Kong, Pokfulam Road, Hong
Kong, China
‡Department of Chemistry, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China
§HKU Shenzhen Institute of Research and Innovation, Shenzhen 518053, China
S
* Supporting Information
ABSTRACT: The development of highly active and selective
metal catalysts for efficient oxidation of hydrocarbons and
identification of the reactive intermediates in the oxidation catalysis
are long-standing challenges. In the rapid hydrocarbon oxidation
catalyzed by ruthenium(IV) and -(III) porphyrins, the putative
Ru(V)-oxo intermediates remain elusive. Herein we report that
arylruthenium(III) porphyrins are highly active catalysts for
hydrocarbon oxidation. Using catalyst [RuIII(TDCPP)(Ph)(OEt2)]
(H2TDCPP = 5,10,15,20-tetrakis(2,6-dichlorophenyl)porphyrin),
the oxidation of C−H bonds of various hydrocarbons with oxidant
m-CPBA at room temperature gave alcohols/ketones in up to 99%
yield within 1 h; use of [nBu4N]IO4 as a mild alternative oxidant
avoided formation of lactone from cyclic ketone in C−H oxidation, and the catalytic epoxidation with up to 99% yield and high
selectivity (no aldehydes as side product) was accomplished within 5 min. UV−vis, electrospray ionization−mass spectrometry,
resonance Raman, electron paramagnetic resonance, and kinetic measurements and density functional theory calculations lend
evidence for the formation of Ru(V)-oxo intermediate [RuV(TDCPP)(O)(Ph)].
mechanism which incorporates a fast catalytic cycle mediated
by metastable RuIII and RuV-oxo intermediates and a slow
catalytic cycle mediated by [RuIV(Por)(O)] and [RuVI(Por)-
(O)2].4e,j Newcomb and co-workers reported the observation
of [RuV(Por)(O)(X)] (X = OH− or pyridine) by UV−vis
spectroscopy upon laser flash photolysis of [{RuIV(Por)-
(OH)}2O] or in situ generated [RuIII(Por)(pyridine N-
oxide)].6 The identification of a RuV-oxo porphyrin species
by spectroscopic methods, including resonance Raman (RR)
and electron paramagnetic resonance (EPR), remains a
challenge, despite the characterization of a few nonporphyrin
RuV-oxo species.7
INTRODUCTION
■
Alkane oxidation catalyzed by cytochrome P-450 has received
much attention over the past decades; tremendous efforts have
been made to mimic the enzymatic reactions using synthetic
metalloporphyrin catalysts1 since the reports on [FeIII(Por)-
(Cl)] (H2Por = porphyrin)-catalyzed alkane oxidation.2 The
key active species in the cytochrome P-450-catalyzed oxidation
is the FeIV-oxo porphyrin cation radical; the transient FeV-oxo
analogue, [FeV(Por)(O)(ClO4)], has also been reported to
oxidize hydrocarbons at fast rates.3 Similarly, ruthenium
porphyrins were studied and used to mimic the enzymatic
oxidation,4 particularly for the oxidation by dioxo complexes
[RuVI(Por)(O)2].5 Hirobe and co-workers reported that
‘[RuVI(Por)(O)2] or [RuII(Por)(CO)] + 2,6-Cl2pyNO’ (Por
= TMP, TDCPP, TPP) can be transformed into a more
reactive catalyst for alkane oxidation by adding strong mineral
acid (HCl or HBr);4c the active intermediate in the catalytic
system was not identified, with the suggested candidates being
[RuV(Por)(O)(X)].4c For the rapid catalytic oxygenation of
hydrocarbons by ‘[RuII(F20-TPP)(CO)] or [RuIV(F20-TPP)-
(Cl)2] + 2,6-Cl2pyNO’, Groves and co-workers proposed a
Because of the high reactivity of [RuV(Por)(O)(X)] (X =
axial ligand), we envisioned that the use of an electron-rich axial
ligand might stabilize such species for spectroscopic detection
as the nature of axial ligand has been reported to affect the
reactivity of metalloporphyrins significantly. For instance, we
recently found that axial N-heterocyclic carbene (NHC) ligand
Received: April 27, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
can improve the catalytic carbene transfer performance of Ru
porphyrins.8 This beneficial effect of strong carbon σ-donor
ligand prompted us to examine the catalytic properties of
[RuIII(Por)(Ar)(L)] (L = labile solvent), which would directly
serve as a precursor to the RuV-oxo species through oxygen
atom transfer, unlike [RuII(Por)(CO)(L)] and [RuIV(Por)-
(Cl)2] both requiring the pregeneration of Ru(III) precursors
by redox processes.4e,i,j In the literature, studies on aryl- and
alkylruthenium porphyrins are scarce, and this type of
organoruthenium(III) porphyrins has not been applied to
oxidation catalysis, despite the hydrocarbon oxidation catalyzed
by [RuIII(OEP)(Br)(PPh3)] (up to 21% product yield).9
Herein we found that [RuIII(Por)(Ar)(L)] are active catalysts
for alkene epoxidation and C−H oxidation at RT (up to 99%
product yield; limiting amount of substrate). UV−vis, electro-
spray ionization−mass spectrometry (ESI−MS), RR, EPR,
kinetic measurements, and density functional theory (DFT)
calculations point to the formation of [RuV(Por)(O)(Ar)] and
its involvement in the catalytic oxidation.
from the reaction of [RuIII(TDFPP)(Cl)(THF)] (3) and PhLi
(Scheme 1). The solid forms of 1 and 2 are stable in air and
light for at least one month and they can be purified by neutral
alumina using CH2Cl2/hexane as eluent.
Complexes 1−3 were characterized by X-ray crystallography
(Figure 1 and Table S1 in the Supporting Information), ESI−
RESULTS
■
Synthesis and Characterizations of [RuIII(Por)(Ph)(L)].
Organoruthenium porphyrins bearing axial alkyl or aryl ligands
were first prepared by treatment of K2[Ru(TPP)]2 with alkyl
iodide,10a or by reaction of [Ru(OEP)(Br)2] with alkyl- or
aryllithium,10b both affording [RuIV(Por)(Y)2] (Por = TPP,
OEP; Y = alkyl, aryl). James, Dolphin, and co-workers reported
that pyrolysis of [RuIV(Por)(Ph)2] afforded [RuIII(Por)(Ph)]
(Por = OEP and TPP) in quantitative yields.10c Leidner and
Seyler found that [RuIII(OEP)(Ph)] can also be obtained as a
major product by treating [RuIV(OEP)(Cl)2] with excess
phenyllithium (PhLi).10d We therefore examined similar
reaction of other [RuIV(Por)(Cl)2] complexes and excess
PhLi for the preparation of [RuIII(Por)(Ph)]. Initially, we
attempted to prepare [RuIII(TPP)(Ph)], but it was oxidized to
[RuIV(TPP)(Ph)]2O by air during chromatographic purifica-
tion within minutes. To circumvent this problem, we turned
our attention to the porphyrin ligands bearing halogen
substituents. Replacement of TPP with a highly fluorinated
porphyrin (F20-TPP) led to unexpected substitution of the
fluorine atoms by a strong nucleophile (PhLi). The chlorinated
porphyrin TDCPP, which is a sterically encumbered porphyrin
ligand, was stable in the presence of excess PhLi. We prepared
[RuIII(TDCPP)(Ph)(OEt2)] (1) in 66% yield by treating
[RuIV(TDCPP)(Cl)2] with excess PhLi (Scheme 1; similar to
the synthesis of [RuIII(OEP)(Ph)(THF)]10d) followed by
column chromatography. Efforts were made to change the
TDCPP ligand to its fluorine counterpart, namely, TDFPP, and
[RuIII(TDFPP)(Ph)(OH2)] (2) was prepared in 73% yield
Figure 1. X-ray crystal structures of 1, 2, and 3 (30% thermal ellipsoid
probability; hydrogen atoms are omitted).
