Performance and characterization of rhenium-modified Rh-Ir alloy catalyst for one-pot conversion of furfural into 1,5-pentanediol

Article abstract of DOI:10.1039/c4cy00161c

One-pot selective conversion of highly concentrated furfural to 1,5-pentanediol (1,5-PeD) was carried out over Rh-added Ir-ReO _x/SiO₂ catalysts through two-step reaction temperatures. Over the optimized catalyst, Rh(0.66 wt%)-Ir-ReO_x/SiO₂, the maximum yield of 1,5-PeD was 71.1% from highly concentrated furfural (50 wt%) and 78.2% from diluted furfural (10 wt%). These values were higher than those obtained with Ir-ReO_x/SiO₂ or Pd-Ir-ReO _x/SiO₂ catalysts. Rh-Ir-ReO_x/SiO₂ showed much higher activity in the hydrogenation of furfural to tetrahydrofurfuryl alcohol intermediate in the low temperature step than Ir-ReO_x/SiO₂, although the hydrogenation activity was lower than that of Pd-Ir-ReO_x/SiO₂. A long reaction time in the low temperature step is necessary to obtain a good 1,5-PeD yield over Rh-Ir-ReO_x/SiO₂ in two-step reaction of furfural. The hydrogenolysis activity of Rh-Ir-ReO_x/SiO₂ for tetrahydrofurfuryl alcohol to 1,5-PeD in the high temperature step was higher than that of Pd-Ir-ReO_x/SiO₂ and was comparable to that of Ir-ReO_x/SiO₂. The characterization results of TPR, XRD, XANES, EXAFS, STEM-EDX and FT-IR of adsorbed CO indicated that Rh-Ir-ReO _x/SiO₂ catalysts showed the structure of Ir-Rh alloy particles partially covered with ReO_x species. The hydrogenation activity of Rh-Ir-ReO_x/SiO₂ for the furan ring was higher than those of the mixture of Rh-Ir/SiO₂ and Ir-ReO _x/SiO₂ or the mixture of Rh-ReO_x/SiO ₂ and Ir-ReO_x/SiO₂. Both Ir-Rh alloy formation and ReO_x modification of alloy particles are essential for the high hydrogenation activity.

Full text of DOI:10.1039/c4cy00161c

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Performance and characterization of rhenium-
modified Rh–Ir alloy catalyst for one-pot conversion
of furfural into 1,5-pentanediol†
Cite this: Catal. Sci. Technol., 2014,
4, 2535
*
Sibao Liu, Yasushi Amada, Masazumi Tamura, Yoshinao Nakagawa
*
and Keiichi Tomishige
One-pot selective conversion of highly concentrated furfural to 1,5-pentanediol (1,5-PeD) was carried
out over Rh-added Ir–ReOx/SiO2 catalysts through two-step reaction temperatures. Over the optimized
catalyst, Rh(0.66 wt%)–Ir–ReOx/SiO2, the maximum yield of 1,5-PeD was 71.1% from highly concentrated
furfural (50 wt%) and 78.2% from diluted furfural (10 wt%). These values were higher than those obtained
with Ir–ReOx/SiO2 or Pd–Ir–ReOx/SiO2 catalysts. Rh–Ir–ReOx/SiO2 showed much higher activity in the
hydrogenation of furfural to tetrahydrofurfuryl alcohol intermediate in the low temperature step than
Ir–ReOx/SiO2, although the hydrogenation activity was lower than that of Pd–Ir–ReOx/SiO2. A long reaction
time in the low temperature step is necessary to obtain a good 1,5-PeD yield over Rh–Ir–ReOx/SiO2 in
two-step reaction of furfural. The hydrogenolysis activity of Rh–Ir–ReOx/SiO2 for tetrahydrofurfuryl alcohol
to 1,5-PeD in the high temperature step was higher than that of Pd–Ir–ReOx/SiO2 and was comparable
to that of Ir–ReOx/SiO2. The characterization results of TPR, XRD, XANES, EXAFS, STEM-EDX and FT-IR
of adsorbed CO indicated that Rh–Ir–ReOx/SiO2 catalysts showed the structure of Ir–Rh alloy particles
partially covered with ReOx species. The hydrogenation activity of Rh–Ir–ReOx/SiO2 for the furan ring
was higher than those of the mixture of Rh–Ir/SiO2 and Ir–ReOx/SiO2 or the mixture of Rh–ReOx/SiO2
and Ir–ReOx/SiO2. Both Ir–Rh alloy formation and ReOx modification of alloy particles are essential for
the high hydrogenation activity.
Received 8th February 2014,
Accepted 20th March 2014
DOI: 10.1039/c4cy00161c
www.rsc.org/catalysis
Hydrogenolysis of THFA into 1,5-PeD has been developed
using heterogeneous catalysts through multistep or one-step
1. Introduction
The sustainable production of transportation fuels and chem-
ical products from biomass and its derived substrates is
attracting more and more attention due to the shortage of
fossil resources and an increasing energy demand.1–8 Furanic
compounds are promising intermediates in the production of
non-petroleum-derived chemicals for their relatively low oxy-
gen content. Furfural is produced by hydration + dehydration
of hemicellulose from agricultural wastes and forest residues
on an industrial scale, and it is considered as one of the top
10 chemical opportunities in biorefinement from carbohy-
drates.9–12 Selective hydrogenation or total hydrogenation of
furfural can give furfuryl alcohol (FFA) and tetrahydrofurfuryl
alcohol (THFA),13–15 which are important building blocks in
the synthesis of diols, especially 1,5-pentanediol (1,5-PeD) that
can be used as monomers of polyesters and polyurethanes.16–19
reactions. Schniepp and Geller proposed a multi-step reaction
requiring the isolation and purification of intermediates and a
70% yield of 1,5-PeD was obtained.20 Tomishige and coworkers
reported one-step hydrogenolysis of THFA into 1,5-PeD using
Rh–ReOx/SiO2, Rh–MoOx/SiO2, Rh–ReOx/C or Ir–ReOx/SiO2
at lower temperature of 373–393 K, 8 MPa.21–27 The highest
yield of 1,5-PeD reached 94%.23 Dumesic et al.28 also
reported hydrogenolysis of THFA with good performance in
the production of 1,5-PeD using Rh–ReOx/C (Vulcan XC-72).
Chatterjee et al.29 achieved about 73% yield of 1,5-PeD using
Rh/MCM-41 in supercritical CO2. Also some researchers have
paid attention to the conversion of furfuryl alcohol into diols.
Adkins et al.30,31 conducted hydrogenolysis of furfuryl alcohol
(FFA) using CuCr2O4 catalyst at 448 K and 10–15 MPa with high
conversion, and 30% of 1,5-PeD and 40% of 1,2-pentanediol
(1,2-PeD) were obtained. Although hydrogenolysis of FFA or
THFA can give a high yield of 1,5-PeD, a step involving the
hydrogenation of furfural is required. So, direct conversion of
furfural into 1,5-PeD with high activity and selectively is more
useful. Lu et al.32 used a Pt–Li/Co2AlO4 catalyst to convert
furfural into 1,5-PeD directly in an alcohol solvent, however
Department of Applied Chemistry, School of Engineering, Tohoku University, 6-6-07,
Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan. E-mail: tomi@erec.che.tohoku.ac.jp,
yoshinao@erec.che.tohoku.ac.jp; Fax: +81 22 795 7214; Tel: +81 22 795 7214
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4cy00161c
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2535–2549 | 2535

View Article Online
Paper
Catalysis Science & Technology
the highest yield of 1,5-PeD was only 34.9%. Recently, we
reported the one-pot conversion of furfural into 1,5-PeD over
a Pd-added Ir–ReOx/SiO2 catalyst through a two-step tempera-
ture reaction and the highest yield of 1,5-PeD was 71.4%.33
Hydrogenation of furfural to THFA proceeds at a lower tem-
perature step and hydrogenolysis of THFA to 1,5-PeD proceeds
at a higher temperature step. However, the hydrogenolysis
activity of Pd–Ir–ReOx/SiO2 was lower than Ir–ReOx/SiO2.
the same procedure as that reported previously.33 All the catalysts
were used in powdered form with a granule size of <100 mesh.
2.2. Activity tests
Activity tests were performed in a 190 ml stainless steel auto-
clave with an inserted glass vessel. The catalyst was put into
an autoclave together with a spinner and an appropriate
amount of water and heated at 473 K with 8 MPa H2 for 1 h
for the reduction pretreatment. After the pretreatment, the
autoclave was cooled down, and hydrogen was removed. Fur-
fural (Wako Pure Chemical Industries, Ltd, >95%) was put
into the autoclave. The reaction mixture was composed of
two layers. The upper layer was furfural-saturated aqueous
solution and the bottom layer was furfural. After sealing the
reactor, the air content was purged by flushing three times
with H2 (1 MPa, 99.99%; Nippon Peroxide Co., Ltd.). The
reactor was pressurized with H2 to 2.0–10.0 MPa and then
heated to a set temperature T1 (303–333 K). The temperature
(T1) was kept for 1–8 h (time 1). After this step (low tempera-
ture step), the reaction temperature was increased to another
set temperature T2 (353–413 K). The temperature increase
took about 30 min. The temperature (T2) was kept for 0–48 h
(time 2). The parameters (T1, time 1, T2, and time 2) are described
for each result. The stirring rate was 500 rpm. After reaction,
the reactor was cooled down and the gases were collected in a
gas bag. The autoclave contents were transferred to a vial and
diluted with a mixture of ethanol and water, and then the
catalyst was separated by centrifugation and filtration. The
products were analyzed using two kinds of gas chromato-
graphs (Shimadzu GC-2014) equipped with FID. A TC-WAX
capillary column (diameter 0.25 mmϕ, 30 m) or an Rtx-1-PONA
capillary column (diameter 0.25 mmϕ, 100 m) were used for
separation. The products were also identified using GC-MS
(QP5050, Shimadzu). The products were furfuryl alcohol
(FFA), tetrahydrofurfuryl alcohol (THFA), 1,2,5-pentanetriol
(1,2,5-PeT), 1,5-pentanediol (1,5-PeD), 1,4-pentanediol (1,4-PeD),
1,2-pentanediol (1,2-PeD), 1-pentanol (1-PeOH), 2-pentanol
(2-PeOH), 2-methyltetrahydrofuran (2-MTHF), n-pentane, C–C
cracking products (such as methane) and other products that
could not be identified (“others”). The amount of gas prod-
ucts (n-pentane and methane) were very small and they were
included in “others”. The products were quantified by the
external standard method using an aqueous solution of each
product with known concentration. The conversion and selec-
tivity were calculated on a carbon basis similar to that reported
previously.35
It is known that furfural can be polymerized easily to form
humins under high reaction temperature and high concentra-
tion of furfural.34 Rearrangement reactions to 1,4-disubstituted
C5 compounds are also well known side-reactions in the reduction
of furfural.17,19 These side reactions decrease the target 1,5-PeD
production remarkably. However, conversion of highly concen-
trated furfural makes large-scale industrial production easier
in view of energy consumption. Therefore, conversion of highly
concentrated furfural into 1,5-PeD with high yield would shed
light on industrial applications.
To improve the hydrogenolysis activity and the yield of
1,5-PeD from highly concentrated furfural, in this work we
reinvestigated the additive effect of noble metal to Ir–ReOx/SiO2
catalyst for the one-pot aqueous phase conversion of furfural
into 1,5-PeD. As a result, it was found that Rh–Ir–ReOx/SiO2 is
more effective than Pd–Ir–ReOx/SiO2. We also characterized
Rh–Ir–ReOx/SiO2 with XRD, TPR, XANES, EXAFS, STEM-EDX
and FT-IR of adsorbed CO.
2. Experimental
2.1. Catalyst preparation
SiO2 (G-6, BET surface area 535 m2 g−1) supplied by Fuji Silysia
Chemical Ltd was used as the catalyst support. Rh–Ir–ReOx/SiO2
catalysts were prepared by a sequential impregnation method.
First, Rh/SiO2 samples were prepared by impregnating with
an aqueous solution of Rh(NO3)3 (Wako Pure Chemical Indus-
tries, Ltd.). After evaporating the solvent and drying at 383 K
for 12 h, they were impregnated with an aqueous solution of
H2IrCl6 (Furuya Metals Co., Ltd). After drying at 383 K for
12 h, the obtained Rh–Ir/SiO2 samples were impregnated with
an aqueous solution of NH4ReO4 (Soekawa Chemical Co., Ltd.).
These catalysts were calcined in air at 773 K for 3 h after drying
at 383 K for 12 h. The loading amounts of Ir and Re were
4.0 and 7.8 wt%, respectively (Re/Ir molar ratio = 2), and the
loading amount of Rh is shown in the parenthesis as weight
percent: i.e. Rh(0.66)–Ir–ReOx/SiO2 contains 0.66 wt% Rh.
Pd(0.66)–Ir–ReOx/SiO2 catalyst with loading amount of 0.66, 4.0
and 7.8 wt% for Pd, Ir and Re, respectively, was prepared with
mol of furfural after reaction
Conversion % 
100


