Alternative selective oxidation pathways for aldehyde oxidation and alkene epoxidation on a SiO2-supported Ru-monomer complex catalyst

Article abstract of DOI:10.1021/ja9079513

We have prepared a novel Ru-mononer complex supported on a SiO₂ surface by using a Rumonomer complex precursor with a p-cymene ligand, which was found to be highly active for the selective oxidation of aldehydes and the epoxidation of alkenes using O₂. The structure of the supported Ru catalyst was characterized by means of FT-IR, solid-state NMR, diffuse-reflectance UV/vis, XPS, Ru K-edge EXAFS, and DFT calculations, which demonstrated the formation of isolatedly located, unsaturated Ru centers behind a p-cymene ligand of the Ru-complex precursor. The site-isolated Ru-monomer complex on SiO₂ achieved tremendous TONs (turnover numbers) for the selective oxidation of aldehydes and alkenes; e.g. TONs of 38,800,000 for selective isobutyraldehyde (IBA) oxidation and 2,100,000 for trans-stilbene epoxidation at ambient temperature, which are among the highest TONs in metal-complex catalyzes to our knowledge. We also found that the IBA sole oxidation with an activation energy of 48 kJ mol^-1 much more facile than the trans-stilbene epoxidation with an activation energy of 99 kJ mol ^-1 was completely suppressed by the coexistence of trans-stilbene. The switchover of the selective oxidation pathways from the IBA oxidation to the trans-stilbene epoxidation was explained in terms of energy profiles for the alternative selective oxidation pathways, resulting in the preferential coordination of trans-stilbene to the Ru-complex at the surface. This aspect gives an insight into the origin of the efficient catalysis for selective epoxidation of alkenes with IBA/O₂.

Full text of DOI:10.1021/ja9079513

Published on Web 12/15/2009
Alternative Selective Oxidation Pathways for Aldehyde
Oxidation and Alkene Epoxidation on a SiO2-Supported
Ru-Monomer Complex Catalyst
Mizuki Tada,*,† Satoshi Muratsugu,† Mutsuo Kinoshita,† Takehiko Sasaki,‡ and
Yasuhiro Iwasawa§
Institute for Molecular Science, 38 Nishigo-Naka, Myodaiji, Okazaki, Aichi 444-8585, Japan,
Department of Complexity Science and Engineering, Graduate School of Frontier Sciences,
The UniVersity of Tokyo, Kashiwanoha, Kashiwa, Chiba 277-8561, Japan, and
Department of Applied Physics and Chemistry, The UniVersity of Electro-Communications,
Chofu, Tokyo 182-8585, Japan
Received September 23, 2009; E-mail: mtada@ims.ac.jp
Abstract: We have prepared a novel Ru-mononer complex supported on a SiO2 surface by using a Ru-
monomer complex precursor with a p-cymene ligand, which was found to be highly active for the selective
oxidation of aldehydes and the epoxidation of alkenes using O2. The structure of the supported Ru catalyst
was characterized by means of FT-IR, solid-state NMR, diffuse-reflectance UV/vis, XPS, Ru K-edge EXAFS,
and DFT calculations, which demonstrated the formation of isolatedly located, unsaturated Ru centers
behind a p-cymene ligand of the Ru-complex precursor. The site-isolated Ru-monomer complex on SiO2
achieved tremendous TONs (turnover numbers) for the selective oxidation of aldehydes and alkenes; e.g.
TONs of 38,800,000 for selective isobutyraldehyde (IBA) oxidation and 2,100,000 for trans-stilbene
epoxidation at ambient temperature, which are among the highest TONs in metal-complex catalyzes to
our knowledge. We also found that the IBA sole oxidation with an activation energy of 48 kJ mol-1 much
more facile than the trans-stilbene epoxidation with an activation energy of 99 kJ mol-1 was completely
suppressed by the coexistence of trans-stilbene. The switchover of the selective oxidation pathways from
the IBA oxidation to the trans-stilbene epoxidation was explained in terms of energy profiles for the alternative
selective oxidation pathways, resulting in the preferential coordination of trans-stilbene to the Ru-complex
at the surface. This aspect gives an insight into the origin of the efficient catalysis for selective epoxidation
of alkenes with IBA/O2.
and highly active for catalytic reactions.4-14 Chemical bonding
between metal complexes and a support surface prevents
gathering the unsaturated active metal species, and hence the
resultant site isolation provides a stabilization effect on such
labile structures. The coordinatively unsaturated metal center
may provide an efficient reaction pathway in the coordination
sphere via adsorption of multiple reactants on the metal center.
Recently, we have succeeded in preparing novel unsaturated
Ru-diamine monomer complexes on a SiO2 surface, which are
active for selective oxidation reactions of alkenes and cycloal-
1. Introduction
Many useful chemicals are produced by heterogeneous solid
catalysts in industrial processes, and the merits of heterogeneous
catalysts are not only the separation of catalysts and products
from reaction media but also their high durability and unique
activities derived from the site-isolation and structure of
catalytically active sites at solid surfaces. Relationship among
the structure, organized environment, and catalytic property of
surface species tells us how to establish a new strategy for
rational design of heterogeneous catalysts.
Supporting organic and inorganic metal complexes on
surfaces is a promising way to produce molecularly regulated
structures of active metal sites at surfaces, and various surface
metal structures can be arranged by stepwise structural trans-
formations in a controllable manner.1-3 Metal complex attach-
ment and subsequent thermal and chemical treatments often
produce novel unsaturated active metal centers, which are unique
(4) Iwasawa, Y. Acc. Chem. Res. 1997, 30, 103.
(5) Thomas, J. M.; Raja, R.; Lewis, D. W. Angew. Chem., Int. Ed. 2005,
44, 6456.
(6) Tada, M.; Iwasawa, Y. Chem. Commun. 2006, 2833.
(7) Notestein, J.; Katz, A. Chem.sEur. J. 2006, 12, 3954.
(8) McMorn, P.; Hutchings, G. J. Chem. Soc. ReV. 2004, 33, 108.
(9) Sheldon, R. A.; Wallau, M.; Arends, I. W. C.; Schuchardt, U. Acc.
Chem. Res. 1998, 31, 485.
(10) Corma, A.; Garcia, H. AdV. Synth. Catal. 2006, 348, 1391.
(11) Coperet, C.; Basset, J. M. AdV. Synth. Catal. 2007, 349, 78.
(12) Fierro-Gonzalez, J. C.; Kuba, S.; Hao, Y. L.; Gates, B. C. J. Phys.
Chem. B 2006, 110, 13326.
† Institute for Molecular Science.
‡ The University of Tokyo.
(13) Yamaguchi, K.; Yoshida, C.; Uchida, S.; Mizuno, N. J. Am. Chem.
Soc. 2005, 127, 530.
§ The University of Electro-Communications.
(1) Iwasawa, Y. AdV. Catal. 1987, 35, 187.
(14) Bianchini, C.; Dal Santo, V.; Meli, A.; Moneti, S.; Moreno, M.;
Oberhauser, W.; Psaro, R.; Sordelli, L.; Vizza, F. J. Catal. 2003, 213,
47.
(2) Hartley, F. R. Supported Metal Complexes; Reidel: Dordrecht, 1985.
(3) Iwasawa, Y. Tailored Metal Catalysts; Reidel: Dordrecht, 1986.
9
10.1021/ja9079513  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 713–724 713