MS, UV−vis and 1H NMR spectroscopy, and cyclic
voltammetry (Figure 2). Diffraction-quality crystals of these
complexes were grown by the procedures described in the
Supporting Information. Both 1 and 2 contain an axial σ-phenyl
ligand bound to the Ru atom. Complex 1 shows a Ru−C(Ph)
distance of 2.053(5) Å, slightly longer than that in [RuIII(OEP)-
(Ph)] (2.005(7) Å).10c For 2, the disordered axial ligands make
precise determination of the Ru−C(Ph) distance difficult. The
axial CPh−Ru-Oaxial bond angle of 1 is 178.7°, and that of 2 is
also close to 180°. In complex 3, there are two independent
molecules in the asymmetric unit, with axial Ru−Cl distances of
2.2899(16) and 2.2940(17) Å comparable to those reported for
[RuIV(Por)(Cl)2] (Por = TDCPP, TMP; 2.288(2)−2.302(4)
Å).4i In addition, the porphyrin planes of 1−3 do not show
significant ruffling or saddling distortions. ESI−MS analyses of
Scheme 1. Preparation of Complexes 1−3 Characterized by
a
1
X-ray Crystallography, H NMR, and ESI−MS
a
The others are known complexes.
B
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Oxidation of Hydrocarbons and Alcohols with m-
a
CPBA at RT Catalyzed by 1 and 2
Figure 2. Cyclic voltammograms of 1−3 in CH2Cl2 containing 0.1 M
[nBu4N]PF6 (scan rate, 100 mV s−1; E, V vs Fc+/Fc).
1−3 revealed prominent cluster peaks at m/z 1067.1, 935.2,
and 893.0 attributed to singly charged ions with formulations of
[Ru(TDCPP)(Ph)], [Ru(TDFPP)(Ph)], and [Ru(TDFPP)-
(Ph)], respectively (see, for examples, Figure S1 in the
Supporting Information).
The electrochemical properties of 1−3 were examined by
cyclic voltammetry (Figure 2). For 1 and 2, their cyclic
voltammograms are reminiscent of that of [Ru(OEP)(Ph)],10d
each showing a reversible oxidation couple (E1/2 + 0.35 V for 1,
+ 0.33 V for 2) assignable to RuIII/RuIV and a reversible
reduction couple (E1/2 −0.85 V for 1, −0.75 V for 2) assignable
to RuIII/RuII. Complex 3 exhibits an analogous cyclic
voltammogram albeit with anodically shifted potentials
(compared with 1 and 2). The reduction potentials of 1−3
(−0.15 to −0.85 V vs Fc+/Fc) are much less cathodic than that
of the porphyrin-centered reductions of ruthenium porphyrins
(e.g., −1.25 to −1.73 V vs Fc+/Fc11).
As 1 and 2 show similar oxidation potentials and a
comparable oxidation potential was found for 3, we used one
of them to study UV−vis spectroelectrochemistry to provide
additional evidence for assignment of the oxidation couples
observed. Upon oxidation of 2 at a constant voltage of +0.5 V,
the Soret and Q bands shifted to shorter wavelengths, without
appearance of a band in the 600−700 nm region (Figure S2).
This supports that the oxidation is ruthenium-centered, that is,
oxidation of Ru(III) to Ru(IV), instead of porphyrin-centered
to generate porphyrin cation radical, as porphyrin cation radical
species are well documented to exhibit a band in the 600−700
nm region.12
C−H Oxidation of Hydrocarbons. Highly efficient and
selective C−H oxidation under mild conditions remains a great
challenge in the field of oxidation chemistry.13 The C−H
oxidation reactions of hydrocarbons catalyzed by 1 or 2 were
performed using limiting amount of hydrocarbon substrates, 3.5
or 2 equiv of m-CPBA, and 2−10 mol % of catalyst at RT,
giving complete substrate conversions. Benzylic hydrocarbons
were oxidized to ketones in 88% to >99% yields (entries 1, 3, 5,
11, Table 1), and <1% yields of alcohols were detected by GC/
NMR. Notably, > 99% ketone yield was obtained for the
oxidation of indan/fluorene with 2 equiv of m-CPBA catalyzed
by 2 mol % of 1 for 0.5 h (entries 5 and 11, Table 1). For
aliphatic hydrocarbons (entries 7, 9, 16, 17, Table 1), their
oxidation catalyzed by 2 (10 mol %) occurred selectively at
tertiary C−H bonds to give tertiary alcohols in 67−80% yields
without oxidation of secondary or primary C−H bonds.
Moreover, the substrate conversions and product yields
obtained for [nBu4N]IO4 (entries 2, 4, 6, 8, 10, 12, Table 1)
are similar to those for m-CPBA (entries 1, 3, 5, 7, 9, 11, Table
a
Reaction conditions: substrate (0.167 mmol), catalyst (2 mol %; 1 for
entries 1, 2, 5, 6, 11−13, 15, 19; 2 for the others), m-CPBA (0.585
mmol), CH2Cl2 (2 mL), under Ar, 0.5 h (entry 11), 1 h (entries 5, 6,
12, 13, 15), 3−5 h (entries 1, 2, 14, 16, 17, 19), overnight (entries 3, 4,
b
c
d
7−10, 18). Conversion (by H NMR or GC). 1H NMR yield.
5
f
1
e
mol % of 1. [nBu4N]IO4 (3.5 equiv) as oxidant instead of m-CPBA. 5
g
h
i
mol % of 2. 10 mol % of 2. Isolated yield. Using m-CPBA (2 equiv).
1). For catalytic oxidation of 1,2,3,4-tetrahydronaphthalene by
‘1 + [nBu4N]IO4’, di- and monoketones were formed in 70%
and 30% yields, respectively, without the formation of lactone
(entry 13, Table 1). Also, primary alkyl alcohol, which is
regarded as a less reactive substrate (compared to secondary
alcohol), was also studied. Oxidation of 1-dodecanol catalyzed
by 2 (5 mol %) for 3 h afforded the carboxylic acid in >99%
yield (entry 14, Table 1). Terpene (L-menthol, entry 18, Table
1) and steroid (estrone derivative, entry 19, Table 1) were
oxidized to give (2S,5R)-8-hydroxymenthone and 6-oxoestrone
in 85% and 87% isolated yields, respectively, showing good
regio-selectivity. The catalytic steroid oxidation by ‘1 +
[nBu4N]IO4’ protocol was completed within 3 h and no
heating was required.
C
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Epoxidation of Terminal and Internal Alkenes. We also
examined the catalytic activity of 1 and 2 toward epoxidation of
alkenes at RT. As epoxidation by m-CPBA is well
documented,14 we used [nBu4N]IO4 as oxidant which gave
negligible background epoxidation. Using 1-decene as a model
substrate, 1 (2 mol %) exhibited a good catalytic activity,
affording 1-decene oxide in 74% yield; with [RuII(TDCPP)-
(CO)], [RuIV(TDCPP)(Cl)2], [RuVI(TDCPP)(O)2], or 3 as a
catalyst (2 mol % at RT or 40 °C), only trace amount of 1-
decene oxide was detected by 1H NMR (Table S2). Change the
substrate to 1-dodecene in the 1-catalyzed reaction gave the
epoxidation product in a similar yield (79% yield; entry 10,
Table 2). It is noteworthy that the reaction time for the
that m-CPBA requires 3 h to finish epoxidation under metal-
free conditions17). The ‘1 + [nBu4N]IO4’ protocol was then
extended to various monosubstituted styrenes (in limiting
amount), which resulted in nearly quantitative substrate
conversion and epoxide yield after reaction at RT for 5 min
(entries 1−8, Table 2). No benzaldehyde was detected in the
crude reaction mixture by NMR and GC. Several other types of
alkenes were also epoxidized by ‘1 + [nBu4N]IO4’ with good-to-
excellent yields and selectivity (entries 11−23, Table 2),
including 1,3-diene (terminal CC bond oxidation in 90%
yield), cis- and trans-5-decene (configuration retention, similar
to alkene epoxidation by [RuVI(TPP)(O)2]18), cis-stilbene, cis-
cyclooctene, and cyclohexene (almost quantitative yield), α,β-
unsaturated ketone (>99% selectivity), and allylic alcohol
(complete consumption after 5 min without oxidizing alcohol
group). Allylic esters were epoxidized in 83−90% yields (8-h
reaction) with erythro/threo (E/T) ratios of (2.6−2.7):1,
similar to that (E/T 2.3:1, 85% total yield) obtained by
‘[RuII(TDCPP)(CO)] + Cl2pyNO’ at 40 °C for 48 h.19 Using 1
as a catalyst, epoxidation of Δ4-3-ketosteroid afforded the
epoxide in 73% isolated yield (based on 46% conversion) after
3 h with an α/β ratio of 9:1, higher than that (α/β 3:1) of a
similar Δ4-3-ketosteroid by ‘[FeII(qpy)](ClO4)2 + Oxone’.20
However, epoxidation of Δ4-3-ketosteroid by ‘2 + [nBu4N]IO4’
at 40 °C for 1 h gave >99% conversion and 91% isolated
epoxide yield albeit with a lower α/β ratio of 1:2.2 (entry 23,
Table 2).
