Total hydrogenation of furfural and 5-hydroxymethylfurfural over supported Pd-Ir alloy catalyst

Article abstract of DOI:10.1021/cs500620b

Hydrogenation of aqueous furfural was conducted with SiO ₂-suported palladium-based bimetallic catalysts. The combination of palladium and iridium gave the best performance for the total hydrogenation to tetrahydrofurfuryl alcohol. Higher H₂ pressure and lower reaction temperature were advantageous to suppress side reactions. The synergy between Pd and Ir in the hydrogenation catalysis is most remarkable for substituted furans as substrates. Furfural was first converted into furfuryl alcohol, which was further converted to tetrahydrofurfuryl alcohol. A small amount of tetrahydrofurfural was formed in the first step (～20% selectivity), and the subsequent hydrogenation of tetrahydrofurfural was much slower. The combined yield of tetrahydrofurfuryl alcohol and tetrahydrofurfural reached 98%. The yield of tetrahydrofurfuryl alcohol reached 94% with larger amount of catalyst. Total hydrogenation of 5-hydroxymethylfurfural was also possible using Pd-Ir/SiO₂ catalyst. The performance of Pd-Ir/SiO₂ catalyst was slightly changed by repeated uses, and the used catalyst can be regenerated by calcination and reduction at 573 K. Characterization results showed that Pd-Ir alloy particles with size of ≤4 nm were formed on the catalyst. Addition of Ir much increased the TOF values as compared with Pd/SiO₂ with similar particle size, especially for C = O hydrogenation. One factor of higher activity of Pd-Ir/SiO₂ than Pd/SiO₂ can be the change of adsorption mode: Ir atom on the surface promotes the adsorption at C = O site, whereas the Pd surface strongly interacts with furan ring.

Full text of DOI:10.1021/cs500620b

Research Article
pubs.acs.org/acscatalysis
Total Hydrogenation of Furfural and 5‑Hydroxymethylfurfural over
Supported Pd−Ir Alloy Catalyst
Yoshinao Nakagawa,* Kana Takada, Masazumi Tamura, and Keiichi Tomishige*
Department of Applied Chemistry, School of Engineering, Tohoku University, 6-6-07 Aoba, Aramaki, Aoba-ku, Sendai 980-8579,
Japan
S
* Supporting Information
ABSTRACT: Hydrogenation of aqueous furfural was conducted with SiO2-suported
palladium-based bimetallic catalysts. The combination of palladium and iridium gave the
best performance for the total hydrogenation to tetrahydrofurfuryl alcohol. Higher H2
pressure and lower reaction temperature were advantageous to suppress side reactions.
The synergy between Pd and Ir in the hydrogenation catalysis is most remarkable for
substituted furans as substrates. Furfural was first converted into furfuryl alcohol, which
was further converted to tetrahydrofurfuryl alcohol. A small amount of tetrahydrofurfural
was formed in the first step (∼20% selectivity), and the subsequent hydrogenation of
tetrahydrofurfural was much slower. The combined yield of tetrahydrofurfuryl alcohol
and tetrahydrofurfural reached 98%. The yield of tetrahydrofurfuryl alcohol reached 94%
with larger amount of catalyst. Total hydrogenation of 5-hydroxymethylfurfural was also
possible using Pd−Ir/SiO2 catalyst. The performance of Pd−Ir/SiO2 catalyst was slightly changed by repeated uses, and the used
catalyst can be regenerated by calcination and reduction at 573 K. Characterization results showed that Pd−Ir alloy particles with
size of ≤4 nm were formed on the catalyst. Addition of Ir much increased the TOF values as compared with Pd/SiO2 with similar
particle size, especially for CO hydrogenation. One factor of higher activity of Pd−Ir/SiO2 than Pd/SiO2 can be the change of
adsorption mode: Ir atom on the surface promotes the adsorption at CO site, whereas the Pd surface strongly interacts with
furan ring.
KEYWORDS: hydrogenation, palladium, iridium, furan, biomass
produced in large scale with Cu−Cr catalyst, and FOL is
used in the production of thermostatic resins.21 Because of the
growing interest in HMF, various systems of reduction of HMF
into BHF or 2,5-dimethylfuran have been reported.22−25 On
the other hand, THFA and BHTHF can be used as green
solvents, because they are degradable, less toxic, and more
stable than unsaturated furan compounds.26 Recently, selective
hydrogenolysis systems have been discovered where THFA or
BHTHF can be converted into useful straight-chain polyols
such as 1,5-pentanediol and 1,6-hexanediol.27−36 These systems
typically use water solvent and Rh or Ir catalysts modified with
reducible metal oxide such as ReOx. Therefore, the importance
of total hydrogenation of furfural and HMF has been growing.
Ni catalysts such as Raney Ni are known to be effective in
total hydrogenation of furfural and HMF.37−43 However, Ni
catalysts are easily leached in liquid-phase systems,42 and to
make matters worse, dissolved Ni species strongly poison Rh or
Ir hydrogenolysis catalysts. The leaching problem can be
avoided by application to a gas-phase system;41 however, very
high boiling point of HMF hinders this approach. Ru catalysts
have been also reported to be active in the total hydrogenation;
however, the selectivity is lower than Ni catalysts, especially
1. INTRODUCTION
Biorefinery, the production of fuels or chemicals from biomass,
has gained much attention due to the limited fossil fuel
resources and global warming issues.1−7 Oxygen-containing
chemicals are more attractive targets than fuels because of the
higher market price of chemicals and the lower amount of
reducing agent (usually hydrogen) consumed in the conversion
process to chemicals.8−11 Furfural and 5-hydroxymethylfurfural
(HMF) are produced from sugars via dehydration and are
promising intermediates in biorefinery.12−18 Furfural and HMF
are highly unsaturated compounds, and therefore, hydro-
genation is the most important process in the conversion of
furfural or HMF.19,20 Scheme 1 shows the reaction pathways of
hydrogenation of furfural and HMF.19 Hydrogenation of
furfural at the aldehyde group and the furan ring gives furfuryl
alcohol (FOL) and tetrahydrofurfural, respectively. Further
hydrogenation of either compound gives tetrahydrofurfuryl
alcohol (THFA). Various side reactions are known such as C−
O hydrogenolysis of side-substituents to 2-methylfuran,
decarbonylation to furan, and rearrangement to levulinic acid.