A R T I C L E S
Tada et al.
kanes using O2.15,16 A p-cymene ligand of the supported Ru-
diamine complex can be extracted to create the coordinatively
unsaturated active Ru center at the surface by two novel methods
that we have developed: (1) the reaction-induced selective
elimination of the p-cymene ligand15 and (2) the photoinduced
selective elimination of the p-cymene ligand.16 Although the
formation of unsaturated metal species is generally endothermic
andnecessitatedbycompulsoryevacuationathightempreture,17-21
we succeeded in producing the unsaturated active Ru center in
conjunction with the exothermic reaction of reactant conversion
and subsequent ligand exchange on a supported Ru complex.15
The exothermic reaction of isobutyraldehyde (IBA) to isobutyric
acid promoted the endothermic elimination of a p-cymene ligand
from the supported Ru complex. The unsaturated Ru complex
thus formed on the surface exhibited high activity and selectivity
for epoxidation of alkenes. On the other hand, the irradiation
of the SiO2-supported Ru complex with UV light brought about
the selective elimination of a p-cymene ligand and the im-
mobilization of the Ru complex at double sites on the SiO2
surface. The doubly immobilized Ru complex was active for
the photooxidation of cycloalkanes using O2 as an oxidant.16
In this contribution, we report the characterization and
alternative selective oxidation catalysis (aldehyde oxidation and
alkene epoxidation) of the SiO2-supported unsaturated Ru-
monomer complex produced by the reactant-induced catalyst
activation. The selective alkene epoxidation using IBA and O2
on the Ru-monomer complex proceeded with a tremendous TON
(turnover number) (e.g., 2.1 × 106 for trans-stilbene epoxidation)
without catalytic IBA oxidation, which may be of interest,
because the catalytic oxidation of IBA itself with O2 in the
absence of trans-stilbene proceeded more rapidly with the larger
TON of 3.9 × 107 and TOF (turnover frequency) of 1.6 × 103
s-1 at ambient temperature, which are both among the highest
values in metal complex catalysis, to our knowledge. The
switchover of the reaction pathways from the aldehyde oxidation
to the alkene epoxidation is also discussed in terms of energy
profiles for the alternative selective oxidation pathways at the
surface to promote the selective epoxidation of alkenes.
C 53.08 (52.88), H 6.24 (6.22), N 12.38 (12.25). 1H NHR (CDCl3):
2.17 (brm, 2H; NH2), 2.81 (m, 2H; CH2), 2.98 (m, 2H; CH2), 5.43
(d, 1H; CHdCH2), 5.87 (d, 1H; CHdCH2), 6.74 (dd, 2H; CH )
CH2), 7.52 (d, 2H; C6H4), 7.82 (d, 2H; C6H4).
Synthesis
of
Ru
Precursor
[(p-Cymene)Ru
{H2NCH2CH2NHSO2C8H7}Cl] (A). The diamine ligand
H2NCH2CH2NHSO2C8H7 (226 mg, 1 mmol) dissolved in CH2Cl2/
CH3OH (1/1 15 mL) and NaOCH3 (1 mmol) in CH3OH (0.5 mL)
were slowly added to a CH2Cl2 solution (20 mL) of Ru2(p-
cymene)2Cl4 (306 mg, 0.5 mmol). The mixture was stirred for 3 h
under N2 atmosphere, and then the solvent was evaporated. The
residue was dissolved in CH2Cl2/diethylether (2/1, 30 mL), and after
filtration the solution was reduced to 2 mL under vacuum. n-Hexane
(15 mL) was added, and the precipitate was dried under vacuum.
Elemental analysis (calculated % for C20H27ClN2O2RuS H2O
1
(observed %)): C 46.73 (46.59), H 5.69 (5.65), N 5.45 (5.39). H
NHR (CDCl3, 298 K): 1.11 (d, 6H; CH(CH3)2), 2.01 (s, 3H; CH3
(p-cymene)), 2.12 (brm, 4H; NCH2), 2.64 (sept, 1H; CH(CH3)2),
5.08 (d, 1H; CHdCH2) 5.13 (brd, 2H; CH (p-cymene)), 5.40 (d,
2H; CH (p-cymene)), 5.64 (d, 1H; CHdCH2), 6.50 (dd, 1H; CH )
CH2), 7.23 (d, 2H; C6H4), 7.67 (d, 2H; C6H4). 13C NMR (CDCl3):
18.56, 22.53 (CH3, p-cymene), 30.39 (CH, p-cymene), 47.27, 48.51
(NCH2), 81.30 (CH, p-cymene), 96.12 (C, p-cymene), 102.21,
115.40, 125.81, 127.47, 135.92, 139.36, 142.80 (C6H4-CH)CH2).
Preparation of a Supported Ru Complex (C). SiO2 (1.5 g,
Aerosil 200; Degussa) was calcined at 673 K for 2 h and refluxed
in a solution of p-styryltrimethoxysilane (0.11 mmol) in anhydrous
toluene at 383 K for 18 h under N2. The modified SiO2 was filtrated
and then dried under vacuum. The amount of grafted p-styryl moiety
was 0.23 nm-2. The prepared Ru precursor (A) (38 mg, 0.074
mmol) was reacted with SiO2 functionalized with p-styryltrimethox-
ysilane in anhydrous CHCl3 (20 mL) in the presence of AIBN,
then refluxed at 333 K for 24 h under N2. The obtained SiO2-
supported catalyst was filtrated, washed with CHCl3, and dried under
vacuum. Loading of Ru complex was 0.4 wt % by XRF. A
supported Ru complex with 2.7 wt % of Ru loading was prepared
with the similar procedure.
Preparation of an Unsaturated Ru Complex on SiO2 (D). The
supported Ru complex (C) (1.5 g, Ru 0.4 wt %) was suspended in
anhydrous CH2Cl2 (20 mL). Neat isobutyraldehyde (0.2 mL) was
added to the suspension, and the mixture was stirred for 3 h under
O2 atmosphere, accompanied with change of the color from orange
to pale black. Then the obtained catalyst was filtrated, washed with
CH2Cl2, and dried under vacuum.
2. Experimental Section
Catalyst Preparation. Preparation of a Diamine Ligand
H2NCH2CH2NHSO2C8H7. A CH2Cl2 solution (15 mL) of 4-vi-
nylbenzensulfonyl chloride (5 mL) was dropped into 5 mL of
ethylene diamine dissolved in 10 mL of CH2Cl2, and the mixture
was stirred for 3 h. CH2Cl2 (200 mL), H2O (100 mL), and HCl/
H2O (2 M, 200 mL) were added to the solution, and the water layer
was washed with CH2Cl2 twice. After filtration, the water layer
was neutralized by KOH solution (2 M, pH 9-10) and extracted
with 150 mL of CH2Cl2 three times. The organic layers were dried
with Na2SO4, and filtrated, and the solvents were evaporated.
Elemental analysis (calculated % for C10H14N2O2S (observed %)):
Catalyst Characterization. XPS. XPS spectra were recorded
on a Rigaku XPS-7000 apparatus at a base pressure of 6 × 10-6
Pa. The X-ray source, voltage, and current were Mg Ka, 10 kV,
and 30 mA, respectively. Binding energies were referred to C 1s
of 284.8 eV.
1
1H and 13C NMR. H and 13C NMR spectra in a liquid phase
were obtained on JNM-AL400 spectrometers at 400 MHz in CDCl3
with TMS of an internal standard. Solid state 13C MAS NMR
spectra (MAS rate ) 6 kHz) were recorded with a Chemagnetics
CMX-300 spectrometer operating at 75.5 MHz. 13C MAS NMR
spectra with cross-polarization (CP) were acquired with contact time
5.0 ms. The rotor spin rate was 6 kHz, with delay time of 2 s.
Hexamethylbenzene (17.17 and 176.46 ppm) was used as an
external standard for the calibration of chemical shifts.
Ru K-edge EXAFS. EXAFS spectra at Ru K-edge were
measured in a transmission mode at 15 K at the NW10A station of
the PF-AR ring at High Energy Accelerator Research Organization
(KEK). The energy and current of electrons in the storage ring were
6.5 GeV and 50-60 mA, respectively. X-rays from the storage
ring were monochromatized by Si(311) channel cut crystals.
Ionization chambers filled with pure Ar and Kr gases were used to
monitor the incident and transmitted X-rays, respectively. EXAFS
(15) Tada, M.; Coquet, R.; Yoshida, M.; Kinoshita, M.; Iwasawa, Y. Angew.
Chem., Int. Ed. 2007, 46, 7220.
(16) Tada, M.; Akatsuka, Y.; Yang, Y.; Sasaki, T.; Kinoshita, M.; Motokura,
K.; Iwasawa, Y. Angew. Chem., Int. Ed. 2008, 47, 9252.
(17) Asakura, K.; Bando, K. K.; Iwasawa, Y.; Arakawa, H.; Isobe, K. J. Am.
Chem. Soc. 1990, 112, 9096.
(18) Izumi, Y.; Chihara, H.; Yamazaki, H.; Iwasawa, Y. J. Phys. Chem.
1994, 98, 594.
(19) Asakura, K.; Noguchi, Y.; Iwasawa, Y. J. Phys. Chem. B 1999, 103,
1051.
(20) Taoufik, M.; Santini, C. C.; Candy, J.-P.; de Mallmann, A.; Basset,
J. M. J. Am. Chem. Soc. 1996, 118, 4167.
(21) Humblot, F.; Didillon, D.; Lepeltier, F.; Candy, J. P.; Corker, J.; Clause,
O.; Bayard, F.; Basset, J. M. J. Am. Chem. Soc. 1998, 120, 137.
9
714 J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010