Table 2. Epoxidation of Alkenes with [nBu4N]IO4 at RT
a
Catalyzed by 1
Mechanistic Studies. Because ‘1 + m-CPBA’ exhibited
high reactivity toward C−H oxidation, we performed a number
of mechanistic experiments to gain insight into the nature of
the active intermediate including high-resolution ESI−MS,
UV−vis, RR, and EPR spectroscopies, DFT calculations, and
kinetic experiments.
UV−vis Spectroscopy. The UV−vis spectrum of 1 in
CH2Cl2 at RT shows a Soret band at 405 nm and a Q-band
at 522 nm (Figure 3a, red curve). After fast-mixing with m-
a
Reaction conditions: alkene (0.167 mmol), 1 (2 mol %), oxidant
(0.300 mmol), CH2Cl2 (2 mL), under Ar; reaction time: 5 min
(entries 1−8, 11, 18), 40 min (entries 9, 10, 12−17), 8 h (entries 19−
b
c
23). Conversion (by GC). Based on conversion and determined by
Figure 3. UV−vis spectral changes. Treatment of 1 with m-CPBA (1.2
equiv) in CH2Cl2 at RT changed the spectrum from (a) to (b) after 3
s. Subsequent treatment of (b) with styrene (4 equiv) changed the
spectrum to (c) via black curves (15-s intervals).
d
e
1H NMR using internal standard. By GC. 1.0 equiv of [nBu4N]IO4.
f
g
1H NMR % conv./yield without internal standard. 3.0 equiv of
h
i
j
[nBu4N]IO4. 2 mol % of 2. Total isolated yield. 40 °C, 1 h.
CPBA or [nBuN4]IO4 (1.2 or 2 equiv), the Soret and Q bands
shifted to 416 and 512 nm, respectively (Figure 3b, blue curve).
No significant absorption band was observed at 600−700 nm
upon addition of the oxidant. Subsequent addition of styrene (4
equiv) to the reaction mixture generated, within minutes, a
spectrum (Figure 3c, green curve), which resembles the
spectrum of 1.
epoxidation of such aliphatic alkenes catalyzed by ‘1 +
[nBu4N]IO4’ was much shorter than that catalyzed by
[RuII(Por)(CO)] and [RuIV(Por)(Cl)2] (40 min vs 24 h),4i,15
while previous epoxidation of 1-octene by ‘[MnIII(TDCPP)-
(OAc)] + [nBu4N]IO4’ for 24 h gave the epoxide in 21%
yield.16 When the epoxidation was catalyzed by ‘1 + m-CPBA’,
the reaction completed within 15 min (it has been reported
D
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
High-Resolution ESI−MS. We performed a double-mixing
experiment using high-resolution ESI−MS (positive ion mode).
To begin with, the ESI−MS spectrum of 1 in CH2Cl2 was
measured, which showed a cluster peak at m/z 1066.8630
attributed to [Ru(TDCPP)(Ph)] (Figure 4a). Subsequently, we
Figure 5. Resonance Raman spectra in CH2Cl2 (excitation wavelength,
416 nm; laser power, 5 mW; laser and solvent signals are denoted by
asterisk). (a) 1, (b) 1+ m-CPBA-16O (1.2 equiv), (c) 1+ m-CPBA-18O
(1.2 equiv). Expanded spectra in the 750−850 cm−1 region are
depicted on the left.
Table 3. Summary of Reported Raman Shifts for RuO
Bond22 and Observed Raman Shifts Assigned to
[RuV(TDCPP)(O)(Ph)] Generated from ‘1 + m-CPBA’
complex
[RuVI(TMP)(16O)2]
Raman shift of ν(RuO) (cm−1
)
811
[RuV(TDCPP)(16O)(Ph)]
[RuVI(TPP)(16O)2]
810 (this work)
808
[RuV(TDCPP)(18O)(Ph)]
[RuVI(TPP)(18O)2]
775 (this work)
764
X-Band EPR Spectroscopy. Analysis of a frozen CH2Cl2
solution of 1 at 40 K by X-band EPR revealed no significant
signal (Figure S5). Upon lowering the temperature to 7 K,
there is still no intense signal in the EPR spectrum (Figure 6a,
Figure 4. High resolution ESI−MS spectra (in CH2Cl2) of (a) 1, (b) 1
+ m-CPBA (1.0 equiv; reaction time: 10 s), (c) b + excess styrene. (d)
Simulated and (e) experimental isotopic patterns for parent ion of
[Ru(TDCPP)(O)(Ph)].
mixed the solution of 1 with m-CPBA (1.0 equiv)21 and
observed that the reaction generated, after 10 s, a new cluster
peak at m/z 1082.8514 assignable to [Ru(TDCPP)(O)(Ph)]
(A) (Figure 4b,d,e). This new cluster peak vanished
immediately upon mixing the reaction mixture with excess
styrene (Figure 4c). In addition, when a solution of
MeCN/18OH2 (5:1) was added to the CH2Cl2 solution
containing A, an isotopic pattern of 5% [Ru(TDCPP)(18O)-
(Ph)] + 95% [Ru(TDCPP)(16O)(Ph)] was observed (Figure
S4), suggesting partial oxygen exchange of the oxo species with
18OH2.
RR Spectroscopy. We measured the RR spectra of 1 (in
CH2Cl2) and its reaction mixtures with m-CPBA-16O or m-
CPBA-18O (Figure 5). Treatment of 1 with m-CPBA-16O (1.2
equiv) showed a new band at 810 cm−1; this new band shifted
to 775 cm−1 upon the replacement of m-CPBA-16O with m-
CPBA-18O. The bands at 810 and 775 cm−1 can be assigned to
Figure 6. X-band EPR spectra of frozen dichloromethane solutions
acquired at 7 K (ν = 9.380 GHz, amplitude = 7 G, power = 1.6 mW):
(a) 1. (b) a + m-CPBA (1.2 equiv). (c) b + styrene (4 equiv).
16
18
ν(Ru O) and ν(Ru O), respectively, and Ru-oxo double
bond is a preferred formulation according to the Raman shifts
reported for Fe- and Ru-oxo porphyrins (Table 3).22 The
experimentally observed isotopic shift value (Δν = 35 cm−1) is
close to the predicted one (Δν = 40 cm−1) obtained by
Hooke’s law of harmonic oscillator. The oxidation state marker
band corresponding to the C−N breathing mode (ν4)22
increased from 1354 to 1363 cm−1 upon mixing 1 with m-
CPBA (16O or 18O).
black curve),23 possibly due to fast relaxation.24 Subsequently, 1
was allowed to warm to room temperature, followed by
addition of m-CPBA (1.2 equiv). X-band EPR measurements of
the frozen reaction mixture at 7 K revealed a strong EPR signal
with g = 2.038, 1.983, 1.890 (Figure 6b, red curve). This EPR
signal vanished upon treatment of the reaction mixture with
styrene (4 equiv), and the resulting EPR spectrum (Figure 6c,
blue curve) resembles that of 1. The findings in these EPR
E
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
measurements are consistent with those in the aforementioned
UV−vis and ESI−MS experiments.
the k2 values for 1-phenylethanol and 1-phenylethanol-d10 (2.48
0.08, 0.78 0.02 M−1 s−1, Table S4 and Figure S6a).
Kinetic Experiments. A number of kinetic measurements
were performed for the hydrocarbon oxidation reactions by ‘1 +
m-CPBA’ at 298 K, including the relative oxidation rates for
para-substituted ethylbenzenes (Table S3), the second order
rate constants (k2, Table S4) for various hydrocarbons, and the
kinetic isotope effects (KIE) for methylcyclohexane and 1-
phenylethanol. For the oxidation of para-substituted ethyl-
benzenes catalyzed by ‘1 + m-CPBA’, the Hammett plot
exhibits a good linearity with ρ+ = −1.6 (Figure 7a and Table
DFT Calculations. Geometric and Electronic Structures.
