Selective Hydrogenation of Crotonaldehyde to Crotyl Alcohol over Metal Oxide Modified Ir Catalysts and Mechanistic Insight

Article abstract of DOI:10.1021/acscatal.6b00400

The scope of metal oxide modified noble metal (M+M′O_x) catalysts was scrutinized in the hydrogenation of crotonaldehyde to crotyl alcohol as a model reaction under mild reaction conditions (303 K, 0.8 MPa, water solvent), demonstrating that MoO_x, WO_x, NbO_x, FeO_x and ReO_x are effective metal oxides for Ir/SiO₂ to enhance both the activity and selectivity, although the optimized (metal oxide)/(Ir metal) molar ratio depends on the metal oxide. MoO_x modified Ir/SiO₂ catalyst (Ir-MoO_x/SiO₂ (Mo/Ir = 1)) was the most efficient, providing a high yield of crotyl alcohol (90%) and a high TOF (217 h^-1). The catalytic activity under such mild reaction conditions is the highest among the reported heterogeneous catalysts. These results showed that modification of active metals with an appropriate amount of metal oxides is an effective method for the development of efficient catalysts for selective hydrogenations. The reaction mechanism over the metal oxide modified Ir catalysts was investigated using Ir-ReO_x/SiO₂ (Re/Ir = 1) as a model catalyst by means of FTIR studies on H₂/D₂ adsorption, crotonaldehyde adsorption, and temperature-programmed desorption of crotonaldehyde, and kinetic studies on effects of H₂ pressure and crotonaldehyde concentration, isotopic effect of hydrogen (V_H2/V_D2), and comparison of reactivities between the aldehyde group and olefin group using various substrates. The reaction proceeds via four steps: (i) adsorption of crotonaldehyde on ReO_x species, (ii) generation of hydride species from H₂ on Ir metal species, (iii) hydride attack to the crotonaldehyde adspecies, and (iv) desorption of the produced crotyl alcohol, and the third step is the rate-determining step. Ir metal plays a role in the generation of hydride (H^-) species from H₂, leading to the high selectivity to crotyl alcohol, and ReO_x plays a role in promotion of crotonaldehyde adsorption, leading to the proximity of crotonaldehyde to the active site and activation of the aldehyde group, which results in high activity and further improvement in the selectivity.

Full text of DOI:10.1021/acscatal.6b00400

Research Article
pubs.acs.org/acscatalysis
Selective Hydrogenation of Crotonaldehyde to Crotyl Alcohol over
Metal Oxide Modified Ir Catalysts and Mechanistic Insight
Masazumi Tamura,* Kensuke Tokonami, Yoshinao Nakagawa, and Keiichi Tomishige*
Department of Applied Chemistry, Graduate School of Engineering, Tohoku University, 6-6-07 Aoba, Aramaki, Aoba-ku, Sendai
980-8579, Japan
*
S Supporting Information
ABSTRACT: The scope of metal oxide modified noble metal (M
+
M′O ) catalysts was scrutinized in the hydrogenation of
x
crotonaldehyde to crotyl alcohol as a model reaction under mild
reaction conditions (303 K, 0.8 MPa, water solvent), demonstrating
that MoO , WO , NbO , FeO and ReO are effective metal oxides for
x
x
x
x
x
Ir/SiO to enhance both the activity and selectivity, although the
2
optimized (metal oxide)/(Ir metal) molar ratio depends on the metal
oxide. MoO modified Ir/SiO catalyst (Ir-MoO /SiO (Mo/Ir = 1))
x
2
x
2
was the most efficient, providing a high yield of crotyl alcohol (90%)
−1
and a high TOF (217 h ). The catalytic activity under such mild
reaction conditions is the highest among the reported heterogeneous
catalysts. These results showed that modification of active metals with
an appropriate amount of metal oxides is an effective method for the
development of efficient catalysts for selective hydrogenations. The
reaction mechanism over the metal oxide modified Ir catalysts was investigated using Ir-ReO /SiO (Re/Ir = 1) as a model
x
2
catalyst by means of FTIR studies on H /D adsorption, crotonaldehyde adsorption, and temperature-programmed desorption of
2
2
crotonaldehyde, and kinetic studies on effects of H pressure and crotonaldehyde concentration, isotopic effect of hydrogen
2
(
V /V ), and comparison of reactivities between the aldehyde group and olefin group using various substrates. The reaction
H2 D2
proceeds via four steps: (i) adsorption of crotonaldehyde on ReO species, (ii) generation of hydride species from H on Ir metal
x
2
species, (iii) hydride attack to the crotonaldehyde adspecies, and (iv) desorption of the produced crotyl alcohol, and the third
−
step is the rate-determining step. Ir metal plays a role in the generation of hydride (H ) species from H , leading to the high
2
selectivity to crotyl alcohol, and ReO plays a role in promotion of crotonaldehyde adsorption, leading to the proximity of
x
crotonaldehyde to the active site and activation of the aldehyde group, which results in high activity and further improvement in
the selectivity.
KEYWORDS: selective hydrogenation, heterogeneous catalysis, iridium, unsaturated aldehyde, unsaturated alcohol
INTRODUCTION
the selective hydrogenation of aldehyde groups. It was also
■
reported that transfer hydrogenation using alcohol reagents can
Selective hydrogenation of unsaturated aldehydes to unsatu-
rated alcohols is a challenging reaction, since hydrogenation of
CC bonds is easier than that of CO bonds thermodynami-
3
be applied to the reaction. However, H is preferable as a
2
hydrogen source to metal hydride and alcohols because it is a
clean and cheap resource. Some homogeneous catalyst
1
cally and kinetically. Crotonaldehyde is often used as a model
4
substrate for unsaturated aldehydes because selective hydro-
genation of crotonaldehyde to crotyl alcohol is comparatively
difficult among the unsaturated aldehydes, due to the absence
of directing groups in the structure (Scheme 1). Convention-
systems using RuCl [(R)-binap](dmf) , [RuH(CO)-
2
n
+
(CH CN)(TPPMS) ][BF ], and [H Ir(phosphine) ] are
3
3
4
2
4
effective for the reaction because the hydrogenation reaction
can be caused by ionic hydrogen species, which are known to
be efficient for polarized double bonds such as carbonyl
2
ally, metal hydrides such as LiAlH and NaBH are used for
4
4
5
groups; however, heterogeneous catalysts are unfavorable for
Scheme 1. Hydrogenation of Crotonaldehyde
the reaction mainly because of the difficulty in generating the
ionic hydrogen species over the heterogeneous catalysts. In fact,
conventional heterogeneous hydrogenation catalysts such as
Received: February 7, 2016
Revised: April 6, 2016
©
XXXX American Chemical Society
3600
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
Research Article
Pd, Rh, Ni, and Pt/C provide butanal or 1-butanol in the
hydrogenation of crotonaldehyde (Scheme 1).
MnO were applied to the hydrogenation of crotonaldehyde to
crotyl alcohol (Table 1). The main products were crotyl
x
6
From environmental and economic viewpoints, heteroge-
neous catalysts are more preferable than metal hydrides and
homogeneous catalysts, and hence, heterogeneous catalysts for
Table 1. Hydrogenation of Crotonaldehyde over Various
Catalysts
a
7
the reaction have been intensely investigated. Effective
selectivity (%)
heterogeneous catalysts are summarized in Table S1 in the
Supporting Information. Among the reported heterogeneous
conversion
(%)
crotyl
alcohol
7a−r
entry
catalyst
butanal 1-butanol
catalysts for hydrogenation of crotonaldehyde,
Ag or Au
1
2
3
4
5
6
7
8
9
Ir-ReO /SiO2
43.3
78.9
14.0
10.1
1.4
95.0
94.2
87.8
85.0
69.3
38.5
3.4
1.0
1.3
4.1
supported heterogeneous catalysts are effective in providing
x
7a−d
x
Ir-MoO /SiO2
high yield and selectivity to crotyl alcohol,
and in particular,
x
Ir-WO /SiO2
8.7
2.0
AgNp@CeO nanocomposite catalyst achieved a high yield of
2
7
a
x
Ir-NbO /SiO2
10.3
30.8
41.2
0.0
9
8%. However, these catalysts suffer from high temperature
x
Ir-FeO /SiO2
0.0
(
≥333 K), high H pressure (≥2.0 MPa), use of organic
2
x
Ir-CoO /SiO2
46.3
<0.1
<0.1
10.9
19.9
solvent, and/or low catalytic activity per total metal (turnover
−1
Ir-VO /SiO2
number (TOF) < 10 h ). Recently, Pt Sn alloy or Ru−Ir
bimetallic catalysts were reported to be efficient for the reaction
to afford a high yield of crotyl alcohol; however, these catalysts
x
x
3
Ir-MnO /SiO2
8
Ir/SiO2
86.9
8.2
3.1
a
also suffer from high H2 pressure (≥2.0 MPa), high
Reaction conditions: crotonaldehyde 3.0 mmol, H O 3.0 g, catalyst
2
(
M′/Ir = 1) 50 mg, 303 K, 1 h, H 0.8 MPa.
temperature (≥333 K), use of organic solvent, and/or low
2
−1
activity (TOF < 10 h ). Modification of hydrogenation-active
metals with other metal species often decreases the original
alcohol, butanal, and 1-butanol, while a small amount of butane
was also produced as a byproduct. Ir-ReO /SiO , Ir-MoO /
7e,f,l,9
activity of the hydrogenation-active metals.
Therefore,
x
2
x
development of effective heterogeneous catalysts without loss
of the activity of the main hydrogenation-active metal species
under mild reaction conditions has been desirable.
We previously reported preliminary results on the selective
hydrogenation of unsaturated aldehydes to unsaturated
SiO , Ir-WO /SiO , and Ir-NbO /SiO catalysts showed high
2
x
2
x
2
selectivity to crotyl alcohol (85.0−95.0%), which is similar to or
higher than the selectivity over Ir/SiO (Table 1, entries 1−4
2
and 9); among these catalysts, Ir-MoO /SiO provided the
x
2
highest conversion of 78.9% (Table 1, entry 2) with a high
alcohols, where ReO -modified Ir/SiO (Ir-ReO /SiO ) catalyst
x
2
x
2
selectivity of 94.2%. Ir-FeO /SiO2 showed a moderate
x
is effective for the reaction, providing a high yield of various
selectivity of 69.3%, although the conversion was very low
(Table 1, entry 5). Ir-CoO /SiO (Table 1, entry 6) showed
1
0
unsaturated alcohols (70−99%).
Modification of ReOx
x
2
species on Ir metal dramatically enhanced both the activity
and selectivity, which can make the reaction conditions milder
much lower selectivity to crotyl alcohol than Ir/SiO . Ir-VO /
2
x
SiO and Ir-MnO /SiO showed almost no activity (Table 1,
2
x
2
such as a low temperature of 303 K, a low H pressure of 0.8
2
entries 7 and 8). These results indicate that CoO , VO , and
x x
MPa, and water solvent. The additive effect on Ir-ReO /SiO is
x
2
MnO are unfavorable modifiers for the selective hydrogenation
x
quite characteristic, considering that the hydrogenation activity
of crotonaldehyde to crotyl alcohol. Therefore, modification of
Ir metal with the appropriate metal oxide species drastically
of metal species in other catalyst systems is often sacrificed to
7e,f,l,9
obtain high selectivity.
On the other hand, it has been
enhanced the activity of Ir/SiO catalyst and improved the
2
reported that metal oxide (M′O ) modified noble metal (M)
x
selectivity over Ir/SiO catalyst. In terms of the selectivity to
2
catalyst (M+M′O catalyst) systems, including Ir-ReO /SiO
x
x
2
crotyl alcohol (>60%), Ir-ReO /SiO , Ir-MoO /SiO , Ir-WO /
x
2
x
2
x
and Ir-MoO /SiO , are efficient for the hydrogenolysis of
x
2
SiO , Ir-NbO /SiO , and Ir-FeO /SiO catalysts were selected
2 x 2 x 2
1
1
12
13
polyols or ethers, hydrogenation of carboxylic acids, and
as catalyst candidates for the selective hydrogenation of
unsaturated aldehydes.
1
4
hydrodeoxygeneation of diols, where the activity and
selectivity were often simultaneously improved in comparison
with only noble-metal catalysts without metal oxide mod-
ification. Therefore, modification of noble metals with metal
oxides will be a promising strategy to realize both high
selectivity and high activity in selective hydrogenation of
unsaturated aldehydes to unsaturated alcohols, and clarification
of the reaction mechanism will provide a clue for development
of effective heterogeneous catalysts for selective hydrogenation
reactions.
15
According to previous literature, the optimized molar ratio
of M′O /M ((metal oxide)/(noble metal)) for achieving high
x
selectivity and activity strongly depends on the metal oxide
modifiers. The effect of M′/Ir atomic ratio was investigated
using Ir-MoO /SiO , Ir-WO /SiO , Ir-NbO /SiO , and Ir-
x
2
x
2
x
2
FeO /SiO catalysts (Table 2). The optimized Re/Ir atomic
x
2
ratio of Ir-ReO /SiO catalyst was reported to be 1 in our
x
2
10
previous work (Table 2, entry 21). Ir-MoO /SiO catalysts
x
2
showed the maximum conversion at an Mo/Ir atomic ratio of 1
(78.9%, Table 2, entry 5), and the atomic ratio of M′/Ir is the
same as that of Ir-ReO /SiO . The selectivity was maintained
Herein, we investigate the scope of M+M′O catalysts in the
x
hydrogenation of crotonaldehyde to crotyl alcohol as a typical
model reaction, and a reaction mechanism is proposed on the
x
2
over 90% at any Mo/Ir ratios (Table 2, entries 1−7). On the
basis of kinetics and spectroscopy using Ir-ReO /SiO catalyst
x
2
other hand, Ir-WO /SiO , Ir-NbO /SiO , and Ir-FeO /SiO
x
2
x
2
x
2
as a model catalyst.
catalysts showed higher conversion and selectivity at a smaller
atomic ratio of M′/Ir, and the conversion and selectivity
RESULTS AND DISCUSSION
Catalyst Screening in the Selective Hydrogenation of
Crotonaldehyde. In order to select effective metal oxide
■
decreased with an increase in the amount of M′O species. At
x
the atomic ratio of M′/Ir = 0.1, Ir-WO /SiO , Ir-NbO /SiO ,
x
2
x
2
and Ir-FeO /SiO catalysts (Table 2, entries 8, 13, and 17)
x 2
modifiers, Ir/SiO catalysts modified with various metal oxides
showed high conversion and selectivity (conversion 31−50%,
2
such as ReO , MoO , WO , NbO , FeO , CoO , VO , and
selectivity 92−94%), which is similar to those over Ir-ReO /
x
x
x
x
x
x
x
x
3
601
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
Research Article
Table 2. Effect of M′/Ir Atomic Ratio in Hydrogenation of
a
Crotonaldehyde over Ir-M′O /SiO Catalysts
x
2
selectivity (%)
conversion
(%)
crotyl 1-
alcohol butanal butanol
entry
catalyst
M′/Ir
1
2
3
4
5
6
7
Ir-MoO /SiO2
0.1
0.3
0.5
0.8
1
35.0
46.8
57.4
62.2
78.9
52.0
45.2
92.7
94.6
94.1
95.9
94.2
92.8
93.8
3.1
2.1
1.7
1.0
1.0
3.3
2.7
3.2
2.8
3.4
2.7
4.1
3.4
2.5
x
1.3
2
8
9
Ir-WO /SiO2
0.1
1
31.0
14.0
17.5
17.6
11.2
94.3
87.8
84.1
82.1
77.8
2.8
8.7
2.2
2.0
0.0
2.2
0.0
x
1
1
1
0
1
2
3
12.3
11.8
15.9
5
7
1
1
1
1
3
4
5
6
Ir-NbO /SiO2
0.1
0.5
1
44.0
14.7
10.1
10.4
92.3
87.0
85.0
86.1
4.0
8.0
2.5
2.0
0.0
0.0
x
10.3
9.5
2
1
1
1
2
7
8
9
0
Ir-FeO /SiO2
0.1
0.5
1
49.8
6.8
93.7
89.3
69.3
50.8
2.3
7.4
3.8
2.4
0.0
0.0
x
1.4
30.8
49.2
2
1.2
2
1
Ir-ReO /SiO2
1
43.3
95.0
3.4
1.3
x
a
Reaction conditions: crotonaldehyde 3.0 mmol, H O 3.0 g, catalyst
2
5
0 mg, 303 K, 1 h, H 0.8 MPa.
2
Figure 1. Time courses in hydrogenation of crotonaldehyde: (a) Ir-
ReO /SiO (Re/Ir = 1); (b) Ir-MoO /SiO (Mo/Ir = 1); (c) Ir-FeO /
x
2
x
2
x
SiO (Fe/Ir = 0.1); (d) Ir-NbO /SiO (Nb/Ir = 0.1); (e) Ir-WO /
SiO (Re/Ir = 1) (Table 2, entry 21). The conversion was
2
x
2
x
2
SiO (W/Ir = 0.1). Reaction conditions: crotonaldehyde 3.0 mmol,
2
slightly lower than that over Ir-MoO /SiO (Mo/Ir = 1), while
x
2
H O 3.0 g, catalyst 50 mg, 303 K, H2 0.8 MPa. Legend: (●)
2
the selectivities were almost the same (Table 2, entries 5, 8, 13,
conversion; (○) selectivity to crotyl alcohol; (□) selectivity to butanal;
and 17). Among Ir-M′O /SiO catalysts, Ir-MoO /SiO (Mo/
x
2
x
2
(△
) selectivity to 1-butanol.
Ir = 1) catalyst showed about 2 times higher activity than the
other optimized Ir-M′O /SiO catalysts, and the TOF was
x
2
−1
calculated to be 227 h under mild reaction conditions such as
high crotyl alcohol yields of 90% (Figure 1a, 8 h) and 92%
(Figure 1b, 8 h), respectively. In contrast, the selectivities at
high conversion were below 90% in the case of the catalysts Ir-
FeO /SiO (Fe/Ir = 0.1) (Figure 1c, 5 h), Ir-NbO /SiO (Nb/
3
03 K, 0.8 MPa H , and water solvent. The TOF was calculated
2
−
1
by the following equation: TOF (h ) = (converted
crotonaldehyde (mmol))/((total Ir metal amount (mmol))-
x
2
x
2
(
reaction time (h))). In addition, TOF per the surface Ir metal
Ir = 0.1) (Figure 1d, 8 h), and Ir-WO /SiO (W/Ir = 0.1)
x 2
sites, TOFs, was calculated by using the adsorbed CO amount
(Figure 1e, 5 h). On the other hand, in the case of the catalyst
Ir-ReO /SiO (Re/Ir = 1) the selectivity to crotyl alcohol was
−1
(CO/(total Ir) = 0.07), giving a value of 3200 h . The catalytic
x
2
activity for the selective hydrogenation of crotonaldehyde is the
highest among the reported catalysts under such mild reaction
conditions (Table S1 in the Supporting Information).
drastically decreased after 100% conversion of crotonaldehyde
(Figure 1a, 24 h), which implies that the reactivity of crotyl
alcohol depends on the presence of crotonaldehyde. The
change in reactivity is discussed in the following section. We
As mentioned in the Introduction, overhydrogenation of the
produced unsaturated alcohols to saturated alcohols will
decrease the selectivity to unsaturated alcohols. In order to
obtain the maximum yield of crotyl alcohol, the time course in
the hydrogenation of crotonaldehyde was investigated using the
optimized catalysts: Ir-ReO /SiO (Re/Ir = 1), Ir-MoO /SiO
2
selected ReO as a model modifier in the following experi-
x
ments.
The support often influences the performance of catalysts
15a,16
due to its acid/base properties or redox properties.
The
effect of supports was investigated using various metal oxides as
x
2
x
(
Mo/Ir = 1), Ir-FeO /SiO (Fe/Ir = 0.1), Ir-NbO /SiO (Nb/
supports (Table 3). In comparison with Ir-ReO /SiO (Re/Ir =
x
2
x
2
x
2
Ir = 0.1), and Ir-WO /SiO (W/Ir = 0.1) (Figure 1). The
1), Ir-ReO /Nb O (Re/Ir = 1) (Table 3, entry 2) showed
x
2
x 2 5
reactions proceeded smoothly over all of the catalysts, and the
selectivities to crotyl alcohol gradually decreased with the
reaction time. A 90% selectivity to crotyl alcohol was
maintained over Ir-ReO /SiO (Re/Ir = 1) and Ir-MoO /
slightly higher conversion (54.2%) with high selectivity
(91.2%), which is slightly lower than the selectivity of Ir-
ReO /SiO (Re/Ir = 1) (Table 3, entry 1, 95.0%). Other
x
2
catalysts (Table 3, entries 3−9) showed comparable or lower
x
2
x
SiO (Mo/Ir = 1) catalysts even at >99% conversion, providing
activity and lower selectivity in comparison with Ir-ReO /SiO
2
2
x
3
602
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
Research Article
Table 3. Effect of Supports in Ir-ReO Catalysts for
Hydrogenation of Crotonaldehyde
state, and the valence of Re is about +3. Ir metal particles are
x
a
10
partially covered with ReO species. These results agree well
x
with the structure of Ir-ReO /SiO (Re/Ir = 1) catalyst in
x
2
selectivity (%)
18
hydrogenolysis of polyols. In addition, Ir-ReO /SiO (Re/Ir =
x
2
conversion
(%)
crotyl
1
) catalyst can be reused without loss of activity and selectivity,
entry
catalyst
alcohol
butanal 1-butanol
and the stability of the catalyst during the reaction was also
confirmed by various analyses such as XRD, BET, and CO
adsorption using Ir-ReO /SiO (Re/Ir = 1) before and after the
reaction. ICP analysis and the leaching test demonstrated that
Ir-ReO /SiO (Re/Ir = 1) works as a heterogeneous catalyst.
x 2
1
2
3
4
5
6
7
8
9
Ir-ReO /SiO2
43.3
54.2
43.2
38.7
25.2
20.8
12.8
11.9
7.4
95.0
91.2
85.3
89.6
82.4
86.1
83.4
69.0
70.6
3.4
4.2
1.3
4.2
6.5
4.9
5.2
6.2
3.4
4.9
5.0
x
Ir-ReO /Nb O
5
x
2
x
2
10
Ir-ReO /TiO2
7.4
x
10
Ir-ReO /ZrO2
4.7
x
Ir-ReO /Al O
12.0
6.7
x
2
3
Ir-MoO /SiO (Mo/Ir = 1) catalyst after the reduction and
x 2
x
Ir-ReO /CeO2
reaction was also characterized by XRD (Figure S1 in the
Supporting Information), TEM (Figure S2 in the Supporting
Information), and XAS (Figures S3−S5 in the Supporting
Information and Tables 3 and 4) analyses. XRD analysis of Ir-
Ir-ReO /Fe O
3
12.5
25.4
23.7
x
2
Ir-ReO /Y O
3
x
2
Ir-ReO /MgO
x
a
Reaction conditions: crotonaldehyde 3.0 mmol, H O 3.0 g, catalyst
2
(
Re/Ir = 1) 50 mg, 303 K, 1 h, H 0.8 MPa.
a
2
Table 4. Comparison of Reactivities of Various Substrates
(
Re/Ir = 1). In particular, strong base metal oxides such as
CeO , Y O , and MgO showed low conversion and selectivity
2
2
3
(
Table 3, entries 6, 8, and 9), indicating that base sites will be
unfavorable for the reaction. This result can be related to the
previous result, where a basic additive is unfavorable for Ir-
ReO /SiO (Re/Ir = 1) catalyst in the hydrogenolysis of
x
2
1
7
glycerol. MO species will not be able to approach the Ir
x
metal surface on basic supports because MO species can be
easily adsorbed on the supports, which decreases the active
sites, the interface between Ir metal and metal oxide species
a
x
Reaction conditions: substrate 3.0 mmol, H O 3.0 g, Ir-ReO /SiO
2 x 2
b
(Re/Ir = 1) 50 mg, 303 K, H 0.8 MPa. TOF was calculated by the
2
−1
following equation: TOF (h ) = (converted substrate (mmol))/
((total Ir metal amount (mmol))(reaction time (h))). The
c
(discussed in the following sections). In addition, the basic sites
of the supports can stabilize IrO to suppress the formation of
calculation of the TOF was tentative because of the high conversion
of substrate.
2
Ir metal, which will also decrease the interface sites. Therefore,
supports without basic sites are preferable for the reaction.
In our previous work on the selective hydrogenation of
MoO /SiO (Mo/Ir = 1) after the reaction showed the signal
x
2
10
unsaturated aldehydes, carbon supported noble-metal cata-
lysts (Pt/C, Ru/C, Rh/C, Pd/C) and M-ReO /SiO (M = Pt,
due to Ir metal, and the particle size of Ir metal was estimated
to be 2.5 nm (Figure S1 in the Supporting Information). On
the other hand, no signals due to Mo species were observed,
indicating that Mo species were highly dispersed on the
catalyst. TEM analysis of Ir-MoO /SiO (Mo/Ir = 1) provided
x
2
Ru, Rh, Pd; Re/M = 0.5) had been applied to the reaction at a
high H2 pressure of 8 MPa, giving low selectivities to
crotonaldehyde of <10% and <35%, respectively. In order to
ascertain that Ir metal is a truly effective metal species for the
x
2
a particle size of 3.3 nm, which is assigned to Ir metal (Figure
reaction, SiO2 supported ReO modified noble-metal (M-
S2 in the Supporting Information). Ir L -edge EXAFS of Ir-
x
3
ReO /SiO , M = Pt, Ru, Rh, Pd) catalysts with various Re/M
MoO /SiO (Mo/Ir = 1) after the reduction and reaction
x
2
x
2
atomic ratios (0.1, 0.5, 1) were applied to the hydrogenation of
showed that Ir was in the metallic state and that the Ir−Ir CN
of ∼10.5 was smaller than 12 (Figure S3 and Table S3 in the
Supporting Information), which is due to the small Ir metal
particles as shown in TEM analysis (Figure S2 in the
Supporting Information). Mo K-edge XANES analysis of Ir-
MoO /SiO (Mo/Ir = 1) after the reduction and reaction
unsaturated aldehydes at a low H pressure of 0.8 MPa (Table
2
S2 in the Supporting Information), because the amount of
metal oxide species drastically influences the selectivity and
activity as discussed above. M-ReO /SiO catalysts showed
x
2
much lower selectivity to crotyl alcohol (≤30%) than Ir-ReO /
x
x
2
SiO (Re/Ir = 1) catalyst, although the selectivity was slightly
showed that the shape of the spectra of Ir-MoO /SiO (Mo/Ir
2
x 2
improved in Ru-ReO /SiO catalysts in comparison with Ru/C.
= 1) seems to be close to that of MoO , indicating that Mo
x
2
2
These results indicate that Ir metal is an effective metal species
for the selective hydrogenation reaction among the tested metal
species.
species of Ir-MoO /SiO (Mo/Ir = 1) were partially reduced
x
2
(Figure S4 in the Supporting Information). Mo K-edge EXAFS
analysis of Ir-MoO /SiO (Mo/Ir = 1) after the reduction and
x
2
From the above results, it can be concluded that Ir metal is
intrinsically effective for the reaction and that modification of Ir
metal with metal oxides is an efficient strategy to obtain high
activity and selectivity at the same time.
reaction provided the presence of Mo−O and Mo−Ir bonds
(Figure S5 and Table S4 in the Supporting Information). The
presence of a Mo−O bond indicates that Mo species were not
fully reduced to the metallic state, which is also supported by
the result of Mo K-edge XANES analysis (Figure S4 in the
Supporting Information). In addition, the state of Ir and MoOx
species in Ir-MoO /SiO (Mo/Ir = 1) was not changed before
Catalyst Structure and Stability. The catalyst Ir-ReOx/
SiO (Re/Ir = 1) had been characterized by XRD, XAFS, TEM,
2
and CO adsorption analyses in our previous work on the
x
2
10
hydrogenation of crotonaldehyde. The state of Ir and ReOx
species and the structure of Ir-ReO /SiO (Re/Ir = 1) have
and after the reaction in XAS analyses, which suggests that Ir-
MoO /SiO2 (Mo/Ir = 1) was stable under the reaction
x
2
x
been proposed as follows: Ir is in the metallic state, and the
conditions. In order to estimate the surface Ir metal amount,
CO adsorption was also carried out using Ir-MoO /SiO (Mo/
particle size is about 2−3 nm. ReO species are in the oxidized
x
x
2
3
603
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
Research Article
Ir = 1). The adsorption amount of CO was calculated to be
(Table 4, entries 3 and 4). On the other hand, 3-heptene, a
simple internal alkene, reacted to afford n-heptane quantita-
tively, but the reactivity is slightly lower than that of 2-heptenal
(Table 4, entries 3 and 5). These results indicate that the
catalyst Ir-ReO /SiO (Re/Ir = 1) originally showed higher
14.5 μmol/(g of cat.), which is about half as small as that of Ir/
SiO (31.4 μmol/(g of cat.)). Considering that the Ir particle
2
18b
size of Ir/SiO (3.6 nm ) is almost the same as that of Ir-
2
MoO /SiO (Mo/Ir = 1), the Ir metal surface will be partially
x
2
x
2
covered by MoO species. From these results, the state of Ir
activity to the aldehyde group than the olefin group. The
reactivity ratio of the simple aldehyde group to the simple
olefin group is calculated to be about 3 (Table 4, entries 4 and
5), which is much lower than that in the case of unsaturated
aldehydes as discussed above (about 15-fold). These results
indicate that the reactivity of the olefin group in α,β-
unsaturated aldehydes drastically decreased by conjugation
between the aldehyde group and olefin group. In general,
electron-withdrawing groups at the α-position of the olefin
group weakens the coordination of the olefin to metal species
by reduction of the electron density in the olefin group via the
conjugation between the aldehyde group and olefin group,
which may be related to a decrease in the reactivity of the olefin
group. In addition, the adsorption state of crotonaldehyde can
also influence the reactivity of these functional groups. If the
adsorption of the aldehyde group on the catalyst Ir-ReO /SiO
x
metal over Ir-MoO /SiO (Mo/Ir = 1) is similar to that of Ir-
x
2
1
0,18
ReO /SiO (Re/Ir = 1),
and the structure is also similar to
x
2
that of Ir-ReO /SiO (Re/Ir = 1) from the viewpoint that the Ir
x
2
metal surface was partially covered by the metal oxide species.
With regard to other catalysts, catalysts similar to Ir-NbOx/
SiO , Ir-FeO /SiO , and Ir-VO /SiO were reported in
2
x
2
x
2
hydrogenation or hydrogenolysis reactions such as Ir/NbOx-
SiO , (Ir 1 wt %, Nb/Ir = 2.3, reduction 473 K, 1 h) in
2
19
hydrogenation of citral, Ir-FeO /SiO (Ir 3 wt %, Fe/Ir = 5,
reduction 573 K, 1 h) in hydrogenation of cinnamaldehyde
and Ir-VO /SiO (Ir 4 wt %, V/Ir = 0.2, reduction 473 K, 2 h)
x
2
2
0
x
2
12a
in hydrogenolysis of tetrahydrofurfuryl alcohol.
In these
reports, Ir was in the metallic state and the particle size of Ir
metal was about 3−5 nm. The detailed structure formed from
Ir metal and metal oxide species was not described, although
the interface between Ir metal and metal oxide species was
x
2
(
Re/Ir = 1) is stronger than that of the olefin group,
12a,20
proposed to be active sites by part of these reports.
addition, we previously reported that the reaction behavior of
Ir-WO /SiO is similar to that of Ir-MoO /SiO in hydro-
In
hydrogenation of the aldehyde group will be more preferred
to that of the olefin group. The adsorption of these functional
groups is discussed later.
x
2
x
2
18b
genolysis of glycerol, which suggests that the state of Ir metal
of Ir-WO /SiO will be similar to that of Ir-MoO /SiO . On
From the time course of the reaction over Ir-ReO /SiO
x
2
x
2
x
2
(Re/Ir = 1), the high selectivity (>90%) of crotyl alcohol was
maintained even at high conversion (99%), but the selectivity
drastically decreased after the conversion reached 100%. The
consideration of these results and facts, the state of Ir metal will
be almost the same in these M+M′O /SiO catalysts, although
x
2
the state of metal oxides depends on the properties of the metal
oxides. Therefore, the state of metal oxides, particularly the
reducibility of metal oxides, will be a key factor in the formation
of the active sites. The relationship between the reducibility of
metal oxides and the formation of the active sites will be
discussed in the last section.
−1
TOF in the case of crotyl alcohol as a substrate is about 48 h
(
Table 4, entry 6), which is half or one-third lower than that of
α,β-unsaturated aldehydes (Table 4, entries 1−3). These results
cannot explain the high selectivity of crotyl alcohol (≥90%) at
high conversion. To elucidate the discrepancy in the reactivities,
a reaction using both crotonaldehyde and crotyl alcohol was
carried out (Scheme 2, third row). The conversion of crotyl
alcohol was calculated using the difference between the
produced amount of 1-butanol (0.07 mmol) and the expected
amount of 1-butanol (0.02 mmol) from the result of
hydrogenation of crotonaldehyde (Scheme 2, the first row).
The details of the calculation method for conversion are
described in the Supporting Information. The crotonaldehyde
We selected Ir-ReO /SiO (Re/Ir = 1) as a model catalyst in
x
2
the following experiments for clarification of the reaction
mechanism.
Kinetic Studies. From the time course of hydrogenation of
crotonaldehyde over Ir-ReO /SiO (Re/Ir = 1) catalyst (Figure
x
2
1a), two competitive reactions take place (Scheme 1): (i)
hydrogenation to crotyl alcohol and (ii) hydrogenation to
butanal. At the initial stage of the reaction over Ir-ReO /SiO2
x
(
Re/Ir = 1) (1 h, Figure 1a), the crotyl alcohol/crotonaldehyde
Scheme 2. Reactivity Comparison of Crotonaldehyde Alone,
Crotyl Alcohol Alone, and Both Crotonaldehyde and Crotyl
Alcohol
selectivity ratio is 15 (95%/5%), meaning that the reactivity of
the aldehyde group in crotonaldehyde is about 15-fold higher
than that of the olefin group, because 1-butanol, an over-
hydrogenated product of crotyl alcohol or butanal, was almost
not observed. In order to ascertain whether the difference in the
reactivity between the aldehyde group and olefin group in
crotonaldehyde is derived from the original reactivities of an
aldehyde group and olefin group, reactions using a simple
aldehyde and olefin as model substrates were carried out (Table
a
4). Since it is difficult to handle butanal or butene owing to
their low boiling points, model compounds with higher boiling
points such as heptanal and 3-heptene were selected to ensure
the reliability of the data (Table 4, entries 4 and 5). At first, the
effect of the main alkyl chain length in α,β-unsaturated
aldehydes was investigated (Table 4, entries 1−3). The
influence of the alkyl chain length on the reactivity is very
small, and the selectivities were also similar. Heptanal, a simple
aldehyde, reacted to afford 1-heptanol with high selectivity, and
the reactivity is about 2 times higher than that of 2-heptenal
a
Reaction conditions: substrate, H O 3.0 g, Ir-ReO /SiO (Re/Ir = 1)
2
x
2
50 mg, 303 K, H 0.8 MPa, 1 h. (*) The detailed calculation method is
described in the Supporting Information.
2
3
604
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
Research Article
conversion was almost the same as that in the case of
crotonaldehyde alone (Scheme 2, first row), but the conversion
of crotyl alcohol was only 2%, which was much lower than that
in the case of hydrogenation of crotyl alcohol alone (Scheme 2,
second row). This result indicates that the hydrogenation of
crotonaldehyde preferably took place and that hydrogenation of
crotyl alcohol was suppressed, which result is in good
accordance with the high selectivity of crotyl alcohol in
hydrogenation of crotonaldehyde. In addition, a reaction with
both crotonaldehyde and pentanal was also carried out
influence on activation of H under the reaction conditions. In
addition, the isotopic effect was studied in the hydrogenation of
2
crotonaldehyde over Ir-ReO /SiO (Re/Ir = 1) catalyst using
x
2
H and D (Figure 3). The reaction rate with H was higher
2
2
2
(Scheme 2, fourth row). The crotonaldehyde conversion is
slightly lower than that of crotonaldehyde alone, the conversion
of pentanal is lower than that expected, and the TOF was
−1
calculated to be 130 h , which is much lower than that in the
case of hydrogenation of a saturated aldehyde (about 1/3-fold,
Table 4, entry 4). This result indicates that the hydrogenation
of crotonaldehyde is more preferable to that of pentanal. From
these results, the coexistence of crotonaldehyde drastically
suppressed the hydrogenation of crotyl alcohol and butanal to
1
-butanol in the hydrogenation of crotonaldehyde, although the
Figure 3. Isotopic effect of hydrogenation of crotonaldehyde with H2
reactivity of crotonaldehyde was nearly maintained even in the
presence of other compounds, which probably suggests that
crotonaldehyde can preferentially access the surface of the
catalyst.
and D . Reaction conditions: crotonaldehyde 3.0 mmol, H O 3.0 g, Ir-
2
2
ReO /SiO (Re/Ir = 1) 50 mg, 303 K, H or D 0.8 MPa.
x
2
2
2
than that with D , providing the reaction rate ratio V /V
D2
≈
2
H2
The effect of the substrate concentration and H pressure
2
1
.5. This also indicates that a hydrogen species is involved in
was investigated in the hydrogenation of crotonaldehyde at 303
K using Ir-ReO /SiO (Re/Ir = 1) and Ir/SiO catalysts (Figure
the rate-determining step, which agrees with the result of the
x
2
2
effect of H pressure (Figure 2b). According to previous works
2
2). The reaction orders with respect to the crotonaldehyde
on hydrogenolysis of polyols over Ir-ReO /SiO (Re/Ir =
x
2
12e,18b
1
),
the reaction order of about +1 with respect to H2
pressure means that H is heterolytically dissociated to generate
2
+
−
H and H species on Ir-ReO /SiO (Re/Ir = 1) and the
nucleophilic addition of the active H will be the rate-
determining step.