Switchover of Selective Oxidation Pathways on Ru Catalyst
A R T I C L E S
spectra were analyzed with the UWXAFS package.22 Threshold
energy E0 was tentatively set at the inflection point of the absorption
edge, and background was subtracted by the AUTOBK program.
k3-Weighted EXAFS oscillations were Fourier transformed into
R-space. Curve-fitting analysis was carried out using the FEFFIT
program in the R-space. k-Range for the Fourier transformation
was 30-140 nm-1. Fitting parameters were coordination numbers
(CN), interatomic distances (R), Debye-Waller factors (σ), and a
correction-of-edge energy (∆E0). The same ∆E0 was used for all
the shells in a sample. Phase shifts and backscattering amplitudes
for Ru-C, Ru-N, and Ru-Cl bonds were calculated by the FEFF8
code.23 The coefficient of multiphoton effect (S0 ) was fixed as 1.0
2
for the fitting.
DFT Calculations. Structural optimizations based on DFT were
performed using the Gaussian03 package.24 The PBE0 hybrid
functional25 was employed with the DGDZVP basis set.26 The
DGDZVP basis set is equivalent to Pople’s Hartree-Fock optimized
6-31G* basis set.27 Unlike the popular B3LYP,28 PBE0 is a
nonempirical hybrid GGA functional and provides remarkable
results for a number of physicochemical observables without any
adjustable parameters.29,30 Relative energies of singlet and triplet
spin states were assessed to obtain the lowest energy structures. It
was found that the system with the Ru complex interacting with
O2, IBA, and trans-stilbene favors the triplet state throughout the
reaction paths. Transition states were optimized by the synchronous
transit-guided quasi-Newton (STQN) method, and they were
confirmed by the existence of a single negative frequency.
Catalytic Oxidation Reactions. Selective Catalytic
Oxidation of Aldehydes. Each catalyst was suspended in a CH2Cl2
solution under 101.3 kPa of oxygen in a closed glass reactor
equipped with a mechanical stirrer. Aldehyde oxidation reactions
were carried out at 288 K by the addition of neat aldehyde to the
suspension (Ru/aldehyde ) 1/200 in a molar ratio). Reactants and
products were analyzed by GC (Shimadzu GC-17A) and GC-MS
(Shimadzu GC-2010) at an appropriate interval. Activation energies
(Ea) were calculated using reaction rates in the temperature range
of 281-313 K, and reaction orders were examined in the aldehyde
concentration range of 0.025-0.4 M.
Figure 1. Structures of a SiO2-supported Ru catalyst for selective oxidation.
Catalytic Epoxidation of Alkenes. Each catalyst was suspended
in a CH2Cl2 solution under 101.3 kPa of oxygen in a closed glass
reactor equipped with a mechanical stirrer. Alkene epoxidation
reactions were conducted at 298 K by the addition of an alkene
reactant and IBA to the suspension (Ru/IBA/alkene ) 1/200/200
in a molar ratio). Reactants and products were analyzed by GC
(Shimadzu GC-17A) and GC-MS (Shimadzu GC-2010) at an
appropriate interval. Activation energies (Ea) were estimated in the
temperature range of 283-314 K, and the reaction orders to
aldehyde and alkene were also determined.
IBA ) 1/200/200) under the atmospheric pressure of O2 at 288 K.
After 3 and 24 h, the products of R-epoxide and ꢀ-epoxide were
analyzed by 1H NMR (H6 of R-epoxide: 2.9 ppm and that of
ꢀ-epoxide: 3.1 ppm).31
Iodometry Analysis. NaI was dissolved in 2-propanol and 10%
acetic acid was added to the solution. The supported Ru catalyst
(D) was immediately added to the mixture after its catalytic reaction
and subsequent filtration, and refluxed at 333 K for 0.5 h. The
mixture was titrated using a 0.1 N sodium thiosulfate solution.
Stereoselective Epoxidation of Cholesterol Benzoate. The
epoxidation of 5,6-double bond in cholesteryl benzoate (0.5 mmol)
was carried out using m-CPBA (0.5 mmol) in CH2Cl2 5 mL under
N2 atmosphere at 288 K. The same epoxidation was performed on
the supported Ru catalyst (D) (Ru 0.4 wt %, 60 mg) with cholesteryl
benzoate (0.5 mmol) in CH2Cl2 5 mL (Ru/cholesteryl benzoate/
3. Results and Discussion
3.1. Catalytically Active Structure of the Supported
Ru-Monomer Complex on a SiO2 Surface. The SiO2-supported
unsaturated Ru-monomer complex was prepared via step-by-
step reactions in three steps as illustrated in Figure 1. The Ru
complexes synthesized/transformed on the surface have been
characterized by 13C and 29Si solid-state NMR, FT-IR, diffuse-
reflectance UV/vis, Ru 3d5/2 XPS, Ru K-edge EXAFS, and
hybrid DFT calculations. Here, the structure of the active Ru
complex supported on SiO2 is discussed, summarizing our
previous results and presenting new data.15 p-Styryltrimethox-
ysilane was tethered on the surface of SiO2, whose density at
the surface was controlled to be a given density in the range of
0.1 - 0.8 nm2. 13C and 29Si solid-state NMR revealed that the
p-styryl moiety was tethered on the SiO2 surface (species B in
Figure 1; and in a and d of Figure 2). The appearance of new
(22) Stern, E. A.; Newville, M.; Ravel, B.; Yacoby, Y.; Haskel, D. Physica
B 1995, 208&209, 117.
(23) Ankudinov, A. L.; Ravel, B.; Rehr, J.; Conradson, S. D. Phys. ReV. B
1998, 58, 7565.
(24) Frisch, M. J., et al., GAUSSIAN 03, Rev. C. 02; Gaussian Inc.:
Wallingford, CT, 2004. Complete ref available in the Supporting
Information.
(25) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. ReV. Lett. 1996, 77,
3865.
(26) Sosa, C.; Andzelm, J.; Elkin, B. C.; Wimmer, E.; Dobbs, K. D.; Dixon,
D. A. J. Phys. Chem. 1992, 96, 6630.
(27) Hariharan, P. C.; Pople, J. A. Chem. Phys. Lett. 1972, 66, 217.
(28) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(29) Adamo, C.; Scuseria, G. E.; Barone, V. J. Chem. Phys. 1999, 111,
2889.
(30) Scalmani, G.; Bredas, J. L.; Barone, V. J. Chem. Phys. 2000, 112,
1178.
(31) Yamada, T.; Imagawa, K.; Mukaiyama, T. Chem. Lett. 1992, 2109.
9
J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010 715