DFT calculations on [RuI I I (TDCPP)(Ph)] and
[RuV(TDCPP)(O)(Ph)] (A) were conducted to gain insight
into their electronic structures and the mechanisms of
hydrocarbon oxidations by A. The geometrical parameters of
the DFT-optimized structure of [RuIII(TDCPP)(Ph)], as
compared with those determined by X-ray crystal analysis of
1, are presented in Table 4. The bond distances of Ru−CPh and
Table 4. Comparison of Key Structural Parameters of
[RuIII(TDCPP)(Ph)] between DFT-Optimized and X-ray
Crystal Structures
bond distance (Å)/angle
(deg)
DFT-optimized
structure
X-ray crystal
a
structure
Ru−CPh
Ru−Oether
1.991
2.307
2.053(5)
2.308(4)
Ru−NPor (average)
CPh−Ru-Oether
2.021
2.031(4)
175.1
178.70(17)
a
Data of complex 1 in this work.
Ru−Oether from the DFT calculations are ∼2.0 and 2.3 Å,
respectively, which are in good agreement with the
corresponding bond distances in the X-ray crystal structure.
Similarly, the calculated and experimental values for CPh−Ru-
Oether angle and average Ru−NPor bond distance are also
consistent with each other (175.1° vs 178.7°; 2.021 vs 2.031 Å).
For A, the calculated ground state is doublet (2A, S = 1/2) with
RuV-oxo bond distance of 1.721 Å; this ground state is 9.7 kcal
4
mol−1 lower in free energy than the quartet excited state A.
The spin density distribution for Ru:oxo was computed to be
0.46:0.72 (Figure 8). The MO diagram of 2A (Figure 9) reveals
an electronic configuration of (dxy)2(dπ)1 in accord with a
Ru(V) oxidation state.
Figure 7. (a) Hammett plot for the C−H oxidation of para-substituted
25
ethylbenzenes and (b) plot of log (k′) versus BDEC−H of
hydrocarbons (k′ = k2/(number of reactive hydrogen atoms)). All
reactions were catalyzed by ‘1 + m-CPBA’ in CH2Cl2 at 298 K.
Figure 8. Spin density distribution of [RuV(TDCPP)(O)(Ph)] (2A)
(Ru:oxo = 0.46:0.72).
S3). The rate constants (k2) for the oxidation of various
saturated hydrocarbons were used to plot log (k′) versus C−H
bond dissociation energy (BDEC−H); this plot shows a
reasonable linearity (Figure 7b), which can be attributed to
hydrogen atom abstraction by RuV-oxo species.25 Notably, the
k2 for ethylbenzene oxidation (1.99 0.04 M−1 s−1, Table S4
and Figure S6b) is four orders of magnitude faster than that
reported for oxidation of the same hydrocarbon by
[RuVI(TPP)(O)2] (k2 = 2.21 × 10−4 M−1 s−1).18 On the
basis of the k2 values for methylcyclohexane and methyl-
cyclohexane-d14 (0.220 0.006, 0.050 0.003 M−1 s−1, Table
S4 and Figure S6a), the KIE value for methylcyclohexane
oxidation was determined to be 4.4 0.4. Similarly, a KIE value
of 3.2 0.2 for oxidation of 1-phenylethanol was obtained from
C−H Oxidation Reactions. In DFT calculations on the
mechanisms of the C−H oxidation reactions by A, only the
low-spin d3 doublet state of A (S = 1/2, 2A) was considered, as
DFT calculations, and also EPR measurements, suggested that
2A is the ground state. From the computed free energy surface
for the hydroxylation of ethylbenzene by 2A to form 1-
phenylethanol (Figure 10, left), the reaction would overcome a
free energy barrier of 16.6 kcal mol−1 via transition state TS1
(Figures 10 and 11) and the reaction is highly exergonic (ΔG =
−40.3 kcal mol−1). In the transition state TS1 (Figure 10), the
F
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
hydroxylation takes place via a side-on overlap of C2−H bond
2
with the Ru-oxo bond. Ongoing from A to TS1, there is a
reorganization of the electron density between Ru and oxo, and
the spin distribution of Ru:oxo computed for TS1 (0.95:0.26)
is an inverse of that for 2A. Subsequent oxidation of 1-
phenylethanol by 2A (Figure 10, right) occurs through
transition state TS2 with an activation free energy of 10.6
2
kcal mol−1, which is reasonably lower than that of TS1. A
initiates this oxidation process by abstracting a hydrogen atom
of 1-phenylethanol on carbon rather than oxygen atom, and
acetophenone and water are produced by releasing free energy
of 88.0 kcal mol−1. The Ru−O and Ru−C1 bond distances of
TS2 are quite close to that of TS1, and TS2 shows a spin
density distribution (0.94 on Ru and 0.27 on O) very similar to
that calculated for TS1. The O−H−C2 angle in TS2 is
considerably smaller than in TS1 (140.7 vs 171.7°), probably
due to the hydrogen bond formed between oxo and the
hydroxyl group of 1-phenylethanol. The calculated KIE for the
oxidation of 1-phenylethanol is 3.2, matching the experimental
value of 3.2 0.2 described above.
Figure 9. MO diagram of [RuV(TDCPP)(O)(Ph)] (2A, β spin
orbitals; four Gouterman molecular orbitals (π1/π2, π1*/π2*) are
indicated with red rectangles).
For the methylcyclohexane hydroxylation by 2A, the
computed free energy surface (Figure S16) revealed that the
reaction proceeds via a transition state of H-abstraction (TS3)
followed by a barrierless rebound step. The transition state TS3
(Figure S16) is similar to that of TS1 (Figure 10), but with
computed activation barrier of 18.4 kcal mol−1, which is slightly
higher than that for ethylbenzene hydroxylation (16.6 kcal
mol−1). The calculated KIE value of 6.0 for the methylcyclo-
hexane hydroxylation is not far from the experimental value
(KIE 4.4 0.4, vide supra).
DISCUSSION
■
Ruthenium(III) porphyrins bearing axial aryl ligands have a few
previously reported examples including [Ru(OEP)-
(Ph)],10c,d,f,26a,b [RuIII(OEP)(Ph)(THF)],10d [RuIII(OEP)(p-
X-C6H4)] (X = Me, OMe, F, Cl),26b [RuIII(TPP)(Ph)],10e
and [RuIII2(DPB)(Ar)2] (Ar = p-Tol, 3,5-(CF3)2C6H3; DPB =
diporphyrinatobiphenylene).26c The catalytic properties of this
class of organometallic ruthenium complexes remain under-
developed; a literature example involves examination of
[Ru2(DPB)(Ar)2] as electrocatalysts for proton reduction on
Hg-pool electrodes, which gave catalytic turnover of ∼1.7 over
20 min.26c In this work, we found that [RuIII(TDCPP)(Ph)-
(OEt2)] (1) and [RuIII(TDFPP)(Ph)(OH2)] (2) are active
catalysts for oxidation of hydrocarbons at RT using m-CPBA as
a terminal oxidant. Considering that m-CPBA can overoxidize
carbonyl compounds via Baeyer−Villiger oxidation in which
cyclic/acyclic ketones can be transformed into lactones/
esters,27 to improve the practicality, we also employed a milder
2-electron oxidant, [nBu4N]IO4, as an alternative terminal
oxidant for the C−H oxidation of ethylbenzene, diphenyl-
methane, fluorene, 1,2,3,4-tetrahydronaphthalene, methylcyclo-
hexane, and 2-methyloctane, which afforded the same products
in similar yields (Table 1). The resemblance of [nBu4N]IO4 to
m-CPBA as oxidant for these oxidation reactions encouraged us
to expand the substrate scope to other hydrocarbons (Tables 1
and 2). Importantly, we observed that the 1- or 2-catalyzed C−
H oxidation is quite regioselective, as many of the products in
Table 1 were obtained as major products with >70% yield. The
regio-selectivity in the C−H oxidation is mainly controlled by
the electronic factor instead of the steric factor, as the oxidation
took place selectively at the weaker C−H bonds (benzylic/
Figure 10. Free energy surface of hydroxylation and subsequent
alcohol oxidation of ethylbenzene by [RuV(TDCPP)(O)(Ph)] (2A).
