Surface functionalization of supported Mn clusters to produce robust Mn catalysts for selective epoxidation

Article abstract of DOI:10.1021/cs400053f

A robust heterogeneous Mn catalyst for selective epoxidation was prepared by the attachment of a Mn₄ oxonuclear complex [Mn₄O ₂(CH₃COO)₇(bipy)₂](ClO ₄)·3H₂O (1) on SiO₂ and the successive stacking of SiO₂-matrix overlayers around a supported Mn cluster. The structures of supported Mn catalysts were characterized by means of FT-IR spectroscopy, diffuse-reflectance UV/vis spectroscopy, Raman spectroscopy, X-ray photoelectron spectroscopy, and Mn K-edge X-ray absorption fine structure. A ligand exchange reaction between the CH₃COO ligand of 1 and surface silanol group produced a SiO₂-supported Mn cluster (2), whose coordination structure was similar to 1. Subsequent heating of 2 under vacuum yielded supported Mn clusters (3, 4) through the partial elimination of CH ₃COO ligands. The surface-attached Mn clusters of 2, 3, and 4 were easily released to a reaction solution under epoxidation conditions (Mn leaching: approximately 50%), although they were active for epoxidation of trans-stilbene (the conversion of trans-stilbene, 99%, and the selectivity of trans-stilbene epoxid, 96%, for 6 h on 3). We found that the functionalization of the supported Mn cluster on 2 with surface SiO₂-matrix overlayers altered the reactivity of the supported Mn cluster. Dimeric Mn species (5c) with reduced Mn oxidation state and coordination numbers was formed together with a reaction nanospace surrounded by the SiO₂-matrix overlayers. By optimizing the stacking manner of the SiO₂-matrix overlayers, the durability of the Mn catalyst was remarkably improved from leaching (the Mn leaching reached the minimum value of 0.01%), and active and stable epoxidation performances were successfully achieved in the heterogeneous phase (the conversion of trans-stilbene, 97%, and the selectivity of trans-stilbene epoxide, 91%, for 31 h on 5c).

Full text of DOI:10.1021/cs400053f

Research Article
pubs.acs.org/acscatalysis
Surface Functionalization of Supported Mn Clusters to Produce
Robust Mn Catalysts for Selective Epoxidation
Satoshi Muratsugu,*,†,‡ Zhihuan Weng,†,∥ and Mizuki Tada*,†,‡,§
†Institute for Molecular Science, 38 Nishigo-naka, Myodaiji, Okazaki, Aichi 444-8585, Japan
§
‡Department of Chemistry, Graduate School of Science, and Research Center for Materials Science, Nagoya University, Furo-cho,
Chikusa, Nagoya 464-8602, Japan
∥Department of Applied Physics and Chemistry, The University of Electro-Communication, Chofu, Tokyo 182-8585, Japan
S
* Supporting Information
ABSTRACT: A robust heterogeneous Mn catalyst for selective epoxidation was prepared by the attachment of a Mn4
oxonuclear complex [Mn4O2(CH3COO)7(bipy)2](ClO4)·3H2O (1) on SiO2 and the successive stacking of SiO2-matrix
overlayers around a supported Mn cluster. The structures of supported Mn catalysts were characterized by means of FT-IR
spectroscopy, diffuse-reflectance UV/vis spectroscopy, Raman spectroscopy, X-ray photoelectron spectroscopy, and Mn K-edge
X-ray absorption fine structure. A ligand exchange reaction between the CH3COO ligand of 1 and surface silanol group produced
a SiO2-supported Mn cluster (2), whose coordination structure was similar to 1. Subsequent heating of 2 under vacuum yielded
supported Mn clusters (3, 4) through the partial elimination of CH3COO ligands. The surface-attached Mn clusters of 2, 3, and
4 were easily released to a reaction solution under epoxidation conditions (Mn leaching: approximately 50%), although they were
active for epoxidation of trans-stilbene (the conversion of trans-stilbene, 99%, and the selectivity of trans-stilbene epoxid, 96%, for
6 h on 3). We found that the functionalization of the supported Mn cluster on 2 with surface SiO2-matrix overlayers altered the
reactivity of the supported Mn cluster. Dimeric Mn species (5c) with reduced Mn oxidation state and coordination numbers was
formed together with a reaction nanospace surrounded by the SiO2-matrix overlayers. By optimizing the stacking manner of the
SiO2-matrix overlayers, the durability of the Mn catalyst was remarkably improved from leaching (the Mn leaching reached the
minimum value of 0.01%), and active and stable epoxidation performances were successfully achieved in the heterogeneous phase
(the conversion of trans-stilbene, 97%, and the selectivity of trans-stilbene epoxide, 91%, for 31 h on 5c).
KEYWORDS: Mn cluster, SiO2 particles, surface functionalization, SiO2-matrix overlayers, selective epoxidation
active metal species.2,3 The site isolation effect on solid surfaces
prevents metal complexes and clusters from undergoing
1. INTRODUCTION
Creating a new catalytically active and stable species from
aggregation that will result in deactivation, and makes them
transition metal complexes and clusters remains a fundamental
easier to separate from reaction media for recycling.2,3
challenge in catalysis. The catalytic properties of homogeneous
However, the chemical grafting of a metal complex on a
transition metal complexes can be fine-tuned by using various
combinations of metal ions and ligands at a molecular level.1
support does not always give a stable catalytically active metal
species, and leaching of the metal species from the surface often
However, coordination structures are likely to decompose
proceeds under catalytic reaction conditions. This phenomenon
during the homogeneous catalytic reaction conditions and tend
to lose catalytic activity, for instance, through aggregation of
metal complexes forming inactive species.
The heterogenization of homogeneous metal complexes and
clusters by grafting or anchoring to a solid surface, or by
immobilizing in polymers or porous materials, has been widely
adopted to produce more stable metal complexes during
catalytic reaction as well as to produce unique catalytically
occurs when chemical bonding between a metal complex and
an oxide surface is relatively weak and does not last sufficiently
long under the catalytic conditions.
Received: January 25, 2013
Revised: July 18, 2013
© XXXX American Chemical Society
2020
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
One notable example is a catalyst system using 3d-metal
complexes (Mn and Fe) under oxidation reaction conditions.
Mn complexes and clusters,4 such as Mn porphyrines,5 Mn
salens,6 and Mn 1,4,7-triaza-1,4,7-trimethylcyclononane
(tmtacn) complexes,7 were reported to be promising oxidation
catalyst precursors because their derivatives exhibit high activity
and selectivity for oxidation of alkenes and alcohols. In some
cases, they could be used with mild oxidants such as O2 and
H2O2. Mn oxonuclear clusters with acetate and phosphate
ligands have attracted considerable attention because they were
potential models of metal enzyme active sites,4b,8 such as
catalase9 and the water oxidation center in Photosystem II.10
The activity for oxidation of alcohols,11 epoxidation and cis-
dihydroxylation of alkenes,7h−j,11b,12 sulfoxidation,11b,13 and
oxidation of alkanes11b,14 were reported. In spite of these
intriguing properties, Mn complexes and clusters tend to
decompose and lose their catalytic activities in homogeneous
conditions.
The heterogenization of Mn complex catalysts has also been
investigated with the aim to increase catalytic activity and
catalyst stability. Grafting of Mn porphyrine and salen
complexes inside mesoporous materials was among the first
methods used to prepare heterogeneous Mn complex catalysts.
Various supports were used for this purpose, including
zeolites,15 mesoporous SiO2,16 clays,17 zinc−aluminum double
hydroxides,18 and carbon.19 However, Mn species sometimes
leached into the solution phase. Covalent anchoring of Mn
salen complexes on oxide surfaces or inside mesoporous
materials via organic linkers was reported to be effective for
dramatically enhancing their stability and recyclability in
oxidation reactions without changing the structure of Mn
complexes.20 Coordination of Mn ions to surface-functionalized
ligands yielded site-isolated supported Mn complex catalysts
that were highly stable for epoxidation reactions.21 Grafting of a
Mn(tmtacn) dinuclear cluster on a carboxylate-functionalized
oxide surface enabled in situ production of a catalytically active
Mn dinuclear cluster with a degree of stability.22 Other efficient
reported approaches for heterogenization were the incorpo-
ration of Mn complexes inside polymers, ionic liquids, and
metal−organic frameworks.23−25 The grafted- or supported Mn
complexes introduced here had basically similar structures to
precursor complexes; therefore, the generation of totally new
catalytically active Mn species from the original Mn complexes
was quite rare.