A R T I C L E S
Tada et al.
Figure 2. 13C liquid and solid-state NMR spectra of (a) p-styryltrimethoxysilane, (b) N-sulfonyl-1,2-ethylenediamine, (c) the Ru precursor complex (A), (d)
the immobilized p-styryltrimethoxysilane on SiO2 (B), (e) the supported Ru complex (C), and (f) the supported Ru catalyst after the removal of p-cymene
(D). Dotted lines represent peaks attributed to p-cymene.
Table 1. Ru Loadings, Released p-Cymene Amounts, and XPS Binding Energies of the Homogeneous Ru Precursor Complex (A), the
Supported Ru Catalysts (C) and (D)
XPS binding energy
d
catalyst
Ru wt %a
Ru wt %a,b
detected p-cymene/Ruc
O 1sd
Si 2pd
Ru 3d5/2
Ru complex (A)
-
-
-
531.7
532.9
532.9
533.0
533.0
-
281.6
281.6
281.9
281.5
282.0
supported catalyst (C)
supported catalyst (D)
supported catalyst (C)
supported catalyst (D)
0.4
0.4
2.7
2.7
-
-
103.4
103.4
103.4
103.4
0.4
-
2.7
0.88
-
0.89
a Loading of Ru was measured by XRF. b Loading of Ru after the catalytic epoxidation. c Free p-cymene in the solution of IBA detected by GC.
d Binding energies were referred to C 1s of 284.8 eV.
peaks at -62 and -54 ppm in 29Si solid-state NMR for species
(B) in Figure 1 indicated that two or three methoxy moieties of
p-styryltrimethoxysilane made the interfacial bonds as -Si(-OSi)3
(30%) and -Si(-OSi)2(-OCH3) (70%) as illustrated in Figure
1(species B).15
Then the Ru-precursor complex with a styryl moiety (A),
synthesized from the reaction of N-surfonyl-1,2-ethylenediamine
with Ru2(p-cymene)2Cl4,32 was grafted by a reaction with the
styryl moiety of the functionalized SiO2 (B). The reduction of
a FT-IR peak intensity at νCdC at 1625 cm-1 and 13C solid-
state NMR peaks at 112.8 ppm and 134.6 ppm (Figure 2) for
the styryl CdC groups confirm that the Ru-precursor complex
(A) was immobilized on the surface via the consumption of
the styryl moieties of species B as shown in Figure 1 (C). UV/
vis spectra of (A) and (C) were observed at 326 nm (Ru d f
Ph π) and 447 nm (Cl p f Ru d), indicating that the Ru complex
(A) was immobilized with retention of its original local
coordination structure on the Ru2+ site as illustrated in Figure
1 (C). XPS binding energies of Ru 3d5/2 for species A (Ru2+
)
and species (C) were 281.6 and 281.5 eV, respectively, which
are almost the same (Table 1). 13C solid-state NMR in Figure
2 and Ru K-edge EXAFS (Table 2) also evidence the similarity
of the local coordination structures of the supported Ru complex
(C) and the precursor complex (A).
We found that the coordinating p-cymene ligand of the
supported complex (C) selectively released in the presence of
isobutyraldehyde (IBA) and O2.15 Hybrid DFT calculations
revealed that the simple extraction of the p-cymene ligand from
(C) was endothermic (66 kJ mol-1) and we could not spontane-
ously produce the unsaturated Ru complex by the elimination
of the p-cymene ligand by heating under vacuum and N2.
However, when IBA and O2 existed, the exothermic reaction
of IBA and O2 to form isobutyric acid was concertedly devoted
to extract the p-cymene ligand of (C). The concerted reaction
of the energy-gaining IBA conversion and the subsequent
p-cymene ligand elimination produced the unsaturated active
Ru complex (D) with aid of the exothermic reaction of 398 kJ
(32) Haak, K. J.; Hashiguchi, S.; Fujii, A.; Ikariya, T.; Noyori, R. Angew.
Chem. 1997, 109, 297.
15
mol-1
.
The p-cymene ligand stoichiometrically released and
9
716 J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010

Switchover of Selective Oxidation Pathways on Ru Catalyst
A R T I C L E S
Table 2. Curve-fitting analysis for the Ru K-edge EXAFS data measured at 15 K
shell
CN
distance/nm
σ2/nm2
Ru Precursor (A)a
Ru-Cl
Ru-N
Ru-C
Supported Complex (C)b
Ru-Cl
Ru-N
Ru-C
Supported Complex (D)c
Ru-Cl
Ru-N
(D) after trans-stilbene epoxidationd
Ru-Cl
Ru-N
Supported Complex (C)e
Ru-Cl
Ru-N
Ru-C
Supported Complex (D)f
Ru-Cl
Ru-N
1.0 (fixed)
2.0 (fixed)
6.0 (fixed)
0.236 ( 0.003
0.213 ( 0.002
0.227 ( 0.021
(2 ( 3) × 10-5
(-1 ( 2) × 10-5
(2 ( 24) × 10-5
Ru 2.7 wt %
1.0 (fixed)
2.0 (fixed)
6.0 (fixed)
0.236 ( 0.006
0.214 ( 0.002
0.226 ( 0.023
(3 ( 6) × 10-5
(0 ( 2) × 10-5
(13 ( 23) × 10-5
Ru 2.7 wt %
1.3 ( 1.5
2.2 ( 1.5
0.234 ( 0.004
0.207 ( 0.003
(5 ( 8) × 10-5
(1 ( 2) × 10-5
Ru 2.7 wt %
1.5 ( 2.3
2.2 ( 1.8
0.232 ( 0.006
0.206 ( 0.003
(4 ( 12) × 10-5
(8 ( 2) × 10-5
Ru 0.4 wt %
1.0 (fixed)
2.0 (fixed)
6.0 (fixed)
0.238 ( 0.002
0.214 ( 0.002
0.230 ( 0.037
(1 ( 2) × 10-5
(0 ( 2) × 10-5
(20 ( 43) × 10-5
Ru 0.4 wt %
1.3 ( 1.5
2.0 ( 1.7
0.236 ( 0.004
0.209 ( 0.003
(2 ( 9) × 10-5
(0 ( 3) × 10-5
b
a k ) 30-140 nm-1, R ) 0.10-0.22 nm, ∆E0 ) 10 ( 8, Rf ) 1.74%. CNs of Ru-Cl, Ru-N, and Ru-C were fixed. k ) 30-140 nm-1, R )
c
0.10-0.22 nm, ∆E0 ) 10 ( 9, Rf ) 2.31%. CNs of Ru-Cl, Ru-N, and Ru-C were fixed. k ) 30-140 nm-1, R ) 0.13-0.24 nm, ∆E0 ) 10 ( 3, Rf
d
e
) 1.29%. k ) 30-140 nm-1, R ) 0.13-0.24 nm, ∆E0 ) 9 ( 5, Rf ) 1.01%. k ) 30-140 nm-1, R ) 0.10-0.22 nm, ∆E0 ) 9 ( 15, Rf ) 2.42%.
f
CNs of Ru-Cl, Ru-N, and Ru-C were fixed. k ) 30-140 nm-1, R ) 0.13-0.24 nm, ∆E0 ) 11 ( 1, Rf ) 1.14%.
89% (2.7 Ru wt% catalyst) and 88% (0.4 Ru wt% catalyst) of
the released p-cymene were detected in the solution (Table 1).
No release of p-cymene occurred under N2 atmosphere instead
of O2 atmosphere where the exothermic oxidation of IBA could
not proceed.
bonds of Ru-Cl (coordination number (CN) ) 1.5 ( 2.3, bond
distance (d) ) 0.232 ( 0.006 nm) and Ru-N (CN ) 2.2 (
1.8, d ) 0.206 ( 0.003 nm) were observed, and these structural
parameters were almost the same as those for the 3-coordinated
unsaturated structure of (D). Such kind of unsaturated species
easily aggregate in solutions and indeed the elimination of
p-cymene caused aggregation of the formed unsaturated Ru
species in the case of homogeneous Ru-precursor complex (A).
Low stability of unsaturated metal complexes leading to
aggregation or decomposition in solutions prevents from
exploiting metal-complex catalysis. The metal-complex attach-
ment on solid surfaces for isolation of the unsaturated metal
centers is promising as a powerful platform not only for the
dramatic improvement of catalytic activity but also for a better
understanding of key issues for selective catalysis by the metal-
complex structures. The coordinatively unsaturated metal center
can provide an efficient reaction pathway in the coordination
sphere via multiple adsorption of plural reactant molecules on
the metal center.
3.2. Selective Oxidation of Aldehydes to Corresponding
Carboxylic Acids. It was found that the supported Ru complex
(D) showed tremendous activities for selective aldehyde oxida-
tion to corresponding acids. Table 3 shows the catalytic
performances of the IBA oxidation with O2. At 288 K, the IBA
oxidation was hard to proceed in the absence of any Ru catalysts
and the support SiO2 gave no positive effect on the oxidation.
RuO2, RuCl3, Ru2(p-cymene)2Cl4, and Ru precursor complex
(A) were inactive or less selective for the IBA oxidation as
shown in Table 3. Ru2(p-cymene)2Cl4 converted 94% of IBA
to isobutyric acid after 24 h with the acid selectivity of 72%,
however the reuse of this homogeneous complex did not produce
any products at all. On the other hand, the supported Ru catalyst
(D) after the elimination of p-cymene efficiently promoted the
IBA oxidation with good conversion (93%) and selectivity
(94%) in 4 h.
After the release of the coordinating p-cymene ligand, Ru
K-edge EXAFS for the supported Ru complex (D) showed that
there are three chemical bonds (two Ru-N and one Ru-Cl):
Ru-N bonds (coordination number (CN) ) 2.2 ( 1.5) at 0.207
( 0.003 nm and Ru-Cl bond (CN ) 1.3 ( 1.5) at 0.234 (
0.004 nm as shown in Table 2. No Ru-Ru bonding was
observed in agreement of a Ru-monomer complex structure
immobilized at the surface. All peaks of 13C NMR attributed to
the p-cymene ligand disappeared as shown in Figure 2 (dotted
lines) and there were no significant changes in the chemical
shifts of the diamine ligand and its branch before and after the
selective elimination of p-cymene. On the supported catalyst
(D) after the removal of p-cymene, O2 was suggested to adsorb
on the unsaturated Ru site.15 When the CH2Cl2 solution of trans-
stilbene and IBA was added to (D) under N2 atmosphere, the
formation of trans-stilbene oxide (70% of the stoichiometric
quantity to Ru) in the solution was observed, indicating that O2
had been coordinated on the unsaturated Ru site. Furthermore,
when we exposed (D) to other aldehyde molecules in CH2Cl2,
no isobutyric acid derived from coordinated IBA was detected
in the solution. Thus, it is suggested that O2 alone coordinates
to the unsaturated Ru site of (D), but there are neither IBA nor
its corresponding peracid species remained as ligands on the
unsaturated Ru species (D). The modeling by DFT calculations
suggested that the side-on coordination structure (D) illustrated
in Figure 1 was more stable than the coordination structure of
end-on-type O2. Ru 3d5/2 XPS binding energy was observed as
282.0 eV, which is assigned as Ru3+ (Table 1), indicating that
the coordinating O2 on Ru is negatively charged on (D).
After the catalytic oxidations on the supported Ru catalyst
(D), the structure of the unsaturated Ru complex (D) remained
unchanged as shown in Table 2. The two kinds of chemical
Interestingly, although the Ru homogeneous complex (A) and
the supported catalyst (D) showed similar color changes under
9
J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010 717