Figure 11. (a) Molecular structure of TS1. (b) Molecular orbital of
TS1 with hyperconjugation effect (depicted with the same orientation
as that in a).
ethylbenzene substrate is oriented to induce an almost linear
O−H−C2 angle (171.7°). The O−H and C2−H bond
distances are comparable, with the C2−H bond (1.243 Å)
being partially broken and the O−H bond (1.248 Å) being
partially formed. The partial formation of O−H bond lengthens
the Ru-oxo bond distance from 1.721 Å (2A) to 1.796 Å (TS1).
The Ru−O−H angle in TS1 is 123.9°, which is close to that for
an sp2-hybridized oxygen atom and this suggests that the C−H
G
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
tertiary), which are adjacent to bulkier groups than the more
accessible C−H bonds.
CPBA” and the catalytic reaction of 1-decene by “1 +
[nBu4N]IO4” (under the same reaction conditions as that
shown in Table 1 and Table 2, respectively), which showed that
the Ru-phenyl porphyrin is the major ruthenium species
present after the catalysis.
In screening various ruthenium porphyrins as catalyst for the
epoxidation of 1-decene with [nBu4N]IO4 (Table S2), it is
conspicuous that [RuII(TDCPP)(CO)], [RuIV(TDCPP)(Cl)2],
and 3 are not effective catalysts. A possible reason is that the
axial ligands (CO or Cl) in these complexes are not strongly
donating (when compared with the σ-aryl ligand in 1 or 2),
hampering the I−O bond cleavage to generate Ru-oxo species.
We also found that ‘[RuVI(TDCPP)(O)2] + [nBu4N]IO4’
cannot catalyze the 1-decene epoxidation (Table S2), which
disfavors the involvement of [RuVI(TDCPP)(O)2] as active
intermediate in the catalytic cycle for the 1-catalyzed
epoxidation. Notably, the alkene oxidation catalyzed by 1
generates epoxide as a major product without formation of
aldehyde, in contrast to the oxidation of styrenes catalyzed by
‘[RuIV(Por)(Cl)2] + Cl2pyNO’28 or ‘[FeIII(Por)]OTf + H2O2’29
affording benzaldehydes in up to 99% yield via epoxidation-
isomerization reaction. This could be ascribed to greater Lewis
acidity of [RuIV(Por)(Cl)2] (or the proposed intermediate
[RuIII(Por)(Cl)];4j Cl− is a weaker donor than σ-Ph−) and
[FeIII(Por)]OTf, rendering these complexes active catalysts for
epoxide isomerization (for treatment of 1 with styrene oxide or
1-decene oxide, no aldehyde(s) was detected in the reaction
mixture). Other notable features of the 1-catalyzed epoxidation
reactions include retention of configuration and good chemo-,
resio-, and diastereo-selectivities (Table 2).
The results from RR spectra (Figure 5) support the
formulation of A as a Ru-oxo species. In the literature, the
Raman shifts of ν(RuO) for [RuVI(TMP)(16O)2] and
[RuVI(TPP)(16O)2] were reported to be 811 and 808 cm−1,
respectively (Table 3);22 ν(RuV O) Raman shifts of 800−
16
812 cm−1 for nonporphyrin ruthenium complexes were also
reported.31,32a In the case of ‘1 + m-CPBA-16O’, ν(RuO) was
determined to be 810 cm−1, which falls within the range of
16
ν(Ru O) values reported for high-valent ruthenium
porphyrins bearing Ru-oxo bond(s).22 According to Table 3,
we assign the Ru-oxo bond in A as a double bond. The isotopic
shift (Δν) of 35 cm−1 upon replacing one 16O with 18O isotope
for A is similar to that reported for [RuVI(TPP)(16O)2] (34
cm−1).22
EPR measurements provide evidence that A belongs to a low
spin d3 (S = 1/2) RuV-oxo complex. As described in the
literature, RuV-oxo species (S = 1/2) characteristically exhibit
an EPR signal with all the components of g-tensor being close
to g = 2.0.32 A comparison of the g and giso values between low
spin (S = 1/2) mononuclear RuV-oxo and RuIII complexes is
depicted in Table S5. The experimentally measured g values (g
= 2.038, 1.983, 1.890) of A (generated by oxidation of 1 with
m-CPBA) in CH2Cl2 correspond to giso of 1.970. These g and
giso values are similar to those of a structurally characterized
mononuclear d3 (S = 1/2) RuV-oxo complex (g = 2.076, 1.977,
1.910; giso = 1.988; from X-band EPR spectrum in CH2Cl2)
reported by Griffith and co-worker.7d Similar g and giso values
have been obtained by EPR studies on other d3 (S = 1/2)
mononuclear RuV-oxo complexes (giso: 1.973−1.993, Table
S5).7e,h,l For the RuIII porphyrin complexes [RuIII(F20-TPP)-
(OEt)] (g = 2.53, 2.12, 1.89),4j [RuIII(F20-TPP)(OD)(X)] (OD
= oxygen donor; X = OD or other ligands, g = 2.55, 2.55,
2.05),4e and [RuIII(OEP)(PPh3)(Br)] (g = 2.31, 2.31, 1.98),33
at least one component of the g-tensor is markedly larger than
that for the RuV-oxo species; their giso values of 2.18−2.38 are
also considerably larger than that for the above-mentioned RuV-
oxo species. In the case of RuIII-aryl porphyrin complex, given
the absence of well-defined EPR signals in the spectrum of 1
(Figure 6a), we employed DFT calculations to compute the g
values, using a literature method,34 for [RuIII(TDCPP)(Ph)],
and also for A, both being in low spin (S = 1/2) state. The
computed g values are 2.831, 2.253, 2.037 (giso = 2.374) for
[RuIII(TDCPP)(Ph)] and 2.042, 2.015, 1.946 (giso = 2.001) for
A (Table S5); the former values are markedly different from,
whereas the latter ones compare well with, the experimental g
and giso values assigned to A. On the other hand, the
disappearance of the EPR signal assigned to A upon treatment
with styrene is in agreement with oxidation of styrene by the
RuV-oxo species, as also found from the UV−vis and ESI−MS
analyses.
The remarkable catalytic activity of arylruthenium(III)
porphyrins such as 1 for oxidation of hydrocarbons with m-
CPBA prompted us to identify the active intermediate in the
catalysis. UV−vis, ESI−MS, and EPR measurements consis-
tently revealed that treatment of 1 with m-CPBA generated a
species which is reactive with styrene. This reactive species is
formulated as [RuV(TDCPP)(O)(Ph)] (A) based on a number
of experimental studies together with DFT calculations.
In the UV−vis spectrum depicted in Figure 3b for the
reaction of 1 with m-CPBA, a band at 600−700 nm
characteristic of the porphyrin cation radical12 was not
observed, suggesting that the oxidation is metal-centered, in
accord with formation of A. This spectrum (Figure 3b; λmax
:
416, 512 nm) is considerably different from those of
[RuVI(TDCPP)(O)2] (λmax: 420, 516 nm)30a and
[RuIV(TDCPP)(OH)2] (λmax: 416, 527 nm).30b The strong
resemblance of the spectrum in Figure 3c, generated by
addition of styrene to the reaction mixture, to the spectrum of 1
(Figure 3a) showed that the Ru(III) species is highly
recoverable. In the reaction mixture after addition of styrene,
∼60% yield of styrene oxide was detected by GC, which can be
attributed to oxygen atom transfer from A to styrene.