2. EXPERIMENTAL SECTION
2-1. Preparation of [Mn4O2(CH3COO)7(bipy)2](ClO4)·
3H2O (1). All reagents were purchased from Sigma-Aldrich,
STREM, Wako Chemicals, and Tokyo Kasei Kogyo, and used
as received unless otherwise noted. [Mn3O(CH3COO)6(py)3]-
(ClO4) was prepared by the literature method28 and
[Mn4O2(CH3COO)7(bipy)2](ClO4)·3H2O (1) was synthe-
sized by literature reported procedure with modifications.26a
Under N2 atmosphere, 2,2′-bipyridine (bipy) (1.15 g, 7.36
mmol) was added to a stirred MeCN solution (50 mL) of
[Mn3O(CH3COO)6(py)3](ClO4) (2.06 g, 2.36 mmol). The
solution was stirred at room temperature for 1 h, and the
resulting deep red solution was mixed with a 1:1 mixture of
hexane and Et2O (50 mL) and additional Et2O (20 mL). The
powder was collected by filtration, washed with Et2O, and
dried; yield 70.9% (from Mn basis). Compound 1 was obtained
as a dark red powder in trihydrate form. Anal. Calcd for
C34H43N4O23ClMn4: C, 36.11; H, 3.83; N, 4.95;. Found: C,
1
36.29; H, 3.65; N, 5.14. H NMR (CD2Cl2): −130.5, −77.0,
−29.7, −19.3, −18.8, −11.5, 20.2, 22.8, 28.0, 33.2, 36.4, 45.2.
ESI-TOF MS (CH2Cl2): m/z 976.9320 [M]+ (Calcd for
C34H37O16N4Mn4, 976.9727) (Supporting Information, Figure
S1).
2-2. Preparation of SiO2-Supported Mn4 Cluster (2).
SiO2 (1 g, Aerosil 200, Degussa, surface area: 200 m2 g−1) was
calcined at 673 K for 2 h and kept under vacuum for 0.5 h prior
to the grafting of 1. The precalcined SiO2 (1 g) was suspended
in CH2Cl2 (20 mL) under N2, and a CH2Cl2 solution (10 mL)
of 1 (0.114 g, 1.01 × 10−4 mol) was added. After stirring for 0.5
h, the solvent was removed under reduced pressure, and the
resulting solid was allowed to dry under vacuum for another 1−
2 h. Loading of Mn was evaluated by X-ray fluorescence (XRF)
to be 2.2 wt %, which corresponded to the surface density of 1
to be 0.3 nm−2 (approximately one-third of the maximum
surface density). Catalyst 2 with low Mn loading (surface
density of 1: 0.2 and 0.1 nm−2) was prepared in a similar way.
Catalysts heated for 3 h at 373 and 413 K under vacuum were
also prepared (3 and 4).
2-3. Stacking of SiO2-Matrix Overlayers. A SiO2 matrix
protecting Mn catalyst (5a) was prepared as follows.
Tetramethoxysilane (TMOS, Si(OCH3)4) (0.67 g (0.65 mL))
and distilled H2O (0.27 mL) were placed in glass reactors
connected to a two-necked flask containing a supported Mn
cluster (2, 0.2 g). After the whole system was evacuated to
prepare a closed system with reduced pressure, the glass
reactors were heated to 393 K and chemical vapor deposition
(CVD) was used to vaporize TMOS and H2O and deposit
them on a supported Mn cluster catalyst. During the CVD
process, the catalyst was vigorously stirred at room temper-
ature. Then, the two-necked flask was separated from the glass
reactor while maintaining the closed system, and hydrolysis−
polymerization of TMOS was performed at 358 K for 24 h. The
whole closed system was subsequently evacuated at 373 K for 4
h. A Mn catalyst protected by a SiO2 matrix (5b) was prepared
in a similar way with double the amounts of TMOS (1.34 g)
and H2O (0.54 mL). A Mn catalyst protected by a SiO2 matrix
(5c) was prepared by twice performing the procedure of
stacking SiO2-matrix overlayers of TMOS (0.67 g) and H2O
(0.27 mL). Weight gain after the hydrolysis-polymerization was
measured after evacuation at 373 K, and the ratio of
polycondensed TMOS was estimated.
Here we report a new catalytically active Mn species through
functionalization of a Mn cluster-attached SiO2 surface toward
robust epoxidation catalysis in a heterogeneous phase. A Mn
oxotetranuclear cluster, [Mn4O2(CH3COO)7(bipy)2](ClO4)·
3H2O (1),26a which has been prepared as a series of model
complexes for Photosystem II,26,27 was chemically attached on
the surface of SiO2 particles by reaction with surface Si−OH
groups. However, the chemical bond attaching the complex at
the interface was found to be easily cleaved under epoxidation
reaction conditions. We functionalized the Mn cluster-attached
SiO2 surface by stacking SiO2-matrix overlayers through the
method we have developed, and the space around the
supported Mn cluster was filled by the matrix overlayers.
This surface functionalization remarkably changed the epox-
idation performances of the Mn catalyst on the SiO2 surface,
and the robust epoxidation performances were achieved on the
Mn catalyst in a heterogeneous phase.
2021
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
2-4. Preparation of SiO2-Supported MnOx Catalysts. 2
was heated for 3 h at 673 K under vacuum to prepare catalyst 6.
Catalyst 7 was prepared by the stacking of SiO2-matrix
overlayers on 6 similar to the preparation of 5c.
protected by SiO2-matrix overlayers were measured in
fluorescence mode, and an ionization chamber filled with N2
and a Lytle detector filled with Ar were used to monitor the
incident and fluorescence X-rays, respectively. All samples were
measured at 20 K under vacuum. Because of low Mn loading on
5c, Aerosil 300 was also used as a support for the XAFS
measurements of 5c.
2-5. Catalyst Characterization. XRF was measured on a
JEOL JSX-3400RII spectrometer. The loading of Mn on the
solid samples was determined by using standard curves of Mn
Kα and Si Kα. The amount of CH3COO ligands released from
Mn clusters during the attachment to SiO2 was determined as
follows. After the impregnation of SiO2 with 1 in a Schlenk tube
under N2 atmosphere, the resulting supernatant was decanted
from the suspension, and the released CH3COOH was
dissolved and quantified by gas chromatography using a flame
ionization detector (FID-GC; Shimadzu GC-14B, CHIRAL-
DEX B-DM column). The number of CH3COO ligands in 2, 3,
4, and 5c were determined as follows. The catalysts were
treated with excess amount of CF3COOH in CH2Cl2 solution,
and the removed CH3COOH was quantified by FID-GC
(Shimadzu GC-14B, CHIRALDEX B-DM column).
FT-IR spectra were recorded in transmission mode with 4
cm−1 resolution on a JEOL JIR-7000 spectrometer. A KBr disk
of Mn4 cluster (1) was evacuated under vacuum for 1 h in an in
situ IR cell equipped with NaCl windows before FT-IR
measurements. FT-IR spectra of 2, 3, and 4 were measured
under in situ conditions using a neat disk of 2 prepared under
N2 atmosphere. The sample was evacuated for 1 h, and the
spectrum of 2 was recorded. Then the sample was heated under
vacuum for 1 h at 373 K and that of 3 was recorded. Finally, it
was heated under vacuum for 1 h at 413 K and that of 4 was
recorded.
Transmittance and diffuse-reflectance (DR) UV/vis spectra
were recorded on a JASCO V-550-DS spectrometer equipped
with an integrating sphere. 1 was dissolved in CH2Cl2. Solid 2,
3, 4, and 5 were placed in quartz cells under Ar atmosphere.
29Si solid-state magic angle spinning (MAS) NMR spectra
were recorded with a JEOL JNM-ECX400 spectrometer
operating at 79.5 MHz. A single-pulse detection method was
used in 29Si NMR measurements, and a sample was set in a
zirconia rotor of 4 mm in diameter and rotated at 10 kHz with a
relaxation delay time of 3 s. Sodium 3-(trimethylsilyl)-
propionate-2,2,3,3,-d4 (29Si: 0 ppm) was used as an external
standard for the calibration of chemical shifts. Accumulation
numbers were fixed at about 10,000.
Mn K-edge XAFS spectra were analyzed with the ATHENA
and ARTEMIS programs using the IFEFFIT suite (ver.
1.2.11).29 Mn K-edge XANES spectra were calibrated with
Mn foil to the first inflection point of the edge, an energy of
6539.0 eV.30 For Mn K-edge EXAFS analysis, the threshold
energy was tentatively set at the first inflection point of the Mn
K edge, and background was subtracted by the Autobk
method.29a,31 k3-Weighted extended EXAFS oscillations were
Fourier transformed into R-space, and single-scattering curve-
fitting analysis was performed in R-space with coordination
number (CN), interatomic distance (R), Debye−Waller factor
(σ2), and correction-of-edge energy (ΔE0) as fitting parameters.
Phase shifts and backscattering amplitudes for Mn···Mn and
Mn−O(N) were calculated with the FEFF8 code32 from the
coordinates of 1 obtained from X-ray single-crystal structural
data.26c For the curve fitting of 1, CNs and bond distances were
fixed as the average values obtained from X-ray single crystal
structural analysis.
2-6. Catalytic Epoxidation of Alkenes. Each catalyst
(Mn4: 0.67 × 10−6 mol) was suspended in a CH2Cl2 solution
(3 mL) under 101.3 kPa of O2 in a closed glass reactor
equipped with a mechanical stirrer. Alkene epoxidation
reactions were conducted at 283 K by the addition of an
alkene reactant, biphenyl as an internal standard, and
isobutyraldehyde (IBA) to the suspension (Mn4/reactant/
biphenyl/IBA = 1/500/250/1500 in molar ratio). To monitor
the reaction, reactants and products were analyzed by GC
(Shimadzu GC-14B, CHIRALDEX G-TA column) using an
internal standard and by GC-MS (Shimadzu GC-2010,
CHIRALDEX B-DM column) at appropriate intervals.
Inductively coupled plasma (ICP) measurements to estimate
Mn amounts in solutions were performed with a plasma
spectrometer (SPS 7800, SII Seiko Instrument Inc.). Activation
energy (Ea) was estimated in the temperature range of 283−
303 K. After epoxidation of cholesteryl benzoate, the product
33a
1
ratio of α-epoxide/β-epoxide was estimated by H NMR.
X-ray photoelectron spectroscopy (XPS) was measured on a
VG ESCALAB 220i-XL apparatus at a base pressure of 1 × 10−7
Pa. The X-ray source, voltage, and current were Mg-Kα, 15 kV,
and 20 mA, respectively. Binding energies were referenced to
those of C 1s (284.8 eV) for 1 and Si 2p (103.4 eV) for 2-5.
Nitrogen adsorption was measured on a Micromeritics
ASAP-2020 analyzer (Shimadzu) at 77 K. Each sample (100
mg) was degassed at 373 K for 2 h under vacuum before an
adsorption measurement, and the dead volume was estimated
by using He.
X-ray absorption fine structure (XAFS) spectra at the Mn K-
edge were measured at the BL-9C and BL-12C stations of the
Photon Factory at KEK-IMSS. The energy and current of the
electrons in the storage ring were 2.5 GeV and 450 mA,
respectively. X-rays from the storage ring were monochromat-
ized by Si(111) double crystals. Mn K-edge XAFS of 1 and
SiO2-supported Mn catalysts were measured in transmission
mode, and ionization chambers filled with pure N2 gas and a
mixture of N2 and Ar gases (15/85, v/v) were used to monitor
the incident and transmitted X-rays, respectively. Mn catalysts
3. RESULTS AND DISCUSSION
3-1. Preparation and Characterization of Supported
Mn Clusters on a SiO2 Surface. Chemical grafting of metal
complexes allows for easy separation of catalyst from reagents,
and prevents the aggregation of metal complexes that lead to
deactivation.3,20−22 However, the simple chemical grafting of
metal complexes on a support does not always give a stable
metal site as catalytically active species, and leaching of the
metal species from the surface often proceeds under catalytic
reaction conditions.15−19 We have proposed the design of
molecularly imprinted metal-complex catalysts on SiO2 surfaces
using SiO2-matrix overlayers for imprinting a ligand of a
supported metal complex.34 The hydrolysis−polymerization of
TMOS deposited on the surface with attached metal complexes
can produce surface SiO2-matrix overlayers on the SiO2 surface,
whose height can be easily regulated.34 Significant improve-
ments in catalytic performance have been observed on
molecularly imprinted metal complex catalysts on SiO2.34
2022
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
Figure 1. (A) Schematics of the attachment of the Mn4O2 cluster (1) on the surface of SiO2 and surface functionalization with SiO2-matrix
overlayers to produce nanopores protecting the supported Mn cluster. (B) HRTEM images of a SiO2 particle of 5c with the SiO2 matrix. The
average particle size of Aerosil 200 support was 12 nm (reported by the product company), but it was difficult to visualize SiO2 overlayers on such
small particles, which gathered each other.
Motivated by these researches, this technique was employed
to try to stabilize a Mn cluster supported on the surface of SiO2
particles. The SiO2-supported Mn cluster catalyst was active for
epoxidation but easily leached to reaction solutions. The surface
of the SiO2 particles with the Mn cluster was functionalized by
stacking surface SiO2-matrix overlayers to suppress the leaching
of the supported Mn cluster under the reaction conditions.
Figure 1 shows schematics of the functionalization of the
supported Mn clusters with surface SiO2-matrix overlayers. A
Mn4O2 precursor [Mn4O2(CH3COO)7(bipy)2](ClO4)·3H2O
(1) was chemically attached on SiO2 particles (Aerosil 200)
(2), and subsequently SiO2-matrix overlayers were stacked
surrounding the supported Mn cluster by hydrolysis-polymer-
ization of TMOS (5).
The attachment of [Mn4O2(CH3COO)7(bipy)2](ClO4)·
3H2O (1) onto the surface of SiO2 particles was performed
by a wet impregnation method, and one of the CH3COO
ligands was observed to be reacted with surface Si−OH groups
in the grafting process. Subsequent treatments under vacuum at
different temperatures produced different supported Mn
clusters (2, 3, and 4), whose structures were characterized by
quantitative analysis of CH3COO ligands, FT-IR, DR-UV/vis,
and Mn K-edge XAFS. During the impregnation of 1 with SiO2,
0.9 equiv. of CH3COOH to 1 was detected in a CH2Cl2
solution, and a SiO2-attached Mn cluster (2) was obtained. The
amount of CH3COO ligands remaining on 2 was observed to
be 5.9 equiv. per 1 by a reaction with CF3COOH, which was
consistent with the released CH3COOH ligand of 1.
which were similar to those of 1 (∼1404, ∼1610, and 1336
cm−1, respectively) (Supporting Information, Figure S2 (A)).
The shoulders shown on ν(CH3COO)sym and ν(CH3COO)asym
peaks of 2 were lower than 1, suggesting the release of a
CH3COO ligand during the attachment. On the other hand,
the release of bipy ligands of 1 was not observed during the first
impregnation process. Three peaks attributed to the bipy
ligands of 2 were observed by FT-IR at 1450, 1472, and 1500
cm−1, which were similar to those of 1 (1448, 1473, and 1499
cm−1) (Supporting Information, Figure S2 (A)). There was no
significant change in the peak intensities of the three peaks
attributed to bipy, indicating that the bipy ligands still
coordinated to the Mn cluster on 2.
Two peaks attributed to the π−π* transition of the bipy
ligands (234 and 294 nm) and two peaks attributed to the d-π*
transition from Mn to the bipy ligands at 442 and 500 nm were
observed in the transmittance UV/vis spectrum of 1 in CH2Cl2
(Figure 2). After the attachment of 1 to the SiO2 surface, two
characteristic peaks attributed to the π−π* transition of the
bipy ligands were still observed at 242 and 303 nm, and two
characteristic peaks in the visible region were also observed at
454 and 499 nm. There were no significant changes in the UV/
vis region during the first impregnation step.
Edge energy at the Mn K-edge is considered to be relative to
the actual oxidation state of Mn species, and the edge energy of
the Mn species in 2 was observed to be 6550.4 eV, which was
similar to that of 1 (6550.8 eV; Supporting Information, Figure
S3, Table S1). The formal oxidation states of four Mn ions were
confirmed to be +3 by the ESI-TOF MS spectrum of 1
(Supporting Information, Figure S1). The local coordination
structure of 2 was almost similar to that of 1 from Mn K-edge
Three broad FT-IR peaks of 2 attributed to ν(CH3COO)sym
,
ν(CH3COO)asym, and ν(CH3COO) of the CH3COO ligands
were observed at ∼1420, 1599, and 1345 cm−1, respectively,
2023
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
Figure 3. k3-Weighted Mn K-edge EXAFS Fourier transforms for 1, 2,
3, and 5c, measured at 20 K.
Figure 2. (A) UV/vis spectrum of 1 in CH2Cl2 and DR-UV/vis
spectra of 2, 3, 4, and 5c. (B) Magnified view from 350 to 700 nm.
the Mn4 butterfly framework of 1 stuck out, leading to the
distortion of the Mn4 cluster framework with the decrease in
the CN of the long Mn−Mn interaction at 0.335 nm. A
possible model of 2 is illustrated in Figure 1. The Mn−O−Mn
bonds in 1 were negligible in Raman spectroscopy, as is the
case in 2.
EXAFS (vide infra). Therefore, the Mn oxidation state of 2 was
suggested to be +3.
The curve-fitting analysis of EXAFS suggested the local
coordination structure of the SiO2-supported Mn cluster (2) as
shown in Table 1, Figure 3 and Supporting Information, Figure
We investigated further the treatment of 2 under vacuum at
different temperatures to release extra CH3COO ligands from
the supported Mn complex. When 2 was heated at 373 K, in
situ FT-IR measurements suggested that one of the six
CH3COO ligands on 2 was released, and the peak intensities
attributed to the CH3COO ligands remaining in 3 further
decreased (Supporting Information, Figure S2(B)). The
amount of CH3COO ligands on 3 was estimated by GC to
be 5.1 equiv relative to Mn4. Further treatment under vacuum
at 413 K produced 4 with 2.8 equiv of CH3COO ligands
relative to Mn4. Mn K-edge XANES and EXAFS revealed that
the Mn K-edge energy in 3 was similar to that in 2 (6550.5 eV;
Supporting Information, Figure S3, Table S1); however, the
CN of the Mn−O(N) interaction became smaller than that of 2
(Table 1). The edge energy suggested that the Mn oxidation
state of 3 is +3. However, edge energy varies (∼2 eV) when the
coordination structures of metal complexes and the type of
ligands are different even if they have the same formal oxidation
states,35 and it is difficult to conclude the details of the
coordination structure of 3 from the obtained parameters from
the curve-fitting analysis of EXAFS. The obvious fact is that
bridging CH3COO ligands were partially released by the heat
treatment at 373 K (3), and the trend was enhanced at 413 K
(4). DR-UV/vis spectra of 3 and 4 (Figure 2) showed two
characteristic peaks in the visible region which were also
observed at 454 and 499 nm, suggesting bipy ligands were still
coordinated to the Mn center. It might be possible that other
silanol groups coordinated from the surface to open
coordination sites of the Mn cluster.
Table 1. Structural Parameters of 1, 2, 3, and 5c Analyzed by
Mn K-edge EXAFS Curve Fitting
shell
CN
R /nm
ΔE0
σ2/105 nm2
a
1: [Mn4O2(CH3COO)7(C10H8N2)2]ClO4·3H2O
Mn···Mn
Mn···Mn
Mn−O/N
Mn−O/N
0.5
0.287 0.001
0.335 0.001
0.191 0.001
0.219 0.001
10
10
8
1
1
2
5
7
7
1
1
1
1
2
3.9 0.2
2.1 0.2
2
2
8
b
2: SiO2-supported Mn cluster (Mn: 2.2 wt %)
Mn···Mn
Mn···Mn
Mn−O/N
Mn−O/N
0.8 1.0
1.0 1.2
4.0 2.4
1.4 2.4
0.285 0.010
0.335 0.007
0.189 0.004
0.220 0.004
1
18
5
4
8
8
8
8
4
4
14 12
4
4
7
7
c
3: SiO2-supported Mn cluster (Mn: 2.2 wt %)
Mn···Mn
Mn···Mn
Mn−O/N
1.3 0.6
0.9 0.5
2.7 0.3
0.289 0.003
0.336 0.003
0.187 0.001
5
4
9
6
5
4
4
1
17
3
5
2
d
5c: SiO2-matrix functionalized Mn cluster (Mn: 0.7 wt %)
Mn···Mn
Mn−O
Mn−O
0.7 0.3
1.4 1.7
1.0 1.8
0.301 0.008
0.185 0.006
0.221 0.013
14 11
9
2
2
3
9
8
8
14 21
a
k = 30−140 nm−1, R = 0.09−0.34 nm, Rf = 1.0%. S0 was fitted to be
b
c
0.82. k = 30−120 nm−1, R = 0.08−0.34 nm, Rf = 0.5%. S0 = 0.82. k =
d
30−120 nm−1, R = 0.08−0.34 nm, Rf = 0.4%. S0 = 0.82. k = 30−110
nm−1, R = 0.07−0.30 nm, Rf = 1.6%. S0 = 0.82.
S4. Two peaks of Mn···Mn at 0.285 and 0.335 nm were
observed on 2. The contribution of the long Mn···Mn
interaction decreased to half, while the short Mn···Mn
interaction remained: two Mn···Mn interactions at 0.285
3-2. Surface Functionalization of SiO2-Matrix Over-
layers on the Surface of SiO2 Particles Attaching the Mn
Clusters. SiO2-matrix overlayers were prepared on the SiO2-
supported Mn clusters (2) by CVD and subsequent
hydrolysis−polymerization of tetramethoxysilane (TMOS)
and H2O. The stacking procedure and manner of the SiO2-
matrix overlayers on SiO2 surfaces so as to surround supported
metal complexes have previously been reported.34 We prepared
three SiO2-matrix overlayers on 2 to produce three Mn catalysts
(5a, 5b, and 5c). The SiO2-matrix overlayers of 5a were
prepared by the hydrolysis−polymerization of TMOS in a step
0.010 nm (coordination number (CN) = 0.8
1.0) and at
0.335 0.007 nm (CN = 1.0 1.2) were fitted, together with
Mn−O(N) at 0.189 0.004 nm (CN = 4.0 2.4) and 0.220
0.004 nm (CN = 1.4 2.4). Similar UV/vis and Mn K-edge
XANES spectra of 1 and 2 indicated that the coordination
structure of the Mn cluster of 2 was similar to that of 1.
However, the structural parameters obtained by Mn K-edge
EXAFS suggested that two Mn atoms at the edge position of
2024
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
using 0.67 g of TMOS and 0.27 mL of H2O, respectively (for
0.2 g of 2); that of 5b was prepared using 1.34 g of TMOS and
0.54 mL of H2O (twice amounts of 5a), respectively, in one
step CVD/hydrolysis−polymerization process. 5c was prepared
by performing the CVD/hydrolysis−polymerization processes
twice, using 0.67 g of TMOS and 0.27 mL of H2O, respectively,
in each process. In the case of the hydrolysis−polymerization of
TMOS for 5a, almost 70% of the used TMOS was actually
polymerized under the conditions.
small as 12 nm and each SiO2 particles gathered each other. It
was difficult to estimate the actual average heights of the SiO2-
matrix overlayers by TEM. Transmission electron microscopy
(TEM) revealed the morphological differences of 2 and 5c
(Figure 1(B) and Supporting Information, Figure S6). The
surface of 5c with the SiO2-matrix overlayers was rougher after
the stacking of SiO2-matrix overlayers than that of 2, and
surface thin layers were observed as shown in Figure 1(B).
The ratios of the three Si species on 5c were estimated to be
4% (Q2), 34% (Q3), and 62% (Q4), respectively (Supporting
Information, Figure S5). The major species being Q4 indicated
that Si−O−Si networks were formed by the hydrolysis−
polymerization of TMOS. On the other hand, the Si species on
5b were 10% (Q2), 42% (Q3), and 48% (Q4), respectively. The
amount of TMOS used for the hydrolysis-polymerization
affected not only the amount of stacked SiO2 matrix but also
the ratio of these compositions.
Figure 4(A) shows 29Si solid-state MAS NMR spectra of 5a,
5b, and 5c with SiO2-matrix overlayers, and three Si
Significant differences were also observed in the surface areas
of 2, 5a, 5b, and 5c. N2 adsorption isotherms at 77 K and their
t-plots are presented in Supporting Information, Figure S7. The
N2 adsorption/desorption isotherms suggest increases in
surface areas after the stacking of the SiO2-matrix overlayers
as well as the formation of micropores in the matrix overlayers.
Brunauer−Emmett−Teller (BET) surface areas and external
surface areas were respectively estimated to be 182 and 164 m2
g−1 for 2, 347 and 286 m2 g−1 for 5a, 578 and 529 m2 g−1 for
5b, and 294 and 197 m2 g−1 for 5c. Note that the external
surface area of 5c was similar to that of 2, while the external
surface areas of 5a and 5b were relatively higher than that of 2.
These results implied that the hydrolysis−polymerization
produced a SiO2 matrix with micropores and that repetition
of the process tended to fill up the rough structures of the SiO2
matrix, retaining the external surface area of the SiO2 particles.
The t-plot of 5c had a gentle break at 0.37 nm, suggesting the
formation of micropores with an average diameter of 0.74 nm.
Figure 4(B) shows Mn 2p3/2 and 2p1/2 XPS spectra of 2, 5a,
5b, and 5c. Mn 2p3/2 and 2p1/2 peaks for 1 were clearly
observed at 641.7 and 653.4 eV, respectively, and those for 2 on
SiO2 were observed at 642.2 and 653.9 eV, respectively. The
slight shifts in the Mn 2p binding energies would reflect the
coordination of surface oxygen to the Mn complex. After the
stacking of the SiO2-matrix overlayers, the Mn 2p peaks were
observed to get smaller. Even the Mn 2p peaks of 5a entirely
disappeared, as did those of 5b and 5c (Figure 4(B)).
Considering the escape depth of Mn 2p, these results clearly
indicated that the supported Mn clusters on the surface of the
SiO2 particles were closely surrounded by the produced SiO2-
matrix overlayers.
It was found out that the structure of 5c after stacking of the
SiO2-matrix overlayers was different from that of 2. Mn K-edge
EXAFS found the local coordination structure of the supported
Mn species on 5c: Mn···Mn interaction length of 0.301 0.008
nm (CN = 0.7 0.3) and two Mn−O interactions at 0.185
0.006 nm (CN = 1.4 1.7) and 0.221 0.013 nm (CN = 1.0
1.8) (Table 1, Figure 3 and Supporting Information, Figure
S4). The amount of remaining CH3COO ligands of 5c was
estimated to be 2.0 equiv to Mn4. In the DR-UV/vis spectrum
of 5c, the characteristic peaks attributed to d-π* transition from
Mn to the bipy ligands disappeared (Figure 2 (B)). These
results indicated that the partial release of the coordinating
ligands brought about the transformation of the supported Mn
cluster to dimeric Mn species in the SiO2-matrix overlayers.
Therefore, a possible composition of dimeric Mn species on 5c
Figure 4. (A) 29Si solid-state MAS NMR of 2, 5a, 5b, and 5c, and (B)
Mn 2p XPS spectra of 1, 2, 5a, 5b, and 5c.
components of Q4 (Si(OSi)4), Q3 (Si(OSi)3(OH)), and Q2
(Si(OSi)2(OH)2) at around −111, −102, and −92 ppm,
respectively, were observed. Difference between the spectra of 5
and 2 is attributed to Si species of the SiO2-matrix overlayers of
5, and the sum of areas of three Q4, Q3, and Q2 peaks in a
difference spectrum (5 − 2) is relative to the amount of stacked
Si species by the hydrolysis−polymerization (Supporting
Information, Figure S5). Although there is a possibility that
the quantitative peak area analyses of 29Si NMR spectra might
be affected by paramagnetic Mn cluster complex (some Si sites
might not be observed by 29Si NMR), the effect of
paramagnetic species was considered to be not significant in
the current case since NMR of Mn4 cluster (1) was measurable.
The ratio of the sum of the peak areas of SiO2-matrix overlayers
on 5a, 5b, and 5c was estimated to be 1/1.1/2.2. It was found
that the repetition of the hydrolysis-polymerization cycles was
efficient to increase the amount of SiO2-matrix overlayers (5c),
while the simple increase in the amounts of used TMOS and
H2O for the hydrolysis-polymerization process was inefficient
(5b).
When 70% of used TMOS was polycondensed, the Si atomic
ratio of the SiO2-matrix overlayers to the SiO2 support was
0.92. The Si atomic ratio calculated by using the peak
intensities (the summation of the peak intensities of the three
Si species) of 29Si NMR in Figure 4A was 0.93, which agreed
with the value calculated for mass gain. Assuming that the SiO2-
matrix overlayers spread on the surface of the SiO2 support
were of quartz structure, which exhibits the largest density of
SiO2, the calculated height of the observed amount of the SiO2-
matrix overlayers on 5a was 2.2 nm on the used SiO2 support
(Aerosil 200; surface area: 200 m2 g−1). In the case of 5c, the
calculated height of the SiO2-matrix overlayers was 4.3 nm. The
average particle size of the SiO2 support (Aerosil 200) was as
2025
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
a
Table 2. Catalytic Performances of Mn Catalysts for Epoxidation of Alkenes
a
Reaction conditions: Mn4, 0.67 × 10−6 mol; Mn4/reactant/biphenyl/IBA = 1/500/250/1500; CH2Cl2, 3 mL; 283 K; O2, 101.3 kPa. Biphenyl was
b
c
used as an internal standard. Conversion % = (Initial mole of reactant − Residual mole of reactant)/Initial mole of reactant × 100%. Selectivity %
d
= Mole of epoxide product/(Initial mole of reactant − Residual mole of reactant) × 100%. The leaching of Mn was detected by ICP spectroscopy
e
f
g
of the reaction mixture filtrate after the reaction. The amount of SiO2 was 25 mg. Benzaldehyde was formed as a byproduct. The reaction was
h
i
conducted under N2 atmosphere. α,β-Unsatrated ketone was formed as a byproduct. 1-Heptanal and 3-octenone were formed as a byproduct. n.d.:
not determined.
could be written as [Mn2(CH3COO)]2x, although the exact
detailed structure was still controversial at this moment. The
edge energy obtained from Mn K-edge XANES in 5c (6548.7
eV; Supporting Information, Figure S3, Table S1) was smaller
than that in 2, suggesting the reduction of the Mn species.
3-3. Catalytic Performance in Alkene Epoxidation.
Alkene epoxidation using O2 and IBA (a sacrificial reagent)33
was examined on the supported Mn clusters (2−4 and 5a−c).
The results are summarized in Table 2. The homogeneous
solution of 1 promoted the epoxidation of trans-stilbene with
high selectivity of 94% (entry 3), whereas no reaction occurred
without 1 (entry 1). The SiO2-supported Mn clusters, 2−4,
showed activity and high selectivity over 95% (entries 4−6),
whereas SiO2 without Mn was inactive (entry 2). After a
reaction was completed, the solid catalysts were separated from
the solution phases, and the amounts of leached Mn species
were determined by ICP measurements. It was found out that
leaching of the supported Mn complexes was substantial: 45.8%
for 2, 53.4% for 3, and 47.3% for 4. These results suggested that
these supported Mn clusters that were simply attached to the
surface were not durable under heterogeneous epoxidation
conditions.
for 5b, and 91% for 5c at ∼90% conversion for 24−33 h
(entries 7−9). Although the epoxidation reaction became
slower, the leaching of the active Mn catalyst was significantly
suppressed on 5a−c. The amounts of leached Mn species were
0.41% for 5a and negligible for 5b (0.02%) and 5c (0.01%).
When the reaction was run to a conversion of ∼50%, the
reaction solution was filtered from solid 5c and showed no
further progress in the catalytic reaction (Figure 5(A)). On the
other hand, the filtration of 3 was significantly active. These
results clearly show that the supported Mn catalysts embedded
in the SiO2-matrix overlayers, in particular 5c, were highly
stable, and that the catalytic reaction on 5c proceeded in the
heterogeneous phase without unfavorable leaching of Mn into
solution. In contrast, dissolved Mn in the solution was active on
3. After the 500 cycle of epoxidation, no Mn−Mn interaction
was observed on 5c (Supporting Information, Figure S4 and
Table S2). The decrease in the amount of Mn in the catalyst
was not observed during the reaction, suggesting that transient
leaching and re-deposition of the Mn species would not
contribute to the reaction.36
Figures 5(B), 5(C), and Supporting Information, Figure S8
show the conversion of trans-stilbene, the selectivity of trans-
stilbene oxide, and the amount of Mn species remaining on the
SiO2 particle surface on the catalysts 3 and 5c after they were
recycled 5 times. On 3 without the SiO2-matrix overlayers, the
conversion decreased significantly because of the extensive
leaching of Mn species into solution. After 2 cycles of
The preparation of the SiO2-matrix overlayers on the SiO2-
supported Mn clusters imparted the catalyst with remarkable
stability (Table 2). The supported Mn catalysts with the SiO2-
matrix overlayers (5a, 5b, and 5c) promoted the epoxidation
with high selectivity for trans-stilbene epoxide: 91% for 5a, 88%
2026
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
products, α-epoxide and β-epoxide (Scheme 1). In the case of
autoxidation using peroxo species (e.g., m-CPBA), α-epoxide is
Scheme 1. Epoxidation of Cholesteryl Benzoate
preferably produced. Analogously, if a free peroxo species were
to be formed on the Mn site, but reacted independently with an
alkene reactant in solution (in a similar way as autoxidation),
the selective formation of α-epoxide would also be observed.
On the other hand, if the epoxidation step were to proceed on a
metal site, the steric hindrance between the cholesteryl
benzoate at 5 and 6-position and the metal center is expected
to change the selectivity of the products. It was reported that
the selectivity of the cholesteryl benzoate epoxidation changed
(β-epoxide was preferably obtained) when catalytic amounts of
Mn(II) complexes were used in the O2/IBA system.33a For
instance, the α-epoxide/β-epoxide product ratio on bis-
(acetylacetonato)manganese(II) was reported to be 20/80
and that on bis(dipivaloylmethanato)manganese(II) was 18/
82.33a
We performed the epoxidation of the cholesteryl benzoate
with m-CPBA and 5c. With m-CPBA without the Mn catalysts,
the α-epoxide was preferably formed with the α-epoxide/β-
1
epoxide product ratio of 73/27 (6.5 h; determined by H
NMR), which agreed with reported results.33a On the other
hand, the product ratio of α-epoxide/β-epoxide was 20/80 (24
h) on 5c. The reaction mechanism and intermediate structures
on 5c are not clear at the moment, but the opposite epoxide
selectivity for the epoxidation of the cholesteryl benzoate
suggests that the Mn species on 5c surrounded by the SiO2-
matrix overlayers indeed are involved in the epoxidation step.
It is to be noted that the activation energy of catalytic
epoxidation on 5c differed widely by alkene. Cyclooctene
epoxidation had similar activation energies on 3 (15 kJ mol−1)
and 5c (16 kJ mol−1), and the turnover frequencies (TOFs)
were 4.0 × 10−2 and 1.9 × 10−2 s−1, respectively. On the other
hand, the TOFs and activation energies for the epoxidation of
the larger reactant trans-stilbene on 5c were 6.7 × 10−3 s−1 and
31 kJ mol−1, whereas those on 3 were 2.5 × 10−2 s−1 and 15 kJ
mol−1. The small activation energies around 15 kJ mol−1 on 3
might be related to mass transfer. The smaller reaction rate of
trans-stilbene on 5c is due to the diffusion resistance of the
reactants inside the SiO2-matrix overlayers. The 2-fold higher
activation energies for the trans-stilbene epoxidation on 5c
would result from the different transition state structure
produced from the reactant, the oxidant, and the Mn species
on 5c inside the SiO2-matrix overlayers. Since the structure of
Mn species on 5c was totally different from that on 3, the loss
of CH3COO and bipy ligands, the addition of surface silanol
groups, and the existence of SiO2-matrix overlayers, the
structures of transition state and intermediate responsible for
determining activation energy would be different. The steric
hindrance for the coordination and subsequent oxidation of
trans-stilbene at the Mn site inside the surface-matrix overlayers
would also be one of the reasons to produce the different
transition state/intermediate structures.34a Similar trends were
observed for the cholesteryl benzoate epoxidation (14 kJ mol−1
on 3; 74 kJ mol−1 on 5c), supporting the above discussion.
Figure 5. (A) Test of the heterogeneous nature of the catalysts in
trans-stilbene epoxidation. ((B), (C)) Recycling tests of 3 (for 6 h)
and 5c (for 30 h) in trans-stilbene epoxidation. (B) Conversion of each
cycle; (C) selectivity for epoxide; reaction conditions: Mn4, 0.67 ×
10−6 mol; Mn4/trans-stilbene/IBA = 1/500/1500; CH2Cl2, 3 mL; 283
K; O2, 101.3 kPa.
epoxidation, 80% of the supported Mn clusters were lost
(Supporting Information, Figure S8). On the other hand, the
leaching of the Mn catalyst on 5c with the SiO2-matrix
overlayers was negligible after 6 epoxidation cycles, and the
conversion of trans-stilbene remained >90%, keeping up with
the high epoxide selectivity. The differences were obvious
between 3 and 5c, demonstrating the positive effect of the
surface SiO2-matrix overlayers on the stability of the supported
Mn epoxidation catalysts.
Other alkenes were smoothly converted to epoxides on 5c
(Table 2). Observed major byproducts were aldehydes from
CC bond cleavage and ketones from oxidation at the α-
position. The reactions of internal alkenes such as cyclohexene,
cyclooctene, and norbornene produced epoxides with high
selectivity, whereas the selectivity of terminal alkene such as 1-
octene was low, forming 1-heptanal and 3-octenone as
byproducts.
We also investigated the epoxidation of cholesteryl benzoate
using m-chloroperbenzoic acid (m-CPBA) and the Mn catalyst
5c. The experiment suggests whether a Mn center contributes
to the oxidation step in epoxidation cycle or not.33a The
epoxidation of cholesteryl benzoate provides two epoxide
2027
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
Notes
We also prepared catalyst 6 by releasing the ligands of 2 at
673 K so that 6 had MnOx as supported species. Catalyst 7 was
prepared by subsequent stacking of SiO2-matrix overlayers on 6
with MnOx species. It is to be noted that 6 and 7 exhibited
different epoxidation performances as shown in Table 2.
Catalyst 6 exhibited a similar behavior to catalysts 2, 3, and 4
(conversion: 99%, epoxide selectivity: 97%) for the trans-
stilbene epoxidation, and significant leaching was observed
(38.6% for 12 h) (entry 12). On the contrary, the epoxidation
was suppressed on 7, and the corresponding epoxide was not
observed even after 36 h (entry 13). These results suggested
that the stacking of the SiO2-matrix overlayers on the MnOx
species without the coordinating ligands did not provide
reaction space above the Mn site, resulting in large loss of the
catalytic activity.
Although the coordination structure of 5c decomposed from
the Mn4 clusters (1 and 2) during the preparation step of the
SiO2-matrix overlayers, the catalyst 5c prepared on 2 with the
Mn cluster with the several coordinating ligands cannot be
obtained from the simple supported MnOx species on the SiO2
surface. Indeed, 5c, whose complete local coordination
structure is not clear at the moment, exhibited robust
epoxidation performances under the identical conditions. The
integration of the active Mn site and the surface SiO2-matrix
overlayers would cooperatively achieve robust catalysis in the
heterogeneous phase.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Prof. Kentaro Tanaka and Prof. Yasuyuki
Yamada at Nagoya University for measuring the ESI-TOF MS
spectrum of 1. This work was supported by the JSPS Funding
Program for Next Generation World-Leading Researchers
(GR090), MEXT/JSPS KAKENHI Grants Nos. 22108534
and 23750068, NEDO, and JST Research Seeds Program.
XAFS measurements were performed with the approval of PF-
PAC (Nos. 2010G021 and 2011G176).
REFERENCES
■
(1) (a) Shibasaki, M.; Yamamoto, Y., Eds.; Multimetallic Catalysts in
Organic Synthesis; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim,
Germany; 2004. (b) Christmann, M.; Brase, S., Eds.; Asymmetric
Synthesis−The Essentials; Wiley-VCH Verlag GmbH & Co. KGaA:
Weinheim, Germany; 2007.
̈
(2) Reviews for supported metal complex catalysts are as follows:
(a) Sheldon, R. A.; Wallau, M.; Arends, I. W. C. E.; Schuchardt, U. Acc.
Chem. Res. 1998, 31, 485−493. (b) McMorn, P.; Hutchings, G. J.
Chem. Soc. Rev. 2004, 33, 108−122. (c) Thomas, J. M.; Raja, R.; Lewis,
D. W. Angew. Chem., Int. Ed. 2005, 44, 6456−6482. (d) Notestein, J.
M.; Katz, A. Chem.Eur. J. 2006, 12, 3954−3965. (e) Corma, A.;
García, H. Adv. Synth. Catal. 2006, 348, 1391−1412. (f) Tada, M.;
́
Iwasawa, Y. Chem. Commun. 2006, 2833−2844. (g) Coperet, C.;
Basset, J. -M. Adv. Synth. Catal. 2007, 349, 78−92. (h) Wegener, S. L.;
Marks, T. J.; Stair, P. C. Acc. Chem. Res. 2012, 45, 206−214.
4. CONCLUSIONS
The functionalization of the surface of SiO2 particles with
attached [Mn4O2(CH3COO)7(bipy)2](ClO4)·3H2O (1) and
the stacking of SiO2-matrix overlayers were successfully used to
prepare a new and robust heterogeneous Mn catalyst for the
selective epoxidation of alkenes. 1 was chemically attached to a
SiO2 particle surface, and the attached structure was
characterized by FT-IR, UV/vis, Raman spectroscopy, XPS,
and Mn K-edge XAFS. The stacking of surface SiO2-matrix
overlayers surrounding the supported Mn clusters of 2 with a
similar structure to the precursor 1 brought about a structural
change in the original Mn cluster to an active dimeric Mn
species on 5c together with the formation of a reaction space in
the close proximity of the Mn species inside the SiO2-matrix
overlayers. 5c exhibited different catalytic activity, especially in
the reaction rate and activation energy compared with 2 and 3,
and the durability of the Mn catalysts under the epoxidation
reaction conditions was remarkably improved. The leaching of
the supported Mn catalysts on 5c was prevented. The surface
functionalization through the SiO2-matrix overlayers to provide
the composite of an active metal site and a reaction space is
promising to achieve a robust catalysis on a supported metal
catalyst in a heterogeneous phase.
(3) Examples for supported metal complex catalysts are as follows:
(a) Tada, M.; Shimamoto, M.; Sasaki, T.; Iwasawa, Y. Chem. Commum.
2004, 2562−2563. (b) Le Roux, E.; Chabanas, M.; Baudouin, A.; de
́
Mallmann, A.; Coperet, C.; Quadrelli, E. A.; Thivolle-Cazat, J.; Basset,
J.-M.; Lukens, W.; Lesage, A.; Emsley, L.; Sunley, G. J. J. Am. Chem.
Soc. 2004, 126, 13391−13399. (c) Riollet, V.; Quadrelli, E. A.;
́
Coperet, C.; Basset, J.-M.; Andersen, R. A.; Kohler, K.; Bottcher, R.-
̈ ̈
M.; Herdtweck, E. Chem.Eur. J. 2005, 11, 7358−7365. (d) Rendon
N.; Berthoud, R.; Blanc, F.; Gajan, D.; Maishal, T.; Basset, J.-M.;
Coperet, C.; Lesage, A.; Emsley, L.; Marinescu, S. C.; Singh, R.;
Schrock, R. R. Chem.Eur. J. 2009, 15, 5083−5089. (e) Villaverde,
G.; Corma, A.; Iglesias, M.; Sanchez, F. ChemCatChem 2011, 3, 1320−
1328. (f) Goodrich, P.; Hardacre, C.; Paun, C.; Ribeiro, A.; Kennedy,
S.; Lourenco, M. J. V.; Manyar, H.; Nieto de Castro, C. A.; Besnea, M.;
Parvulescu, V. I. Adv. Synth. Catal. 2011, 353, 995−1004. (g) Sievers,
C.; Jimenez, O.; Knapp, R.; Lin, X.; Muller, T. E.; Turler, A.;
́
,
́
́
̧
̂
́
̈
̈
Wierczinski, B.; Lercher, J. A. J. Mol. Catal. A: Chem. 2008, 279, 187−
199. (h) Caplan, N. A.; Hancock, F. E.; Bulman Page, P. C.;
Hutchings, G. J. Angew. Chem., Int. Ed. 2004, 43, 1685−1688. (i) Hara,
K.; Akiyama, R.; Takakusagi, S.; Uosaki, K.; Yoshino, T.; Kagi, H.;
Sawamura, M. Angew. Chem., Int. Ed. 2008, 47, 5627−5630. (j) Tada,
M.; Coquet, R.; Yoshida, J.; Kinoshita, M.; Iwasawa, Y. Angew. Chem.,
Int. Ed. 2007, 46, 7220−7223. (k) Tada, M.; Akatsuka, Y.; Yang, Y.;
Sasaki, T.; Kinoshita, M.; Motokura, K.; Iwasawa, Y. Angew. Chem., Int.
Ed. 2008, 47, 9252−9255. (l) Tada, M.; Muratsugu, S.; Kinoshita, M.;
Sasaki, T.; Iwasawa, Y. J. Am. Chem. Soc. 2010, 132, 713−724.
(m) Muratsugu, S.; Weng, Z.; Nakai, H.; Isobe, K.; Kushida, Y.; Sasaki,
T.; Tada, M. Phys. Chem. Chem. Phys. 2012, 14, 16023−16031.
(4) (a) Brinksma, J.; de Boer, J. W.; Hage, R.; Feringa, B. L. In
ASSOCIATED CONTENT
■
S
* Supporting Information
An ESI-TOM MS spectrum, FT-IR spectra, Mn K-edge
XANES and EXAFS spectra and their analyses, 29Si NMR
spectra, N2 adsorption, TEM images, and residual Mn ratio in
recycling tests. This material is available free of charge via the
Internet at http://pubs.acs.org.
Modern Oxidation Methods; Backvall, J.-E.; Ed; Wiley-VCH Verlag
̈
GmbH & Co. KGaA: Weinheim, Germany; 2004; pp 295−326.
(b) Dismukes, G. C.; Brimblecombe, R.; Felton, G. A. N.; Pryadun, R.
S.; Sheats, J. E.; Spiccia, L.; Swiegers, G. F. Acc. Chem. Res. 2009, 42,
1935−1943.
(5) (a) Collman, J. P.; Lee, V. J.; Kellen-Yuen, C. J.; Zhang, X.; Ibers,
J. A.; Brauman, J. I. J. Am. Chem. Soc. 1995, 117, 692−703.
(b) Shimazaki, Y.; Nagano, T.; Takesue, H.; Ye, B.-H.; Tani, F.;
Naruta, Y. Angew. Chem., Int. Ed. 2004, 43, 98−100. (c) Rebelo, S. L.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: smura@ims.ac.jp (S.M.), mtada@ims.ac.jp (M.T.).
2028
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
H.; Pereira, M. M.; Simoes, M. M. Q.; Neves, M. G. P. M. S.;
Cavaleiro, J. A. S. J. Catal. 2005, 234, 76−87. (d) Srour, H.; Le Maux,
P.; Simonneaux, G. Inorg. Chem. 2012, 51, 5850−5856.
(c) Kuzniarska-Biernacka, I.; Silva, A. R.; Carvalho, A. P.; Pires, J.;
́
̃
Freire, C. Langmuir 2005, 21, 10825−10834. (d) Cardoso, M.; Silva,
A. R.; de Castro, B.; Freire, C. Appl. Catal., A 2005, 285, 110−118.
(18) (a) Bhattacharjee, S.; Anderson, J. A. Chem. Commun. 2004,
554−555. (b) Bhattacharjee, S.; Dines, T. J.; Anderson, J. A. J. Catal.
2004, 225, 398−407. (c) Bhattacharjee, S.; Anderson, J. A. Adv. Synth.
Catal. 2006, 348, 151−158.
(6) (a) Irie, R.; Noda, K.; Ito, Y.; Matsumoto, N.; Katsuki, T.
Tetrahedron Lett. 1990, 31, 7345−7348. (b) Zhang, W.; Loebach, J. L.;
Wilson, S. R.; Jacobsen, E. N. J. Am. Chem. Soc. 1990, 112, 2801−2803.
(c) Kureshy, R. I.; Khan, N. H.; Abdi, S. H. R.; Singh, S.; Ahmed, I.;
Shukla, R. S.; Jasra, R. V. J. Catal. 2003, 219, 1−7. (d) Shitama, H.;
Katsuki, T. Chem.Eur. J. 2007, 13, 4849−4858. (e) Zhang, J.-L.;
Garner, D. K.; Liang, L.; Barrios, D. A.; Lu, Y. Chem.Eur. J. 2009, 15,
7481−7489.
(19) Silva, A. R.; Vital, J.; Figueiredo, J. L.; Freire, C.; de Castro, B.
New. J. Chem. 2003, 27, 1511−1517.
(20) (a) Bigi, F.; Moroni, L.; Maggi, R.; Sartori, G. Chem. Commun.
2002, 716−717. (b) Zhang, H.; Xiang, S.; Xiao, J.; Li, C. J. Mol. Catal.
A: Chem. 2005, 238, 175−184. (c) Zhang, H.; Wang, Y. M.; Zhang, L.;
Gerritsen, G.; Abbenhuis, H. C. L.; van Santen, R. A.; Li, C. J. Catal.
2008, 256, 226−236. (d) Jutz, F.; Grunwaldt, J.-D.; Baiker, A. J. Mol.
Catal. A: Chem. 2008, 279, 94−103.
(7) (a) Wieghardt, K.; Bossek, U.; Nuber, B.; Weiss, J.; Bonvoisin, J.;
Corbella, M.; Vitols, S. E.; Girerd, J. J. J. Am. Chem. Soc. 1988, 110,
7398−7411. (b) Hage, R.; Iburg, J. E.; Kerschner, J.; Koek, J. H.;
Lempers, E. L. M.; Martens, R. J.; Racherla, U. S.; Russell, S. W.;
Swarthoff, T.; van Vliet, M. R. P.; Warnaar, J. B.; van der Wolf, L.;
Krijnen, B. Nature 1994, 369, 637−639. (c) Quee-Smith, V. C.;
DelPizzo, L.; Jureller, S. H.; Kerschner, J. L.; Hage, R. Inorg. Chem.
1996, 35, 6461−6465. (d) De Vos, D.; Bein, T. Chem. Commun. 1996,
917−918. (e) De Vos, D. E.; Sels, B. F.; Reynaers, M.; Subba Rao, Y.
V.; Jacobs, P. A. Tetrahedron Lett. 1998, 39, 3221−3224. (f) De Vos,
D. E.; de Wildeman, S.; Sels, B. F.; Grobet, P. J.; Jacobs, P. A. Angew.
Chem., Int. Ed. 1999, 38, 980−983. (g) Burdinski, D.; Bothe, E.;
Wieghardt, K. Inorg. Chem. 2000, 39, 105−116. (h) de Boer, J. W.;
Brinksma, J.; Browne, W. R.; Meetsma, A.; Alsters, P. L.; Hage, R.;
Feringa, B. L. J. Am. Chem. Soc. 2005, 127, 7990−7991. (i) de Boer, J.
W.; Browne, W. R.; Brinksma, J.; Alsters, P. L.; Hage, R.; Feringa, B. L.
Inorg. Chem. 2007, 46, 6353−6372. (j) de Boer, J. W.; Alsters, P. L.;
Meetsma, A.; Hage, R.; Browne, W. R.; Feringa, B. L. Dalton Trans.
2008, 6283−6295.
(8) (a) Yagi, M.; Wolf, K. V.; Baesjou, P. J.; Bernasek, S. L.;
Dismukes, G. C. Angew. Chem., Int. Ed. 2001, 40, 2925−2928.
(b) Brimblecombe, R.; Swiegers, G. F.; Dismukes, G. C.; Spiccia, L.
Angew. Chem., Int. Ed. 2008, 47, 7335−7338.
(9) (a) Kanyo, Z. F.; Scolnick, L. R.; Ash, D. E.; Christianson, D. W.
Nature 1996, 383, 554−557. (b) Boelrijk, A. E. M.; Dismukes, G. C.
Inorg. Chem. 2000, 39, 3020−3028. (c) Wu, A. J.; Penner-Hahn, J. E.;
Pecoraro, V. L. Chem. Rev. 2004, 104, 903−938. (d) Jackson, T. A.;
Brunold, T. C. Acc. Chem. Res. 2004, 37, 461−470.
(21) (a) Terry, T. J.; Stack, T. D. P. J. Am. Chem. Soc. 2008, 130,
4945−4953. (b) Tang, J.; Zu, Y.; Huo, W.; Wang, L.; Wang, J.; Jia, M.;
Zhang, W.; Thiel, W. R. J. Mol. Catal. A: Chem. 2012, 355, 201−209.
(22) (a) Schoenfeldt, N. J.; Korinda, A. W.; Notestein, J. M. Chem.
Commun. 2010, 46, 1640−1642. (b) Schoenfeldt, N. J.; Ni, Z.;
Korinda, A. W.; Meyer, R. J.; Notestein, J. M. J. Am. Chem. Soc. 2011,
133, 18684−18695. (c) Schoenfeldt, N. J.; Notestein, J. M. ACS Catal.
2011, 1, 1691−1701.
(23) (a) Song, C. E.; Roh, E. J.; Yu, B. M.; Chi, D. Y.; Kim, S. C.; Lee,
K.-J. Chem. Commun. 2000, 615−616. (b) Nielsen, M.; Thomsen, A.
H.; Jensen, T. R.; Jakobsen, H. J.; Skibsted, J.; Gothelf, K. V. Eur. J.
Org. Chem. 2005, 342−347. (c) Amarasekara, A. S.; McNeal, I.;
Murillo, J.; Green, D.; Jennings, A. Catal. Commun. 2008, 9, 2437−
2440.
(24) Tan, R.; Yin, D.; Yu, N.; Zhao, H.; Yin, D. J. Catal. 2009, 263,
284−291.
(25) (a) Cho, S.-H.; Ma, B.; Nguyen, S. T.; Hupp, J. T.; Albrecht-
Schmitt, T. E. Chem. Commun. 2006, 2563−2565. (b) Cho, S.-H.;
Gadzikwa, T.; Afshari, M.; Nguyen, S. T.; Hupp, J. T. Eur. J. Inorg.
Chem. 2007, 4863−4867. (c) Song, F.; Wang, C.; Falkowski, J. M.; Ma,
L.; Lin, W. J. Am. Chem. Soc. 2010, 132, 15390−15398.
(d) Bhattacharjee, S.; Yang, D.-A.; Ahn, W.-S. Chem. Commun. 2011,
47, 3637−3639.
(26) (a) Vincent, J. B.; Christmas, C.; Chang, H. R.; Li, Q.; Boyd, P.
D. W.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. J. Am. Chem.
Soc. 1989, 111, 2086−2097. (b) Christou, G. Acc. Chem. Res. 1989, 22,
328−335. (c) Tamane, T.; Tsubomura, T.; Sakai, K. Acta Crystallogr.
1996, C52, 777−779.
(27) (a) Libby, E.; McCusker, J. K.; Schmitt, E. A.; Folting, K.;
Hendrickson, D. N.; Christou, G. Inorg. Chem. 1991, 30, 3486−3495.
(b) Wemple, M. W.; Tsai, H.-L.; Wang, S.; Claude, J. P.; Streib, W. E.;
Huffman, J. C.; Hendrickson, D. N.; Christou, G. Inorg. Chem. 1996,
35, 6437−6449. (c) Albela, B.; El Fallah, M. S.; Ribas, J.; Folting, K.;
Christou, G.; Hendrickson, D. N. Inorg. Chem. 2001, 40, 1037−1044.
(d) Karotsis, G.; Jones, L. F.; Papaefstathiou, G. S.; Collins, A.;
Parsons, S.; Nguyen, T. D.; Evangelisti, M.; Brechin, E. K. Dalton
Trans. 2008, 4917−4925.
(28) Vincent, J. B.; Chang, H.-R.; Folting, K.; Huffman, J. C.;
Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1987, 109, 5703−
5711.
(29) (a) Ravel, B.; Newville, M. J. Synchrotron Rad. 2005, 12, 537−
541. (b) Newville, M.; Ravel, B.; Haskel, D.; Rehr, J. J.; Stern, E. A.;
Yacoby, Y. Phys. B 1995, 208−209, 154−156.
(30) Thompson, A.; Attwood, D.; Gullikson, E.; Howells, M.; Kim,
K.-J.; Kirz, J.; Kortright, J.; Lindau, I.; Liu, Y.; Pianetta, P.; Robinson,
A.; Scofield, J.; Underwood, J.; Williams, G.; Winick, H. X-ray Data
Booklet; Lawrence Berkeley National Laboratory, University of
California: Berkeley, CA, 2009.
(10) (a) Yagi, M.; Kaneko, M. Chem. Rev. 2001, 101, 21−35.
(b) Mukhopadhyay, S.; Mandal, S. K.; Bhaduri, S.; Armstrong, W. H.
Chem. Rev. 2004, 104, 3981−4026. (c) Yano, J.; Yachandra, V. K.
Inorg. Chem. 2008, 47, 1711−1726. (d) Pushkar, Y.; Long, X.; Glatzel,
P.; Brudvig, G. W.; Dismukes, G. C.; Collins, T. J.; Yachandra, V. K.;
Yano, J.; Bergmann, U. Angew. Chem., Int. Ed. 2010, 49, 800−803.
(e) Umena, Y.; Kawakami, K.; Shen, J.-R.; Kamiya, N. Nature 2011,
473, 55−61.
(11) (a) Zondervan, C.; Hage, R.; Feringa, B. L. Chem. Commun.
1997, 419−420. (b) Carrell, T. G.; Cohen, S.; Dismukes, G. C. J. Mol.
Catal. A: Chem. 2002, 187, 3−15.
(12) (a) Kang, B.; Kim, M.; Lee, J.; Do, Y.; Chang, S. J. Org. Chem.
2006, 71, 6721−6727. (b) Kilic, H.; Adam, W.; Alsters, P. L. J. Org.
Chem. 2009, 74, 1135−1140.
(13) Barker, J. E.; Ren, T. Tetrahedron. Lett. 2004, 45, 4681−4683.
(14) Bennur, T. H.; Sabne, S.; Deshpande, S. S.; Srinivas, D.;
Sivasanker, S. J. Mol. Catal. A: Chem. 2002, 185, 71−80.
(15) (a) Knops-Gerrits, P.-P.; De Vos, D.; Thibault-Starzyk, F.;
Jacobs, P. A. Nature 1994, 369, 543−546. (b) Sabater, M. J.; Corma,
A.; Domenech, A.; Fornes
́
, V.; García, H. Chem. Commun. 1997,
1285−1286. (c) Alcon
́
, M. J.; Corma, A.; Iglesias, M.; Sanchez, F. J.
Mol. Catal. A: Chem. 2002, 178, 253−266. (d) Chavan, S. A.; Srinivas,
D.; Ratnasamy, P. J. Catal. 2002, 212, 39−45.
́
(16) (a) Serrano, D. P.; Aguado, J.; Vargas, C. Appl. Catal., A 2008,
335, 172−179. (b) Saikia, L.; Srinivas, D. Catal. Today 2009, 141, 66−
71.
̌
(31) Newville, M.; Lıviņs, P.; Yacoby, Y.; Rher, J. J.; Stern, E. A. Phys.
̅
Rev. B 1993, 47, 14126−14131.
(17) (a) Kureshy, R. I.; Khan, N. H.; Abdi, S. H. R.; Ahmad, I.; Singh,
(32) Ankudinov, A. L.; Ravel, B.; Rehr, J. J.; Conradson, S. D. Phys.
Rev. B 1998, 58, 7565−7576.
S.; Jasra, R. V. J. Catal. 2004, 221, 234−240. (b) Kuzniarska-Biernacka,
́
I.; Silva, A. R.; Ferreira, R.; Carvalho, A. P.; Pires, J.; de Carvalho, M.
(33) (a) Yamada, T.; Imagawa, K.; Mukaiyama, T. Chem. Lett. 1992,
2109−2112. (b) Haber, J.; Mlodnicka, T.; Poltowicz, J. J. Mol. Catal.
B.; Freire, C.; de Castro, B. New. J. Chem. 2004, 28, 853−858.
2029
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030

ACS Catalysis
Research Article
1989, 54, 451−461. (c) Rodgers, K. R.; Arafa, I. M.; Goff, H. M. J.
Chem. Soc., Chem. Commun. 1990, 1323−1324. (d) Yamada, T.; Takai,
T.; Rhode, O.; Mukaiyama, T. Chem. Lett. 1991, 1−4. (e) Takai, T.;
Hata, E.; Yamada, T.; Mukaiyama, T. Bull. Chem. Soc. Jpn. 1991, 64,
2513−2518. (f) Murahashi, S.-I.; Oda, Y.; Naota, T. J. Am. Chem. Soc.
1992, 114, 7913−7914. (g) Mukaiyama, T.; Yamada, T.; Nagata, T.;
Imagawa, K. Chem. Lett. 1993, 327−330. (h) Imagawa, K.; Nagata, T.;
Yamada, T.; Mukaiyama, T. Chem. Lett. 1994, 527−530. (i) Kaneda,
K.; Ueno, S.; Imanaka, T. J. Mol. Catal. A: Chem. 1995, 102, 135−138.
(34) (a) Tada, M.; Sasaki, T.; Iwasawa, Y. Phys. Chem. Chem. Phys.
2002, 4, 4561−4574. (b) Tada, M.; Sasaki, T.; Iwasawa, Y. J. Catal.
2002, 211, 496−510. (c) Tada, M.; Sasaki, T.; Shido, T.; Iwasawa, Y.
Phys. Chem. Chem. Phys. 2002, 4, 5899−5909. (d) Tada, M.; Sasaki, T.;
Iwasawa, Y. J. Phys. Chem. B 2004, 108, 2918−2930. (e) Yang, Y.;
Weng, Z.; Muratsugu, S.; Ishiguro, N.; Ohkoshi, S.; Tada, M. Chem.
Eur. J. 2012, 18, 1142−1153. (f) Muratsugu, S.; Tada, M. Acc. Chem.
Res. 2013, 46, 300−311.
(35) (a) Sauer, K.; Yachandra, V. K.; Britt, R. D.; Klein, M. P. In
Manganese Redox Enzymes; Pecoraro, V. L., Ed.; VCH: New York,
1992; Chapter 8. (b) Penner-Hahn, J. E.; Fronko, R. M.; Pecoraro, V.
L.; Yocum, C. F.; Betts, S. D.; Bowlby, N. R. J. Am. Chem. Soc. 1990,
112, 2549−2557. (c) Stemmler, T. L.; Sossong, T. M.; Goldstein, J. I.;
Ash, D. E.; Elgren, T. E.; Kurts, D. M., Jr.; Penner-Hahn, J. E.
Biochemistry 1997, 36, 9847−9858. (d) Kuzek, D.; Pace, R. J. Biochim.
́
Biophys. Acta 2001, 1503, 123−137. (e) Fernandez, G.; Corbella, M.;
Alfonso, M.; Stoeckli-Evans, H.; Castro, I. Inorg. Chem. 2004, 43,
6684−6698.
(36) (a) Phan, N. T. S.; van der Sluys, M.; Jones, C. W. Adv. Synth.
Catal. 2006, 348, 609−679. (b) Arends, I. W. C. E.; Sheldon, R. A.
Appl. Catal., A 2001, 212, 175−187.
2030
dx.doi.org/10.1021/cs400053f | ACS Catal. 2013, 3, 2020−2030