Further hydrogenation of these byproducts gives more kinds of
byproducts. Hydrogenation of HMF gives 2,5-bis-
(hydroxymethyl)furan (BHF), 5-hydroxymethyltetrahydro-
furfural and 2,5-bis(hydroxymethyl)tetrahydrofuran (BHTHF)
as products of C−O, furan-ring and total hydrogenation,
respectively. Among these products, only FOL has been
Received: March 8, 2014
Revised: June 16, 2014
Published: July 9, 2014
© 2014 American Chemical Society
2718
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726

ACS Catalysis
Research Article
Scheme 1. Reaction Routes of Hydrogenation of Furfural and HMF
when pure water is used as a solvent.43,44 Very recently, Biradar
et al. reported an effective total hydrogenation system of
furfural with Pd/MFI catalyst and isopropyl alcohol solvent.45
Tuteja et al. reported a hydrogenation/hydrogenolysis system
of HMF into 1,6-hexanediol in ethanol with Pd catalyst and
formic acid as a hydrogen source.46 In water solvent, various
side reactions such as rearrangement easily proceed during the
reduction of furfural or HMF.19,47−49 Development of stable
total hydrogenation catalysts is still a challenging task. A typical
method to improve the performance of hydrogenation catalysts
is the addition of secondary metal element to active metal (Ni,
Cu or noble metal).50−52 The secondary metal element is
usually Sn or first-row transition metal such as Fe. However,
this type of bimetallic catalysts is generally effective in selective
CO hydrogenation of unsaturated aldehydes including
furfural and HMF, and it is not suitable to total hydrogenation.
The selectivity is explained by the presence of cationic species
of an additive element that strongly interacts with the CO
group. In addition, these additive elements are easily leached
into aqueous media. Therefore, in this study we focused on the
combination of noble metals, which are stable in the metallic
state in water. We tested various SiO2-supported bimetallic
noble metal catalysts for total hydrogenation of aqueous
furfural, and we found that the combination of Pd−Ir was
remarkably effective.
373 K. The reduced catalysts were passivated with 2% O2/He
at room temperature.
2.2. Activity Tests. Furfural (Wako) was used as received
or after distillation. 5-Hydroxymethylfurfural (Sigma-Aldrich)
was purified with column chromatography (Merck silica gel 60;
methanol/diethyl ether = 2/98).42 Activity tests were
performed in a 190 mL stainless steel autoclave with an
inserted glass vessel. An aqueous solution of substrate (0.5 M)
was put into the autoclave together with a spinner and reduced
catalyst. After sealing the reactors, autoclaves were warmed with
a heater (for 313 K reaction) or cooled with a water bath (for
275 or 296 K reaction). After the temperature was stabilized,
the gas phase was quickly replaced with hydrogen, and the
reaction was started. After an appropriate reaction time, the
autoclave contents were transformed to vials, and the catalysts
were separated by filtration. The gas phase was also collected in
a gas bag. The products were analyzed using a gas
chromatograph equipped with FID. A TC-WAX or an InertCap
5 capillary column was used for the separation. Products were
also identified using GC-MS (Shimadzu QP 5050), and the
details of identification are shown in Supporting Information.
Conversions were calculated by the consumption of substrates
except in the case of furan hydrogenation where the sum of
products was regarded as “conversion”. Selectivities were
defined as the molar ratio of the product to converted
substrate. The kinetic fitting was conducted by minimizing the
sum of absolute differences of conversion and selectivities to
FOL and THFA in percent between experimental and
calculated data. The error range was estimated by changing
each parameter while optimizing the other parameter until the
sum of absolute differences becomes two times as its minimum
value.
2. EXPERIMENTAL SECTION
2.1. Catalyst Preparation. SiO2 (Fuji silysia G-6; BET
surface area 523 m2 g−1) was used as a support. Catalysts were
prepared by impregnating SiO2 with a mixed aqueous solution
of metal precursors. The used precursors were PdCl2 (Kanto
Chemical), RuCl3·nH2O (Kanto Chemical), RhCl3 (Soekawa
Chemical), H2IrCl6 (Furuya Metals), and H2PtCl6 (Kanto
Chemical). For monometallic Pd catalyst, the one using
Pd(NH3)4(NO3)2 precursor (Sigma-Aldrich) was also prepared
(denoted by Pd/SiO2(N)). After impregnation, the catalysts
were dried in air at 383 K overnight and then calcined at 773 K
for 3 h. The catalyst samples were stored in the calcined form.
Reduction was conducted in a glass tube reactor with flowing
H2 for 0.5 h. The reduction temperature was 573 K except in
the cases of Pd/SiO2 and Pd/SiO2(N) which were reduced at
2.3. Catalyst Characterization. Temperature-pro-
grammed reduction (TPR) with H2 was performed in a fixed-
bed flow reactor. The TPR profile of each sample (∼50 mg)
was recorded from room temperature to 1073 K at a heating
rate of 10 K min−1 under a flow of 5% H2/Ar (30 mL min−1).
The amount H2 in the effluent gas after passing through a
frozen acetone trap was monitored with a thermal conductivity
detector. XRD patterns of the catalysts were recorded by a
Rigaku MiniFlex 600 diffractometer. TEM images were taken
with HITACHI HD-2700 or HITACHI HF-2000. EDX
analysis was conducted with HITACHI HD-2700. The average
2719
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726

ACS Catalysis
Research Article
a
Table 1. Hydrogenation of Furfural over Various Catalysts
selectivity (%)
Pd loading (wt %) M/Pd ratio catalyst amount (mg) time (h) conv. (%) THFA FOL tetrahydro- furfural others
entry
catalyst
Pd/SiO2
1
2
2
2
2
2
2
50
50
1
1
1
1
1
1
1
1
1
1
2
4
6
6
6
25
63
67
14
19
27
31
23
13
5
69
33
47
49
77
79
96
85
47
67
16
<1
<1
<1
<1
7
28
19
20
3
5
13
3
b
Pd/SiO2(N)
3
Pd−Ir/SiO2
1
1
1
1
10
4
Pd−Ru/SiO2
Pd−Rh/SiO2
Pd−Pt/SiO2
Ir/SiO2
10
8
5
10
5.7
2.5
14
7
6
2
10
<1
<1
1
16
4
c
7
4
50
<1
10
27
19
63
80
80
91
94
8
Pd/SiO2+Ir/SiO2
Pd−Ir/SiO2
Pd−Ir/SiO2
Pd−Ir/SiO2
Pd−Ir/SiO2
Pd−Ir/SiO2
Pd−Ir/SiO2
Pd−Ir/SiO2
2
3
1
2
2
2
2
2
1
1/3
3
25 + 25
10
26
30
4
9
21
11
20
18
18
6
5
10
11
12
13
14
15
10
61
3
1
10
99
2
1
10
>99
>99
>99
>99
1
1
10
2
1
50
4
1
150
1
5
a
b
c
Reaction conditions: Furfural (5 mmol), water (9 g), catalyst (10−150 mg), H2 (8 MPa), 275 K. Prepared using Pd(NH3)4(NO3)2 precursor. Ir
loading amount.
size was calculated by ∑nidi3/∑nidi2 (di: particle size, ni:
number of particle with di). The amount of CO adsorption was
measured in a homemade high-vacuum system by a volumetric
method (dead volume: 43 cm3). The calcined sample (∼0.15 g)
in the measurement cell was reduced with H2 at 573 K for 0.5 h
and evacuated at 573 K. After cooling to room temperature,
CO was introduced and the adsorption amount was measured.