High-resolution ESI−MS measurements (Figure 4) revealed
that addition of m-CPBA (1 equiv) to 1 generated a reactive
species assignable to the formulation [Ru(TDCPP)(O)(Ph)]
(A), which is stable up to minutes in the absence of styrene but
vanished immediately upon treatment with excess styrene,
suggesting a rapid oxygen atom transfer from A to styrene
rather than self-decomposition of A. The partial oxygen
exchange of A with 18OH2 also supported that it is a Ru-oxo
species. From the ESI−MS spectra depicted in Figure 4a−c, it
is evident that, after oxidation by m-CPBA followed by
reduction with styrene, 1 can be recovered without significant
decomposition, similar to the finding from UV−vis measure-
ments. We also examined the fate of 1, by ESI−MS analysis,
after the catalytic reactions of indan or tetralin by “1 + m-
The kinetic experiments lend further insight into the nature
and reactivity of A. The negative ρ+ value in the linear
Hammett plot for the C−H oxidation by ‘1 + m-CPBA’ (ρ+ =
−1.6, Figure 7a) is consistent with the involvement of an
electrophilic RuV-oxo species.4j This ρ+ value associated with A
is less negative than that for the C−H oxidation by ‘[RuII(F20-
TPP)(CO)] + 2,6-Cl2pyNO’ (ρ+ = −2.0) involving the
[RuV(F20-TPP)(O)(X)] (X = Cl− or OH−) species proposed
H
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
by Groves and co-workers,4j which can be ascribed to a higher
electrophilicity of the latter RuV-oxo species in which the trans
axial ligand (Cl− or OH−) is a weaker donor than the σ-Ph−
ligand. Also, the KIE value obtained in this work for
than that of O−H bond in the same molecule (e.g., for
CH3OH: BDECH = 96.1 kcal mol−1, BDEOH = 104.6 kcal
mol−1).38 In the case of hydroxylation of methylcyclohexane via
transition state TS3 (Figure S16), the activation barrier
obtained by DFT calculations (18.4 kcal mol−1) is also close
to that inferred from experimental k2 value (18.3 kcal mol−1);
these values are larger than that of ethylbenzene oxidation
involving transition state TS1. In TS1, the molecular orbital
indicating electron delocalization between the σ bond of C2−H
and the π nonbonding orbital of the aromatic ring can be easily
captured (Figure 11). Accordingly, the C2−H bond breaking
and O−H bond forming could be facilitated by the hyper-
conjugation effect, rendering TS1 more stable than TS3 and
thus resulting in a lower activation barrier. The ethylbenzene
and methylcyclohexane components in TS1 and TS3,
respectively, are oriented to induce almost linear O−H−C2
angles (171.7° in Figure 10, 169.9° in Figure S16), which are
similar to the virtually linear O−H−C angle in the transition
state for the hydrogen atom abstraction of methane by
[RuV(Por)(O)(SH)]36a and comparable to the computed
approaching angles in the C−H abstraction of cyclohexane
and methane by [RuV(N4O)(O)]2+ (O−H−C 170.2° and
177.6°, respectively).37 For the hydroxylation of ethylbenzene
and methylcyclohexane by 2A, the computed free energy
surfaces (Figure 10 and Figure S16) revealed that the rebound
process (following hydrogen atom abstraction) is barrierless,
which is reminiscent of the barrierless rebound revealed by the
computed energy profiles for the low-spin process of methane
hydroxylation by [RuV(Por)(O)(SH)].36b
methylcyclohexane oxidation (kH/kD 4.4
0.4) is close to
that for cyclohexane oxidation by ‘[RuIV(D4-Por*)(Cl)2] +
Cl2pyNO’ (kH/kD 4.2) in which RuV-oxo species may be
involved.4k In addition, the good linearity of the log (k′) versus
BDEC−H plot (Figure 7b) is in accord with the hydrogen atom
abstraction by metal-oxo species in the rate-determining step.25
We estimated the BDEO−H of [RuIV(TDCPP)(OH)(Ph)] by
using the k2 for the oxidation of DHA; the interpolated
BDEO−H value is 88 kcal mol−1 (Figure S18) based on the
Brønsted-Evans−Polanyi relationship.12b,35 This BDEO−H value
is larger than the BDEC−H values of the reactive C−H bonds of
the hydrocarbons such as ethylbenzene, fluorene, indan,
tetralin, xanthene, and estrone 3-methyl ether (BDEC−H 75.5
to 87.0 kcal mol−1),25 which were oxidized by ‘1 + m-CPBA’
with complete conversions and high-to-excellent yields (see
Table 1). On the basis of the k2 values for the oxidation of
various substrates (Table S4), the observed oxidation rates
follow a trend of 1-decene > 1-phenylethanol > ethylbenzene.
The high ketone-to-alcohol ratio for the oxidation of
ethylbenzene is consistent with the higher k2 for 1-phenyl-
ethanol than for its precursor (ethylbenzene).
According to the DFT calculations, the ground state of
[RuV(TDCPP)(O)(Ph)] (A) adopts a low spin d3 config-
uration with S = 1/2 doublet state (2A), consistent with the
finding by EPR studies. The quartet excited state (4A) is higher
in energy by 9.7 kcal mol−1. Previous DFT calculations by
Groves, Shaik, and co-workers36a for the hypothetical species
[RuV(Por)(O)(SH)] (the ruthenium analogue of the high-
valent metal-oxo species of cytochrome P450) also revealed a
doublet ground state (d3, S = 1/2) with the quartet state lying
considerably higher than the ground state. Similar findings have
also been obtained by DFT calculations on mononuclear
nonporphyrin RuV-oxo complexes including [RuV(bpy)2(O)-
(OH)]2+ (bpy = 2,2′-bipyridine),7h [RuV(bpy)(tpy)(O)]3+ (tpy
= 2,2′;6′,2″-terpyridine),32b and [RuV(N4O)(O)]2+ (N4OH =
bis(2-(2-pyridyl)ethyl)(2-hydroxy-2-(2-pyridyl)ethyl)amine).37
The RuV-oxo bond distance calculated for the six-coordinate 2A
(1.721 Å) is similar to that calculated for the six-coordinate
species [RuV(Por)(O)(SH)] (1.779 Å),36a [RuV(bpy)2(O)-
(OH)]2+ (1.715 Å),7h [RuV(bpy)(tpy)(O)]3+ (1.76 Å),32b and
[RuV(N4O)(O)]2+ (1.720 Å);37 these RuV-oxo distances are
longer than, yet comparable to, that in the X-ray crystal
structures of five-coordinate trigonal bipyramidal complexes
(Pr4N)[RuV(O2CC(O)Et2)2(O)] (1.687 Å; HO2CC(OH)Et2
= 2-hydroxy-2-ethylbutanoic acid)7b and (Pr4N)[RuV(PHAB)-
(O)] (1.702(3) Å; H4PHAB = 1,2-bis(2,2-diphenyl-2-
hydroxyethanamido)benzene).7e For the hydroxylation of 1-
We also attempted to detect [RuV(Por)(O)(Cl)] from the
oxidation of [RuIII(Por)(Cl)] (using 3 as an example) with
[nBu4N]IO4 or m-CPBA but without success. Some insight
could be gained into the role of axial Ph− ligand in A by
comparing the catalytic behavior of [RuIII(Por)(X)] (X = Ph−:
1, 2; Cl−: 3). While 1 catalyzed the oxidation of 1-decene with
[nBu4N]IO4 to give the epoxide in 75% yield, no reaction was
observed upon changing the catalyst from 1 (1.5 mol %) to 3
(2 mol %) under otherwise the same conditions (Table S2).
We further examined the catalytic behavior of 2 and 3 for the
oxidation of estrone 3-methyl ether with m-CPBA (2 equiv);
complete conversion with 80% yield of ketone (product in
entry 19, Table 1) was observed using 2 but only a 3% yield of
ketone (conversion 26%) was found using 3. Possibly,
compared with [RuV(Por)(O)(Cl)], [RuV(Por)(O)(Ph)] is
more readily generated in the reactions examined.
CONCLUSIONS
■
Arylruthenium(III) porphyrin complexes [RuIII(TDCPP)(Ph)-
(OEt2)] (1) and [RuIII(TDFPP)(Ph)(OH2)] (2) exhibit high
catalytic activity and selectivity for oxidation of hydrocarbons
with m-CPBA or [nBu4N]IO4 at room temperature, resulting in
C−H oxidation, or alkene epoxidation, of hydrocarbons in
limiting amounts with >70−99% product yields (27 examples).