A R T I C L E S
Tada et al.
Table 3. Catalytic Performances of Various Ru Compounds for the IBA Oxidationa to Isobutyric Acid in CH2Cl2
catalyst
IBA/Ru
time/h
conversion %
selectivity %
b
SiO2
2
6
3
24
3
24
1
24
24
24
4
0
0.5
30
99
18
93
33
94
0
0
60
58
61
65
64
72
72
0
c
RuCl3
50
50
50
50
50
50
50
200
200
50
c
RuO2
c
Ru2(p-cymene)2Cl4
Ru2(p-cymene)2Cl4 (reuse)c
Ru precursor (A)d
0
93
90
0
94
94
supported Ru catalyst (D) (Ru 0.4 wt %)d
supported Ru catalyst (D) (Ru 2.7 wt %)c
2
c
d
a Reagents and conditions: 288 K, CH2Cl2 3 mL, O2 101.3 kPa. b SiO2 35 mg. Ru 6.0 × 10-6 mol. Ru 1.5 × 10-6 mol.
Table 4. Catalytic Performances for Aldehyde Oxidationa to Corresponding Carboxylic Acids at 288 K on the SiO2-Supported Ru-Complex
Catalysts (Ru 0.4 wt %) after the Removal of p-Cymene
reactant
[aldehyde] (mol/L)
time/h
TON
selectivity/%
isobutyraldehyde
butyraldehyde
1-heptanal
1-octanal
2-ethylhexanal
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
neat
neat
4.5
4
4
24
24
4
186
192
186
182
194
184
184
22
200
190
122
166
150
136
30
94
100
100
100
87
98
88
100
41
90
100
85
100
100
68
94
98
85
1-decanal
1-dodecanal
acetaldehyde
24
24
48
5
24
24
24
24
24
24
12
48
48
glutaraldehydeb
cyclohexanecarboxaldehyde
benzaldehyde
3-phenylpropionaldehyde
phenylglyoxal
p-chlorobenzaldehyde
2,6-dichlorobenzaldehyde
isobutyraldehydec
1-heptanalc
38,800,000
31,500,000
1,530,000
3-phenylpropionaldehyded
a Reagents and conditions: Ru 1.5 × 10-6 mol, 288 K, O2 101.3 kPa. b In the oxidation of glutaraldehyde with two aldehyde moieties mono- and
c
d
di-acids were observed, and the summation of the selectivity of the both products was 95%. Ru 4.2 × 10-7 mol, 288-300 K. Ru 4.2 × 10-7 mol.
the epoxidation conditions in the presence of IBA and O2, their
catalytic performances were completely different (Table 3). The
supported Ru complex (C) transformed to the active unsaturated
Ru complex (D) under the catalytic reaction conditions, while
the homogeneous Ru precursor (A) aggregated to form precipi-
tates under the catalytic reaction conditions, leading to a loss
of its catalytic activity. The addition of SiO2 into the homoge-
neous phase of (A) did not improve the deactivation of (A),
which indicates that the chemical grafting on the SiO2 surface
and site-isolation are the key issue to preserve the active species
for the selective oxidation.
We carried out the catalytic oxidations of various aldehyde
compounds on the supported Ru catalyst (D), whose results are
summarized in Table 4. 1-Butanal (n-butyraldehyde), 1-heptanal,
1-octanal, 1-decanal, and 1-dodecanal, which were linear
aldehydes, selectively converted to their corresponding car-
boxylic acids with the selectivity of 88-100%. The TON of a
conjugated benzaldehyde was 122 after 24 h with the acid
selectivity of 100%. Both p-chlorobenzaldehyde and 2,6-
dichlorobenzaldehyde with electron-withdrawing groups were
less active than no-substituent benzaldehyde.
and 1,530,000, respectively. Thus, the synthesized Ru complex
on SiO2 (D) was tremendously active for the selective oxidation
of aldehyde compounds using O2. Heteropolyacids with Ni, Co,
Mn, and Cu,33 supported Au catalysts,34 and Y complexes35
were reported to be so active for the aldehyde oxidation, but
their TONs were less than 1000. Our supported Ru-monomer
complex (D) provides the tremendous catalysis for the selective
oxidation of aldehydes, showing remarkable TONs and TOFs.
Furthermore, the SiO2-supported catalyst (D) could be recycled
by simple filtration though there was a drop of 6% in the
selectivity between the first and second runs, while the conver-
sion did not decrease as shown in Figure 3a. The reason of the
drop is not clear at moment, but no leaching of the Ru complex
was observed on the supported Ru catalyst (D) under the present
reaction conditions and after the five-times recycled reactions,
as characterized by XRF and ICP. Both the conversion and
selectivity were still constant for runs 2-5 (Figure 3a), while
the oxidation in the solution phase of the filtration did not occur.
3.3. Selective Epoxidation of Alkenes Using IBA/O2. Highly
efficient epoxidation of alkenes using a combination system of
molecular oxygen and aldehyde on Ni2+ complex was reported
Huge turnover numbers could be achieved on the supported
Ru catalyst (D) as shown in Table 4. The 38,800,000 equivalent
of IBA (1.11 kg) was oxidized with the selectivity of 94% using
10 mg of the supported Ru catalyst (D) (TON was so
tremendous as 38,800,000). TONs for 1-heptanal and 3-phe-
nylpropionaldehyde oxidations were also so large as 31,500,000
(33) Sloboda-Ronzner, D.; Neimann, K.; Neumann, R. J. Mol. Catal. 2007,
262, 109.
(34) Biella, S.; Prati, L.; Rossi, M. J. Mol. Catal. 2003, 197, 207.
(35) Roesky, P. W.; Canseco-Melchor, G.; Zulys, A. Chem. Commun. 2004,
738.
9
718 J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010

Switchover of Selective Oxidation Pathways on Ru Catalyst
A R T I C L E S
surfaces often creates efficient catalysis at the surfaces44-46 and
the site-isolation of the Ru complex was critical to maintain
the catalytic activity for the selective epoxidation. The reaction
with H2O2 on the surface Ru complex (D) was performed under
a two-phase reaction condition, but we could not find any
activity for the oxidation reaction. A reason may be due to the
immiscible solvent condition. In the case of tert-butyl hydro-
peroxide (TBHP) in a n-heptane solution, conversion was similar
to that of m-CPBA and epoxide selectivity was less than 40%.
It is to be noted that the exclusive catalytic oxidation of IBA
itself was suppressed during the selective alkene epoxidation.
During the catalytic epoxidation of alkenes, coexisting IBA
converted to isobutyric acid stoichiometrically, and the facile
oxidation of sole IBA was not observed; that is, the amount of
the acid product was stoichiometric to that of the epoxide
product as shown in Figure 4. This means that the coexisting
alkenes completely suppressed the facile catalytic oxidation of
IBA itself with O2.
The supported Ru complex (D) exhibited the good catalytic
performances for the alkene epoxidation with 83-90% selectiv-
ity to epoxides under the O2/IBA reaction conditions as shown
in Table 5. Benzaldehyde (10-15%) was observed as a
byproduct, and its diol and ester compounds were not observed.
Other oxidants PhIO and m-chloroperbenzoic acid (m-CPBA)
were poor oxidants with low conversion (30 and 46%, respec-
tively) and low epoxide selectivity (16 and 41%, respectively).
In the case of cis-stilbene epoxidation, cis-stilbene epoxide was
only the product and no trans-stilbene epoxide was observed
on the Ru catalyst (D). Benzaldehyde (24%) was observed as a
byproduct. Cyclopentene, cyclohexene, and cyclooctene epoxi-
dation also proceeded readily, and 1-hexene, 1-octene, and
1-decene, which are terminal alkenes, also converted to their
corresponding epoxides on (D) (Table 5).
We also investigated large-scale catalytic epoxidation on the
supported Ru catalyst (D), whose results are also listed in Table
5. In the case of trans-stilbene, we found a large TON of
2,100,000 with the selectivity of 90% after 168 h at 298 K,
consuming the equivalent amount of IBA. The selective
epoxidation of cyclopentene and norbornene showed TONs of
1,020,000 and 234,000, respectively (Table 5). The supported
Ru catalyst (D) can be reused without any significant loss of
the catalytic activity (100% conversion) seven times as shown
in Figure 3b, indicating the high durability of the unsaturated
active Ru complex supported on SiO2 (D). There was a drop of
6% in the selectivity between the first and second runs, while
the conversion of 100% did not decrease as shown in Figure
3b. The reason for the drop is not clear at moment, but no
leaching of the Ru complex was observed on the supported Ru
catalyst (D) under the present reaction conditions after the seven
times recycled reactions, as characterized by XRF and ICP, and
no change in the structural parameters for the supported Ru
complexes was detected by XAFS analysis. Both the conversion
and selectivity were still constant for runs 2-7 (Figure 3b),
while the oxidation in the solution phase of the filtration did
not occur.
Figure 3. Conversion (O) and selectivity (∆) in the recycling of the
supported Ru catalyst (D). (a) IBA oxidation (Ru 0.4 wt %, Ru 6.0 × 10-6
mol, CH2Cl2 12 mL, 288 K, O2 101.3 kPa, Ru/IBA ) 1/200, 4 h). (b) trans-
Stilbene epoxidation (Ru 2.7 wt %, Ru 1.0 × 10-5 mol, CH2Cl2 5 mL, 298
K, O2 101.3 kPa, Ru/trans-stilbene/IBA ) 1/100/100, 48 h).
by T. Yamada et al.36 and selective oxidation (e.g., epoxidation,
Baeyer-Villiger oxidation) using a combination system of
molecular oxygen and aldehyde was well studied on various
catalysts.37-43 The supported Ru complex (D) with the unsatur-
ated Ru site acted as an active site for the selective catalytic
epoxidation of trans-stilbene using IBA and O2 as reported
previously as a communication.15
It was found that the surface Ru complex (D) exhibited unique
catalytic performances for the selective epoxidation of not only
trans-stilbene but also other alkenes with IBA/O2. On the other
hand, homogeneous Ru(NH3)6Cl3, Ru(bpy)Cl2 with both nitro-
gen and chlorine ligands, Ru2(p-cymene)2Cl4, and RuCl3 never
catalyzed the epoxidation of trans-stilbene with any oxidants
such as O2, H2O2, PhIO, or O2/IBA. The Ru-precursor complex
(A) produced a small amount of trans-stilbene oxide with PhIO
or O2/IBA, but the epoxidation stopped in a little while, forming
black precipitates. The addition of SiO2 into the homogeneous
solution of (A) did not give any positive effect for the catalytic
performance of the epoxidation (Table 5). Site-isolation on
(36) Yamada, T.; Takai, T.; Rhode, O.; Mukaiyama, T. Chem. Lett. 1991,
1.
(37) Habor, J.; Mlodnicka, T.; Poltowicz, J. J. Mol. Catal. 1989, 54, 451.
(38) Rodgers, K. R.; Arafa, I. M.; Goff, H. M. J. Chem. Soc., Chem.
Commun. 1990, 1323.
(39) Takai, T.; Hata, E.; Yamada, T.; Mukaiyama, T. Bull. Chem. Soc.
Jpn. 1991, 64, 2513.
(44) Joubery, J.; Delbecq, F.; Sautet, P.; Le Roux, E.; Taofik, M.; Thieuleux,
C.; Blanc, F.; Coperet, C.; Thivolle-Cazat, J.; Basset, J. M. J. Am.
Chem. Soc. 2006, 128, 9157.
(40) Murahashi, S.-I.; Oda, Y.; Naota, T. J. Am. Chem. Soc. 1992, 114,
7913.
(41) Mukaiyma, T.; Yamada, T.; Nagata, T.; Imagawa, K. Chem. Lett. 1993,
(45) Barbaro, P.; Bianchini, C.; Dal Santo, V.; Meli, A.; Moneti, S.; Psaro,
R.; Scaffidi, A.; Sordelli, L.; Vizza, F. J. Am. Chem. Soc. 2006, 128,
7065.
327.
(42) Imagawa, K.; Nagata, T.; Yamada, T.; Mukaiyama, T. Chem. Lett.
1994, 527.
(46) Kruithof, C. A.; Casado, M. A.; Guilena, G.; Egmond, M. R.; van der
Kerk-van Hoof, A.; Heck, J. R. A.; Gebbink, J. M. K.; van Koten, G.
Chem.sEur. J. 2005, 11, 6869.
(43) Kaneda, K.; Ueno, S.; Imanaka, T. J. Mol. Catal. A: Chem. 1995,
102, 135.
9
J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010 719

A R T I C L E S
Tada et al.
Table 5. Catalytic Performances for Alkene Epoxidationa on the Ru-Complex Catalysts (A) and (D)
catalyst
reactant
[alkene]/M
Ru/alkene/IBA
oxidantb
time/h
TON
epoxide selectivity/%
(A)
(A) + SiO2
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
(D)
trans-stilbene
trans-stilbene
trans-stilbene
trans-stilbene
trans-stilbene
trans-stilbene
trans-stilbeneb
trans-stilbeneb
trans-stilbeneb, f
trans-stilbeneg
cis-stilbene
cyclopentene
cyclopenteneg
cyclohexene
cyclooctene
styrene
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.8
0.8
0.1
0.1
5.0
0.1
0.1
0.1
0.1
0.1
0.1
0.1
4.0
1/50/50
1/50/50
1/50/0
1/50/0
1/50/0
IBA/O2
IBA/O2
O2
72
72
72
72
72
48
4
5
5
0
0
15
71
66
0
c
d e
,
H2O2
0
PhIOd
16
41
83
88
90
90
73
82
80
65
81
38
67
69
82
74
87
1/50/0
1/50/50
m-CPBAd
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
IBA/O2
23
50
1/100/100
1/1000/1000
1/2 × 107/2 × 107
1/50/50
12
4
100
1000
2100000
45
43
1020000
27
49
25
29
27
168
12
48
240
48
48
48
96
96
96
48
384
1/50/50
1/2 × 106/2 × 106
1/50/50
1/50/50
1/50/50
1/50/50
1/50/50
1-hexene
1-octene
1-decene
1/50/50
29
50
234000
norbornene
1/50/50
norborneneg
1/2 × 106/2 × 106
a Reagents and conditions: Ru 1.0 × 10-5 mol, CH2Cl2 5 mL, 298 K, O2 101.3 kPa. Catalyst (D) Ru 2.7 wt %. b In the trans-stilbene epoxidation by
IBA/O2 benzaldehyde was a major byproduct (about 10-15%). Other minor byproducts were benzylalcohol and benzoic acid. Diol and ester were not
detected. In the use of PhIO and m-CPBA as oxidants, lots of benzaldehyde and benzoic acid were produced as byproducts. In the case of cyclohexene
epoxidation, cyclohexenone (15%) and cyclohexenol (20%) were observed as byproducts. c SiO2 40 mg. d trans-Stilbene/oxidant ) 1/1. e Two-phase
g
reaction. f CH2Cl2 12 mL. Catalyst (D) Ru 0.4 wt %, Ru 4.2 × 10-7 mol, 288 K, O2 101.3 kPa.
was found to be 21/79 (3 h, conv. 16%) and 22/78 (24 h, conv.
67%) on the supported Ru catalyst (D). Judging from the
ꢀ-selective epoxidation of cholesteryl benzoate, we can say with
safety that the trans-stilbene epoxidation is not autooxidation
using free radical species in the solution and the supported Ru
species positively participates in the catalytic epoxidation on
the surface.
3.4. Reaction Mechanisms for the Alternative Selective
Oxidation Pathways. We have found that the supported Ru
complex (D) is highly active for both the IBA oxidation with
O2 and the trans-stilbene epoxidation with IBA/O2 as mentioned
above. The reaction rate of the IBA oxidation to isobutyric acid
is much faster than that of trans-stilbene epoxidation to trans-
stilbene oxide under the identical IBA/O2 concentration for the
both oxidations. Indeed, when the concentration of IBA was
0.1 M in CH2Cl2, the TOF (turnover frequency: equivalent to
reaction rate) of the catalytic IBA oxidation with O2 in the
absence of trans-stilbene (IBA + O2 f 2 isobutyric acid) was
138 h-1, while the TOF of the catalytic trans-stilbene epoxi-
dation with O2 using IBA (0.1 M) (trans-stilbene + IBA + O2
f trans-stilbene oxide + isobutyric acid) was 12 h-1, where
no catalytic oxidation of IBA alone with O2 proceeded. The
origin and mechanism of the both selective oxidation pathways
and the switchover of the reaction pathways will be discussed
below.
Figure 4. TONs of trans-stilbene oxide and isobutyric acid for the
epoxidation of trans-stilbene on the supported Ru catalyst (D) (Ru 2.7 wt
%) at 303 K. O: trans-Stilbene and ∆: IBA. Ru 2.7 × 10-6 mol, CH2Cl2 3
mL, O2 101.3 kPa, Ru/trans-stilbene/IBA)1/100/100.
The reaction mechanism of epoxidation using the combination
system of molecular oxygen and aldehyde was investigated, and
several reaction mechanisms and active oxygen species were
proposed.47-51 Autooxidation in the absence of metal species
proceeds, and the stereoselective epoxidation of 5,6-double bond
in cholesteryl benzoate (eq 1) clearly shows whether metal
species is concerned in epoxidation.31
It is to be noted that the IBA/O2 molecules were consumed
for the epoxidation of trans-stilbene (trans-stilbene + IBA +
O2 f trans-stilbene oxide + isobutyric acid) and the much faster
oxidation of IBA itself (2 IBA + O2 f 2 isobutyric acid) in
(47) Nishida, Y.; Fujimoto, T.; Tanaka, N. Chem. Lett. 1992, 1291.
(48) Simmons, K. E.; van Sickle, D. E. J. Am. Chem. Soc. 1973, 95, 7759.
(49) Diaz, R. R.; Selby, K.; Waddington, D. J. J. Chem. Soc., Perkin Trans.
2 1975, 758.
We investigated the epoxidation (eq 1) using m-chloroper-
benzoic acid (m-CPBA) and the supported Ru catalyst (D),
whose steroselectivities were opposite. In the case of m-CPBA,
autooxidation pathway takes place in epoxidation, R-epoxide
was preferably produced and the product ratio of R-epoxide/
ꢀ-epoxide was 70/30, which agreed with the reported result.31
On the other hand, the product ratio of R-epoxide/ꢀ-epoxide
(50) Simandi, L. I., Ed. Catalytic ActiVation of Dioxygen by Metal
Complexes; Kluwer Academic Publishers: Dordrecht, The Netherlands,
1992; pp 318-331.
(51) Sheldon, R. A., Kochi, J. K., Eds. Metal-Catalyzed Oxidations of
Organic Compounds; Academic Press: New York, 1981; pp 359-
363.
9
720 J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010

Switchover of Selective Oxidation Pathways on Ru Catalyst
A R T I C L E S
Figure 5. (a) Reaction order to IBA concentration in the IBA oxidation on (D) (Ru 0.4 wt %, Ru 4.2 × 10-7 mol, 288 K). (b) Reaction order to IBA
concentration in the trans-stilbene epoxidation on (D) (Ru 0.4 wt %, Ru 4.2 × 10-7 mol, 314 K, [trans-stilbene] ) 0.1 M). (c) Reaction order to trans-
stilbene concentration in the trans-stilbene epoxidation on (D) (Ru 0.4 wt %, Ru 4.2 × 10-7 mol, 314 K, [IBA] ) 0.1 M). (d) Arrhenius plots for the IBA
oxidation (O, [IBA] ) 0.1 M) and the trans-stilbene epoxidation (b, [IBA] ) 0.1 M and [trans-stilbene] ) 0.1 M) on (D) (Ru 2.7 wt %, Ru 2.7 × 10-6
mol). (e) Amount of adsorbed trans-stilbene on (D) against the concentration of trans-stilbene at 298 K (Ru 2.7 wt %, Ru 2 × 10-5 mol, [IBA] ) 0.1 M,
N2 atmosphere).
the presence of trans-stilbene was completely suppressed. Indeed
the produced trans-stilbene epoxide and isobutyric acid were
equivalent to each other from the initial stage of the epoxidation
through the whole reaction period as shown in Figure 4. The
addition of trans-stilbene switched over the reaction pathway
of the much faster aldehyde oxidation to the much slower alkene
epoxidation. Even if the excess amount of IBA compared to
the amount of trans-stilbene existed in the reaction solution,
the IBA sole oxidation without the alkene epoxidation did not
proceed in the presence of alkene.
The kinetic parameters of the both selective oxidations on
the supported Ru catalyst (D) were investigated to understand
the alternative selective oxidation mechanism. As shown in
Figure 5a, the reaction order to the IBA concentration was 1.04
for the IBA sole oxidation without any alkenes, and the
activation energy was estimated to be 48 kJ mol-1 (Figure 5d).
There was no peroxide species observed by the iodometric
titration of the steady-state reaction solution after the filtration
of the solid catalyst. The species (26% equivalent to Ru) capable
of iodo-oxidation (e.g., peroxyacid, peroxide, metal-oxo, etc.)52
was detected on the supported Ru catalyst separated from the
reaction solution. These results indicate that IBA was oxidized
to form active intermediate species with alkene oxidation activity
on the supported Ru complex and IBA was not autooxidized in
the reaction solution. The amount of the active intermediate
species on the supported Ru complex (D) decreased to 12% of
Ru quantity after the completion of the IBA oxidation reaction,
indicating that the intermediates are labile.
In the case of the trans-stilbene epoxidation, the similar results
on iodometric titration were observed and the opposite selectivity
in the stereoselective epoxidation of cholesteryl benzoate (R-
epoxide and ꢀ-epoxide) was observed as mentioned in section
3.3, which exclude the possibility of the autooxidation mech-
anism on the supported Ru catalyst.31 The reaction orders with
respect to IBA and trans-stilbene were estimated to be 1.25
and 0.15, respectively, as shown in b and c of Figure 5. The
activation energy of the epoxidation was 99 kJ mol-1 (Figure
5d), which was about twice that of the IBA oxidation (48 kJ
mol-1). The activation energies of the epoxidation of limonene
with a terminal CdC bond and an internal CdC bond were 81
kJ mol-1 and 61 kJ mol-1, respectively. The different activation
energies for the terminal and internal CdC bond in limonene
and the small reaction order to trans-stilbene demonstrate that
trans-stilbene as a strong adsorbed species on the Ru center is
(52) Cook, G. K.; Mayer, J. M. J. Am. Chem. Soc. 1994, 116, 1855.
9
J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010 721

A R T I C L E S
Tada et al.
Figure 6. Reaction pathway of the selective IBA oxidation and the intermediate structures calculated by hybrid DFT. Acid: isobutyric acid. Blue: Ru, red:
O, green: Cl, yellow: S, gray: C, and white: H.
involved in the rate-determining step of the epoxidation, where
most of Ru sites are coordinated by trans-stilbene at the steady-
state reaction.
Indeed, trans-stilbene preferentially adsorbed on the unsatur-
ated active Ru complex (D) in the presence of the equivalent
amount of IBA. Figure 5e shows the amounts of adsorbed trans-
stilbene per Ru against the concentration of trans-stilbene in
the presence of IBA under N2 atmosphere. It was found that
trans-stilbene preferably adsorbed on the unsaturated Ru centers
(D) and the adsorption saturated at the nearly equivalent amount
to Ru as shown in Figure 5e. The preferable coordination of
alkenes onto the Ru sites may induce the switchover of the
selective oxidation pathways from the facile aldehyde oxidation
to the hard epoxidation.
9
722 J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010

Switchover of Selective Oxidation Pathways on Ru Catalyst
A R T I C L E S
Figure 7. Reaction pathway of the trans-stilbene epoxidation and the intermediate structures calculated by hybrid DFT. Stb: trans-stilbene, Stbox: trans-
stilbene oxide, and Acid: isobutyric acid. Blue: Ru, red: O, green: Cl, yellow: S, gray: C, and white: H.
We investigated the energy profiles of the reaction pathways
for the IBA sole oxidation with O2 in the absence of trans-
stilbene and the IBA-assisted trans-stilbene epoxidation with
O2 on the surface Ru complex (D) by DFT calculations. The
calculated energy profiles of the IBA sole oxidation with O2
and the trans-stilbene epoxidation with IBA/O2 are shown in
Supporting Information 1, where the relative energy profiles in
the same spin states would be valid to discuss on the reaction
pathways. The structures of the reaction intermediates and the
transition states are also illustrated in Supporting Information
2, 3 and 4, respectively. On the basis of the energy profiles the
reaction pathways for the catalytic IBA oxidation and the
catalytic trans-stilbene oxidation are shown in Figure 6 and
Figure 7, respectively.
In the IBA sole oxidation, both IBA and O2 coordinate to
the unsaturated Ru complex (1) to form species (2) with an
energy barrier of 57 kJ mol-1 as shown in Figure 6 and
Supporting Information 1. The first-order kinetics to the IBA
concentration with the activation energy of 48 kJ mol-1 is
attributed to the rate-limiting O2 addition to the aldehyde carbon
to form species (3) with elongated O2, which is followed by
the transformation of (3) to an intermediate (5) with RudO via
species (4) with RudO. The species (5) further reacts with the
second IBA to produce the corresponding acids via species (6),
9
J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010 723

A R T I C L E S
Tada et al.
(7), and (8). The RudO groups were successfully detected by
iodometry in the steady-state oxidation (26% of the stoichio-
metric amount of Ru).
species active for both selective aldehyde oxidation (TON of
IBA oxidation: 38,800,000) and alkene epoxidation (TON of
trans-stilbene epoxidation: 2,100,000). This aspect gives insight
into the origin of the efficient catalysis for selective epoxidation
of alkenes with IBA/O2.
On the other hand, we observed the nearly stoichiometric
adsorption of trans-stilbene on the Ru site (D) in Figure 5e,
which implies that the majority of Ru sites exists as the species
coordinated with trans-stilbene in the selective oxidation of
trans-stilbene. The adsorption energy of trans-stilbene on Ru
sites was estimated to be 93 kJ mol-1 by DFT calculations, and
the obtained Ru(trans-stilbene) species is further coordinated
with O2 and IBA to form species (10) in Figure 7. Alternatively,
trans-stilbene coordinates to species (2) to form the species (10).
The species (10) transforms to species (11) and to an intermedi-
ate (12) with RudO via double cascades. The epoxidation occurs
by the rate-limiting addition of the RudO oxygen to the CdC
bond of trans-stilbene with an energy barrier of 98 kJ mol-1 in
the step from (12) to (13). This agrees with the observed
activation energy of 99 kJ mol-1 for the catalytic trans-stilbene
epoxidation with IBA/O2. The energy profile of the reaction
pathways may explain the nearly stoichiometric adsorption of
trans-stilbene on the Ru site as species (12), and it is also
compatible with the kinetics proportional to the IBA concentra-
tion and nearly independent of the trans-stilbene concentration.
It should be noted that the present DFT calculations for the
transition states such as TS(10-11) and TS(11-12) may involve
certain values of error bars.
Thus, we can document how the alternative selective oxida-
tion pathways are switched over from the facile rapid aldehyde
oxidation with the lower activation energy to the harder alkene
epoxidation with the higher activation energy by the existence
of alkenes. The important issue for the selective catalytic
epoxidation of trans-stilbene with IBA/O2 without the catalytic
IBA oxidation with O2 is the coordination of trans-stilbene on
the Ru center, which makes it possible to efficiently rearrange
the bonds of the three coordinated reactants (trans-stilbene +
IBA + O2) on the Ru center to form the species (12) via the
species (10) and (11) in the coordination sphere. As a result,
the IBA sole oxidation is suppressed due to a negligible
concentration of the species (2). The coordination of the second
IBA to the species (2) to form species (9) decreases the IBA
oxidation rate by self-poisoning due to appearance of the higher
energy barrier for the IBA oxidation (Supporting Information
1). The unsaturated Ru-complex structure (D) immobilized at
the SiO2 surface, which is obtained by p-cymene elimination
in couple with the exothermic IBA + O2 reaction, is crucial to
provide the efficient reaction pathways involving the Ru4+)O
4. Conclusions
The SiO2-supported unsaturated Ru-monomer complex, which
was prepared the energy-gaining way by using IBA and O2,
exhibited tremendous TONs and TOFs for the selective oxida-
tion catalyzes of the aldehyde oxidation and the alkene
epoxidation at ambient temperature. The TONs of 38,800,000
for the IBA oxidation (TOF: 1600 s-1) and 2,100,000 for the
trans-stilbene epoxidation (TOF: 6 s-1) were achieved, and the
active Ru-complex catalyst (D) could be reusable without any
significant loss of its catalytic activities. We found that the
addition of alkenes switched over the selective oxidation
pathways from the fast IBA sole oxidation to the harder slow
alkene epoxidation, suppressing the facile IBA sole oxidation,
due not only to the occupation of a vacant site on the Ru center
by trans-stilbene but also to the facile conformation and bond
rearrangements of IBA, O2, and alkene in the coordination
sphere. The unsaturated Ru-complex structure isolatedly at the
SiO2 surface was crucial to provide the selective oxidation
catalysis via the adsorption of plural reactants in the same
coordination sphere.
Acknowledgment. The work was supported by a Grant-in-Aid
for Scientific Research on Priority Areas (No. 18065003) and a
Grant-in-Aid for Scientific Research (S) (No. 18106013) from
MEXT, an Inoue Science Research Award from Inoue Foundation
for Science, and a JCII Subsidy System for Advanced Innovation
in Chemical Industry. XAFS measurements were performed by the
approval of PF-PAC (Nos. 2005G209 and 2008G154).
Supporting Information Available: Energy diagram for the
selective oxidation reactions calculated by hybrid DFT, the
intermediate structures in the IBA oxidation calculated by hybrid
DFT, the intermediate structures in the trans-stilbene oxidation
calculated by hybrid DFT, the structures of the transition states
calculated by hybrid DFT, and complete reference 24. This
information is available free of charge via the Internet at http://
pubs.acs.org.
JA9079513
9
724 J. AM. CHEM. SOC. VOL. 132, NO. 2, 2010