FTIR Studies. In order to confirm the formation of H and
H species from H over Ir-ReO /SiO (Re/Ir = 1), adsorption
x
2
−
−
+
2
x
2
of H and D on Ir-ReO /SiO (Re/Ir = 1) after reduction was
2
2
x
2
examined by in situ FTIR (Figure 4). The large peaks around
Figure 2. (a) Crotonaldehyde concentration dependence and (b) H2
pressure dependence in hydrogenation of crotonaldehyde over Ir/SiO2
and Ir-ReO /SiO (Re/Ir = 1). Reaction conditions: (a) crotonalde-
x
2
hyde 1.0−5.0 mmol, H O 3.0 g, Ir-ReO /SiO (Re/Ir = 1) 50 mg, 303
2
x
2
K, H 0.8 MPa; (b) crotonaldehyde 3 mmol, H O 3.0 g, Ir-ReO /
2
2
x
SiO (Re/Ir = 1) 50 mg, 303 K, H 0.8−8 MPa. Detailed results for (a)
2
2
are shown in Tables S5 and S6 in the Supporting Information, and
those for (b) are shown in Tables S7 and S8 in the Supporting
Information.
concentration over Ir-ReO /SiO (Re/Ir = 1) and Ir/SiO
2
x
2
catalysts were calculated to be 0.0 and +0.6, respectively
Figure 2a), which indicates that the addition of ReO species
decreases the reaction order and that the adsorption of
(
x
Figure 4. FTIR spectra of (a) H and (b) D adsorption on Ir-ReO /
2
2
x
SiO (Re/Ir = 1) at 7.0 kPa of H or D pressure and 303 K.
2
2
2
crotonaldehyde is saturated on the surface of Ir-ReO /SiO2
x
(
Re/Ir = 1) under the reaction conditions. With regard to the
3745 cm− are due to the free silanol group of SiO . FTIR
1
effect of H pressure (Figure 2b), the reaction orders with
2
2
respect to H pressure over Ir-ReO /SiO (Re/Ir = 1) and Ir/
spectra of H and D adsorption on the reduced Ir-ReO /SiO
(Re/Ir = 1) catalyst showed bands at 3676 and 3499 cm due
2
x
2
2
2
x
2
−
1
SiO catalysts were calculated to be +0.9 and +0.8, respectively,
2
which indicates that a hydrogen species is involved in the rate-
determining step. The reaction orders over Ir-ReO /SiO (Re/
to the ν(OH) species (Figure 4a) and bands at 2717 and 2551
−1
cm due to the ν(OD) species (Figure 4b), respectively. Ir−H
x
2
Ir = 1) and Ir/SiO are almost the same, which probably
and Ir−D species cannot be detected clearly because of the
2
suggests that the addition of ReO species has almost no
change in the SiO absorption. The position of the observed
x
2
3
605
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
Research Article
−
1
bands due to the ν(OD) species are in good accordance with
crotonaldehyde (∼40 cm ), which indicates that crotonalde-
−1
−1
the calculated positions (2675 and 2546 cm ) from the 3676
hyde was adsorbed on SiO . The last band (1642 cm ) is due
2
and 3499 cm− bands due to the ν(OH) species. Therefore, H
1
to ν(CC), which was only slightly shifted to a smaller
2
+
−
−1
will be heterolytically dissociated to produce H and H species
on Ir-ReO /SiO (Re/Ir = 1), which supports the reaction
wavenumber (∼5 cm ) from that of gaseous crotonaldehyde.
These results indicate that crotonaldehyde is adsorbed on SiO
x
2
2
order with respect to H pressure (about +1, Figure 2b).
at the carbonyl group. On the other hand, Ir-ReO /SiO (Re/Ir
2
x
2
Taking into consideration that ionic hydrogenation using
= 1), ReO /SiO , and Ir/SiO showed similar bands around
x 2 2
−
−1
hydride (H ) species is generally effective for hydrogenation of
1682 ± 1, 1669 ± 1, and 1642 ± 1 cm (Figure 5b−d). The
first two bands can be assigned to ν(CO) of crotonaldehyde
5
polar double bonds in the organic synthesis, the comparatively
−1
high selectivity to crotyl alcohol over Ir-ReO /SiO (Re/Ir = 1)
adsorbed on the catalyst, and the last band (1642 ± 1 cm )
can be assigned to ν(CC) of crotonaldehyde. The band
shapes of crotonaldehyde adspecies on Ir-ReO /SiO (Re/Ir =
x
2
+
(
H
95.0%) can be explained mainly by the generation of H and
−
species on the catalyst surface. The comparatively high
x
2
selectivity (86.9%) over Ir/SiO can also be explained by the
1), ReO /SiO , and Ir/SiO are different from those on SiO ,
indicating that crotonaldehyde can be adsorbed on the Ir metal
2
x 2 2 2
same interpretation, because the reaction order with respect to
H pressure over Ir/SiO was almost +1 (Figure 2b).
and ReO species at the aldehyde group, considering that the
2
2
x
As discussed in the comparison of reactivities of various
substrates (Table 4 and Scheme 2) and the effect of the
substrate concentration (Figure 2a), the adsorption state of
crotonaldehyde on the catalyst can largely influence the activity
and selectivity. The adsorption states of crotonaldehyde on Ir-
ReO /SiO (Re/Ir = 1), Ir/SiO , ReO /SiO , and SiO were
band shift due to ν(CO) is larger than that due to ν(CC).
The similar shapes of the ν(CO) bands on ReO /SiO and
x
2
Ir/SiO may be attributed to the similar adsorption states of the
2
−1
crotonaldehyde. In addition, in the region of 2700−3100 cm ,
the fingerprint region of the C−H stretching vibration, the
2
2
band positions of the aliphatic and olefinic ν(C−H) at 3049,
x
2
2
x
2
2
3007, 2984, 2955, and 2926 cm− were almost the same on all
1
studied at 303 K by in situ FTIR. Since crotonaldehyde can be
22
adsorbed on the SiO support, introduction of a large amount
catalysts, and the band positions of the aldehydic ν(C−H) at
2
2763 cm− on Ir-ReO /SiO (Re/Ir = 1), ReO /SiO , and Ir/
1
of crotonaldehyde is inappropriate for the assignment of
crotonaldehyde adspecies on the catalysts. Thus, a small
x
2
x
2
SiO showed higher wavenumber in comparison to those on
2
−3
amount of crotonaldehyde (∼0.2 μL, ∼2.9 × 10 mmol),
SiO2 (Figure S6 in the Supporting Information), which
which is adequately lower than the loaded metal amount of the
supports that crotonaldehyde is adsorbed not on SiO but on
2
−3
catalyst disk (∼25 mg, amount of Ir (Re) 5.2 × 10 mmol)
was adopted in the experiments to precisely estimate the
adsorption state of crotonaldehyde. Figure 5 shows the
Ir or ReO species at the aldehyde group.
x
In order to elucidate which site of Ir metal or ReO species
x
on Ir-ReO /SiO2 (Re/Ir = 1) is the adsorption site for
x
crotonaldehyde, crotonaldehyde adsorbed on all the catalysts
was desorbed by increasing the temperature from 303 to 373 K,
and the change in the bands was monitored by FTIR. Figure 6a
Figure 6. (a) Change in FTIR spectra of crotonaldehyde adspecies on
Ir-ReO /SiO (Re/Ir = 1) at 303−373 K (a, 303 K; b, 313 K; c, 323 K;
x
2
d, 333 K; e, 343 K; f, 353 K; g, 363 K; h, 373 K). (b) Band area change
of crotonaldehyde adspecies on (○) Ir-ReO /SiO (Re/Ir = 1), (□)
x
2
ReO /SiO , (◇) Ir/SiO , and (△) SiO . The band area (%) was
x
2
2
2
Figure 5. FTIR spectra of gaseous crotonaldehyde and crotonaldehyde
normalized by the area at 303 K.
adspecies on Ir-ReO /SiO (Re/Ir = 1), ReO /SiO , Ir/SiO , and SiO
x
2
x
2
2
2
−1
at 303 K in the region 1550−1800 cm : (a) gas phase; (b) Ir-ReO /
shows the band change of the crotonaldehyde adspecies on the
catalyst Ir-ReO /SiO (Re/Ir = 1) as a model case. The
x
SiO (Re/Ir = 1); (c) ReO /SiO ; (d) Ir/SiO ; (e) SiO .
2
x
2
2
2
x
2
intensity of the band gradually decreased with an increase in the
temperature, maintaining the shape of the band. This result
indicates that crotonaldehyde was desorbed by heating without
decomposition of crotonaldehyde. The change in the band area
crotonaldehyde adspecies on these catalysts in the region
−1
1
550−1800 cm , the fingerprint region of CO and CC
stretching vibrations of crotonaldehyde. Gaseous crotonalde-
−1
−1
hyde showed bands at 1711, 1724, and 1647 cm ; the first two
bands are due to ν(CO), and the last band is due to ν(C
in the range 1550−1750 cm on Ir-ReO /SiO (Re/Ir = 1),
x
2
ReO /SiO , Ir/SiO , and SiO was plotted as a function of
x
2
2
2
2
1
C) (Figure 5a). SiO showed new bands at 1673, 1689, and
temperature versus area percent of the corresponding band at
2
1
642 cm− (Figure 5e). The first two bands on SiO can be
1
303 K (Figure 6b). The band area on SiO decreased more
2
2
assigned to ν(CO), and the bands were largely shifted to
easily in comparison to those on Ir-ReO /SiO (Re/Ir = 1),
ReO /SiO , and Ir/SiO , indicating that the adsorption
x 2 2
x
2
lower wavenumbers in comparison to those of gaseous
3
606
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
Research Article
strength of crotonaldehyde on SiO is weaker than that on the
on ReO . (iv) The alkoxide adspecies is protonated to provide
2
x
other catalysts. Ir-ReO /SiO (Re/Ir = 1) and ReO /SiO
showed similar desorption profiles, and the adsorption strength
of the crotonaldehyde adspecies on these catalysts is larger than
crotyl alcohol. The rate-determining step is step iii. The high
selectivity can be explained by the formation of the hydride
x
2
x
2
species at the interface between Ir metal and ReO species and
x
that on Ir/SiO , which can be related to the slightly lower
activation of the aldehyde group on ReO species. Hydro-
2
x
wavenumbers of ν(CO) on Ir-ReO /SiO (Re/Ir = 1) and
x
2
genation by the hydride species refers to ionic hydrogenation.
The high activity can be explained by the activation of the
aldehyde group and effective proximity of crotonaldehyde to
the active hydride species on the catalyst by virtue of the
ReO /SiO in comparison to that on Ir/SiO , as shown in
x
2
2
Figure 5b−d. From these results, crotonaldehyde was adsorbed
more strongly on ReO species than on Ir metal, which is
x
supported by the results of the effect of the crotonaldehyde
concentration (Figure 2a). Therefore, crotonaldehyde will be
adsorbed on ReO species of Ir-ReO /SiO (Re/Ir = 1) catalyst
adsorption of crotonaldehyde on ReO species.
x
In the reaction mechanism, the active sites of Ir-MOx
catalysts will be the interface between Ir metal and metal
oxide species. Therefore, the catalytic performance will depend
on the state or amount of the interface, and considering that the
x
x
2
at the aldehyde group, and the aldehyde group will be more
activated on ReO than on Ir metal. As above, ReO species
x
x
work as the adsorption site of crotonaldehyde and strengthen
the adsorption of crotonaldehyde on the catalyst in comparison
states of Ir species in Ir-MO catalysts are almost the same, the
x
formation of the active interface sites can be changed by the
properties of the metal oxides. ReOx and MoOx are
comparatively reducible and tend to form partially reduced
metal oxide species, which favors the formation of the interface
between Ir metal and metal oxide species probably due to the
high mobility of ReO and MoO species. In contrast, other
with Ir/SiO . This strong interaction of crotonaldehyde with
2
ReO species will place the crotonaldehyde close to the active
x
hydride species on the catalyst efficiently, leading to the high
activity of Ir-ReO /SiO (Re/Ir = 1). In addition, activation of
x
2
crotonaldehyde on ReO species will increase the reactivity of
x
x
x
the aldehyde group, leading to improvement in selectivity to
crotyl alcohol (Ir/SiO 86.9% (Table 1, entry 9) to Ir-ReO /
metal oxides will not be reduced at all or will be reduced to the
metallic state. On the other hand, in the case of small amounts
of WO , FeO and NbO , high conversion and selectivity were
2
x
SiO (Re/Ir = 1) 95.0% (Table 1, entry 1)).
2
x
x
x
Reaction Mechanism and Active Sites of Ir-MO /SiO
x
2
obtained (Table 2) in spite of the low reducibility of these
metal oxides. At present, a detailed structure of these catalysts
has not been clarified. In the future, we will address the
characterization of these catalysts in order to clarify the role of
these metal oxides and the structure of the catalysts.
Catalyst. From the above results, the plausible reaction
mechanism of the hydrogenation of crotonaldehyde to crotyl
alcohol over Ir-ReO /SiO (Re/Ir = 1) catalyst is proposed in
x
2
Scheme 3. The model of the catalyst Ir-ReO /SiO (Re/Ir = 1)
x
2
Scheme 3. Proposed Reaction Mechanism of Hydrogenation
of Crotonaldehyde to Crotyl Alcohol over Ir-ReO /SiO
CONCLUSION
x
2
■
(Re/Ir = 1)
Selective hydrogenation of crotonaldehyde to crotyl alcohol
was investigated as a typical reaction by using various metal
oxide modified noble metal (M+M′O ) catalysts. Ir is the most
x
effective active metal for the selective hydrogenation, and ReO ,
x
MoO , WO , NbO , and FeO are effective metal oxide
x
x
x
x
modifiers for Ir/SiO . These metal oxide modified Ir/SiO (Ir-
2
2
M′O /SiO ) catalysts improved both activity and selectivity to
x
2
crotyl alcohol, providing high conversion and selectivity (92−
5%). Among these catalysts, Ir-MoO /SiO (Mo/Ir = 1)
9
x
2
catalyst showed the highest activity while maintaining high
−1
selectivity (>90%), and the TOF was 227 h . In order to
clarify the reaction mechanism, various kinetics such as the
isotopic effect of the hydrogen source (H /D ), dependence of
2
2
H pressure and crotonaldehyde concentration on the reaction
2
rate, comparison of the reactivities between the aldehyde group
and olefin group using various substrates, and FTIR
spectroscopic studies such as adsorption of H /D , adsorption
2
2
of crotonaldehyde, and temperature-programmed desorption of
crotonaldehyde were performed using Ir-ReO /SiO (Re/Ir =
x
2
1
) as a model catalyst. The reaction proceeds via hydride attack
was proposed in the previous report on hydrogenolysis of
at the carbonyl carbon of crotonaldehyde adsorbed on ReOx
species as the rate-determining step, and the hydride species is
produced from H by heterolytic dissociation on the Ir metal
near ReO species. The formation of the hydride species brings
about the high selectivity probably thanks to the ionic
hydrogenation, and the strong adsorption of crotonaldehyde
on ReO species leads to high activity thanks to the proximity
of crotonaldehyde to the active sites and activation of the
aldehyde group of crotonaldehyde, which also improves the
selectivity.
glycerol, where ReO species are three-dimensional clusters,
x
18
and they are located on the surface of Ir metal species. For
simplicity, only metal species (Ir metal and ReO ) are depicted
2
x
in Scheme 3. The reaction proceeds in four steps as follows. (i)
x
Crotonaldehyde is adsorbed on ReO species at the aldehyde
x
group, and the aldehyde group is activated by ReO . (ii)
x
−
Hydrogen is dissociated heterolytically to hydride (H ) and
x
+
proton (H ) at the interface between Ir metal and ReO
x
species. (iii) The produced hydride species attack the carbonyl
carbon of crotonaldehyde adspecies to form alkoxide adspecies
3
607
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
EXPERIMENTAL SECTION
Research Article
genation for Organic Synthesis; Nishimura, S., Ed.; Wiley: New York,
001.
2) (a) Abdel-Magid, A. F. In Reductions in Organic Synthesis, Recent
Advances and Practical Applications; Brown, H. C., Ramachandran, P.
V., Eds.; American Chemical Society: Washington, DC, 1996; ACS
Symposium Series, Chapter 1, pp 1−30. (b) Brown, H. C.;
Krishnamurthy, S. Tetrahedron 1979, 35, 567−607.
■
2
(
Typical Procedure for Hydrogenation of Crotonalde-
hyde to Crotyl Alcohol. Activity tests were performed in a
90 mL stainless steel autoclave with an inserted glass vessel. Ir-
ReO /SiO (Re/Ir = 1) (50 mg, 0.010 mmol of Ir) and water
1
x
2
(
3.0 g) were put into an autoclave together with a spinner and
heated at 473 K with 8 MPa H for 1 h for the reduction
(3) (a) Chen, S.-j.; Lu, G.-p.; Cai, C. RSC Adv. 2015, 5, 13208−
13211. (b) Sakaguchi, S.; Yamaga, T.; Ishii, Y. J. Org. Chem. 2001, 66,
4710−4712.
2
pretreatment. After the pretreatment, the autoclave was cooled
and hydrogen was removed; then, 3 mmol of crotonaldehyde
(
4) (a) Ohkuma, T.; Ooka, H.; Ikariya, T.; Noyori, R. J. Am. Chem.
(
Tokyo Chemical Industry Co., Ltd., >99%) was put into the
Soc. 1995, 117, 10417−10418. (b) Melean, L. G.; Rodriguez, M.;
Gonzalez, A.; Gonzalez, B.; Rosales, M.; Baricelli, P. J. Catal. Lett.
autoclave. After the reactor was sealed, the air content was
quickly purged by flushing three times with 1 MPa hydrogen
2
011, 141, 709−716. (c) Farnetti, E.; Pesce, M.; KaSpar, J.;
Spogliarich, R.; Graziani, M. J. Chem. Soc., Chem. Commun. 1986, 10,
46−747.
5) (a) Fagan, P. J.; Voges, M. H.; Bullock, R. M. Organometallics
010, 29, 1045−1048. (b) Song, J. S.; Szalda, D. J.; Bullock, R. M.
Angew. Chem., Int. Ed. Engl. 1992, 31, 1233−1235. (c) Schlaf, M.
Dalton Trans. 2006, 4645−4653. (d) Guan, H.; Iimura, M.; Magee, M.
P.; Norton, J. R.; Zhu, G. J. Am. Chem. Soc. 2005, 127, 7805−7814.
(6) (a) Fujita, S.; Sano, Y.; Bhanage, B. M.; Arai, M. J. Catal. 2004,
225, 95−104. (b) Ghosh, A.; Kumar, R. Microporous Mesoporous Mater.
2005, 87, 33−44.
7) Selected papers: (a) Mitsudome, T.; Matoba, M.; Yamamoto, M.;
Mizugaki, T.; Jitsukawa, K.; Kaneda, K. Chem. Lett. 2013, 42, 660−662.
b) Mitsudome, T.; Matoba, M.; Mizugaki, T.; Jitsukawa, K.; Kaneda,
(
99.99%; Nippon Peroxide Co., Ltd.). The autoclave was then
heated to 303 K, and the H pressure was increased to 0.8 MPa.
2
7
(
2
After the reaction, the gas phase was collected in a gas bag. The
liquid phase in the autoclave was diluted with 10 g of ethanol
and transferred to a vial. The catalyst was separated by
filtration. Details of the reaction conditions are described in
each result. The products were analyzed using GC and GC-MS.
In Situ FTIR Spectroscopy. In situ FTIR spectra were
recorded with a Nicolet 6700 spectrometer (Thermo Scientific)
equipped with a liquid nitrogen cooled MCT (HgCdTe)
−1
(
detector (resolution 4 cm ), using an in situ IR cell with CaF
2
windows, which was connected to a conventional gas flow
system. The sample (about 30 mg of Ir-ReO /SiO , Ir/SiO ,
(
x
2
2
K. Chem. - Eur. J. 2013, 19, 5255−5258. (c) Tian, L.; Yang, Q.; Jiang,
Z.; Zhu, Y.; Pei, Y.; Qiao, M.; Fan, K. Chem. Commun. 2011, 47,
ReO /SiO , and SiO ) was pressed into a self-supporting wafer
x
2
2
(
20 mm diameter) and mounted in the IR cell. Adsorption of
6
168−6170. (d) Zhu, Y.; Tian, L.; Jiang, Z.; Pei, Y.; Xie, S.; Qiao, M.;
crotonaldehyde was carried out by the following method. The
Fan, K. J. Catal. 2011, 281, 106−118. (e) Rodiansono; Khairi, S.; Hara,
T.; Ichikuni, N.; Shimazu, S. Catal. Sci. Technol. 2012, 2, 2139−2145.
catalyst was reduced at 473 K under H (30 mL/min) flow for
2
1
0 min. Then the catalyst was cooled to 303 K under a He flow
(
f) Taniya, K.; Jinno, H.; Kishida, M.; Ichihashi, Y.; Nishiyama, S. J.
(60 mL/min). The baseline spectrum was obtained before
Catal. 2012, 288, 84−91. (g) Chen, P.; Lu, J.; Xie, G.; Zhu, L.; Luo, M.
React. Kinet., Mech. Catal. 2012, 106, 419−434. (h) Zhu, Y.; Qian, H.;
Drake, B. A.; Jin, R. Angew. Chem., Int. Ed. 2010, 49, 1295−1298.
(i) Hidalgo-Carrillo, J.; Aramendia, M. A.; Marinas, A.; Marinas, J. M.;
Urbano, F. J. Appl. Catal., A 2010, 385, 190−200. (j) Galloway, E.;
Arbruster, M.; Kovnir, K.; Tikhov, M. S.; Lambert, R. M. J. Catal.
009, 261, 60−65. (k) Mertens, P. G. N.; Vandezande, P.; Ye, X.;
Poelman, H.; Vankelecom, I. F. J.; De Vos, D. E. Appl. Catal., A 2009,
55, 176−183. (l) Yang, Q.-Y.; Zhu, Y.; Tian, L.; Xie, S.-H.; Pei, Y.; Li,
H.; Li, H.-X.; Qiao, M.-H.; Fan, K.-N. Appl. Catal., A 2009, 369, 67−
6. (m) Gebauer-Henke, E.; Farbotko, J.; Touroude, R.; Rynkowski, J.
injection of crotonaldehyde. Crotonaldehyde (0.2 μL) was
injected into the gas line heated at 423 K under a He flow,
which was fed to the in situ IR cell. Then, IR measurement was
carried out at 303 K. Desorption of crotonaldehyde was
performed by gradually heating the IR cell from 303 to 373 K.
The adspecies of crotonaldehyde at different temperatures was
monitored.
2
3
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.6b00400.
■
7
*
S
J. Kinet. Catal. 2008, 49, 599−606. (n) Gebauer-Henke, E.; Grams, J.;
Szubiakiewicz, E.; Farbotko, J.; Tourude, R.; Rynkowski, J. J. Catal.
007, 250, 195−208. (o) Pei, Y.; Guo, P.; Qiao, M.; Li, H.; Wei, S.;
2
Detailed experimental methods for the catalyst prepara-
tion, reagents, and analyses and Tables S1−S8 and
Figures S1−S6 (PDF)
He, H.; Fan, K. J. Catal. 2007, 248, 303−310. (p) Ando, C.; Kurokawa,
H.; Miura, H. Appl. Catal., A 1999, 185, L181−L183. (q) Sokolski, V.
D.; Zharmagambetova, K. A.; Anisilmova, V. N.; Ualikhanova, A. Dokl.
Akad. Nauk SSSR 1983, 273, 151−153. (r) Wang, M.-M.; He, L.; Liu,
Y.-M.; Cao, Y.; He, H.-Y.; Fan, K.-N. Green Chem. 2011, 13, 602−607.
(s) Gilkey, M. J.; Panagiotopoulou, P.; Mironenko, A. V.; Jenness, G.
R.; Vlachos, D. G.; Xu, B. ACS Catal. 2015, 5, 3988−3944. (t) Wang,
S.; Vorotnikov, V.; Vlachos, D. G. ACS Catal. 2015, 5, 104−112.
(u) Salnikov, O. G.; Kovtunov, K. V.; Barskiy, D. A.; Khudorozhkov,
A. K.; Inozemtseva, E. A.; Prosvirin, I. P.; Bukhtiyarov, V. I.; Koptyug,
I. V. ACS Catal. 2014, 4, 2022−2028. (v) Panagiotopoulou, P.; Martin,
N.; Vlachos, D. G. J. Mol. Catal. A: Chem. 2014, 392, 223−228.
AUTHOR INFORMATION
Corresponding Authors
E-mail for M.T.: mtamura@erec.che.tohoku.ac.jp.
E-mail for K.T.: tomi@erec.che.tohoku.ac.jp.
■
*
*
Notes
The authors declare no competing financial interest.
(
w) Maity, P.; Yamazoe, S.; Tsukuda, T. ACS Catal. 2013, 3, 182−185.
ACKNOWLEDGMENTS
The present work was supported by “Grant for Research” of
The Japan Petroleum Institute.
■
(x) Maclennan, A.; Banerjee, A.; Hu, Y.; Miller, J. T.; Scott, R. W. J.
ACS Catal. 2013, 3, 1411−1419. (y) Wu, B.; Huang, H.; Yang, J.;
Zheng, N.; Fu, G. Angew. Chem., Int. Ed. 2012, 51, 3440−3443. (z) Ide,
W. S.; Hao, B.; Neurock, M.; Davis, R. J. ACS Catal. 2012, 2, 671−683.
REFERENCES
(8) (a) Shi, J.; Zhang, M.; Du, W.; Ning, W.; Hou, Z. Catal. Sci.
■
(
1) (a) Saudan, L. A. Acc. Chem. Res. 2007, 40, 1309−1319.
Technol. 2015, 5, 3108−3112. (b) Yu, Q.; Zhang, X.; Li, B.; Lu, J.; Hu,
G.; Jia, A.; Luo, C.; Hong, Q.; Song, Y.; Luo, M. J. Mol. Catal. A: Chem.
2014, 392, 89−96.
(
b) Mak
̈
i-Arvela, P.; Haj
́
ek, J.; Salmi, T.; Murzin, D. Y. Appl. Catal., A
2
005, 292, 1−49. (c) Handbook of Heterogeneous Catalytic Hydro-
3
608
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609

ACS Catalysis
Research Article
(
9) (a) Claus, P. Top. Catal. 1998, 5, 51−62. (b) Galloway, E.;
Arbruster, M.; Kovnir, K.; Tikhov, M. S.; Lambert, R. M. J. Catal.
009, 261, 60−65. (c) Ramos-Fernandez, E. V.; Samaranch, B.;
Ramírez de la Piscina, P.; Homs, N.; Fierro, J. L. G.; Rodríguez-
Reinoso, F.; Sepulveda-Escribano, A. Appl. Catal., A 2008, 349, 165−
69. (d) Huidobro, A.; Sepulveda-Escribano, A.; Rodríguez-Reinoso,
(16) (a) Stakheev, A. Y.; Kustov, L. M. Appl. Catal., A 1999, 188, 3−
35. (b) Somorjai, G. A.; Park, J. Y. Angew. Chem., Int. Ed. 2008, 47,
9212−9228.
2
́
(17) Nakagawa, Y.; Ning, X.; Amada, Y.; Tomishige, K. Appl. Catal.,
A 2012, 433−434, 128−134.
́
(18) (a) Amada, Y.; Watanabe, H.; Tamura, M.; Nakagawa, Y.;
1
́
Okumura, K.; Tomishige, K. J. Phys. Chem. C 2012, 116, 23503−
3514. (b) Amada, Y.; Shinmi, Y.; Koso, S.; Kubota, T.; Nakagawa, Y.;
Tomishige, K. Appl. Catal., B 2011, 105, 117−127.
19) Borda, G.; Rojas, H.; Murcia, J.; Fierro, J. L. D.; Reyes, P.;
Oportus, M. React. Kinet. Catal. Lett. 2007, 92, 369−376.
20) Martinez, J. J.; Rojas, H.; Castaneda, C.; Diaz, G.; Gomez-
Cortes, A.; Arenas-Alatorre, J. Curr. Org. Chem. 2012, 16, 2791−2796.
21) (a) Lochar, V.; Smolakova, L. React. Kinet. Catal. Lett. 2009, 96,
17−123. (b) de Jesus, J. C.; Zaera, F. Surf. Sci. 1999, 430, 99−115.
22) (a) Durig, J. R.; Brown, S. C.; Kalasinsky, V. F. Spectrochim. Acta
tke, T.
F. J. Catal. 2002, 212, 94−103. (e) Bailie, J. E.; Hutchings, C. J. Chem.
Commun. 1999, 2151−2152.
(
2
10) Tamura, M.; Tokonami, K.; Nakagawa, Y.; Tomishige, K. Chem.
(
Commun. 2013, 49, 7034−7036.
11) (a) Zhu, S.; Gao, X.; Zhu, Y.; Li, Y. J. Mol. Catal. A: Chem. 2015,
98, 391−398. (b) Falcone, D. D.; Hack, J. H.; Klyushin, A. Y.; Knop-
Gericke, A.; Schlogl, R.; Davis, R. J. ACS Catal. 2015, 5, 5679−5695.
c) Liu, S.; Okuyama, Y.; Tamura, M.; Nakagawa, Y.; Imai, A.;
(
3
(
̈
(
̌
́
́
(
1
́
Tomishige, K. Green Chem. 2016, 18, 165−175. (d) Liu, S.; Okuyama,
Y.; Tamura, M.; Nakagawa, Y.; Imai, A.; Tomishige, K. ChemSusChem
(
1
976, 32, 807−813. (b) McInroy, A. R.; Uhl, A.; Lear, T.; Klapo
̈
2
015, 8, 628−635. (e) Zhu, S.; Gao, X.; Zhu, Y.; Cui, J.; Zheng, H.; Li,
M.; Shaikhutdinov, S.; Schauermann, S.; Rupprechter, G.; Freund, H.-
Y. Appl. Catal., B 2014, 158−159, 391−339. (f) Liu, S.; Tamura, M.;
Nakagawa, Y.; Tomishige, K. ACS Sustainable Chem. Eng. 2014, 2,
J.; Lennon, D. J. Chem. Phys. 2011, 134, 214704.
1
819−1827. (g) Arundhathi, R.; Mizugaki, T.; Mitsudome, T.;
Jitsukawa, K.; Kaneda, K. ChemSusChem 2013, 6, 1345−1347.
h) Chen, K.; Tamura, M.; Yuan, Z.; Nakagawa, Y.; Tomishige, K.
(
ChemSusChem 2013, 6, 613−621. (i) Amada, Y.; Watanabe, H.; Hirai,
Y.; Kajikawa, Y.; Nakagawa, Y.; Tomishige, K. ChemSusChem 2012, 5,
1
́
991−1999. (j) Chia, M.; Pagan-Torres, Y. J.; Hibbitts, D.; Tan, Q.;
Pham, H. N.; Datye, A. K.; Neurock, M.; Davis, R. J.; Dumesic, J. A. J.
Am. Chem. Soc. 2011, 133, 12675−12689. (k) Nakagawa, Y.; Shinmi,
Y.; Koso, S.; Tomishige, K. J. Catal. 2010, 272, 191−194. (l) Amada,
Y.; Koso, S.; Nakagawa, Y.; Tomishige, K. ChemSusChem 2010, 3,
7
28−736. (m) Shinmi, Y.; Koso, S.; Kubota, T.; Nakagawa, Y.;
Tomishige, K. Appl. Catal., B 2010, 94, 318−326.
12) (a) Pholjaroen, B.; Li, N.; Huang, Y.; Li, L.; Wang, A.; Zhang, T.
(
Catal. Today 2015, 245, 93−99. (b) Guan, J.; Peng, G.; Cao, Q.; Mu,
X. J. Phys. Chem. C 2014, 118, 25555−25566. (c) Wang, Z.;
Pholjaroen, B.; Li, M.; Dong, W.; Li, N.; Wang, A.; Wang, X.; Cong,
Y.; Zhang, T. J. Energy Chem. 2014, 23, 427−434. (d) Chen, K.; Mori,
K.; Watanabe, H.; Nakagawa, Y.; Tomishige, K. J. Catal. 2012, 294,
1
71−183. (e) Koso, S.; Watanabe, H.; Okumura, K.; Nakagawa, Y.;
Tomishige, K. Appl. Catal., B 2012, 111−112, 27−37. (f) Buntara, T.;
Noel, S.; Phua, P. H.; Melian-Cabrera, I.; de Vries, J. G.; Heeres, H. J.
́
Angew. Chem., Int. Ed. 2011, 50, 7083−7087. (g) Koso, S.; Nakagawa,
Y.; Tomishige, K. J. Catal. 2011, 280, 221−229. (h) Chen, K.; Koso,
S.; Kubota, T.; Nakagawa, Y.; Tomishige, K. ChemCatChem 2010, 2,
5
47−555. (i) Koso, S.; Ueda, N.; Shinmi, Y.; Okumura, K.; Kizuka, T.;
Tomishige, K. J. Catal. 2009, 267, 89−92.
13) (a) Takeda, Y.; Tamura, M.; Nakagawa, Y.; Okumura, K.;
(
Tomishige, K. ACS Catal. 2015, 5, 7034−7047. (b) Takeda, Y.; Shoji,
T.; Watanabe, H.; Tamura, M.; Nakagawa, Y.; Okumura, K.;
Tomishige, K. ChemSusChem 2015, 8, 1170−1178. (c) Tamura, M.;
Tamura, R.; Takeda, Y.; Nakagawa, Y.; Tomishige, K. Chem. Commun.
2
014, 50, 6656−6659. (d) Wang, Z.; Li, G.; Liu, X.; Huang, Y.; Wang,
A.; Chu, W.; Wang, X.; Li, N. Catal. Commun. 2014, 43, 38−41.
e) Corbel-Demailly, L.; Ly, B.-K.; Minh, D.-P.; Tapin, B.; Especel, C.;
Epron, F.; Cabiac, A.; Guillon, E.; Besson, M.; Pinel, C. ChemSusChem
013, 6, 2388−2395. (f) Manyar, H. G.; Paun, C.; Pilus, R.; Rooney,
D. W.; Thompson, J. M.; Hardacre, C. Chem. Commun. 2010, 46,
279−6281.
14) (a) Ota, N.; Tamura, M.; Nakagawa, Y.; Okumura, K.;
Tomishige, K. Angew. Chem., Int. Ed. 2015, 54, 1897−1900.
b) Amada, Y.; Ota, N.; Tamura, M.; Nakagawa, Y.; Tomishige, K.
(
2
6
(
(
ChemSusChem 2014, 7, 2185−2192. (c) Ota, N.; Tamura, M.;
Nakagawa, Y.; Okumura, K. ACS Catal. 2016, 6, 3213−3226.
(
15) (a) Tamura, M.; Tamura, R.; Takeda, Y.; Nakagawa, Y.;
Tomishige, K. Chem. - Eur. J. 2015, 21, 3097−3107. (b) Takeda, Y.;
Nakagawa, Y.; Tomishige, K. Catal. Sci. Technol. 2012, 2, 2221−2223.
(
c) Nakagawa, Y.; Mori, K.; Chen, K.; Amada, Y.; Tamura, M.;
Tomishige, K. Appl. Catal., A 2013, 468, 418−425.
3
609
DOI: 10.1021/acscatal.6b00400
ACS Catal. 2016, 6, 3600−3609