Mechanistic studies by a number of spectroscopic measure-
ments including UV−vis, ESI−MS, EPR, and RR, along with
the kinetic studies and DFT calculations, support the formation
of a highly reactive low-spin d3 Ru(V)-oxo species,
[RuV(TDCPP)(O)(Ph)] and point to the involvement of
[RuV(TDCPP)(O)(Ph)] as the active intermediate in the 1-
catalyzed hydrocarbon oxidation reactions. The present work
demonstrated that an electron-rich axial ligand, such as σ-aryl
ligand, can significantly stabilize a highly reactive RuV-oxo
porphyrin species to make it detectable on the time scales of
2
ethylbenzene by A to give 1-phenylethanol via transition state
TS1 (Figure 10, left), the free energy barrier of 16.6 kcal mol−1
obtained from the DFT calculations is close to that (17.0 kcal
mol−1) inferred from the experimental k2 value (1.99 M−1 s−1,
Table S4) via transition state theory. The lower energy barrier
(10.6 kcal mol−1) calculated for the oxidation of 1-phenyl-
ethanol to acetophenone (Figure 10, right) could account for
the formation of acetophenone, instead of 1-phenylethanol, as
the dominant product in catalytic ethylbenzene oxidation (e.g.,
entry 1, Table 1). The C−H, other than O−H, hydrogen atom
abstraction of 1-phentylethanol by 2A (Figure 10, right) can be
rationalized by the smaller dissociation energy of C−H bond
I
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
typical spectroscopic methods. Insight gained into the RuIII and
RuV-oxo porphyrin system would be useful for the design of
new highly efficient Ru catalysts for hydrocarbon oxidation
reactions, including the asymmetric ones, under mild
conditions, and for extension to RuV-imido, the nitrogen-
analogue of RuV-oxo, to develop highly active and selective Ru
porphyrin catalysts for amination of C−H bonds.
(6) (a) Vanover, E.; Huang, Y.; Xu, L.; Newcomb, M.; Zhang, R. Org.
Lett. 2010, 12, 2246. (b) Zhang, R.; Vanover, E.; Luo, W.; Newcomb,
M. Dalton Trans. 2014, 43, 8749.
(7) For examples, see: (a) Che, C.-M.; Wong, K.-Y.; Mak, T. C. W. J.
Chem. Soc., Chem. Commun. 1985, 988. (b) Dengel, A. C.; Griffith, W.
P.; O’Mahoney, C. A.; Williams, D. J. J. Chem. Soc., Chem. Commun.
1989, 1720. (c) Che, C.-M.; Yam, V. W.-W.; Mak, T. C. W. J. Am.
Chem. Soc. 1990, 112, 2284. (d) Dengel, A. C.; Griffith, W. P. Inorg.
Chem. 1991, 30, 869. (e) Fackler, N. L. P.; Zhang, S.; O’Halloran, T.
V. J. Am. Chem. Soc. 1996, 118, 481. (f) Concepcion, J. J.; Jurss, J. W.;
Brennaman, M. K.; Hoertz, P. G.; Patrocinio, A. O. T.; Iha, N. Y. M.;
Templeton, J. L.; Meyer, T. J. Acc. Chem. Res. 2009, 42, 1954. (g) Liu,
C.-G.; Guan, W.; Yan, L.-K.; Su, Z.-M. Dalton Trans 2011, 40, 2967.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.8b04470.
(h) Planas, N.; Vigara, L.; Cady, C.; Miro
́
, P.; Huang, P.;
Abbreviations, experimental procedures, characterization
of compounds, computational details, Tables S1−S5,
Figures S1−S18 (PDF)
Hammarstrom, L.; Styring, S.; Leidel, N.; Dau, H.; Haumann, M.;
̈
Gagliardi, L.; Cramer, C. J.; Llobet, A. Inorg. Chem. 2011, 50, 11134.
(i) Duan, L.; Bozoglian, F.; Mandal, S.; Stewart, B.; Privalov, T.;
Llobet, A.; Sun, L. Nat. Chem. 2012, 4, 418. (j) Moonshiram, D.;
Alperovich, I.; Concepcion, J. J.; Meyer, T. J.; Pushkar, Y. Proc. Natl.
Acad. Sci. U. S. A. 2013, 110, 3765. (k) Liu, Y.; Ng, S.-M.; Yiu, S.-M.;
Lam, W. W. Y.; Wei, X.-G.; Lau, K.-C.; Lau, T.-C. Angew. Chem., Int.
Ed. 2014, 53, 14468. (l) Lebedev, D.; Pineda-Galvan, Y.; Tokimaru, Y.;
Crystallographic data for structure 1 (CIF)
Crystallographic data for structure 2 (CIF)
Crystallographic data for structure 3 (CIF)
AUTHOR INFORMATION
Corresponding Author
*cmche@hku.hk
ORCID
David Lee Phillips: 0000-0002-8606-8780
Xiao-Yong Chang: 0000-0003-1394-6164
Chi-Ming Che: 0000-0002-2554-7219
■
Fedorov, A.; Kaeffer, N.; Coper
2018, 140, 451.
́
et, C.; Pushkar, Y. J. Am. Chem. Soc.
(8) Chan, K.-H.; Guan, X.; Lo, V. K.-Y.; Che, C.-M. Angew. Chem.,
Int. Ed. 2014, 53, 2982.
(9) (a) Dolphin, D.; James, B. R.; Leung, T. Inorg. Chim. Acta 1983,
79, 25. (b) Leung, T.; James, B. R.; Dolphin, D. Inorg. Chim. Acta
1983, 79, 180.
(10) (a) Collman, J. P.; Brothers, P. J.; McElwee-White, L.; Rose, E. J.
Am. Chem. Soc. 1985, 107, 6110. (b) Sishta, C.; Ke, M.; James, B. R.;
Dolphin, D. J. Chem. Soc., Chem. Commun. 1986, 787. (c) Ke, M.;
Rettig, S. J.; James, B. R.; Dolphin, D. J. Chem. Soc., Chem. Commun.
1987, 1110. (d) Seyler, J. W.; Leidner, C. R. Inorg. Chem. 1990, 29,
3636. (e) Ke, M.; Sishta, C.; James, B. R.; Dolphin, D.; Sparapany, J.
W.; Ibers, J. A. Inorg. Chem. 1991, 30, 4766. (f) Collman, J. P.; Rose,
E.; Venburg, G. D. J. Chem. Soc., Chem. Commun. 1994, 11.
(g) Alexander, C. S.; Rettig, S. J.; James, B. R. Organometallics 1994,
13, 2542.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Hong Kong Research Grants
Council (HKU 17303815 and 17301817), the National Key
Basic Research Program of China (No. 2013CB834802), and
B a s i c R e s e a r c h P r o g r a m o f S h e n z h e n ( N o .
JCYJ20160229123546997, JCYJ20170412140251576, and
JCYJ20170818141858021).
(11) Li, Y.; Huang, J.-S.; Xu, G.-B.; Zhu, N.; Zhou, Z.-Y.; Che, C.-M.;
Wong, K.-Y. Chem. - Eur. J. 2004, 10, 3486.
(12) (a) Groves, J. T.; Haushalter, R. C.; Nakamura, M.; Nemo, T.
E.; Evans, B. J. J. Am. Chem. Soc. 1981, 103, 2884. (b) Bell, S. R.;
Groves, J. T. J. Am. Chem. Soc. 2009, 131, 9640.
REFERENCES
■
(1) Selected reviews: (a) Meunier, B. Chem. Rev. 1992, 92, 1411.
(b) Groves, J. T. Cytochrome P450: Structure, Mechanism, and
Biochemistry; Ortiz de Montellano, P. R., Ed.; Kluwer Academic/
Plenum Publishers: New York, 2005, 3rd ed., p 1. (c) Nam, W. Acc.
Chem. Res. 2007, 40, 522.
(2) (a) Groves, J. T.; Nemo, T. E.; Myers, R. S. J. Am. Chem. Soc.
1979, 101, 1032. (b) Chang, C. K.; Kuo, M.-S. J. Am. Chem. Soc. 1979,
101, 3413. (c) Chang, C. K.; Ebina, F. J. Chem. Soc., Chem. Commun.
1981, 778.
́
(13) (a) Mas-Balleste, R.; Que, L., Jr. Science 2006, 312, 1885.
(b) Zhou, M.; Crabtree, R. H. Chem. Soc. Rev. 2011, 40, 1875.
(c) Costas, M. Coord. Chem. Rev. 2011, 255, 2912. (d) Talsi, E. P.;
Bryliakov, K. P. Coord. Chem. Rev. 2012, 256, 1418. (e) White, M. C.
Science 2012, 335, 807. (f) Sun, M.; Zhang, J.; Putaj, P.; Caps, V.;
Lefebvre, F.; Pelletier, J.; Basset, J.-M. Chem. Rev. 2014, 114, 981.
(g) Wang, V. C.-C.; Maji, S.; Chen, P. P.-Y.; Lee, H. K.; Yu, S. S.-F.;
Chan, S. I. Chem. Rev. 2017, 117, 8574. (h) Grant, J. T.; Venegas, J.