The pressure at adsorption equilibrium was about 1 kPa. The
physisorption amount of CO on the support was determined by
repeating the adsorption measurement after evacuation at room
temperature, and the amount was subtracted from that in the
first measurement to determine the chemisorption amount.
The chemisorption amount was represented as the molar ratio
to total noble metal atoms (CO/(Pd + Ir)). FT-IR spectra of
adsorbed CO were recorded by a Nicolet 6700 FT-IR with a
resolution of 4 cm−1 in a transition mode using an in situ cell
with CaF2 windows. The calcined sample was pressed into a
self-supporting disk (7 mm diameter, ∼30 mg). The sample
disk was put into the IR cell connected to the closed circulation
system. The samples were reduced with flowing H2 at 573 K for
0.5 h. After evacuation and cooling, CO (6.7 kPa) was
introduced and rested for 1 h. The remaining CO was
evacuated, and then FT-IR spectra were recorded. The FT-IR
spectra of adsorbed species were obtained by subtracting the
spectra of the catalysts before exposing CO.
catalyst was even higher than that of Pd/SiO2(N). Mono-
metallic Ir/SiO2 catalyst showed very low activity and THFA
yield (entry 7), showing the synergy between Pd and Ir.
Physical mixture of Pd/SiO2 and Ir/SiO2 gave almost no
enhancing effect in the catalytic performance (entry 8). Various
Pd−Ir/SiO2 catalysts were tested with different Ir/Pd ratio and
constant total metal amount (entries 3, 9, 10). The catalyst with
Ir/Pd = 1 showed the best performance.
There have been limited reports of Pd−Ir bimetallic catalysts.
Mahmoud et al. reported that Pd−Ir/SiO2 catalyst prepared by
sol−gel method showed higher activity in hydrogenation of
cinnamaldehyde in toluene than Pd/SiO2 with the same Pd
amount.54 However, the increase of activity was rather small
(<1.3 times) when compared on the basis of the Pd + Ir
amount. The promotion by Ir addition in our system was more
drastic than this literature system. Rocha et al. reported Pd−Ir
catalysts supported on H−Y zeolite for hydrogenation of
tetralin to decaline with fixed-bed flow system at high
temperature (523 K).55 The catalyst with the molar ratio of
Ir/Pd = 1 showed the highest activity in catalysts with the same
Pd + Ir amount, although the difference from monometallic Pd
́
́
catalyst was very small. Lopez-De Jesus et al. reported the
catalysis of Pd−Ir/γ-Al2O3 for hydrogenation of benzonitrile to
benzylamine, dibenzylamine, or toluene in ethanol.56 The
catalyst with the ratio of Ir/Pd = 1 showed higher activity,
especially for the formation of benzylamine, than those with Ir/
Pd = 4 or 1/4. However, the reported TOF of Pd−Ir/γ-Al2O3
(Ir/Pd = 1) based on the surface atoms was lower to that of
Pd/γ-Al2O3. Ziasi-azad et al. have recently reported Pd−Ir/γ-
Al2O3 catalysts with homogeneous alloy or core(Pd)-shell(Ir)
structure and the application to ring-opening reaction of
Indane.57 Pd works only as an inert surface diluting metal in
this ring-opening reaction. Above all, the remarkable synergy
between Pd and Ir like this furfural hydrogenation has not been
reported.
3. RESULTS AND DISCUSSION
3.1. Optimization of the Catalyst Components. Table 1
shows the results of activity tests of various bimetallic or
monometallic noble metal catalysts for hydrogenation of
aqueous furfural. Pd was selected as a component because of
the high reported hydrogenation activity.32,39,45,52 We used
PdCl2 as a precursor of Pd because precursors of other noble
metals were also chlorides. Addition of another noble metal
enhanced the catalytic activity (entries 1, 3−6). In particular,
addition of Ir gave by far the largest promotion and the highest
yield of THFA (entry 3). We also tested Pd/SiO2 prepared
with Pd(NH3)4(NO3)2 as a precursor (denoted by Pd/
SiO2(N); entry 2), because it is known that more dispersed
Pd metal particles can be formed from Pd(NH3)4(NO3)2
precursor than from PdCl2.53 The activity of Pd−Ir/SiO2
3.2. Performance of Pd−Ir/SiO2 (Pd/Ir = 1). The time
course of hydrogenation of furfural over Pd−Ir/SiO2 (Pd/Ir =
1) is also shown in Table 1. FOL was the main product at
shorter reaction time, and later FOL was converted to THFA,
similarly to other reported systems of furfural hydrogenation to
THFA.41−45 Smaller amount of tetrahydrofurfural was also
2720
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726

ACS Catalysis
Research Article
a
Table 2. Hydrogenation of HMF over Pd−Ir/SiO2 Catalyst
selectivity (%)
entry
catalyst amount (mg)
time (h)
conv. (%)
BHTHF
BHF
5-hydroxymethyl- tetrahydrofurfural
others
1
2
10
150
1
4
>99
>99
85
95
<1
<1
14
1
1
4
a
Reaction conditions: HMF (5 mmol), water (9 g), Pd−Ir/SiO2 (Pd 2 wt %, Ir/Pd = 1) 10 or 150 mg, H2 (8 MPa), 275 K.
formed, and the hydrogenation of tetrahydrofurfural was
relatively slow. At 4 h, FOL was totally converted, and the
yield of THFA reached 80% (entry 14). At that time, the yield
of tetrahydrofurfural was 18%. Higher yield of THFA can be
obtained with larger amount of catalyst, which effectively
increases reaction progress at the same reaction time. With 0.15
g catalyst (15 times larger amount), 94% yield of THFA was
obtained (entry 15).
This catalytic system was further applied to HMF hydro-
genation. The results were similar to those for furfural
hydrogenation (Table 2). Mixture of BHTHF and 5-hydroxy-
methyltetrahydrofurfural was obtained in short reaction time
(entry 1), and high yield of BHTHF (95%) was obtained with
larger amount of catalyst (entry 2). The reactivity of HMF was
higher than furfural over this Pd−Ir/SiO2 catalyst. This is in
contrast to the case of Ni−Pd/SiO2 catalyst where the reactivity
of HMF is minimally higher than furfural.42
The effect of reaction conditions on the catalytic perform-
ance are shown in Figure 1. The activity was increased with
increasing H2 pressure for both Pd−Ir/SiO2 and monometallic
catalysts. Higher H2 pressure was also advantageous in terms of
selectivity: byproducts from other reactions than hydrogenation
were suppressed by higher H2 pressure. The substrate
concentration had little effect in the catalysis. The reaction
temperature affected the selectivity as well as activity: higher
reaction temperature promoted side reactions. The increase of
activity by higher reaction temperature was relatively small,
especially for the Ir/SiO2 catalyst. After all, higher H2 pressure
and lower reaction temperature are advantageous to this
hydrogenation system.
The reusability of Pd−Ir/SiO2 was tested (Figure 2). First,
the used catalyst was collected by centrifugation, and reused
without reactivation. In this case, the selectivity was gradually
changed by subsequent reuse: the hydrogenation activity of the
furan ring was decreased. The activity during the third run was
slightly lower than that of fresh catalyst. After three uses, the
catalyst was calcined at 573 K and then reduced again at 573 K.
The regenerated catalyst showed almost the same performance
as fresh one. The used catalyst might be poisoned by some
byproducts such as polymer or CO, which can be removed by
calcination.
3.3. Catalyst Characterization. The TPR profile of
calcined Pd−Ir/SiO2 (Pd/Ir = 1) is shown in Figure 3a.
Consumption signal was observed between 350 and 550 K, and
the molar ratio of consumed H2 to Ir was 1.2. This value was
smaller than the value for the reduction of IrO2,58,59 and a part
of Ir in addition to Pd was already reduced at room
temperature. A small desorption peak was observed at 350 K,
similarly to the pattern of Pd/SiO2 (Figure 3b), also indicating
that metallic phase was already formed at room temperature.
The consumption signal of Pd−Ir/SiO2 (peak top: 460 K) was
slightly shifted to lower temperature than Ir/SiO2 (peak top:
510 K; Figure 3e59), indicating that reduction of Ir was
promoted by addition of Pd. The TPR profiles of catalysts after
reduction and passivation showed no apparent H2 consumption
Figure 1. Hydrogenation of furfural in various reaction conditions.
Standard reaction conditions: catalyst (10 mg for Pd−Ir/SiO2 (Ir/Pd
= 1), 50 mg for Pd/SiO2 or Ir/SiO2), furfural (5 mmol), water (9 g),
H2 (8 MPa), 275 K, 1 h.
signal above room temperature (Figure 3c,d), suggesting that
the catalysts in the reaction media were totally reduced. The
XRD patterns of catalysts are shown in Figure 4. Reduced
monometallic catalysts (Pd/SiO2, Pd/SiO2(N), and Ir/SiO2)
clearly showed the peaks of the metallic phase in addition to the
broad peak of SiO2 support (∼22°). The particle size calculated
by the width of the XRD peak of (111) reflection at around 40°
was 4 nm for both Pd/SiO2(N) and Ir/SiO2 and 7 nm for Pd/
SiO2. Reduced Pd−Ir/SiO2 at 573 K, which was used for
2721
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726

ACS Catalysis
Research Article
showed the (111) peak (40.4°) at the midmost positions
between the simple substances (40.1 and 40.7°). The pattern
suggests the formation of alloy phase; however, the overlap of
two different monometallic phases cannot be ruled out. Figure
5 show the TEM image of reduced Pd−Ir/SiO2 (Ir/Pd = 1).
Figure 2. Reuses of Pd−Ir/SiO2 catalyst. Reaction conditions: Pd−Ir/
SiO2 (Ir/Pd = 1; 10 mg), furfural (5 mmol), water (9 g), H2 (8 MPa),
275 K, 1 h. The catalyst after the third run was calcined at 573 K for 3
h and then reduced with H2 at 573 K for 0.5 h before the fourth run.
Figure 5. TEM image of reduced Pd−Ir/SiO2 (Ir/Pd = 1).
Metal nanoparticles were observed with narrow distribution of
size, and the average size was 3.4 nm. The particle size was
similar to those of Pd/SiO2(N) (TEM images of Pd/SiO2 and
Pd/SiO2(N) are shown in Supporting Information) and Ir/
SiO2 (3.6 nm).59 The EDX analysis was conducted to
determine the distribution of each metal. Although the element
distribution in one particle could not be determined because of
the small particle size, the element distribution among particles
was determined to be narrow and centered at the ratio of the
whole sample (Pd/Ir = 1:1; Figure 6), showing the formation
Figure 3. TPR profiles of catalysts. (a) Pd−Ir/SiO2 (Ir/Pd = 1), (b)
Pd/SiO2, (c) Pd−Ir/SiO2 (Ir/Pd = 1) after reduction and passivation,
(d) Ir/SiO2 after reduction and passivation, and (e) Ir/SiO2. The
broken line is the baseline for integration. The pattern of (e) is
reproduced from ref 59. Copyright 2011 Elsevier.
Figure 6. EDX analysis of particles on reduced Pd−Ir/SiO2 (Ir/Pd =
1).
of bimetallic particles. The particle size was also calculated by
the width of the XRD peak of (111) reflection at around 40°
(Figure 4d). The calculated value was 3.5 nm. The particle size
of bimetallic particles calculated from XRD line width is
generally underestimated by the broadening derived from the
heterogeneity of alloy compositions. This agreement in particle
sizes determined by TEM and XRD suggests the formation of
uniform Pd−Ir alloy particles. The XRD peak positions of Pd−
Ir/SiO2 (Ir/Pd = 1) also supports the formation of uniform
alloy. We measured the FT-IR spectra of adsorbed CO on the
reduced catalysts in order to see the property of metal surface
Figure 4. XRD patterns of reduced catalysts. (a) Pd/SiO2(N), (b) Pd/
SiO2, (c) Pd−Ir/SiO2 (Ir/Pd = 1/3), (d) Pd−Ir/SiO2 (Ir/Pd = 1), (e)
Pd−Ir/SiO2 (Ir/Pd = 1) after reaction, (f) Pd−Ir/SiO2 (Ir/Pd = 3),
(g) Ir/SiO2.
catalytic reaction, showed the peaks located between those of
Pd metal and Ir metal. In particular, the catalyst with Ir/Pd = 1
2722
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726

ACS Catalysis
Research Article
(Figure 7). On Ir/SiO2 catalyst, one strong peak at 2079 cm−1
assignable to linear CO on Ir was observed. On the Pd−Ir/SiO2
of linear CO on Pd by alloy formation has been observed for
other Pd−M systems such as Pd−Ag,61 and this phenomenon
can be explained by decrease of PdsurfacePdsurface bonds.
Anyway, the spectral patterns of CO on Pd−Ir/SiO2 were
different from the simple sum of those on Pd metal and Ir
metal, supporting the formation of Pd−Ir alloy.
Above all results show that reduced Pd−Ir/SiO2 (Ir/Pd = 1)
contains Pd−Ir alloy particles with the size of ≤4 nm. The CO
adsorption amount of Pd/SiO2(N), Ir/SiO2, and Pd−Ir/SiO2
(Ir/Pd = 1) was similar in the ratio to total metal atoms: CO/
(Pd + Ir) = 0.19, 0.17, and 0.16, respectively. The higher
activity of Pd−Ir/SiO2 than Pd/SiO2 (and Pd/SiO2(N)) and
Ir/SiO2 indicates that Pd−Ir alloy phase has high catalytic
activity than Pd or Ir monometallic phases.
In the literature, limited characterizations have been reported
for Pd−Ir catalysts except size-related techniques such as TEM
and adsorption amounts.54−56 On the other hand, Ziaki-azad et
al. reported the formation of Pd−Ir random alloy particles
prepared by reduction of mixed solution of PdCl2 and IrCl3
with refluxing ethanol.57 They characterized the particles with
various techniques such as XPS, XANES, and ion-scattering
spectroscopy. However, both literature data and our data
cannot strictly decide whether completely uniform alloy phase
was formed or not; that is, we do not exclude the possibility
that heterogeneity of compositions exists in each particle. The
decision will be another work.
The used Pd−Ir/SiO2 catalyst (Ir/Pd = 1) was also
characterized with XRD (Figure 4e) and TEM (Figure S2 in
Supporting Information). The obtained data was essentially
identical to the fresh, unused catalyst, also confirming the
stability of the catalyst.
3.4. Reaction Mechanism. Table 3 shows the comparison
of reactivity between furfural, FOL, and furan over Pd−Ir/SiO2,
Ir/SiO2, Pd/SiO2, and Pd/SiO2(N). Pd−Ir/SiO2 showed
higher activity than Ir/SiO2 for all these substrates. However,
the difference of activity between Pd−Ir/SiO2 and mono-
metallic Pd catalyst was varied with substrates. Pd−Ir/SiO2
showed higher activity than Pd/SiO2 and Pd/SiO2(N) in
furfural hydrogenation, as discussed in section 3.1. The
promoting effect of Ir addition was smaller in FOL hydro-
Figure 7. FT-IR spectra of adsorbed CO on the catalysts. (a) Pd/
SiO2(N), (b) Pd−Ir/SiO2 (Ir/Pd = 1/3), (c) Pd−Ir/SiO2 (Ir/Pd = 1),
(d) Pd−Ir/SiO2 (Ir/Pd = 3), (e) Ir/SiO2.
catalyst with high Ir/Pd ratio (Ir/Pd = 3), the peak was slightly
shifted to lower wavenumber (2067 cm−1). With decreasing Ir/
Pd ratio, the signal for linear CO was splitted into two peaks,
and one broad signal below 2000 cm−1 appeared. The latter
signal can be assigned to bridge CO on Pd, and this type of
signal was the main band for monometallic Pd/SiO2(N)
catalyst (Figure 7a). The presence of a small amount of bridge
CO on the bimetallic Pd−Ir particles means the presence of Pd
atom on the surface as well as Ir. It should be noted that
metallic Pd was formed at much lower temperature than
metallic Ir and that the number of atoms on the particle surface
was smaller than that of the total Ir atoms. Therefore, a portion
of the Ir0 atoms formed during the reduction was moved into
the inside of particles. The signal for linear CO on Pd particle
in Pd/SiO2(N) was located at 2083 cm−1, and in the literature,
the signal for linear CO on Pd surface is reported to be located
at around 2090 cm−1.60 These wavenumber values are larger
than that of linear CO on Ir. Therefore, the peaks with lower
and higher wavenumber for linear CO on bimetallic catalyst
(2062 and 2079 cm−1 for Ir/Pd = 1 sample) can be tentatively
assigned to CO on Ir and Pd, respectively. Increasing amount
a
Table 3. Hydrogenation of Furanic Compounds over Pd/SiO2, Pd−Ir/SiO2 (Ir/Pd = 1), and Ir/SiO2
entry
catalyst
catalyst amount (mg)
conv. (%)
products (selectivity (%))
furfural hydrogenation (furfural = 5 mmol)
1
2
3
4
5
Pd/SiO2
50
50
10
50
50
25
63
67
14
94
THFA (19), FOL (69), tetrahydrofurfural (7)
THFA (27), FOL (33), tetrahydrofurfural (28)
THFA (31), FOL (47), tetrahydrofurfural (19)
FOL (96)
Pd/SiO2(N)
Pd−Ir/SiO2
Ir/SiO2
b
Ir/SiO2
FOL (99)
FOL hydrogenation (FOL = 5 mmol)
6
7
8
9
Pd/SiO2
50
50
10
50
11
84
46
2
THFA (99)
THFA (96)
THFA (99)
THFA (64)
Pd/SiO2(N)
Pd−Ir/SiO2
Ir/SiO2
furan hydrogenation (furan = 50 mmol)
10
11
12
13
Pd/SiO2
50
10
10
50
3.1
5.8
5.8
0.2
tetrahydrofuran (99)
Pd/SiO2(N)
Pd−Ir/SiO2
Ir/SiO2
tetrahydrofuran (99)
tetrahydrofuran (99)
tetrahydrofuran (39), 1-butanol (61)
a
b
Other conditions: water (9 g), H2 (8 MPa), 275 K, 1 h. 313 K, 24 h.
2723
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726

ACS Catalysis
Research Article
Table 4. Kinetic Parameters in the Hydrogenation of Furfural under 8 MPa H2 at 275 K
catalyst
Pd−Ir/SiO2 (Ir/Pd = 1)
Pd/SiO2
Pd/SiO2(N)
60
Ir/SiO2
SFOL (%)
78
88
96
a
k1 (mmol gcat min−1
)
)
11 (9−13)
3.9
0.7 (0.3−1.8)
0.18
1.4 (1.2−1.6)
1.4
0.2
−1
k2 (mmol gcat min−1
0.03
≫1
0.21
−1
a
Kfurfural/KFOL
0.5 (0.36−0.6)
0.38
0.3 (0.1−0.7)
0.19
1.0 (0.7−1.4)
0.19
−1
total metal (mmol gcat
)
b
c
particle size (nm)
3.5
7
4.1
3.6
c
CO adsorption (CO/(Ir + Pd))
0.16
0.04
0.19
0.23
d
TOFCO (104 h−1
)
1.1 (0.9−1.3)
0.4
0.6 (0.2−1.4)
0.1
0.2 (0.20−0.27)
0.2
0.03
e
TOFCC (104 h−1
)
0.005
a
Determined by fitting. The numbers in parentheses are the error range (see section 2.2 for the determination of the error range in the fitting).
b
c
d
Determined by XRD. Ref 59. Calculated by (k1)/[(total metal){CO/(Ir + Pd)}]. The numbers in parentheses are the error range derived from
e
fitting. Calculated by (k2)/[(total metal){CO/(Ir + Pd)}].
genation. For furan, the activities of Pd−Ir/SiO2 and Pd/
SiO2(N) was quite similar.
furfural concentration (Figure 1), the Kfurfural value should be
large (Kfurfural[furfural] ≫ 1). The values of k2 can be obtained
from the FOL hydrogenation data in Table 3, entries 6−9 on
the assumption that KFOL value is also large: KFOL[FOL] ≫ 1
and [furfural] = 0; eq 4 is simplified into d[FOL]/dt = −(W/
V) k2. We calculated k1 and Kfurfural/KFOL values by fitting to the
conversion and selectivity patterns (Table 4 and Figure 8).
When both furfural and FOL exist in the reaction media, the
relative reactivity can be changed by the relative adsorption
ability of these substrates on the catalyst surface. We analyzed
the selectivity change in the time course based on Langmuir−
Hinshelwood kinetic model. Different activation sites for
organic molecules and hydrogen were assumed because of
the positive reaction order with respect to H2 pressure and low,
not negative reaction order with respect to furfural concen-
tration (Figure 1). We have used similar model for the kinetic
analysis for gas-phase furfural hydrogenation over Ni/SiO2.41
We considered two reactions: furfural hydrogenation to FOL
(eq 1) and FOL hydrogenation to THFA (eq 2).
furfural + H2 → FOL + other products (tetrahydrofurfural + others)
(1)
FOL + 2H2 → THFA
(2)
On the basis of the model, the following equations for
concentration changes can be derived.
d[furfural]/dt = −(W/V)k1Kfurfural[furfural]
/(1 + Kfurfural[furfural] + KFOL[FOL])
(3)
d[FOL]/dt = (W/V)
{k1SFOL[furfural]/(1 + Kfurfural[furfural] + KFOL[FOL])
− k2KFOL[FOL]/(1 + Kfurfural[furfural] + KFOL[FOL])}
Figure 8. Fitted profiles of furfural hydrogenation. Markers:
experimental data (Table 1 and additional data for longer reaction
time); lines: calculated curves using kinetic parameters shown in Table
4
(4)
d[THFA]/dt = (W/V)k2KFOL[FOL]
/(1 + Kfurfural[furfural] + KFOL[FOL])
(5)
Here, k1 and k2 are the rate constants of furfural conversion
and FOL hydrogenation, respectively. W and V means the
catalyst weight and volume of reaction solution, respectively.
The term of the number of active sites per unit catalyst weight
is included in k1 and k2. The term of hydrogen pressure is also
included to these rate constants, because hydrogen pressure
was almost constant (8 MPa) in these reaction tests. Kfurfural and
For Pd/SiO2 with large crystal size, low Kfurfural/KFOL value
(∼0.3) was calculated, showing that FOL is adsorbed on Pd
surface considerably stronger than furfural. The stronger
adsorption of FOL than furfural on Pd(111) surface has been
already reported.62 According to the DFT calculations, all the
carbon atoms in the furan ring of furfural or FOL are bonded to
surface Pd atoms.63 The electron-withdrawing nature of the
carbonyl group may weaken the adsorption of the furan ring of
furfural on Pd.
On the other hand, Pd/SiO2(N) with smaller particle size
showed larger Kfurfural/KFOL ratio (1.0) than Pd/SiO2. The
difference between k1 values of these monometallic Pd catalysts
was rather small, and, because of the difference of the number
of surface atoms, the TOF value for CO hydrogenation of
Pd/SiO2(N) was not higher than that of Pd/SiO2. The lower
K
FOL are the equilibrium constant of adsorption of furfural and
FOL, respectively. SFOL and (1 − SFOL) are the selectivities to
FOL and (tetrahydrofurfural + others), respectively, in the first
step of the conversion of furfural.
The selectivity to (tetrahydrofurfural + others) of the total
system will be constant (1 − SFOL) during the time course in
this model, and therefore, SFOL can be obtained from the data in
Table 1. On the basis of the low reaction order with respect to
2724
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726

ACS Catalysis
Research Article
activity of Pd/SiO2 than Pd/SiO2(N) in furfural hydrogenation
was due to the fewer active sites and the blocking of active site
by produced FOL. In contrast, k2 and TOF for FOL
hydrogenation of Pd/SiO2(N) were surely larger than those
of Pd/SiO2. The high activity of Pd/SiO2(N) in furan ring
hydrogenation was also observed in hydrogenation of furan
itself (Table 3, entries 10 and 11). Hydrogenation activity is
known to be strongly related to the adsorption structure. The
sites directly bonded to the surface metal atoms are generally
reacted. However, too strong adsorption can rather decrease
the reactivity. The high activity of Pd/SiO2(N) in furan ring
hydrogenation may be due to weaker adsorption of furan ring.
In the case of Ir/SiO2, hydrogenation of FOL to THFA took
place to some extent when FOL alone was used as substrate,
whereas THFA was not produced at all in the hydrogenation of
ASSOCIATED CONTENT
■
S
* Supporting Information
Details of identification of products, TEM images of Pd/SiO2,
Pd/SiO2(N), and used Pd−Ir/SiO2 (Ir/Pd = 1), and the
histograms of size distributions. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: yoshinao@erec.che.tohoku.ac.jp. Fax/Tel.: +81-22-
795-7215.
*E-mail: tomi@erec.che.tohoku.ac.jp. Fax/Tel.: +81-22-795-
7214.
Notes
furfural even at high conversion. These data mean that Kfurfural
/
The authors declare no competing financial interest.
KFOL was very large. Similar weaker adsorption of FOL has
been reported for Cu and Ni surfaces, where the adsorption
with only the CO site of furfural is proposed.41,64,65 The
reactivity of adsorbed species on Ir (k1, k2, and TOF values)
was much lower than that on Pd. Reyes et al. reported that Ir/
TiO2 adsorbs crotonaldehyde at the positively charged Ir center
as [CO···surface] complex.66
ACKNOWLEDGMENTS
■
This work was in part supported by the Cabinet Office,
Government of Japan through it is “Funding Program for Next
Generation World-Leading Researchers”. We thank Prof. T.
Kizuka (University of Tsukuba) and Technical Division, School
of Engineering, Tohoku University for TEM observation.
In the case of Pd−Ir/SiO2, a moderate value (0.5) of Kfurfural
/
K
FOL ratio and very large k1 (or TOF for CO hydrogenation)
REFERENCES
was calculated. The k2 value or TOF for FOL hydrogenation
was also larger than those of monometallic Pd catalysts. In light
of the above considerations, it is believed that the addition of Ir
may promote the adsorption of CO site of furfural and
weaken the adsorption of furan ring. The schematic picture of
furfural adsorption on the catalysts is shown in Figure 9. Both
■
(1) Corma, A.; Ibbora, S.; Velty, A. Chem. Rev. 2007, 107, 2411−
2502.
(2) Chheda, J. N.; Hubewr, G. W.; Dumesic, J. A. Angew. Chem., Int.
Ed. 2007, 46, 7164−7183.
(3) Fernando, S.; Adhikari, S.; Chandrapal, C.; Murali, N. Energy
Fuels 2006, 20, 1727−1737.
(4) Ragauskas, A. J.; Williams, C. K.; Davison, B. H.; Britovsek, G.;
Cairney, J.; Eckert, C. A.; Frederick, W. J., Jr.; Hallett, J. P.; Leak, D. J.;
Liotta, C. L.; Mielenz, J. R.; Murphy, R.; Templer, R.; Tschaplinski, T.
Science 2006, 311, 484−489.
(5) Demirbas, M. F.; Balat, M. Energy Convers. Manage. 2006, 47,
2371−2381.
(6) Marshall, A.-L.; Alaimo, P. J. Chem.Eur. J. 2010, 16, 4970−
Figure 9. Proposed adsorption structure of furfural on catalyst surface.
4980.
(7) Besson, M.; Gallezot, P.; Pinel, C. Chem. Rev. 2014, 114, 1827−
1870.
of these effects on adsorption can increase the reaction rate: the
adsorption of CO site can increase k1, and the weakened
adsorption of furan ring can increase k2. The particle size may
also affect the adsorption strength of furan ring and k2 value.
(8) Schlaf, M. Dalton Trans. 2006, 4645−4653.
(9) Rinaldi, R.; Schuth, F. Energy Environ. Sci. 2009, 2, 610−626.
̈
(10) Kobayashi, H.; Ohta, H.; Fukuoka, A. Catal. Sci. Technol. 2012,
2, 869−883.
4. CONCLUSIONS
(11) Ruppert, A. M.; Weinberg, K.; Palkovits, R. Angew. Chem., Int.
Ed. 2012, 51, 2564−2601.
Simple coimpregnation of Pd and Ir on SiO2 gives Pd−Ir alloy
particles after reduction with H2 at 573 K. The Pd−Ir alloy
particles show very high catalytic activity in total hydrogenation
of furfural and 5-hydroxymethylfurfural in water. The
remarkable synergy between Pd and Ir is specific to
hydrogenation of substituted furans. High hydrogen pressure
and low reaction temperature are advantageous to obtain good
yields. The surface of Pd−Ir alloy catalyst has much higher
TOF than monometallic Pd catalyst with similar particle size,
especially for CO hydrogenation of furfural. Monometallic Ir
catalyst has much lower activity or TOF for both CO
hydrogenation of furfural and CC hydrogenation of furan
ring than Pd and Pd−Ir catalysts. In addition, in the presence of
furfural, the hydrogenation of furfuryl alcohol over Ir is totally
suppressed, suggesting the strong adsorption of CO group
on surface Ir atom. The adsorption of CO group on surface
Ir atom might also exist in Pd−Ir alloy catalyst, which can cause
the higher CO hydrogenation activity than Pd catalysts.
(12) Tong, X.; Ma, Yang; Li, Y. Appl. Catal., A 2010, 385, 1−13.
(13) Zakrzewska, M. E.; Bogel-Łukasik, E.; Bogel-Łukasik, R. Chem.
Rev. 2011, 111, 397−417.
(14) Karinen, R.; Vilonen, K.; Niemela, M. ChemSusChem 2011, 4,
̈
1002−1016.
(15) Lange, J.-P.; van der Heide, E.; van Buijtenen, J.; Price, R.
ChemSusChem 2012, 5, 150−166.
(16) Moreau, C.; Belgacem, M. N.; Gandini, A. Top. Catal. 2004, 27,
11−30.
(17) Lewkowski, J. Arkivoc 2001, 1, 17−54.
(18) Van Putten, R.-J.; van der Waal, J. C.; de Jong, E.; Rasrendra, C.
B.; Heeres, H. J.; de Vries, J. G. Chem. Rev. 2013, 113, 1499−1597.
(19) Nakagawa, Y.; Tamura, M.; Tomishige, K. ACS Catal. 2013, 3,
2655−2668.
(20) Medlin, J. W. ACS Catal. 2011, 1, 1284−1297.
(21) Wu, J.; Shen, Y.; Liu, C.; Wang, H.; Geng, C.; Zhang, Z. Catal.
Commun. 2005, 6, 633−637.
(22) Tamura, M.; Tokonami, K.; Nakagawa, Y.; Tomishige, K. Chem.
Commun. 2013, 49, 7034−7036.
2725
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726

ACS Catalysis
Research Article
(23) Ohyama, J.; Esaki, A.; Yamamoto, Y.; Arai, S.; Satsuma, A. RSC
Adv. 2013, 3, 1033−1036.
(58) Nakagawa, Y.; Shinmi, Y.; Koso, S.; Tomishige, K. J. Catal. 2010,
272, 191−194.
(59) Amada, Y.; Shinmi, Y.; Koso, S.; Kubota, T.; Nakagawa, Y.;
Tomishige, K. Appl. Catal., B 2011, 105, 117−127.
(60) Satsuma, A.; Osaki, K.; Yanagihara, M.; Ohyama, J.; Shimizu, K.
Appl. Catal., B 2013, 132−133, 511−518.
́
(24) Roman-Leshkov, Y.; Barrett, C. J.; Liu, Z. Y.; Dumesic, J. A.
Nature 2007, 447, 982−986.
(25) Nishimura, S.; Ikeda, N.; Ebitani, K. Catal. Today 2014, 232,
89−98.
(61) Hirasawa, S.; Watanabe, H.; Kizuka, T.; Nakagawa, Y.;
Tomishige, K. J. Catal. 2013, 300, 205−216.
(26) Chen, X.; Sun, W.; Xiao, N.; Yan, Y.; Liu, S. Chem. Eng. J. 2007,
126, 5−11.
(62) Pang, S. H.; Medlin, J. W. ACS Catal. 2011, 1, 1272−1283.
(63) Vorotnikov, V.; Mpourmpakis, G.; Vlachos, D. G. ACS Catal.
2012, 2, 2496−2504.
(27) Koso, S.; Furikado, I.; Shimao, A.; Miyazawa, T.; Kunimori, K.;
Tomishige, K. Chem. Commun. 2009, 2035−2037.
(28) Koso, S.; Ueda, N.; Shinmi, Y.; Okumura, K.; Kizuka, T.;
Tomishige, K. J. Catal. 2009, 267, 89−92.
(64) Sitthisa, S.; Sooknoi, T.; Ma, Y.; Balbuena, P. B.; Resasco, D. E.
J. Catal. 2011, 277, 1−13.
(29) Chen, K.; Koso, S.; Kubota, T.; Nakagawa, Y.; Tomishige, K.
ChemCatChem 2010, 2, 547−555.
(65) Sitthisa, S.; Pham, T.; Prasomsri, T.; Sooknoi, T.; Mallinson, R.
G.; Resasco, D. E. J. Catal. 2011, 280, 17−27.
́
(30) Chia, M.; Pagan-Torres, Y. J.; Hibbitts, D.; Tan, Q.; Pham, H.
́ ́
(66) Reyes, P.; Aguirre, M. C.; Melian-Cabrera, I.; Lopez Granados,
N.; Datye, A. K.; Neurock, M.; Davis, R. J.; Dumesic, J. A. J. Am. Chem.
Soc. 2011, 133, 12675−12689.
M.; Fierro, J. L. G. J. Catal. 2002, 208, 229−237.
(31) Chen, K.; Mori, K.; Watanabe, H.; Nakagawa, Y.; Tomishige, K.
J. Catal. 2012, 294, 171−183.
(32) Liu, S.; Amada, Y.; Tamura, M.; Nakagawa, Y.; Tomishige, K.
Green Chem. 2014, 16, 617−626.
́
(33) 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.
(34) Buntara, T.; Noel, S.; Phua, P. H.; Melian-Cabrera, I.; de Vries,
J. G.; Heeres, H. J. Top. Catal. 2012, 55, 612−619.
(35) Buntara, T.; Melian-Cabrera, I.; Tan, Q.; Fierro, J. L. G.;
́
́
Neurock, M.; de Vries, J. G.; Heeres, H. J. Catal. Today 2013, 210,
106−116.
(36) Chatterjee, M.; Kawanami, H.; Ishizaka, T.; Sato, M.; Suzuki, T.;
Suzuki, A. Catal. Sci. Technol. 2011, 1, 1466−1471.
(37) Seo, G.; Chon, H. J. Catal. 1981, 67, 424−429.
(38) Rodiansono, S.; Khairi, T.; Hara, T.; Ichikuni, N.; Shimazu, S.
Catal. Sci. Technol. 2012, 2, 2139−2145.
(39) Sitthisa, S.; Resasco, D. E. Catal. Lett. 2011, 141, 784−791.
(40) Connolly, T. J.; Considine, J. L.; Ding, Z.; Fordatz, B.; Jennings,
M. N.; MacEwan, M. F.; McCoy, K. M.; Place, D. W.; Sharma, A.;
Sutherland, K. Org. Process Res. Dev. 2010, 14, 459−465.
(41) Nakagawa, Y.; Nakazawa, H.; Watanabe, H.; Tomishige, K.
ChemCatChem 2012, 4, 1791−1797.
(42) Nakagawa, Y.; Tomishige, K. Catal. Commun. 2010, 12, 154−
156.
(43) Merat, N.; Godawa, C.; Gaset, A. J. Chem. Technol. Biotechnol.
1990, 48, 145−159.
́
(44) Alamillo, R.; Tucker, M.; Chia, M.; Pagan-Torres, Y.; Dumesic,
J. Green Chem. 2012, 14, 1413−1419.
(45) Biradar, N. S.; Hengne, A. M.; Birajdar, S. N.; Niphadkar, P. S.;
Joshi, P. N.; Rode, C. V. ACS Sustainable Chem. Eng. 2014, 2, 272−
281.
(46) Tuteja, J.; Choudhary, H.; Nishimura, S.; Ebitani, K.
ChemSusChem 2014, 7, 96−100.
(47) Shiavo, V.; Descotes, G.; Mentech, J. Bull. Soc. Chim. Fr. 1991,
128, 704−711.
̌
́ ́
B.; Metelko, B.; Sunjic, V. Tetrahedron Lett.
(48) Horvat, J.; Klaic,
1985, 26, 2111−2114.
(49) Patil, S. K. R.; Lund, C. R. F. Energy Fuels 2011, 25, 4745−4755.
(50) Ponec, V. Appl. Catal., A 1997, 149, 27−48.
(51) Maki-Arvela, P.; Haj
́
ek, J.; Salmi, T.; Murzin, D. Y. Appl. Catal.,
̈
A 2005, 292, 1−49.
(52) Lee, J.; Kim, Y. T.; Huber, G. W. Green Chem. 2014, 16, 708−
718.
(53) Zou, W.; Gonzalez, R. D. Catal. Lett. 1992, 12, 73−86.
(54) Mahmoud, S.; Hammoudeh, A.; Gharaibeh, S.; Melsheimer, J. J.
Mol. Catal. A 2002, 178, 161−167.
(55) Rocha, A. S.; Moreno, E. L.; da Silva, G. P. M.; Zotin, J. L.; Faro,
A. C., Jr. Catal. Today 2008, 133−135, 394−399.
́
́
(56) Lopez-De Jesus, Y. M.; Johnson, C. E.; Monnier, J. R.; Williams,
C. T. Top. Catal. 2010, 53, 1132−1137.
(57) Ziaei-azad, H.; Yin, C.-X.; Shen, J.; Hu, Y.; Karpuzov, D.;
Semagina, N. J. Catal. 2013, 300, 113−124.
2726
dx.doi.org/10.1021/cs500620b | ACS Catal. 2014, 4, 2718−2726