mol of furfural charged
molproduct  C atoms in product
Yield of products %  C =
100


molfurfural charged  C atoms in furfural
2536 | Catal. Sci. Technol., 2014, 4, 2535–2549
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Yield of unidentified products [% − C] = (Conversion −
and Re as a scattering atom. Analyses of EXAFS data were
performed using a computer program (REX2000, ver. 2.5.9;
Rigaku Corp.). Error bars for each parameter were esti-
mated by stepping each parameter, while optimizing the other
parameters, until the residual factor becomes two times its
minimum value.40
the sum of yield of products) × 100
product formation rate mol h1


TOF CO h1
=




adsorption of CO mol


Transmission electron microscope (TEM) photographs and
energy dispersive X-ray (EDX) analysis were taken by means of
an instrument (JEM-2010F, JEOL) operated at 200 kV. The sample
powders after the reduction were dispersed in 2-propanol by a
supersonic wave and put on Cu grids for TEM observation
under atmosphere.
The used catalyst was washed with excess water and dried
in air, and then calcined at 573 K or 773 K for 3 h. A slight
loss (<8% in weight) was observed during the recovery pro-
cess and was compensated with fresh catalyst in each reuse
experiment. The amount of eluted metal during the reaction
was analyzed by inductively coupled plasma atomic emission
spectrometry (ICP-AES, Thermo Scientific iCAP 6500).
FT-IR spectra were recorded by a Nicolet 6700 FT-IR with
a resolution of 4 cm−1 in diffuse reflectance mode, using an
in situ cell (Thermo Spectra-Tech) equipped with a sample cup,
heater and ZnSe windows. The catalyst powder was placed in
the sample cup. The cell was connected to the gas flow system.
Pretreatment was carried out in H2 flow at 473 K for 1 h. CO
was introduced by pulses with N2 as a carrier gas at room
temperature until the CO coverage reached the saturation
level. The FT-IR spectra were collected at room temperature.
The amount of CO chemisorption was measured in a high-
vacuum system using a volumetric method in the vacuum line.
Before adsorption measurements, the catalysts were reduced
in H2 at 473 K for 1 h. Subsequently the adsorption equilib-
rium, the sample weight and the dead volume of the appara-
tus was about 1.1 kPa, 0.15 g and 43 cm3, respectively. The
amount of CO adsorption is represented as the molar ratio to
Rh (Rh/SiO2), Ir (Ir–ReOx/SiO2) and Ir + Rh (Rh–Ir–ReOx/SiO2).
2.3. Catalyst characterization
Temperature-programmed reduction (TPR) was carried out in
a fixed-bed reactor equipped with a thermal conductivity
detector using 5% H2 diluted with Ar (30 ml min−1). The
amount of catalyst was 0.05 g, and the temperature was
increased from room temperature to 1123 K at a heating rate
of 10 K min−1. X-Ray diffraction (XRD) patterns were recorded
by a diffractometer (Rigaku MiniFlex 600) under air.
The extended X-ray absorption fine structure (EXAFS) spec-
tra were measured at the BL01B1 station at SPring-8 with the
approval of the Japan Synchrotron Radiation Research Insti-
tute (JASRI; proposal no. 2013B1037). The storage ring was
operated at 8 GeV, and a Si (111) single crystal was used to
obtain a monochromatic X-ray beam. Two ion chambers for
I0 and I were filled with 85% N2 + 15% Ar and 50% N2 + 50% Ar,
respectively, for Re L3-edge and Ir L3-edge measurements,
and 100% Ar and 100% Ar for Rh K-edge measurement. We
prepared the samples after the reduction. The reduction pro-
cess was carried out in an autoclave. The standard reduction
conditions were the same as in activity tests. After cooling,
the catalyst powder was transferred to the measurement cell
in a glove bag filled with nitrogen. The thickness of the cell
filled with the powder was 2 mm to give an edge jump of 1.2,
1.7 and 0.2 for Ir L3-edge, Re L3-edge and Rh K-edge measure-
ment, respectively, in the case of Rh–Ir–ReOx/SiO2. The
EXAFS data were collected in transmission mode. For EXAFS
analysis, the oscillation was first extracted from the EXAFS
data using a spline smoothing method.36 Fourier transforma-
tion of the k3-weighted EXAFS oscillation from the k space to
the r space was performed to obtain a radial distribution
function. The inversely Fourier filtered data were analyzed
using a usual curve fitting method.37,38 For curve fitting anal-
ysis, the empirical phase shift and amplitude functions for
the Re–O, Ir–Ir and Rh–Rh bonds were extracted from data
for NH4ReO4, Ir powder and Rh foil, respectively. Theoretical
functions for the Ir–Rh, Re–Ir, Re–Rh and Rh–Ir bond were
calculated using the FEFF8.2 program.39 The Re–Ir, Ir–Ir and
Rh–Ir bonds are represented by the Re–Ir (or –Re), Ir–Ir (or –Re)
and Rh–Ir (or –Re), respectively, in the curve fitting results.
This is because it is very difficult to distinguish between Ir
3. Results and discussion
3.1 Conversion of furfural into 1,5-PeD over different catalysts
In our previous research, catalyst screening was conducted
using diluted furfural (10 wt%) and one-step temperature con-
ditions at 393 K.33 At those conditions, the Pd–Ir–ReOx/SiO2
catalyst showed higher yields of 1,5-PeD than other noble-
metal-added Ir–ReOx/SiO2 catalysts including Rh–Ir–ReOx/SiO2.
However, the highest 1,5-PeD yield was obtained with a two-
step reaction, 313 K for 8 h and 373 K for 72 h. In this study we
re-evaluated the additive effect of noble metal to Ir–ReOx/SiO2
using highly concentrated furfural (50 wt%) and a two-step
reaction (313 K for 8 h and 373 K for 24 h).
First, we investigated the reactions during the low tempera-
ture step and the heating to T2 over various catalysts (Table 1).
The Ir–ReOx/SiO2 catalyst gave FFA very selectively during
the low temperature step (entry 1).13 During the heating from
T1 to T2, the hydrogenation of furan rings didn't proceed over
Ir–ReOx/SiO2 catalyst (entry 8), and side-reactions such as
polymerization began to start, as reported earlier.33 On the other
hand, both Rh(0.66)–Ir–ReOx/SiO2 and Pd(0.66)–Ir–ReOx/SiO2
showed a good yield of THFA after the low-temperature step
(Table 1, entries 2–3). Compared to the Pd(0.66)–Ir–ReOx/SiO2
catalyst, which has been determined as the best catalyst in
the previous study,33 a larger amount of FFA remained over
Rh(0.66)–Ir–ReOx/SiO2, indicating that Rh(0.66)–Ir–ReOx/SiO2
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2535–2549 | 2537

View Article Online
Paper
Catalysis Science & Technology
Table 1 Conversion of furfural over different catalysts at low temperature step
Yield (%)
1,5- 1,4- 1,2- 1-
Conversion (%) PeD PeD PeD PeOH PeOH THFA FFA MTHF PeT
Time
T1/K 1/h
Time
T2/K 2/h
2-
2-
1,2,5-
Entry Catalysts
Others
1
2
3
4
Ir–ReOx/SiO2
313
Rh(0.66)–Ir–ReOx/SiO2 313
Pd(0.66)–Ir–ReOx/SiO2 313
8
8
8
8
—
—
—
—
—
—
—
—
>99.9
>99.9
>99.9
>99.9
0.0
0.4
0.0
0.2
0.0
0.2
1.6
0.1
0.3
0.4
0.3
0.5
0.2
0.2
0.1
0.2
0.0
0.0
0.0
0.0
0.5 96.9 0.6
58.2 14.4 1.4
0.0 <2.0
4.8 19.8
6.5 21.0
0.0 5.3
66.8
0.0 3.6
Rh(0.66)/SiO2 +
313
6.6 86.6 0.7
21.5 55.7 1.2
1.9 81.7 0.9
a
Ir–ReOx/SiO2
5
6
7
Rh(0.66)–Ir/SiO2 +
313
8
8
8
—
—
—
—
—
—
>99.9
>99.9
>99.9
0.3
0.0
0.1
0.7
0.0
0.5
0.5
0.0
0.3
0.4
0.0
0.1
0.0
0.0
0.0
0.0 19.8
0.0 15.6
1.4 36.7
b
Ir–ReOx/SiO2
Rh(0.66)–ReOx/SiO2 + 313
c
Ir–ReOx/SiO2
Pd(0.66)/SiO2 +
313
313
49.3
8.8 2.8
d
Ir–ReOx/SiO2
8
9
10
11
Ir–ReOx/SiO2
Rh(0.66)–Ir–ReOx/SiO2 313
Pd(0.66)–Ir–ReOx/SiO2 313
8
8
8
8
373
373
373
373
0
0
0
0
>99.9
>99.9
>99.9
>99.9
0.0
0.6
0.2
0.2
0.1
0.7
3.1
0.1
0.6
0.5
0.3
0.7
0.3
0.2
0.1
0.2
0.0
0.0
0.0
0.0
0.9 87.9 1.9
0.0 8.3
11.1 17.7
11.5 11.8
0.0 25.1
67.3
69.7
0.3 1.5
0.0 3.2
Rh(0.66)/SiO2 +
313
11.0 61.3 1.4
31.2 22.2 1.6
4.8 59.9 2.7
a
Ir–ReOx/SiO2
12
13
14
Rh(0.66)–Ir/SiO2 +
313
8
8
8
373
373
373
0
0
0
>99.9
>99.9
>99.9
0.4
0.1
0.3
0.8
0.6
2.9
0.7
0.5
0.4
0.5
0.4
0.1
0.0
0.0
0.1
1.4 41.2
0.0 30.9
6.8 29.7
b
Ir–ReOx/SiO2
Rh(0.66)–ReOx/SiO2 + 313
c
Ir–ReOx/SiO2
Pd(0.66)/SiO2 +
313
56.4
0.1 3.4
d
Ir–ReOx/SiO2
PeD: pentanediol; PeOH: pentanol; THFA: tetrahydrofurfuryl alcohol; FFA: furfuryl alcohol; 2-MTHF: 2-methyltetrahydrofuran; PeT: pentanetriol;
others: unidentified products. Pretreatment: 473 K, H2 (8 MPa), 1 h. Reaction conditions: furfural (3 g), H2Ir = 2; 300 mg/Ir = 2; 300 mg), initial H2
b
(6 MPa). a Physical mixture of Rh(0.66)/SiO2 and Ir–ReOx/SiO2 with the same metal amount as Rh(0.66)–Ir–ReOx/SiO2. Physical mixture of
c
d
Rh(0.66)–Ir/SiO2 (300 mg) and Ir–ReOx/SiO2 (300 mg). Physical mixture of Rh(0.66)–ReOx/SiO2 (300 mg) and Ir–ReOx/SiO2 (300 mg). Physical
mixture of Pd(0.66)/SiO2 and Ir–ReOx/SiO2 with the same metal amount as Pd(0.66)–Ir–ReOx/SiO2.
has lower hydrogenation activity. The lower performance of
Rh–Ir–ReOx/SiO2 than Pd–Ir–ReOx/SiO2 in the previous report33
may be due to the too-short reaction time at low temperature
in the activity tests for catalyst selection. When heated
to 373 K, almost all the FFA was converted into THFA over
Rh(0.66)–Ir–ReOx/SiO2 (entry 9). However, the selectivities
to other products of Rh(0.66)–Ir–ReOx/SiO2 were lower than
those of Pd(0.66)–Ir–ReOx/SiO2, and the THFA yields were simi-
lar for both catalysts (entry 10). GC-MS analysis suggested that
the “others” of entries 2 and 3 contained significant amounts
of 5-hydroxy-2-pentanone and 1,5-dihydroxy-2-pentanone. These
compounds are precursors of 1,4-pentanediol (1,4-PeD) and
1,2,5-pentanetriol (1,2,5-PeT). The physical mixtures of
Rh(0.66)/SiO2 with Ir–ReOx/SiO2 showed only 6.6% yield
of THFA after the low temperature step (entry 4). The yield of
THFA was still very low even after the reaction temperature
reached T2 (entry 11). On the other hand, the physical mixtures
of Pd(0.66)/SiO2 with Ir–ReOx/SiO2 showed a much higher
yield of THFA than that of physical mixture of Rh(0.66)/SiO2
with Ir–ReOx/SiO2 (entries 4,7, 11 and 14). From these results,
it could be concluded that the synergistic interaction among
Rh, Ir and ReOx in Rh(0.66)–Ir–ReOx/SiO2 enhanced the
hydrogenation rate of the furan ring.
was obtained when the Rh loading amount was 0.66 wt%. In
these conditions, over Pd(0.66)–Ir–ReOx/SiO2 catalyst, a signifi-
cant amount of THFA intermediate remained in the reaction
mixtures (entry 9). When the reaction time was prolonged to
48 h, the yield of 1,5-PeD over Pd(0.66)–Ir–ReOx/SiO2 catalyst
was still lower than that over Rh(0.66)–Ir–ReOx/SiO2 catalyst
and a larger amount of 1-pentanol was formed (entry 10). The
physical mixture of Rh(0.66)/SiO2 and Ir–ReOx/SiO2 showed only
2.1% of 1,5-PeD (entry 11) while the mixture of Pd(0.66)/SiO2
and Ir–ReOx/SiO2 showed 35.0% yield of 1,5-PeD (entry 12).
Compared to Pd(0.66)–Ir–ReOx/SiO2, the synergistic interac-
tion among Rh, Ir and ReOx in Rh(0.66)–Ir–ReOx/SiO2 was
more obvious.
The catalysts were prepared with the sequential impregna-
tion method, and drying at 383 K was conducted between the
impregnation steps. We also prepared Rh(0.66)–Ir–ReOx/SiO2
samples with different preparation procedures: (A) successive
impregnation with calcination after each drying process; and
(B) co-impregnation. However, the activity tests showed that
both of samples A and B had lower hydrogenation activity
than the catalyst prepared by standard method (Table S1†).
It should be noted that a black precipitate was formed when
the three precursors were mixed together in water. Prepara-
tion by these procedures ((A) or (B)) was unsuitable to develop
synergistic interactions.
The reaction results after the high temperature step are listed
in Table 2. Rh(0.66)–Ir–ReOx/SiO2 catalyst showed a higher
1,5-PeD yield than Pd(0.66)–Ir–ReOx/SiO2 (entry 5). Different Rh
loading amount (entries 1–8) were tested. The best performance
Conversion of diluted furfural (10 wt%) was also conducted
over Rh(0.66)–Ir–ReOx/SiO2 and Pd(0.66)–Ir–ReOx/SiO2 catalysts.
2538 | Catal. Sci. Technol., 2014, 4, 2535–2549
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Table 2 Conversion of furfural over different catalysts from low temperature step to high temperature step
Yield (%)
Time
T1/K 1/h
Time
T2/K 2/h
1,5-
1,4-
1,2-
1-
2-
2-
1,2,5-
Entry Catalysts
Conversion (%) PeD PeD PeD PeOH PeOH THFA FFA MTHF PeT Others
1
2
3
4
5
6
7
8
Ir–ReOx/SiO2
313
Rh(0.1)–Ir–ReOx/SiO2 313
Rh(0.3)–Ir–ReOx/SiO2 313
Rh(0.5)–Ir–ReOx/SiO2 313
Rh(0.66)–Ir–ReOx/SiO2 313
Rh(0.75)–Ir–ReOx/SiO2 313
8
8
8
8
8
8
8
8
8
8
8
373 24
373 24
373 24
373 24
373 24
373 24
373 24
373 24
373 24
373 48
373 24
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
1.4
1.1 0.3
0.4 0.0
0.7 0.3
7.3 0.5
7.2 0.3
7.5 0.3
8.0 0.3
5.4 0.2
6.8 0.3
3.3 0.3
13.2 0.7
0.4 0.1
0.4 9.7 1.2
20.3 0.4 3.5
8.3 0.0 2.2
7.8 0.0 2.3
7.1 0.0 2.4
9.3 0.0 2.0
14.7 0.0 2.1
10.3 0.0 2.7
20.9 0.0 3.7
5.0 0.0 5.0
14.5 0.3 2.4
0.0 85.4
14.4 44.5
9.4 7.3
6.5 6.3
3.6 9.0
2.8 10.7
3.7 10.6
2.8 19.3
4.6 16.0
3.3 5.4
2.6 12.2 1.1
58.6
5.2 1.2
4.1 0.9
3.9 0.6
3.6 0.7
3.4 0.7
3.5 0.6
6.1 0.5
5.4 0.4
64.4
65.8
62.7
59.3
53.6
44.5
61.5
Rh(1)–Ir–ReOx/SiO2
Rh(2)–Ir–ReOx/SiO2
Pd(0.66)–Ir–ReOx/SiO2 313
Pd(0.66)–Ir–ReOx/SiO2 313
313
313
9
10
11
Rh(0.66)/SiO2 +
313
313
313
2.1 12.7 0.6
5.7 61.1
a
Ir–ReOx/SiO2
12
13
14
Pd(0.66)/SiO2 +
8
8
8
373 24
373 24
373 24
>99.9
>99.9
>99.9
35.0
4.7 0.3
2.2 0.2
0.9 0.3
0.4 0.0
16.4 0.0 3.8
31.4 0.0 4.8
7.4 3.6 2.5
4.0 33.5
11.0 29.7
0.0 75.9
b
Ir–ReOx/SiO2
Rh(0.66)–Ir/SiO2 +
8.0 12.9 1.1
c
Ir–ReOx/SiO2
Rh(0.66)–ReOx/SiO2 + 313
0.1
9.4 0.6
d
Ir–ReOx/SiO2
15e
16e
17e
Rh(0.66)–Ir–ReOx/SiO2 313
Rh(0.66)–Ir–ReOx/SiO2 313
Pd(0.66)–Ir–ReOx/SiO2 313
8
8
8
373 24
373 32
373 24
>99.9
>99.9
>99.9
72.4
78.2
55.6
2.8 0.7
2.8 1.2
5.8 0.8
6.3 0.2
9.5 0.2
3.5 0.2
11.8 0.0 1.4
6.4 0.0 1.4
24.6 0.0 2.5
3.5 <2.0
2.0 <2.0
3.8 3.4
PeD: pentanediol; PeOH: pentanol; THFA: tetrahydrofurfuryl alcohol; FFA: furfuryl alcohol; 2-MTHF: 2-methyltetrahydrofuran; PeT: pentanetriol;
others: unidentified products. Pretreatment: 473 K, H2 (8 MPa), 1 h. Reaction conditions: furfural (3 g), H2Ir = 2; 300 mg/Ir = 2; 300 mg), ini-
b
tial H2 (6 MPa). a Physical mixture of Rh(0.66)/SiO2 and Ir–ReOx/SiO2 with the same metal amount as Rh(0.66)–Ir–ReOx/SiO2. Physical mix-
c
ture of Rh(0.66)–Ir/SiO2 (300 mg) and Ir–ReOx/SiO2 (300 mg). Physical mixture of Rh(0.66)–ReOx/SiO2 (300 mg) and Ir–ReOx/SiO2 (300 mg).
d
e
Physical mixture of Pd(0.66)/SiO2 and Ir–ReOx/SiO2 with the same metal amount as Pd(0.66)–Ir–ReOx/SiO2. Furfural (1 g), H2O (9 g), catalyst
(Re/Ir = 2; 100 mg), initial H2 (6 MPa).
The yields of 1,5-PeD over both Rh(0.66)– and Pd(0.66)–Ir–ReOx/SiO2
catalysts were higher and those of unidentified products were
lower than in the case of highly concentrated furfural, for the
intrinsic polymerization property of highly concentrated fur-
fural (entries 15 and 17). After 24 h high-temperature step
reaction, the yield of 1,5-PeD over Rh(0.66)–Ir–ReOx/SiO2 was
higher than that over Pd(0.66)–Ir–ReOx/SiO2 (entries 15 and 17).
When the reaction time was prolonged to 32 h, the highest
yield of 1,5-PeD over Rh(0.66)–Ir–ReOx/SiO2 reached 78.2% in the
low concentration of furfural conversion (entry 16). This value
was higher than the maximum yield over Pd(0.66)–Ir–ReOx/SiO2
catalyst (71.4%). In addition, the reaction time is much
shorter than the time required to obtain the highest yield over
Pd(0.66)–Ir–ReOx/SiO2 (72 h).
Pd(0.66)–Ir–ReOx/SiO2 than that over Rh(0.66)–Ir–ReOx/SiO2.
The high activity of the Pd-added catalyst may be due to the
strong interaction between the Pd metal and π electrons of
Fig. 1 shows the effect of reaction times on the conversion
of furfural over Rh(0.66)– and Pd(0.66)–Ir–ReOx/SiO2 cata-
lysts. For simplicity, the yields of the main products such as
FFA, THFA, 1,2,5-PeT and 1,5-PeD are displayed. The data in
detail are shown in Tables S2 and S3 (ESI†). In the low-
temperature step, almost all the furfural was converted over
both Rh(0.66)– and Pd(0.66)–Ir–ReOx/SiO2 catalysts in 1 h,
and the main products were FFA and THFA. FFA was further
hydrogenated into THFA, indicating that the formation
of THFA originated from furfural via selective hydrogena-
tion of furfural into FFA and subsequent hydrogenation of
furan rings. The rate of hydrogenation of furan ring as deter-
mined by the slope of THFA yield during 1–3 h was higher over
Fig. 1 Time course of conversion of furfural to 1,5-PeD over Rh(0.66)–
Ir–ReOx/SiO2 or Pd(0.66)–Ir–ReOx/SiO2.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2535–2549 | 2539

View Article Online
Paper
Catalysis Science & Technology
the FFA molecule resulting from rehybridization of sp2 to sp3
of C on the Pd surface.41 The yield of 1,2,5-PeT increased as
the reaction time was prolonged. The production of 1,2,5-PeT
originates from FFA by ring-opening acid hydrolysis reac-
tion and subsequent hydrogenation.42 The hydrogenation
reaction was almost complete over Pd(0.66)–Ir–ReOx/SiO2
catalyst after 8 h, and when heated to 373 K, the product dis-
tributions were almost the same over Pd(0.66)–Ir–ReOx/SiO2.
In the case of Rh(0.66)–Ir–ReOx/SiO2, hydrogenation of FFA
was not complete after 8 h. However, it proceeded quickly
during heating to T2. At the high-temperature step, 1,5-PeD
began to be produced over both catalysts. As the reaction time
increased, the yield of 1,5-PeD increased gradually while
THFA and 1,2,5-PeT decreased, suggesting that both THFA
and 1,2,5-PeT were the intermediates of 1,5-PeD. Based on a
reaction time of 4 h (as time 2), the formation rate of 1,5-PeD
over Rh(0.66)–Ir–ReOx/SiO2 catalyst was 2.3 times higher
than that of the Pd(0.66)–Ir–ReOx/SiO2 catalyst. The differ-
ence was even larger in TOF(CO) values (190 h−1/57 h−1).
These data indicate the higher hydrogenolysis activity of the
Rh(0.66)–Ir–ReOx/SiO2 catalyst. To compare the hydrogenolysis
activity of these catalysts more clearly, catalytic tests using
50 wt% THFA as reactant were also conducted (Table 3).
Indeed, the hydrogenolysis activity of Rh(0.66)–Ir–ReOx/SiO2
was higher than that of Pd(0.66)–Ir–ReOx/SiO2 and was com-
parable to that of Ir–ReOx/SiO2. The selectivities to 1,5-PeD
were very high for all of these three catalysts and no 1,2,5-PeT
was formed over all the catalysts. Therefore, the higher per-
formance of Rh(0.66)–Ir–ReOx/SiO2 originates from high
selectivity in the hydrogenation of furfural and high activity in
the hydrogenolysis reaction. The formation of the 1,2,5-PeT
intermediate was probably due to the hydrolysis and hydroge-
nation of FFA. The possible reaction routes are shown in
Scheme 1.
3.2 Effect of reaction conditions
Table 4 shows the effect of reaction temperature on produc-
tion of 1,5-PeD from furfural over Rh(0.66)–Ir–ReOx/SiO2
under 6 MPa of hydrogen. After the low temperature step, the
yield of THFA increased as T1 increased from 303 K to 323 K
and the yield decreased at 333 K (entries 1–4). At T1 = 333 K,
the selectivity to “others” was very high and the colour of
solution was pale yellow, while at lower T1 the solutions
were almost colourless. This suggests that polymerization of
furfural proceeded at high T1. After the high temperature
step, the yield of 1,5-PeD was lower and that of 1,4-PeD and
1,2,5-PeT was higher when higher T1 was applied (entries 5–8).
This indicated that high T1 leads to a rearrangement side reac-
tion to 1,4-PeD19 and a hydrolysis + hydrogenation reaction to
1,2,5-PeT.42 In addition, as T1 increased, yield of unidentified
products increased even after the high temperature step.
Therefore, lower T1 was preferable for to the production of
1,5-PeD. As the temperature of high-temperature step (T2)
increased from 353 K to 413 K, the yield of 1,5-PeD showed a
volcano-type performance and that of THFA and 1,2,5-PeT,
which are the intermediates of 1,5-PeD, decreased gradually
(entries 9–12). The overhydrogenolysis product, 1-pentanol,
increased quickly as the reaction temperature increased. As a
result, 373 K was optimum. It should be noted that the yield
of 1-pentanol reached 34.6% when reaction temperature was
413 K, while 1-pentanol is very attractive for gasoline compo-
nents for its moderate vapor pressure and low water solubility.43
Table 5 shows the effect of hydrogen pressure on the conver-
sion of furfural over Rh(0.66)–Ir–ReOx/SiO2. A large increases
in the 1,5-PeD yield and conversions of THFA and 1,2,5-PeT
occurred when the H2 pressure was elevated from 2 to 8 MPa,
consistent with the first-order dependence of hydrogenolysis
Table 3 Hydrogenolysis of THFA over different catalystsa
Selectivity (%)
Conversion
Catalysts (Re/Ir = 2)
(%)
1,5-PeD
1,2-PeD
1-PeOH
2-PeOH
Others
Ir–ReOx/SiO2
Pd(0.66)–Ir–ReOx/SiO2
Rh(0.66)–Ir–ReOx/SiO2
63.7
39.0
53.6
92.8
94.7
95.2
0.0
0.0
0.0
6.9
4.8
4.2
0.0
0.0
0.0
0.3
0.5
0.5
a
PeD: pentanediol; PeOH: pentanol; others: 2-methyltetrahydrofuran + 1-butanol + pentane. Pretreatment: 473 K, H2 (8 MPa), 1 h. Reaction
conditions: THFA (3 g), H2O (3 g), catalyst (300 mg), H2 (6 MPa), 373 K, 2 h.
Scheme 1 Possible reaction routes of conversion of furfural.
2540 | Catal. Sci. Technol., 2014, 4, 2535–2549
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Table 4 Conversion of furfural over Rh(0.66)–Ir–ReOx/SiO2 at different temperaturea
Yield (%)
Time
Entry T1/K 1/h
Time
T2/K 2/h
Conversion (%) 1,5-PeD 1,4-PeD 1,2-PeD 1-PeOH 2-PeOH THFA FFA 2-MTHF 1,2,5-PeT Others
1
2
3
4
5
6
7
8
303
313
323
333
303
313
323
333
313
313
313
313
8
8
8
8
8
8
8
8
8
8
8
8
—
—
—
—
—
—
—
—
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
0.2
0.4
0.2
0.5
0.2
0.6
1.0
3.8
3.9
4.7
5.2
3.9
3.9
5.5
3.1
0.0
0.4
0.0
0.0
0.5
0.6
0.7
0.7
0.6
0.6
0.6
0.2
0.3
0.2
0.3
0.4
7.6
7.5
6.8
5.7
1.4
7.5
17.3
34.6
0.0
0.0
0.0
0.0
0.3
0.3
0.4
0.3
0.1
0.3
1.1
1.9
52.5
58.2
62.0
57.6
7.8
8.3
7.5
11.7
31.7
8.3
25.3 0.5
14.4 1.4
13.1 0.8
4.2 1.7
0.0 2.4
0.0 2.4
0.0 2.4
0.0 2.6
0.0 2.1
0.0 2.4
0.0 4.5
0.0 5.5
3.7
4.8
4.0
4.2
4.5
3.2
6.4
7.3
15.6
3.2
1.3
0.0
16.9
19.8
19.0
30.3
6.6
0.5
373 24
373 24
373 24
373 24
353 24
373 24
393 24
413 24
66.4
65.8
60.8
53.8
38.5
65.8
57.3
37.7
8.1
10.4
12.7
6.1
8.1
10.7
15.9
9
10
11
12
1.7
1.0
a
PeD: pentanediol; PeOH: pentanol; THFA: tetrahydrofurfuryl alcohol; FFA: furfuryl alcohol; 2-MTHF: 2-methyltetrahydrofuran; PeT: pentanetriol;
others: unidentified products. Pretreatment: 473 K, H2 (8 MPa), 1 h. Reaction conditions: furfural (3 g), H2Ir = 2; 300 mg/Ir = 2; 300 mg), ini-
tial H2 (6 MPa).
Table 5 Conversion of furfural over Rh(0.66)–Ir–ReOx/SiO2 at different pressurea
Yield (%)
Pressure/
MPa
Conversion
(%)
Entry
1,5-PeD
1,4-PeD
1,2-PeD
1-PeOH
2-PeOH
THFA
FFA
2-MTHF
1,2,5-PeT
Others
1
2
3
4
5
2
4
6
8
10
>99.9
>99.9
>99.9
>99.9
>99.9
1.5
6.9
65.8
71.1
69.8
0.7
12.1
3.9
3.4
3.8
0.1
0.4
0.6
0.5
0.5
0.2
0.6
7.5
10.8
14.0
0.0
0.2
0.3
0.3
0.4
6.0
30.1
8.3
4.8
5.6
0.1
0.0
0.0
0.0
0.0
0.3
3.1
2.4
2.2
2.5
0.0
14.8
3.2
1.8
1.8
91.2
31.7
8.1
5.2
<2.0
a
PeD: pentanediol; PeOH: pentanol; THFA: tetrahydrofurfuryl alcohol; FFA: furfuryl alcohol; 2-MTHF: 2-methyltetrahydrofuran; PeT: pentanetriol;
others: unidentified products. Pretreatment: 473 K, H2 (8 MPa), 1 h. Reaction conditions: furfural (3 g), H2Ir = 2; 300 mg/Ir = 2; 300 mg), T1: 313 K,
time 1: 8 h, T2: 373 K, time 2: 24 h.
rate over Ir–ReOx/SiO2.27 The highest yield of 1,5-PeD was
71.1% when the pressure was 8 MPa (entry 4). The yield of
unidentified products decreased sharply as H2 pressure
increased. This is probably because low H2 pressure decreases
the hydrogenation rate of furfural to THFA, and the reaction
time of low temperature step becomes too short. However, the
hydrogen pressure increased further to 10 MPa, the yield of
1,5-PeD decreased slightly and the overhydrogenolysis product
1-pentanol increased with an increase in pressure (entry 5).
3.3 Catalyst stability
Table 6 lists the results of the activity test of the reused
Rh(0.66)–Ir–ReOx/SiO2. The method of the catalyst reuse was
Table 6 Reuse of Rh(0.66)–Ir–ReOx/SiO2 for conversion of furfural
Yield (%)
Usage Calcined
Re leaching
times temperature/K Conversion (%) 1,5-PeD 1,4-PeD 1,2-PeD 1-PeOH 2-PeOH THFA FFA 2-MTHF 1,2,5-PeT Others amount (%)
1
2
3
2
—
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
65.8
54.6
47.9
54.6
16.9
72.4
61.2
45.3
3.9
3.4
3.4
4.5
4.7
2.8
2.6
2.7
0.6
0.7
0.7
1.1
1.3
0.7
0.6
0.5
7.5
4.0
2.6
5.1
1.1
6.3
3.1
1.6
0.3
0.2
0.2
0.3
0.2
0.2
0.1
0.1
8.3
22.0
24.3
14.9
44.1
11.8
26.3
32.0
0.0 2.4
0.0 1.6
0.2 1.6
0.0 2.1
0.2 1.6
0.0 1.4
0.0 1.1
0.0 1.2
3.2
7.6
8.8
9.4
19.6
3.5
4.1
7.8
8.1
5.9
2.8
0.8
2.1
0.8
5.0
0.4
0.2
0.2
773
773
573
573
—
773
773
10.3
8.0
10.4
<2.0
<2.0
8.7
3
1a
2a
3a
PeD: pentanediol; PeOH: pentanol; THFA: tetrahydrofurfuryl alcohol; FFA: furfuryl alcohol; 2-MTHF: 2-methyltetrahydrofuran; PeT: pentanetriol.
Pretreatment: 473 K, H2 (8 MPa), 1 h. Calcination time before reuse: 3 h. Reaction conditions: furfural (3 g), H2Ir = 2; 300 mg/Ir = 2; 300 mg).
a
Furfural (1 g), H2O (9 g), catalyst (Re/Ir = 2; 100 mg). T1: 313 K, time 1: 8 h, T2: 373 K, time 2: 24 h.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2535–2549 | 2541

View Article Online
Paper
Catalysis Science & Technology
as follows: the catalyst was washed with water, dried at 383 K
for 12 h and calcined in air at 573 K or 773 K for 3 h. Before
the activity test, the reduction pretreatment was carried out in
the same way for the fresh catalyst. First, the used catalyst was
the results including the consumption of H2 and the calcu-
lated valence of Re are listed in Table 7. In the case of the
monometallic catalysts, Rh/SiO2 was reduced at very low tem-
perature (Fig. 2a). Ir/SiO2 was completely reduced at 600 K
(Fig. 2b) while ReOx/SiO2 was partially reduced at 800 K to
the valence of +1.6 (Fig. 2c).44 In the cases of the bimetallic
catalysts, Ir–ReOx/SiO2 showed only one large peak centered
at around 460 K (Fig. 2f).44,45 Rh–ReOx/SiO2 showed two peaks
between 365 K and 436 K (Fig. 2e). Rh–Ir/SiO2 showed one
broad peak at temperature range between Rh/SiO2 and Ir/SiO2
(Fig. 2d). In the case of trimetallic catalysts, Rh–Ir–ReOx/SiO2
showed only one broad signal when the loading amount of
Rh was below 1 wt%. The reduction temperature range was
lower than Ir–ReOx/SiO2 indicating that the presence of Rh
promoted the reduction of Ir–ReOx/SiO2. In the case of the
catalysts with loading amount of Rh ≤ 0.5 wt%, the H2 con-
sumption signal was shifted to the low temperature range
with increasing Rh amount (Fig. 2g–i). The reduction temper-
ature was almost unchanged by more Rh addition (Fig. 2j–l).
When the loading amount of Rh was 2 wt%, three signals
appeared (Fig. 2m). The two signals at lower temperature
may be assigned to Rh–ReOx/SiO2 reduction, suggesting that
Ir-free Rh–ReOx species existed on the Rh(2)–Ir–ReOx/SiO2
catalyst. The peak areas in the profiles of Ir–ReOx/SiO2 and
Rh–Ir–ReOx/SiO2 corresponded to the reduction of Re species
to the valence of 2–5 in addition to the total reduction of
Ir and Rh. The valence of Re in Rh–Ir–ReOx/SiO2 was lower
than that of Ir–ReOx/SiO2, demonstrating that the addition
of Rh promoted the reduction of Re species (Table 8). The
valence of Re species after reduction was also determined
by Re L3-edge XANES (3.1 for Ir–ReOx/SiO2 and 2.3–2.4 for
Rh–Ir–ReOx/SiO2). Similar to the TPR results, the valence of
Re in Rh–Ir–ReOx/SiO2 was lower than that of Ir–ReOx/SiO2.
The difference in the valence from the XANES analysis and
TPR can be explained by the difference of pretreatment condi-
44,45
calcined at 773 K, similar to the cases with Ir–ReOx/SiO2
or Pd–Ir–ReOx/SiO2.33 The 1,5-PeD yield was decreased with
reuse in both cases for concentrated and diluted furfural. This
is in contrast to the case of Pd–Ir–ReOx/SiO2 that showed sim-
ilar activity after reuses.33 To identify the cause of the catalyst
deactivation, inductively coupled plasma (ICP) and XRD char-
acterizations were conducted. ICP analysis of the reaction
solution after the catalyst had been removed by filtration
showed no leaching of Ir and Rh (below detection limit) and
slight leaching of Re (0.8–2.8% for each run, Table 6). The
XRD peaks (Fig. S1, ESI†) of the catalysts after the 3rd run
became sharper than those of the catalysts after reduction and
the 1st reaction, indicating the aggregation of the metal during
regeneration. The aggregation could be the reason for the
decrease in the 1,5-PeD formation. We further conducted the
reuse test with low calcination temperature (573 K) during
the regeneration process in order to suppress aggregation.
However, the 1,5-PeD yield was decreased significantly further
(Table 6). The XRD pattern of reused catalyst showed aggregation
similar to the case with 773 K calcination (Fig. S1†). There-
fore, the Rh–Ir–ReOx/SiO2 catalyst aggregates more easily than
Ir–ReOx/SiO2 and Pd–Ir–ReOx/SiO2, resulting in lower stability.
The more rapid deactivation of used catalyst after 573 K calci-
nation may be due to the formation of carbonaceous deposits
which can be removed by calcination at higher temperature.
3.4 Catalyst characterization
The profiles of temperature-programmed reduction (TPR)
of Rh–Ir–ReOx/SiO2, Ir–ReOx/SiO2, Rh–ReOx/SiO2, ReOx/SiO2,
Ir/SiO2 and Rh/SiO2 after calcination are shown in Fig. 2 and
tions between PH = 8 MPa in water in XANES/EXAFS measure-
2
ment and PH = 0.05 atm in N2 flow in TPR. A similar tendency
2
in the valence difference was also observed in the cases of
Ir–ReOx/SiO2,44–46 Rh–ReOx/SiO2 and Pt–ReOx/SiO2.48,49 The
47
spectra and the detailed analysis of Re L3-edge XANES are
shown in Fig. 3. We have reported that the Re L3-edge XANES
analysis of Pd–Ir–ReOx/SiO2, and the valence of Re was deter-
mined to be 3.1. The activity trend of these catalysts (Table 3);
Ir–ReOx/SiO2 > Rh–Ir–ReOx/SiO2 > Pd–Ir–ReOx/SiO2 for hydro-
genolysis is different from the order of the valence of Re.
Fig. 4 shows XRD patterns of Rh–Ir–ReOx/SiO2 after reac-
tion and Ir–ReOx/SiO2, Rh–ReOx/SiO2, Rh–Ir/SiO2, ReOx/SiO2,
Ir/SiO2 and Rh/SiO2 after reduction with 8 MPa H2 at 473 K
for 1 h. In the cases of monometallic catalysts, Ir/SiO2 showed
peaks at 2θ = 41.3°, 47.3° and 69.1° which were assigned to Ir
metal.50 Rh/SiO2 showed no diffraction peaks, because of the
low Rh amount and/or the high dispersion. ReOx/SiO2 showed
small peaks at 2θ = 37.5°, 40.5° and 42.9° which were assigned
to Re metal.51 It should be noted that most Re species were in
the oxidized state based on the TPR and XANES results. In the
cases of bimetallic catalysts, Rh–ReOx/SiO2 also showed no
Fig. 2 TPR profiles of the catalysts. (a) Rh/SiO2 (Rh = 0.66 wt%), (b)
Ir/SiO2 (Ir = 4 wt%), (c) ReOx/SiO2 (Re = 7.8 wt%), (d) Rh–Ir/SiO2
(Rh = 0.66 wt%, Ir = 4 wt%), (e) Rh–ReOx/SiO2 (Rh = 0.66 wt%,
Re = 7.8 wt%), (f) Ir–ReOx/SiO2 (Ir = 4 wt%, Re/Ir = 2), (g) Rh(0.1)–Ir–ReOx/SiO2,
(h) Rh(0.3)–Ir–ReOx/SiO2, (i) Rh(0.5)–Ir–ReOx/SiO2, (j) Rh(0.66)–Ir–ReOx/SiO2,
(k) Rh(0.75)–Ir–ReOx/SiO2, (l) Rh(1)–Ir–ReOx/SiO2 and (m) Rh(2)–Ir–ReOx/SiO2.
2542 | Catal. Sci. Technol., 2014, 4, 2535–2549
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Table 7 Summary of characterization results
Metal amount/
−1
CO adsorption/ Dispersionb
mmol gcat
TPR
Re L3-edge XANES
Valence of Rea Valence of Re
−1
−1
Catalyst
Rh
Ir
Re
H2 consumption/mmol gcat
mmol gcat
(%)
Ir–ReOx/SiO2
Rh–ReOx/SiO2
—
0.064
0.208 0.416 0.927
— 0.416 1.05
4.5
2.4
3.0
2.4
2.8
3.0
3.4
3.2
3.7
—
3.1
0.037
0.037
0.061
0.062
0.060
0.066
0.068
0.070
0.092
0.032
0.036
18
52
28
26
23
24
24
23
23
49
13
Rh(0.1)–Ir–ReOx/SiO2 0.010 0.208 0.416 1.27
Rh(0.3)–Ir–ReOx/SiO2 0.029 0.208 0.416 1.41
Rh(0.5)–Ir–ReOx/SiO2 0.049 0.208 0.416 1.36
Rh(0.66)–Ir–ReOx/SiO2 0.064 0.208 0.416 1.35
Rh(0.75)–Ir–ReOx/SiO2 0.073 0.208 0.416 1.28
Rh(1)–Ir–ReOx/SiO2
Rh(2)–Ir–ReOx/SiO2
Rh(0.66)/SiO2
2.4
0.097 0.208 0.416 1.36
0.194 0.208 0.416 1.40
0.064
2.3
2.4
—
—
—
Rh(0.66)–Ir/SiO2
0.064 0.208
—
—
—
a
7 − 2 × [(amount of H2 consumed, mol) − 2 × (Ir loading amount, mol) − 3/2 × (Rh loading amount, mol)]/(Re loading amount, mol).
Dispersion (%) = [CO adsorption amount]/[(Ir loading amount) + (Rh loading amount)] × 100.
b
Table 8 Curve fitting results of Ir L3-edge EXAFS of Ir–ReOx/SiO2 and Rh–Ir–ReOx/SiO2 after the reduction
Catalyst
Shells
CNa
R/10−1 nmb
σ/10−1 nmc
ΔE0/eVd
Rf (%)e
f
Ir–ReOx/SiO2
Ir–Ir (or –Re)
Ir–Ir (or –Re)
Ir–Rh
Ir–Ir (or –Re)
Ir–Rh
Ir–Ir (or –Re)
Ir–Rh
Ir–Ir
10.6 0.9
8.6 0.9
2.2 0.4
8.5 0.8
2.4 0.3
7.9 0.7
2.8 0.5
12
2.76 0.01
2.76 0.01
2.72 0.02
2.77 0.01
2.72 0.02
2.77 0.01
2.71 0.01
2.77
0.067 0.004
0.068 0.004
0.081 0.006
0.069 0.003
0.081 0.005
0.069 0.004
0.079 0.005
0.06
−1.0 1.2
−1.2 1.1
9.8 3.5
−0.2 1.2
9.2 3.5
0.6 1.3
9.9 2.7
0
0.9
1.6
Rh(0.66)–Ir–ReOx/SiO2
Rh(1)–Ir–ReOx/SiO2
Rh(2)–Ir–ReOx/SiO2
1.7
2.2
—
Ir powder
a
b
c
d
Coordination number. Bond distance. Debye–Waller factor. Difference in the origin of photoelectron energy between the reference and
the sample. e Residual factor. f After glycerol hydrogenolysis.44 Fourier filtering range: 0.163–0.325 nm.44
Fig. 3 Results of Re L3-edge XANES analysis of Ir–ReOx/SiO2 and Rh–Ir–ReOx/SiO2 after reduction. (I) Re L3-edge XANES spectra, (II) relation between
white line area and valence of Re. (a) Re powder, (b) Ir–ReOx/SiO2 after glycerol hydrogenolysis,44 (c) Rh(0.66)–Ir–ReOx/SiO2, (d) Rh(1)–Ir–ReOx/SiO2,
(e) Rh(2)–Ir–ReOx/SiO2 after the reduction, (f) ReO2, (g) ReO3 and (h) Re2O7.
peaks, suggesting that Rh and Re species were highly dis-
persed. Rh–Ir/SiO2 exhibited a pattern which was similar
to the Ir/SiO2 catalyst but the metal (111) peak was shifted
to a smaller angle indicating the formation of an Rh–Ir alloy.
Ir–ReOx/SiO2 showed a pattern whose peak positions were
almost the same as those of Ir/SiO2 while the peaks were
broadened by the addition of Re, as reported earlier.44 The
peaks corresponding to Re metal or ReOx were not detected at
all. These results indicate that Ir metal and highly dispersed
Re species are present and the average metal particle size of
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2535–2549 | 2543

View Article Online
Paper
Catalysis Science & Technology
existence of an Ir–Rh bond. The bond distance of the Ir–Rh
bond is about 0.271–0.272 nm, which lies between the Ir–Ir
bond (0.276 nm) and Rh–Rh (0.268 nm). At the same time,
the Ir–Ir bond distance in Rh–Ir–ReOx/SiO2 catalysts is almost
as same as that in Ir–ReOx/SiO2 metal. The behavior of these
bond lengths suggests the formation of an Ir–Rh alloy with
predominantly metallic Ir. The coordination number (CN)
of the Ir–Rh bond increased and the CN of the Ir–Ir
bond decreased with increasing Rh amount. However, the
sum of the coordination number (CN) of Ir–Rh and Ir–Ir in
Rh–Ir–ReOx/SiO2 was 10.7–10.9, which is comparable to
that of Ir–ReOx/SiO2, suggesting that the Ir–Rh alloy parti-
cle size of Rh–Ir–ReOx/SiO2 is almost the same as that of
Ir–ReOx/SiO2, which was determined to be 1.9 nm.44
Fig. 6 shows the Re L3-edge EXAFS spectra of Ir–ReOx/SiO2
after catalytic use and Rh–Ir–ReOx/SiO2 after reduction.
Table 9 lists the curve fitting results. The curve fitting analy-
sis indicated the presence of the Re–O, Re–Ir (or –Re) in all
of the samples, with a bond length (R) of 0.203–0.205 and
0.266–0.267 nm, respectively, and CN of 1.0–1.7 and 5.3–6.1,
respectively. The presence of the Re–O bond indicates that
the Re species is not fully reduced to the metallic state and
this is supported by TPR and Re L3-edge XANES analysis.
Although the Re–Rh bond on Rh(0.66)–Ir–ReOx/SiO2 was not
detected, both Rh(1)–Ir–ReOx/SiO2 and Rh(2)–Ir–ReOx/SiO2
showed an Re–Rh bond and the CN increased with increasing
Rh amount, indicating that the interaction between ReOx spe-
cies and Rh species exists. If the curve fitting was carried out
without an Re–Rh bond, the fitting was much worse than
that with Re–Rh bond (Table S5 and Fig. S3, ESI†), which
strongly supported the existence of an Re–Rh bond. The CN
of the Re–Ir (or –Re) bonds decreased with increasing Rh
content, which is explained by the increase in CN of the
Re–Rh bond.
Fig. 4 XRD patterns of catalysts. (a) Ir/SiO2 (Ir = 4 wt%), (b) Rh/SiO2
(Rh = 0.66 wt%), (c) ReOx/SiO2 (Re = 7.8 wt%), (d) Rh–ReOx/SiO2
(Rh = 0.66 wt%, Re = 7.8 wt%), (e) Rh–Ir/SiO2 (Rh = 0.66 wt%, Ir = 4 wt%),
(f) Ir–ReOx/SiO2 (Ir = 4 wt%, Re/Ir = 2) after reduction at 473 K for 1 h.
(g) Rh(0.1)–Ir–ReOx/SiO2, (h) Rh(0.3)–Ir–ReOx/SiO2, (i) Rh(0.5)–Ir–ReOx/SiO2,
(j) Rh(0.66)–Ir–ReOx/SiO2, (k) Rh(0.75)–Ir–ReOx/SiO2, (l) Rh(1)–Ir–ReOx/SiO2
and (m) Rh(2)–Ir–ReOx/SiO2 after reaction. (A) 2 theta range 30–75°,
(B) 2 theta range 38–46°.
Ir was decreased by the addition of Re.44 In the case
of Rh-added Ir–ReOx/SiO2 catalysts, the XRD patterns of
Rh–Ir–ReOx/SiO2 were similar to those of Ir–ReOx/SiO2 and
the peak of metal (111) plane (around 41°) was shifted gra-
dually to a smaller angle with increasing Rh amount, espe-
cially ≥1 wt% (Fig. 4B), from 41.3° (Ir–ReOx/SiO2) to 41.0°
(Rh(2)–Ir–ReOx/SiO2), suggesting that Rh–Ir alloy was formed.
Further characterization of Rh–Ir–ReOx/SiO2 catalysts
were carried out using Ir L3-edge, Re L3 edge and Rh K-edge
EXAFS (Tables 8–10) and the detailed spectra and analyses
are shown in Fig. 5–7. The Ir L3-edge curve fitting results of
Rh–Ir–ReOx/SiO2 after reduction were different from that of
Ir–ReOx/SiO2, and this can be explained by the presence of the
Ir–Rh bond, which suggests an interaction between Ir and Rh.
If the curve fittings were carried out without the Ir–Rh bond,
the fitting was much worse than that with the Ir–Rh bond
(Table S4 and Fig. S2, ESI†), which supported strongly the
The Rh K-edge EXAFS curve fitting analysis of Rh–Ir–ReOx/SiO2
showed that Rh was in the metallic state (Table 10 and Fig. 7).
The presence of both the Rh–Rh bond and Rh–Re (or –Ir) bond
was verified. The CN of the Rh–Rh bond increased with
increasing Rh content. If the curve fitting was carried out
without an Rh–Rh bond, the fitting was much worse than that
Table 9 Curve fitting results of Re L3-edge EXAFS of Ir–ReOx/SiO2 and Rh–Ir–ReOx/SiO2 after the reduction
Catalyst
Shells
CNa
R/10−1 nmb
σ/10−1 nmc
ΔE0/eVd
Rf (%)e
f
Ir–ReOx/SiO2
Re–O
Re–Ir (or –Re)
Re–O
Re–Ir (or –Re)
Re–O
Re–Ir (or –Re)
Re–Rh
Re–O
Re–Ir (or –Re)
Re–Rh
1.7 0.5
6.1 1.1
1.4 0.4
5.9 0.4
1.0 0.2
5.7 0.6
0.4 0.2
1.3 0.4
5.3 0.6
0.7 0.1
4
2.03 0.03
2.68 0.01
2.04 0.02
2.67 0.01
2.04 0.03
2.67 0.01
2.64 0.02
2.05 0.03
2.66 0.01
2.63 0.01
1.73
0.080 0.016
0.082 0.006
0.085 0.010
0.087 0.002
0.085 0.008
0.086 0.004
0.062 0.014
0.086 0.009
0.087 0.003
0.061 0.005
0.06
0.6 5.0
8.1 1.8
−2.1 3.6
6.6 1.2
−3.0 6.0
5.8 1.2
−1.3 6.2
−1.3 4.1
5.1 1.4
−6.0 2.9
0
2.5
Rh(0.66)–Ir–ReOx/SiO2
Rh(1)–Ir–ReOx/SiO2
1.9
2.2
Rh(2)–Ir–ReOx/SiO2
2.3
NH4ReO4
Re = O
—
a
b
c
d
Coordination number. Bond distance. Debye–Waller factor. Difference in the origin of photoelectron energy between the reference and
the sample. e Residual factor. f After glycerol hydrogenolysis.44 Fourier filtering range: 0.138–0.325 nm.
2544 | Catal. Sci. Technol., 2014, 4, 2535–2549
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Table 10 Curve fitting results of Rh K-edge EXAFS of Rh–Ir–ReOx/SiO2 after the reduction
Catalyst
Shells
CNa
R/10−1 nmb
σ/10−1 nmc
ΔE0/eVd
Rf (%)e
Rh(0.66)–Ir–ReOx/SiO2
Rh–Rh
Rh–Ir (or –Re)
Rh–Rh
Rh–Ir (or –Re)
Rh–Rh
Rh–Ir (or –Re)
Rh–Rh
2.0 0.5
7.0 0.6
2.8 0.6
6.2 0.6
4.9 0.5
4.2 0.9
12
2.68 0.02
2.69 0.01
2.68 0.01
2.69 0.01
2.68 0.01
2.69 0.01
2.68
0.087 0.007
0.078 0.003
0.085 0.008
0.082 0.003
0.082 0.004
0.086 0.004
0.06
−1.7 3.9
−1.8 1.1
−4.5 2.6
0.1 1.1
0.0 1.2
2.5 2.2
0
2.2
Rh(1)–Ir–ReOx/SiO2
Rh(2)–Ir–ReOx/SiO2
2.3
2.1
—
Rh foil
a
b
c
d
Coordination number. Bond distance. Debye–Waller factor. Difference in the origin of photoelectron energy between the reference and
the sample. e Residual factor. Fourier filtering range: 0.163–0.310 nm.
Fig. 5 Results of Ir L3-edge EXAFS analysis of Ir–ReOx/SiO2 and Rh–Ir–ReOx/SiO2 after the reduction. (I) k3-Weighted EXAFS oscillations. (II) Fourier
transform of k3-weighted Ir L3-edge EXAFS, FT range: 30–130 nm−1. (III) Fourier filtered EXAFS data (solid line) and calculated data (dotted line),
Fourier filtering range: 0.163–0.325 nm. (a) Ir powder, (b) IrO2, (c) Ir–ReOx/SiO2 after glycerol hydrogenolysis44 and (d–f) Rh–Ir–ReOx/SiO2 after the
reduction ((d) Rh = 0.66 wt%, (e) Rh = 1 wt% and (f) Rh = 2 wt%).
with an Rh–Rh bond (Table S6 and Fig. S4, ESI†), which
strongly supported the existence of an Rh–Rh bond. The coor-
dination number of the Rh–Ir bond and Rh–Rh bond in Rh
K-edge EXAFS gives information on the miscibility. In the
homogeneously mixed alloy, the ratio (relative to the Rh
atom) of the coordination number CNRh–Ir to CNRh–Rh is equal
to the mole fraction ratio nIr/nRh of the elements in the bulk,
it is difficult to determine CNRh–Ir directly from the results
of Rh K-edge EXAFS analysis, because the contribution of Ir
and Re as a backscattering atom is not distinguishable. There-
fore, CNRh–Ir was determined by using the relationship of
CNRh–Ir = nIr/nRh × CNIr–Rh. As a result, CNRh–Ir is calculated to be
7.2, 5.0 and 2.8 for Rh(0.66)–, Rh(1)– and Rh(2)– Ir–ReOx/SiO2,
respectively. In the case of Rh(0.66)– and Rh(1)–Ir–ReOx/SiO2,
the ratios of the coordination numbers CNRh–Ir/CNRh–Rh are
i.e. CNRh–Ir/CNRh–Rh = nIr/nRh.
52 In the case of Rh–Ir–ReOx/SiO2,
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2535–2549 | 2545

View Article Online
Paper
Catalysis Science & Technology
Fig. 6 Results of Re L3-edge EXAFS analysis of Ir–ReOx/SiO2 and Rh–Ir–ReOx/SiO2 after the reduction. (I) k3-Weighted EXAFS oscillations.
(II) Fourier transform of k3-weighted Re L3-edge EXAFS, FT range: 30–120 nm−1. (III) Fourier filtered EXAFS data (solid line) and calculated
data (dotted line), Fourier filtering range: 0.138–0.325 nm. (a) Re powder, (b) NH4ReO4, (c) Ir–ReOx/SiO2 after glycerol hydrogenolysis44 and
(d–f) Rh–Ir–ReOx/SiO2 after the reduction ((d) Rh = 0.66 wt%, (e) Rh = 1 wt% and (f) Rh = 2 wt%).
3.6 and 1.8, respectively, which are almost the same as the
actual mole fraction ratios nIr/nRh of 3.3 and 2.1, respectively.
This finding indicates that metallic Rh atoms can be mixed
well with Ir atoms. However, in the case of Rh(2)–Ir–ReOx/SiO2,
the ratio of the coordination number CNRh–Ir/CNRh–Rh is 0.6
which is much lower than the actual mole fraction ratio
nIr/nRh 1.1, indicating that the miscibility of Ir–Rh alloy was
low. There are two possible structures of Rh(2)–Ir–ReOx/SiO2
to explain the low miscibility of the Ir–Rh alloy. The first one
is that Rh metal particles can be partially formed without
interaction with the Ir–Rh alloy. The other one is the surface
segregation of Rh species in the Ir–Rh alloy particles. On the
other hand, TPR profiles gave three signals which is analo-
gous to Pd–Ir–ReOx/SiO2 which has been reported before,33
suggesting that Ir-free Rh metal particles modified with ReOx
species exist on the Rh(2)–Ir–ReOx/SiO2 catalyst.
particles are also partially covered with ReOx species existing
on the SiO2 support. This property is unique for the case of
Rh addition, judging from the absence of alloy formation in
the case of Pd and Ru addition.33,53
STEM-EDX analysis with element mapping of the
Rh(0.66)–Ir–ReOx/SiO2 catalyst was conducted to investigate
the distribution of the metallic nanoparticles on the silica
(Fig. 8). Fig. 8b and c shows images of the silica support. The
distribution of Si is similar to that of O. The blue area in the
image of Fig. 8d is the Ir metal phase. The green area of
Fig. 8e is attributed to Rh metal species, and the green area
overlaps well the blue one. It is highly likely that the two
metals formed an alloy in the catalyst. For Re (Fig. 8f), it is
difficult to determine the dispersion because ReOx can be
oxidized easily during the preparation of TEM-EDX samples.
The image suggested a higher concentration of Re on the
metal particles. The intensity of Re on the support might be
due to oxidized species.
Based on the above results from the structural analysis,
the Rh(0.66)– and Rh(1)–Ir–ReOx/SiO2 catalyst has a structure
in which the Ir–Rh alloy is formed and ReOx species are
located on the Ir–Rh metals. The miscibility of Ir–Rh alloy
was good. In Rh(2)–Ir–ReOx/SiO2 catalysts, the Ir–Rh alloy is
partially covered with ReOx species and small Rh metal
The FT-IR spectra of adsorbed CO on the in situ reduced
catalysts at room temperature are given in Fig. 9. The IR spec-
trum of Rh(0.66)/SiO2 interacting with CO at room tempera-
ture exhibited a strong band at 2072 cm−1, which can be
2546 | Catal. Sci. Technol., 2014, 4, 2535–2549
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Fig. 7 Results of Rh K-edge EXAFS analysis of Rh–Ir–ReOx/SiO2 after the reduction. (I) k3-Weighted EXAFS oscillations. (II) Fourier transform of
k3-weighted Rh K-edge EXAFS, FT range: 30–130 nm−1. (III) Fourier filtered EXAFS data (solid line) and calculated data (dotted line), Fourier
filtering range: 0.163–0.310 nm. (a) Rh foil, (b) Rh2O3, (c–e) Rh–Ir–ReOx/SiO2 after the reduction ((c) Rh = 0.66 wt%, (d) Rh = 1 wt% and
(e) Rh = 2 wt%).
Fig. 8 TEM image and STEM-EDX mappings of Rh(0.66)–Ir–ReOx/SiO2 after reduction. The maps reflect the measured (b) Si K intensity,
(c) O K intensity, (d) Ir M intensity, (e) Rh L intensity and (f) Re M intensity.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2535–2549 | 2547

View Article Online
Paper
Catalysis Science & Technology
and Rh(0.66)–ReOx/SiO2 + Ir–ReOx/SiO2, respectively). There-
fore, the high performance of Rh(0.66)–Ir–ReOx/SiO2 will be
due to the Ir–Rh alloy modified with ReOx species. This unique
structure improves the total hydrogenation performance in
the low temperature step which is the key step for the
production of 1,5-PeD. It should be noted that Rh catalysts
modified with metal oxide such as ReOx have high activity
in THFA hydrogenolysis, similar to the Ir–ReOx/SiO2 cata-
lyst.21,27 On the other hand, Pd has almost no activity in
THFA hydrogenolysis. The higher hydrogenolysis activity of
Rh–Ir–ReOx/SiO2 than Pd–Ir–ReOx/SiO2 may be due to the
difference in intrinsic activity between Rh and Pd.
Fig. 9 DRIFT spectra of CO adsorbed on the catalysts after reduction
at 473 K. (a) Rh/SiO2 (Rh = 0.66 wt%), (b) Ir–ReOx/SiO2 (Ir = 4 wt%,
4. Conclusions
Re/Ir
= 2), (c) Rh(0.66)–Ir–ReOx/SiO2, (d) Rh(1)–Ir–ReOx/SiO2 and
(e) Rh(2)–Ir–ReOx/SiO2.
In a one-pot two-step temperature reaction, addition of Rh to
Ir–ReOx/SiO2 catalyst (Rh–Ir–ReOx/SiO2) gave a higher yield of
1,5-PeD from furfural in a shorter total reaction time than the
addition of Pd (Pd–Ir–ReOx/SiO2). Rh(0.66)–Ir–ReOx/SiO2 showed
a higher activity for hydrogenolysis of the tetrahydrofurfuryl
alcohol intermediate to 1,5-PeD than Pd(0.66)–Ir–ReOx/SiO2
and the activity was comparable to that of Ir–ReOx/SiO2. For
the total hydrogenation of furfural to tetrahydrofurfural alcohol,
Rh(0.66)–Ir–ReOx/SiO2 has lower activity, but lower selectivity to
undesirable side products than Pd(0.66)–Ir–ReOx/SiO2. The
highest yield of 1,5-PeD was 71.1% from highly concentrated
furfural (50 wt%) and 78.2% from diluted furfural (10 wt%)
when using Rh (0.66 wt%)–Ir–ReOx/SiO2 as catalyst. The lower
performance of Rh–Ir–ReOx/SiO2 than Pd–Ir–ReOx/SiO2 in the
previous report33 may be due to the too-short reaction time for
hydrogenation (at low temperature) in the activity tests. In
view of stability, Rh–Ir–ReOx/SiO2 catalyst can aggregate more
easily than Pd–Ir–ReOx/SiO2.
Characterizations of TPR, XRD, XANES, EXAFS STEM-EDX
and FT-IR of adsorbed CO were performed. The results indi-
cated that the Rh(0.66)– and Rh(1)–Ir–ReOx/SiO2 catalysts
have the structure of an Ir–Rh alloy partially covered with
ReOx species. However, besides ReOx-species-modified Ir–Rh
alloys, small-size Rh particles modified with ReOx species
also existed on the Rh(2)–Ir–ReOx/SiO2 catalysts.
The high performance of Rh(0.66)–Ir–ReOx/SiO2 could be
attributed to its unique structure. The ReOx-species-modified
Ir–Rh alloy enhanced the hydrogenation rate of furfural to
the tetrahydrofurfuryl alcohol (THFA) intermediate in the low
temperature step as compared to Ir–ReOx/SiO2. THFA was
converted to 1,5-pentanediol by hydrogenolysis during the high
temperature step over the same ReOx-modified Ir–Rh alloy.
attributed to linear adsorbed CO; a doublet at 2092 cm−1 and
2034 cm−1, which can be assigned to the symmetric and
asymmetric carbonyl stretching frequencies of gem-dicarbonyl
Rh(I)(CO)2; and a broad peak at 1865 cm−1, which is assigned
to bridge-bonded CO.54 The formation of the dicarbonyl
species could be an indication of highly dispersed Rh since
it is widely accepted that the dicarbonyl species can only
be formed on highly dispersed rhodium.55,56 None of the
Rh–Ir–ReOx/SiO2 showed the peak assigned to the bridge CO
adsorption on Rh metal surface, but they did show similar
spectral patterns to that of Ir–ReOx/SiO2 with a broader
absorption band indicating that Rh was in the metallic state.
The position of the peak top was located between 2066 and
2069 cm−1 which can be assigned to the linear CO adsorbed
on the Ir–Rh alloy surface. In addition, metal Rh is also par-
tially covered with low-valent ReOx species which could not
adsorb CO. In the case of Rh(2)–Ir–ReOx/SiO2, a small adsorp-
tion band at 2099 cm−1 appeared. This peak is assigned to the
symmetric carbonyl stretching frequencies of gem-dicarbonyl
Rh(I)(CO)2, indicating that some Rh metal particles with
smaller size existed on the Rh(2)–Ir–ReOx/SiO2 catalyst.
According to the catalytic performance and the proposed
catalyst model, the formation of Ir–Rh alloy partially covered
with ReOx species on Rh(0.66)–Ir–ReOx/SiO2 catalysts played
an important role in catalytic performance. The physical mix-
ture of Rh(0.66)–Ir/SiO2 and Ir–ReOx/SiO2 catalysts showed
higher yield of THFA in the low temperature step than the
physical mixture of Rh(0.66)/SiO2 and Ir–ReOx/SiO2 catalysts or
the physical mixture of Rh(0.66)–ReOx/SiO2 and Ir–ReOx/SiO2
catalysts, indicating that Ir–Rh alloy enhanced the hydrogena-
tion activity of furan rings (Table 1, entries 4–6 and 11–13).
However, the yield of THFA was still much lower than that of
Rh(0.66)–Ir–ReOx/SiO2 (Table 1, entries 2, 5, 9 and 13). The
yield of the target product 1,5-PeD in the two-step reaction
(Table 2, entries 5 and 13) was also lower. The CO adsorption
amount of Rh(0.66)–Ir–ReOx/SiO2 (0.020 mmol) was similar to
that of these mixed catalysts (0.022, 0.021 and 0.022 mmol for
Rh(0.66)–Ir/SiO2 + Ir–ReOx/SiO2, Rh(0.66)/SiO2 + Ir–ReOx/SiO2
Acknowledgements
This work was in part supported by the Cabinet Office,
Government of Japan through it's “Funding Program for Next
Generation World-Leading Researchers”. Sibao Liu thanks the
China Scholarship Council (CSC) for the financial support.
The authors appreciate Dr Ya Xu for STEM-EDX observation
of the catalyst.
2548 | Catal. Sci. Technol., 2014, 4, 2535–2549
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
29 M. Chatterjee, H. Kawanami, T. Ishizaka, M. Sato, T. Suzuki
and A. Suzuki, Catal. Sci. Technol., 2011, 1, 1466–1471.
30 H. Adkins and R. Connor, J. Am. Chem. Soc., 1931, 53,
1091–1095.
References
1 G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
4044–4098.
2 J. C. Serrano-Ruiz, R. Luque and A. Sepulveda-Escribano,
Chem. Soc. Rev., 2011, 40, 5266–5281.
31 R. Connor and H. Adkins, J. Am. Chem. Soc., 1932, 54,
4678–4690.
3 P. Gallezot, Chem. Soc. Rev., 2012, 41, 1538–1558.
4 M. Schlaf, Dalton Trans., 2006, 4645–4653.
5 H. Kobayashi and A. Fukuoka, Green Chem., 2013, 15, 1740–1763.
6 A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107,
2411–2502.
7 A. M. Ruppert, K. Weinberg and R. Palkovits, Angew. Chem.,
Int. Ed., 2012, 51, 2564–2602.
8 M. Yabushita, H. Kobayashi and A. Fukuoka, Appl. Catal., B,
2014, 145, 1–9.
9 J. J. Bozell and G. R. Petersen, Green Chem., 2010, 12, 539–554.
10 A. S. Mamman, J. M. Lee, Y. C. Kim, I. T. Hwang, N. J. Park,
Y. K. Hwang, J. S. Chang and J. S. Hwang, Biofuels, Bioprod.
Biorefin., 2008, 2, 438–454.
32 W. Xu, H. Wang, X. Liu, J. Ren, Y. Wang and G. Lu, Chem.
Commun., 2011, 47, 3924–3926.
33 S. B. Liu, Y. Amada, M. Tamura, Y. Nakagawa and
K. Tomishige, Green Chem., 2014, 16, 617–626.
34 K. Zeitsch, The Chemistry and Technology of Furfural and its
Many By-Products, Elsevier Science, Amsterdam, 2000.
35 K. Y. Chen, M. Tamura, Z. L. Yuan, Y. Nakagawa and
K. Tomishige, ChemSusChem, 2013, 6, 613–621.
36 J. W. Cook and D. E. Sayers, J. Appl. Phys., 1981, 52, 5024–5031.
37 K. Okumura, J. Amano, N. Yasunobu and M. Niwa, J. Phys.
Chem. B, 2000, 104, 1050–1057.
38 K. Okumura, S. Matsumoto, N. Nishiaki and M. Niwa, Appl.
Catal., B, 2003, 40, 151–159.
11 R. Xing, A. V. Subrahmanyam, H. Olcay, W. Qi, G. P. van Walsum,
H. Pendse and G. W. Huber, Green Chem., 2010, 12, 1933–1946.
12 R. Weingarten, G. A. Tompsett, W. C. Conner and G. W. Huber,
J. Catal., 2011, 279, 174–182.
39 A. L. Ankudinov, B. Ravel, J. J. Rehr and S. D. Conradson,
Phys. Rev. B: Condens. Matter Mater. Phys., 1998, 58, 7565–7576.
40 K. Tomishige, K. Asakura and Y. Iwasawa, J. Catal.,
1994, 149, 70–80.
13 M. Tamura, K. Tokonami, Y. Nakagawa and K. Tomishige,
Chem. Commun., 2013, 49, 7034–7036.
41 R. Alamillo, M. Tucker, M. Chia, Y. Pagán-Torres and
J. Dumesic, Green Chem., 2012, 14, 1413–1419.
14 Y. Nakagawa and K. Tomishige, Catal. Commun., 2010, 12,
154–156.
15 Y. Nakagawa, H. Nakazawa, H. Watanabe and K. Tomishige,
ChemCatChem, 2012, 4, 1791–1797.
42 S. Sitthisa and D. E. Resasco, Catal. Lett., 2011, 141, 784–791.
43 D. E. Resasco, S. Sitthisa, J. F. T. Prasomsri and M. Pilar
ruiz, Heterogeneous in Biomass to Chemicals and Fuels,
Research Signpost, 2012, pp. 155–188.
16 Y. Nakagawa and K. Tomishige, Catal. Surv. Asia, 2011, 15,
111–116.
17 J. P. Lange, E. van der Heide, J. van Buijtenen and R. Price,
ChemSusChem, 2012, 5, 150–166.
44 Y. Amada, Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and
K. Tomishige, Appl. Catal., B, 2011, 105, 117–127.
45 Y. Nakagawa, X. H. Ning, Y. Amada and K. Tomishige, Appl.
Catal., A, 2012, 433–434, 128–134.
18 Y. Nakagawa and K. Tomishige, Catal. Today, 2012, 195,
136–143.
19 Y. Nakagawa, M. Tamura and K. Tomishige, ACS Catal.,
2013, 3, 2655–2668.
20 L. E. Schniepp and H. H. Geller, J. Am. Chem. Soc., 1946, 68,
1646–1648.
21 S. Koso, I. Furikado, A. Shimao, T. Miyazawa, K. Kunimori
and K. Tomishige, Chem. Commun., 2009, 2035–2037.
22 S. Koso, N. Ueda, Y. Shinmi, K. Okumura, T. Kizuka and
K. Tomishige, J. Catal., 2009, 267, 89–92.
46 Y. Amada, H. Watanabe, M. Tamura, Y. Nakagawa, K. Okumura
and K. Tomishige, J. Phys. Chem. C, 2012, 116, 23503–23514.
47 Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and
K. Tomishige, Appl. Catal., B, 2010, 94, 318–326.
48 T. Ebashi, Y. Ishida, Y. Nakagawa, S.-I. Ito, T. Kubota and
K. Tomishige, J. Phys. Chem. C, 2010, 114, 6518–6526.
49 Y. Ishida, T. Ebashi, S.-I. Ito, T. Kubota, K. Kunimori and
K. Tomishige, Chem. Commun., 2009, 5308–5310.
50 X. Hong, B. Li, Y. J. Wang, J. Q. Lu, G. S. Hu and M. F. Luo,
Appl. Surf. Sci., 2013, 270, 388–394.
23 K. Y. Chen, S. Koso, T. Kubota, Y. Nakagawa and
K. Tomishige, ChemCatChem, 2010, 2, 547–555.
24 S. Koso, Y. Nakagawa and K. Tomishige, J. Catal., 2011, 280,
221–229.
25 S. Koso, H. Watanabe, K. Okumura, Y. Nakagawa and
K. Tomishige, Appl. Catal., B, 2012, 111–112, 27–37.
26 S. Koso, H. Watanabe, K. Okumura, Y. Nakagawa and
K. Tomishige, J. Phys. Chem. C, 2012, 116, 3079–3090.
27 K. Y. Chen, K. Mori, H. Watanabe, Y. Nakagawa and
K. Tomishige, J. Catal., 2012, 294, 171–183.
51 H.-D. Kim, H. J. Park, T.-W. Kim, K.-E. Jeong, H.-J. Chae,
S.-Y. Jeong, C.-H. Lee and C.-U. Kim, Catal. Today, 2012, 185,
73–80.
52 http://www.asminternational.org/asmenterprise/apd.
53 M. Tamura, Y. Amada, S. B. Liu, Z. L. Yuan, Y. Nakagawa
and K. Tomishige, J. Mol. Catal. A: Chem., 2013, DOI:
10.1016/j.molcata.2013.09.015.
54 A. C. Yang and C. W. Garland, J. Phys. Chem., 1957, 61,
1504–1512.
55 P. Basu, D. Panayotov and J. T. Yates, J. Phys. Chem.,
1987, 91, 3133–3136.
28 M. Chia, Y. J. P. Torres, D. Hibbitts, Q. Tan, H. N. Pham,
A. K. Datye, M. Neurock, R. J. Davis and J. A. Dumesic, J. Am.
Chem. Soc., 2011, 133, 12675–12689.
56 P. Basu, D. Panayotov and J. T. Yates, J. Am. Chem. Soc.,
1988, 110, 2074–2081.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2535–2549 | 2549