M.; McDermott, W. P.; Hermans, I. Chem. Rev. 2018, 118, 2769.
(14) (a) Srinivasan, R.; Chandrasekharam, M.; Vani, P. V. S. N.;
Chida, A. S.; Singh, A. K. Synth. Commun. 2002, 32, 1853. (b) Capriati,
V.; Florio, S.; Perna, F. M.; Salomone, A.; Abbotto, A.; Amedjkouh,
M.; Nilsson Lill, S. O. Chem. - Eur. J. 2009, 15, 7958.
(3) Pan, Z.; Wang, Q.; Sheng, X.; Horner, J. H.; Newcomb, M. J. Am.
Chem. Soc. 2009, 131, 2621.
(4) For examples, see: (a) Groves, J. T.; Quinn, R. J. Am. Chem. Soc.
1985, 107, 5790. (b) Marchon, J.-C.; Ramasseul, R. J. Chem. Soc.,
Chem. Commun. 1988, 298. (c) Ohtake, H.; Higuchi, T.; Hirobe, M. J.
Am. Chem. Soc. 1992, 114, 10660. (d) Higuchi, T.; Satake, C.; Hirobe,
M. J. Am. Chem. Soc. 1995, 117, 8879. (e) Groves, J. T.; Bonchio, M.;
Carofiglio, T.; Shalyaev, K. J. Am. Chem. Soc. 1996, 118, 8961.
(f) Gross, Z.; Ini, S. Inorg. Chem. 1999, 38, 1446. (g) Yu, X.-Q.;
Huang, J.-S.; Yu, W.-Y.; Che, C.-M. J. Am. Chem. Soc. 2000, 122, 5337.
(h) Ito, R.; Umezawa, N.; Higuchi, T. J. Am. Chem. Soc. 2005, 127,
834. (i) Zhang, J.-L.; Che, C.-M. Chem. - Eur. J. 2005, 11, 3899.
(j) Wang, C.; Shalyaev, K. V.; Bonchio, M.; Carofiglio, T.; Groves, J. T.
Inorg. Chem. 2006, 45, 4769. (k) Zhang, R.; Yu, W.-Y.; Che, C.-M.
Tetrahedron: Asymmetry 2005, 16, 3520.
(15) (a) Zhang, J.-L.; Che, C.-M. Org. Lett. 2002, 4, 1911. (b) Zhang,
J.-L.; Huang, J.-S.; Che, C.-M. Chem. - Eur. J. 2006, 12, 3020.
(16) Mohajer, D.; Karimipour, G.; Bagherzadeh, M. New J. Chem.
2004, 28, 740.
(17) Compared with an analogous alkene with a shorter chain:
Hodgson, D. M.; Chung, Y. K.; Nuzzo, I.; Freixas, G.; Kulikiewicz, K.
K.; Cleator, E.; Paris, J.-M. J. Am. Chem. Soc. 2007, 129, 4456.
(18) Ho, C.; Leung, W.-H.; Che, C.-M. J. Chem. Soc., Dalton Trans.
1991, 2933.
(19) Chan, W.-K.; Wong, M.-K.; Che, C.-M. J. Org. Chem. 2005, 70,
(5) For a review, see: Che, C.-M.; Huang, J.-S. Chem. Commun. 2009,
3996.
4226.
J
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(20) Liu, P.; Liu, Y.; Wong, E. L.-M.; Xiang, S.; Che, C.-M. Chem. Sci.
2011, 2, 2187.
(21) [nBu4N]IO4 was not used as oxidant in the ESI−MS analysis to
avoid hampering the signal by n-tetrabutylammonium cation.
(22) Paeng, I. R.; Nakamoto, K. J. Am. Chem. Soc. 1990, 112, 3289.
(23) We noted the absence of EPR signal in the previously reported
X-band EPR spectrum of the reaction mixture of [RuIV(Por)(Cl)2]
and Zn(Hg) in CH2Cl2 (spectrum recorded after quenched with liquid
N2); this reaction is reported to generate a RuIII(F20-TPP) species (see
ref 4j).
(24) For examples of paramagnetic metal complexes with noninteger
spin states (including S = 1/2) reported to be EPR-silent at low
temperatures, due possibly to reasons including fast relaxation, see:
(a) Ruggiero, C. E.; Carrier, S. M.; Antholine, W. E.; Whittaker, J. W.;
Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 1993, 115, 11285.
(b) Baumann, F.; Kaim, W.; García Posse, M.; Katz, N. E. Inorg. Chem.
1998, 37, 658. (c) Schneider, J. L.; Carrier, S. M.; Ruggiero, C. E.;
Young, V. G., Jr.; Tolman, W. B. J. Am. Chem. Soc. 1998, 120, 11408.
(d) Kundu, T.; Sarkar, B.; Mondal, T. K.; Fiedler, J.; Mobin, S. M.;
Kaim, W.; Lahiri, G. K. Inorg. Chem. 2010, 49, 6565. (e) Vaidya, S.;
̀
Shukla, P.; Tripathi, S.; Riviere, E.; Mallah, T.; Rajaraman, G.;
Shanmugam, M. Inorg. Chem. 2018, 57, 3371.
(25) Bryant, J. R.; Mayer, J. M. J. Am. Chem. Soc. 2003, 125, 10351.
(26) (a) Seyler, J. W.; Leidner, C. R. J. Chem. Soc., Chem. Commun.
1989, 1794. (b) Seyler, J. W.; Safford, L. K.; Leidner, C. R. Inorg.
Chem. 1992, 31, 4300. (c) Collman, J. P.; Ha, Y.; Wagenknecht, P. S.;
Lopez, M.-A.; Guilard, R. J. Am. Chem. Soc. 1993, 115, 9080.
(27) (a) Baeyer, A.; Villiger, V. Ber. Dtsch. Chem. Ges. 1899, 32, 3625.
(b) Hassall, C. H. Org. React. 1957, 9, 73.
(28) Chen, J.; Che, C.-M. Angew. Chem., Int. Ed. 2004, 43, 4950.
(29) Du, Y.-D.; Tse, C.-W.; Xu, Z.-J.; Liu, Y.; Che, C.-M. Chem.
Commun. 2014, 50, 12669.
(30) (a) Liu, C.-J.; Yu, W.-Y.; Che, C.-M.; Yeung, C.-H. J. Org. Chem.
1999, 64, 7365. (b) Dubourdeaux, P.; Tavares, M.; Grand, A.;
̀
Ramasseul, R.; Marchon, J.-C. Inorg. Chim. Acta 1995, 240, 657.
(31) Hurst, J. K.; Zhou, J.; Lei, Y. Inorg. Chem. 1992, 31, 1010.
(32) (a) Murakami, M.; Hong, D.; Suenobu, T.; Yamaguchi, S.;
Ogura, T.; Fukuzumi, S. J. Am. Chem. Soc. 2011, 133, 11605.
(b) Pushkar, Y.; Moonshiram, D.; Purohit, V.; Yan, L.; Alperovich, I. J.
Am. Chem. Soc. 2014, 136, 11938.
(33) James, B. R.; Dolphin, D.; Leung, T. W.; Einstein, F. W. B.;
Willis, A. C. Can. J. Chem. 1984, 62, 1238.
(34) Scheibel, M. G.; Askevold, B.; Heinemann, F. W.; Reijerse, E. J.;
de Bruin, B.; Schneider, S. Nat. Chem. 2012, 4, 552.
(35) (a) Wang, K.; Mayer, J. M. J. Am. Chem. Soc. 1997, 119, 1470.
(b) Mayer, J. M. Acc. Chem. Res. 1998, 31, 441.
(36) (a) Sharma, P. K.; de Visser, S. P.; Ogliaro, F.; Shaik, S. J. Am.
Chem. Soc. 2003, 125, 2291. (b) Ogliaro, F.; de Visser, S. P.; Groves, J.
T.; Shaik, S. Angew. Chem., Int. Ed. 2001, 40, 2874.
(37) Guan, X.; Chan, S. L.-F.; Che, C.-M. Chem. - Asian J. 2013, 8,
2046.
(38) Blanksby, S. J.; Ellison, G. B. Acc. Chem. Res. 2003, 36, 255.
K
DOI: 10.1021/jacs.8b04470
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX