Ir-Re alloy as a highly active catalyst for the hydrogenolysis of glycerol to 1,3-propanediol

Information for Authors .

View Journal

Catalysis

Science &

Technology

Accepted Manuscript

This article can be cited before page numbers have been issued, to do this please use: C. Deng, X.

Duan, J. Zhou, X. Zhou, W. Yuan and S. L. Scott, Catal. Sci. Technol., 2014, DOI: 10.1039/C4CY01285B.

This is an Accepted Manuscript, which has been through the

Royal Society of Chemistry peer review process and has been

accepted for publication.

Accepted Manuscripts are published online shortly after

acceptance, before technical editing, formatting and proof reading.

Using this free service, authors can make their results available

to the community, in citable form, before we publish the edited

article. We will replace this Accepted Manuscript with the edited

and formatted Advance Article as soon as it is available.

You can find more information about Accepted Manuscripts in the

Please note that technical editing may introduce minor changes

to the text and/or graphics, which may alter content. The journal’s

standard T erms & Conditions and the Ethical guidelines still

apply. In no event shall the Royal Society of Chemistry be held

responsible for any errors or omissions in this Accepted Manuscript

or any consequences arising from the use of any information it

contains.

www.rsc.org/catalysis

Catalysis Science & Technology

^{Page 1 of 9}Journal Name

Dynamic Article Links ►

DOI: 10.1039/C4CY01285B

Cite this: DOI: 10.1039/c0xx00000x

ARTICLE TYPE

Ir-Re alloy as a highly active catalyst for the hydrogenolysis of glycerol

to 1,3-propanediol

Chenghao Deng,^aXuezhi Duan,^aJinghong Zhou,*^aXinggui Zhou,^aWeikang Yuan^aand Susannah L.

Scott*^b

Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX

DOI: 10.1039/b000000x

In this work, bimetallic IrꢀRe catalysts supported on KITꢀ6 are prepared by tuning the thermal treatment

procedures, i.e., conventional calcination and reduction (IrꢀRe/KITꢀ6ꢀCR) and modified direct reduction

(IrꢀRe/KITꢀ6ꢀR) after impregnation of two metal precursors. The structure of both catalysts is intensively

10 characterized by H₂ꢀTPR, STEMꢀHAADFꢀEDX, XPS and COꢀDRIFTS. Results indicate that an IrꢀRe

alloy forms on the KITꢀ6 support when direct reduction is employed, which exhibits excellent catalytic

performance in hydrogenolysis of glycerol. The formation rate of 1,3ꢀpropanediol over IrꢀRe/KITꢀ6ꢀR

reaches 25.6 mol_1,3ꢀPDmol_Ir^ꢀ1h^ꢀ1at 63% glycerol conversion with the addition of amberlystꢀ15 under 8

MPa H₂, 393 K and 20 wt% glycerol aqueous solution, almost twice that over IrꢀRe/KITꢀ6ꢀCR. It is

15 revealed that Re species without prior calcination treatment could be fully reduced and therefore couple

with Ir to form IrꢀRe alloy structure with enhanced resistance against particle aggregation, while the

calcination and subsequent reduction leads to the formation of IrꢀReO_xstructure since the rhenium oxide

species generated during the calcination is difficult to be reduced.

effective in commercial application.

For the typical structureꢀsensitive reaction of glycerol

1. Introduction

hydrogenolysis, the catalytic performance is closely related to the

⁵⁰structures of the bimetallic catalysts.^2a,3c,4bDensity functional

theory calculation revealed that surfaceꢀalloyed RhꢀRe and atopꢀ

bonded ReO_xꢀRh clusters would have different properties, e.g.,

acidity, where the alloyed RhꢀRe was expected to exhibit better

performance in CꢀO bond hydrogenolysis reactions compared

⁵⁵with the RhꢀReO_x.¹⁵Inspired by the research above, for glycerol

hydrogenolysis over the bimetallic IrꢀRe based catalyst, where

the catalyst with IrꢀReO_xstructure have been employed and

extensively studied,^4a,4f,5it is reasonable to expect the appearance

of improved performance when alloying Ir with Re metal. It was

60 demonstrated that the lowꢀvalent ReO_xspecies forms after

reduction, and is attached to the Ir metal surface, in forms of Irꢀ

ReO_xstructure.^3a,5The iridium species in the reduced Irꢀ

ReO_x/SiO₂catalyst was reduced to metallic state, while the

oxidation state of rhenium species is ca. 2.⁵The interface between

65 the lowꢀvalent ReO_xand the Ir metal was proposed to be active

site for the coreꢀshell like nanoparticles.^3aSimilar situations were

proposed for RuꢀReO_xand RhꢀReO_xbased catalysts.^3a,3bThe type

of IrꢀReO_x/SiO₂catalyst has also been employed in many other

hydrogenation and hydrogenolysis reactions.³¹Yet, to the best of

70 our knowledge, the synthesis of IrꢀRe alloy, and its application in

glycerol hydrogenolysis has not been reported in the literature.

Therefore, it appears worthwhile to explore the preparation of Irꢀ

Re alloy structures and to exploit the synergy between Ir and Re

in glycerol hydrogenolysis. The relationship between the

75 bimetallic structure and the catalytic performance is also of great

20 Transformation of biomass and biomassꢀderivatives to fuels and

valuable chemicals provides sustainable alternatives to the

processes based on the fossil resources. Glycerol has been

proposed as one of the most valuable biomassꢀderived platform

feedstocks, since it is readily available in biodiesel process.

²⁵Tremendous efforts have been made to convert the biomassꢀ

derived glycerol to valuable fine chemicals.¹Hydrogenolysis of

glycerol to 1,3ꢀpropanediol (1,3ꢀPD) and 1,2ꢀpropanediol (1,2ꢀ

PD) is an effective way of glycerol valorization.^1,2It has been

shown that combination of Re or W species with a more

30 reducible transition metal (e.g., Ir, Pt, Rh or Ru) results in

excellent catalytic performance in the selective hydrogenolysis of

CꢀO bond, e.g., polyols.^3,4,11Specifically, the modification of Re

species on an Ir/SiO₂catalyst greatly improves the activity for

glycerol hydrogenolysis as well as the selectivity to 1,3ꢀPD,

35 which is the most valuable hydrogenolysis product.^4aThe H₂SO₄ꢀ

promoted IrꢀReO_x/SiO₂has shown the highest reported

productivity of 1,3ꢀPD so far (Table S1). However, the activity

for the nobleꢀmetal based catalyst is still far too low for

commercial application. Generally, high catalyst/glycerol loading

⁴⁰ratio and long reaction time are required for the batchꢀwise

reaction to achieve high glycerol conversion, rendering the

process inefficient. To date, several attempts have been made to

optimize the catalytic performance of IrꢀReO_x/SiO₂catalysts,

such as adjusting the Ir/Re composition, screening additives and

45 adding a third noble metal.^4f,4i,4jYet, considerable improvements

are needed to make such nobleꢀmetal based catalysts costꢀ

This journal is © The Royal Society of Chemistry [year]

[journal], [year], [vol], 00–00 | 1

CREATED USING THE RSC ARTICLE TEMPLATE (VER. 3.0) - SEE WWW.RSC.ORG/ELECTRONICFILES FOR DETAILS

Catalysis Science & Technology

Page 2 of 9

ARTICLE TYPE

DOI: 10.1039/C4CY01285B

interest.

temperature controller and a mechanical stirrer. In a standard

procedure, 20 g aqueous glycerol (20 wt% unless stated

otherwise) and 0.15 g preꢀreduced catalyst were loaded into the

In this work, Ir and Re are designed to be supported on KITꢀ6

(an ordered mesoporous silica with a cubic arrangement of

interconnected pores), because the silica support has weak ⁶⁰autoclave. In some cases, amberlystꢀ15 powder was used as

interaction with metal precursors and thus promotes the desired

formation of active IrꢀRe interfaces.^4f,4iThe bimetallic IrꢀRe

structure is manipulated by thermal treatment, obtaining the

target IrꢀRe alloy catalyst and the conventional IrꢀReO_xstructured

externally added acid promoter. The usage amount was 50 mg.

After sealing the reactor, it was purged three times with 2 MPa

H₂, then heated to 120 °C, pressurized with 8 MPa H₂, and stirred

at 500 rpm. After 12 h, the resulting reaction mixture was

catalyst, respectively, which are analyzed by characterizations of ⁶⁵analyzed by UPLC (Waters 2414) equipped with an RI detector

¹⁰H₂ꢀTPR, STEMꢀHAADFꢀEDX, XPS and COꢀDRIFTS.

Comparison between IrꢀRe alloy and IrꢀReO_xstructured catalyst

was performed to investigate the bimetallic structureꢀperformance

relationship in glycerol hydrogenolysis.

and a C18 AQ column. The used catalyst was collected by

centrifugation, washed extensively with deꢀionized water and

dried at 80 °C under vacuum. The carbon balance, representing

the ratio of molar of carbon in all liquid products to molar of

⁷⁰carbon in glycerol charged, was greater than 96% for each run.

Conversion and selectivity were calculated as follows:

2. Experimental

(mol of glycerol converted)

15 2.1 Catalyst preparation

Conversion (%) =

Selectivity (%) =

×100

(mol of glycerol charged)

KITꢀ6 was synthesized using P123 (Aldrich, MW=5800) as the

structureꢀdirecting agent and TEOS (Shanghai Lingfeng

(mol of product)× (number of carbon atoms in the product)

(Sum of carbon −based mol for all liquid products)

×100

Chemical Reagent co., Ltd., 99.0%) as the silica source,

according to previously published reports.^16,

In a typical

2.3 Catalyst characterization

20 preparation procedure, 4.0 g P123 was dissolved in a mixture of

144 g distilled water and 7.9 g 35 wt% HCl solution by stirring at ⁷⁵Before and after being used for glycerol hydrogenolysis, the

35 °C for 3 h. After complete dissolution of P123, 4.0 g nꢀBuOH

was added. Finally, 8.6 g TEOS was added to the mixture after

stirring for 1 h. The mixture was then stirred at 35 °C for a further

²⁵24 h before hydrothermal treatment at 110 °C in a polypropylene

metal loadings of the IrꢀRe based catalysts were determined by

inductively coupled plasma atomic emission spectroscopy (ICPꢀ

AES, Vanan 710). Powder Xꢀray diffraction (XRD) was

performed on a Rigaku D/Max 2550VB/PC diffractometer, using

bottle for another 24 h. The resulting precipitate was collected by ⁸⁰Cu K_αradiation. N₂physisorption measurements were performed

filtration and then calcined at 550 °C in air for 5 h. The physical

properties of the asꢀsynthesized KITꢀ6 are shown in Fig. S1.

Bimetallic IrꢀRe/KITꢀ6 catalysts were prepared by sequential

³⁰impregnation of KITꢀ6 with aqueous solutions of H₂IrCl₆(Sigmaꢀ

at 77 K on an ASAP 2010C (Micromeritics, USA), after outꢀ

gassing the samples at 573 K and 133.3 Pa for 6 h. The total pore

volume was defined as the singleꢀpoint pore volume at a relative

pressure p/p₀= 0.95. NH₃temperatureꢀprogrammed desorption

Aldrich, 99.9%) and NH₄ReO₄(SigmaꢀAldrich, 99.0%). In a 85 (NH₃ꢀTPD) was carried out to analyze the acidity of the asꢀ

typical procedure, Ir/KITꢀ6 was first prepared by incipient

wetness impregnation of KITꢀ6 with an aqueous solution of

H₂IrCl₆, then dried at 120 °C for 12 h. Dried IrꢀRe/KITꢀ6 was

35 then synthesized by subsequent impregnation of the dried Ir/KITꢀ

synthesized samples. Samples were preꢀreduced in hydrogen flow

at 500 ^oC for 3 h before introduction of NH₃.

TEM characterization was carried out using a Tecnai G2 F20

SꢀTwin (FEI) with an accelerating voltage of 200 kV and a point

6 with an aqueous solution of NH₄ReO₄. After drying, IrꢀRe/KITꢀ ⁹⁰resolution of 0.24 nm. The metal particle sizes and their

6 was reduced in flowing H₂(70 mL/min) at 500 °C for 3 h. This

uncalcined material after reduction is designated IrꢀRe/KITꢀ6ꢀR.

As a reference, the bimetallic catalyst was also prepared

⁴⁰following a conventional route. The dried IrꢀRe/KITꢀ6 was first

corresponding size distributions were estimated by measuring

more than 250 randomly selected particles. Images were also

captured in STEM mode with the annular darkꢀfield detector. The

EDX signals of the particles were obtained in STEM mode by

calcined at 500 °C for 3 h in air (designated IrꢀRe/KITꢀ6ꢀC), 95 focusing the electron beam on the selected areas or individual

followed by reduction in flowing H₂(70 mL/min) at 500 °C for 3

h. The resulting material is designated IrꢀRe/KITꢀ6ꢀCR. All

catalysts after reduction were passivated in 1% O₂/Ar for 20 min

45 (30 mL/min) at room temperature and then quickly transferred to

particles and accumulating the spectra. Both reduced and

passivated catalysts were used as samples for TEM observation.

H₂ꢀTemperatureꢀprogrammed reduction (H₂ꢀTPR) was carried

out using a Micromeritics AutoChem 2920, in a Uꢀshaped quartz

the reactor. The metal loadings for IrꢀRe/KITꢀ6, as determined by 100 reactor using 5% H₂diluted with Ar (30 mL/min). The Autochem

ICPꢀAES, are 4.0 wt% Ir and 3.8 wt% Re, which are exactly the

same as the nominal loadings, as shown in Table 1. For

comparison, monometallic Ir/KITꢀ6 and Re/KITꢀ6 catalysts were

⁵⁰also prepared by incipient wetness impregnation of KITꢀ6 with

is equipped with a thermal conductivity detector. The amount of

sample used was 30ꢀ70 mg, and the temperature was increased

from room temperature to 800 °C at the heating rate of 10

°C/min.

aqueous solutions of either H₂IrCl₆or NH₄ReO₄. The nominal 105 IR spectra were recorded on

a Nicolet 6700 FTIR

loading of Ir and Re for monometallic samples was 4.0 wt% and

3.8 wt%, respectively.

spectrophotometer in diffuse reflectance mode, using an inꢀsitu

cell equipped with a sample cup, heater and ZnSe windows. Preꢀ

reduced or passivated samples were placed into the sample cup

and reduced in situ at 350 °C in flowing H₂for 1 h. The sample

110 was cooled to room temperature, then 2% CO/Ar was introduced.

2.2 Glycerol hydrogenolysis

55 The catalyst evaluation was carried out in a 100ꢀmL stainless

steel autoclave (Parr Instruments) equipped with an electronic

Journal Name, [year], [vol], 00–00

This journal is © The Royal Society of Chemistry [year]

Catalysis Science & Technology

^{Page 3 of 9}Journal Name

Dynamic Article Links

►

DOI: 10.1039/C4CY01285B

Cite this: DOI: 10.1039/c0xx00000x

ARTICLE TYPE

After the CO coverage reached its saturation level, IR spectra

were collected at room temperature.

Xꢀray photoelectron spectroscopy (XPS) measurements were

conducted on a Multilab 2000 spectrometer (Thermo VH

Scientific) using Al K_αradiation (1486.6 eV). The aluminum

anode was operated at an accelerating voltage of 15 kV, 15 mA,

surface areas measured by N₂ꢀsorption (Table 1). The mean

³⁰particle size for IrꢀRe/KITꢀ6ꢀR is slightly smaller than that for Irꢀ

Re/KITꢀ6ꢀCR, i.e., 2.5 vs. 3.2 nm. Furthermore, the spectra of

STEMꢀEDX elemental analysis on randomly selected individual

metal particles reveal the coꢀexistence of both Ir and Re in a

single metal particle, indicating the close contact of Ir and Re in

20 V. The pressure in the analysis chamber was maintained in the ³⁵both materials (Figure 1, CꢀD).

range of 5×10^ꢀ9mbar. Binding energies were calibrated using the

binding energy of the C 1s peak (284.6 eV) as a reference.

10 Table 1 Physical properties of the materials

Sample

Ir/Re loading

(wt %) ^a

ꢀ

S_BET

V_p

d_p

d_av.

(m²/g) ^b(cm³/g) ^c(nm) ^c(nm) ^d

KITꢀ6

Ir/KITꢀ6ꢀR

IrꢀRe/KITꢀ6ꢀR

IrꢀRe/KITꢀ6ꢀCR

812

721

580

617

1.34

1.18

0.94

1.08

7.0

ꢀ

4.1

4.0/3.8

6.6 2.3 ± 0.29

6.9 2.5 ± 0.17

7.2 3.2 ± 0.33

Determined by ICPꢀAES.^bPore volume was calculated using the BET

method.^cPore diameter was calculated using the BJH method.^dMean

metal particle size was estimated from TEM observations.

3. Results and discussion

15 3.1 Catalyst characterization

Fig. 2 TPR profiles of uncalcined materials (A), and calcined materials

(B). Conditions: sample mass 30ꢀ70 mg; H₂/Ar 10 % v/v, 30 mL/min;

heating rate 10 °C/min.

H₂ꢀtemperatureꢀprogrammed

reduction

(H₂ꢀTPR)

was

performed to evaluate the influence of thermal treatment on the

reducibility of the metal species (Fig. 2). As shown in Fig. 2A,

the TPR profile of Ir/KITꢀ6 shows three main hydrogen

consumption peaks. The one appears at temperature lower than

⁴⁵200 C corresponds to the iridium species interacts weakly with

the KITꢀ6 support, while the peaks at higher temperature range

(200ꢀ430 ^oC) can be ascribed to iridium species interacts strongly

with the support.²⁶Re/KITꢀ6 displays two hydrogen consumption

peaks at around 300 ^oC and 430 C, ascribing to the the rhenium

⁵⁰species interacts weakly and strongly with the support,

respectively. The reduction peaks of IrꢀRe/KITꢀ6 shift to lower

temperature compared with that of the Re/KITꢀ6, indicating that

the reduction of rhenium species is promoted by Ir due to the

hydrogen spillover effects. On the other hand, as shown in Fig.

55 2B, calcined Ir/KITꢀ6ꢀC and Re/KITꢀ6ꢀC show respective

hydrogen consumption peak centered at 190 ^oC and 320 ^oC,

which can be ascribed to the reduction of IrO₂and rhenium oxide,

respectively.^4a,4fFor the calcined IrꢀRe/KITꢀ6ꢀC material, a shift

of the reduction peak to lower temperature range is observed

60 compared with the TPR profiles of the monometallic materials.

This phenomenon suggests that the close contact of Ir and Re

species promotes the reduction of both metal species.^4a,5

Fig. 1 Typical HAADFꢀSTEM images for IrꢀRe/KITꢀ6ꢀR (A) and Irꢀ

Re/KITꢀ6ꢀCR (B); Spot EDX elemental analysis on randomly selected

individual metal particles (marked by red arrows) for IrꢀRe/KITꢀ6ꢀR (C)

20 and IrꢀRe/KITꢀ6ꢀCR (D). Insets are the corresponding metal particle size

distribution obtained from STEM images.

Fig. 1 shows typical HAADFꢀSTEM images and corresponding

particle size distributions of the reduced IrꢀRe/KITꢀ6 catalysts

without and with calcination treatment, i.e., IrꢀRe/KITꢀ6ꢀR and

25 IrꢀRe/KITꢀ6ꢀCR, respectively. For both bimetallic materials, the

metal nanoparticles are highly dispersed on the KITꢀ6 support.

Some metal particles are observed locating inside the mesoporous

channels, accounting for the decrease in pore volumes and

The mean valence of each metal after reduction is estimated

from the amount of H₂consumed in H₂ꢀTPR, which is

65 summarized in Table 2. Ir precursors are readily reduced to the

metallic state, regardless of whether the material has been

This journal is © The Royal Society of Chemistry [year]

Journal Name, [year], [vol], 00–00 | 3

CREATED USING THE RSC ARTICLE TEMPLATE (VER. 3.0) - SEE WWW.RSC.ORG/ELECTRONICFILES FOR DETAILS

Catalysis Science & Technology

Page 4 of 9

ARTICLE TYPE

DOI: 10.1039/C4CY01285B

calcined or not. However, the reducibility of Re species varies

considerably, depending on which thermal treatment procedure

was employed. For the uncalcined materials (i.e., Re/KITꢀ6 and

to linearly adsorbed CO and bridged CO on the surface Ir,

respectively. For the Re/KITꢀ6ꢀR, a weak adsorption band at

2040 cm^ꢀ1due to the linearly adsorbed CO is observed. Because

IrꢀRe/KITꢀ6), Re species can be fully reduced to metallic state. ⁵⁵Re oxides cannot adsorb CO,¹⁸the result indicates the presence of

This is further supported by the XRD results. The XRD pattern of

Re/KITꢀ6ꢀR exhibits diffraction peaks assignable to the metallic

phase of Re (Fig. S2), suggesting the formation of Re crystal

even without the promotion of Ir by hydrogen spillover.

Furthermore, as shown in Fig. S3, the XRD pattern of the Irꢀ

surface metallic Re in Re/KITꢀ6ꢀR. Yet, Re/KITꢀ6ꢀCR shows no

CO adsorption bands, indicating that the surface Re species is in

oxidation state (not shown here).

¹⁰Re/KITꢀ6 without calcination treatment suggests that the Re

ꢀ

species exists as ReO₄species after drying, which is highly

reducible.

Table 2 Average metal valence, obtained by quantitative analysis of H₂

consumption in TPR profiles.

Sample

Valence of metal

ꢀ

Re ^a

ꢀ

Ir/KITꢀ6ꢀC

Re/KITꢀ6ꢀC

IrꢀRe/KITꢀ6ꢀC

Ir/KITꢀ6

4.7

2.4

ꢀ

Re/KITꢀ6

IrꢀRe/KITꢀ6

Re valence = 7ꢀ2×[(amount of H₂consumed, mol)ꢀ2×(Ir loading,

mol)]/(Re loading, mol).

60 Fig. 3 DRIFT spectra of CO adsorbed onto Ir/KITꢀ6ꢀR, IrꢀRe/KITꢀ6ꢀCR,

IrꢀRe/KITꢀ6ꢀR and Re/KITꢀ6ꢀR. The inset displays enlarged spectra for

IrꢀRe/KITꢀ6ꢀR and Re/KITꢀ6ꢀR.

In contrast, for the materials preꢀcalcined in air (i.e., Re/KITꢀ6ꢀ

C and IrꢀRe/KITꢀ6ꢀC), the extent of Re reduction is limited. For

the IrꢀRe/KITꢀ6ꢀC, the amount of H₂consumed cannot cover the

20 complete reduction of both metals. Since IrO₂is easily reduced to

metallic state in low temperature range, the mean valence of Re is

estimated by eliminating the amount of H₂consumed by

complete reduction of IrO₂to Ir.^4a,4f,24The valence of Re is

estimated to be 2.4 for the IrꢀRe/KITꢀ6ꢀCR, which is lower than

²⁵that of the Re/KITꢀ6ꢀCR (i.e., 4.7) because of the promotion of Ir

by hydrogen spillover. Such a valence of Re suggests the

formation of lowꢀvalent ReO_xspecies upon reduction, analogous

to previous reports for bimetallic IrꢀReO_x,^4a,4fRuꢀReO_x,^3aIrꢀ

For the bimetallic IrꢀRe materials, the band due to the bridged

CO on the surface Ir is absent. This indicates that the continuous

⁶⁵surface Ir ensembles are separated by Re species. For the Irꢀ

Re/KITꢀ6ꢀCR, CO adsorption band is observed at 2070 cm^ꢀ1,

ascribing to the linearly adsorbed CO on Ir.^4fThe band shape and

position are similar to that of the Ir/KITꢀ6ꢀR, but the intensity is

much lower. The decreased band intensity for the IrꢀRe/KITꢀ6ꢀ

⁷⁰CR can be rationalized to the partial coverage of lowꢀvalent ReO_x

on the particle surface.^3a,4f,24After all, if metallic Re was present

on the surface of IrꢀRe/KITꢀ6ꢀCR, the band should be at least

broadened. Accordingly, we conclude that the surface of the Irꢀ

Re/KITꢀ6ꢀCR is probably IrꢀReO_x.

VO_x²⁴, RhꢀReO_xand PtꢀReO_x.³⁰It has been proposed that, for

³⁰supported Re₂O₇catalyst, the Re reducibility was significantly

decreased by calcination treatment due to the formation of Re⁷⁺ꢀ

Oꢀsupport bond.^6ꢀ7As a result of the formation of the Re⁷⁺ꢀOꢀ

support bond, the mobility of Re species also decreased.

Tomishige et al. showed that the lowꢀvalent ReO_xspecies for the

35 preꢀcalcined IrꢀReO_x/SiO₂material was rather stable, and the

ReO_xspecies with the valence of ca. 2 remained almost

unchanged even when the reduction temperature was as high as

895 K.⁵This indicates that the interaction between the trioxo Re

species (ꢀOꢀRe⁷⁺) and support surface or IrO₂surface can be

40 rather strong. Based on these results, it can be concluded that the

For the IrꢀRe/KITꢀ6ꢀR, the CO adsorption band appears at

2055 cm^ꢀ1, an intermediate position between the band positions

for Ir/KITꢀ6ꢀR and Re/KITꢀ6ꢀR, suggesting the formation of Irꢀ

Re alloy. The observed redꢀshift of about 13 cm^ꢀ1of the linearly

adsorbed CO on Ir could be due to the increased electron backꢀ

80 donation from Ir to the adsorbed CO.^23,24This is because alloying

Ir with Re results in electron transfer from Re to Ir, leading to the

electron enrichment of Ir. Moreover, the intensity of the CO

adsorption band for the IrꢀRe/KITꢀ6ꢀR is also lower than that for

the Ir/KITꢀ6ꢀR. This may be a consequence of changed electronic

⁸⁵properties by alloying Ir with Re,²⁵and/or dilution of surface Ir

atoms by Re atoms on the IrꢀRe alloy surface because CO

adsorption ability for Re/KITꢀ6ꢀR is very weak.

The bulk IrꢀRe structure for the IrꢀRe/KITꢀ6ꢀR without

calcination treatment was further investigated using STEMꢀEDX

90 line scans on individual metal particles (Fig. 4). In order to

increase the EDX signal intensity, the metal loadings of the

studied catalyst were doubled, i.e., 8.0 and 7.6 wt% for Ir and Re,

respectively. The spectra shown in Fig. 4 confirm that both Ir and

ꢀ

highly reducible ReO₄species exists in the IrꢀRe/KITꢀ6 prior to

calcination, favoring the formation of IrꢀRe alloy during

reduction. On the other hand, the calcination treatment on Irꢀ

Re/KITꢀ6 immobilized the trioxo Re species on the support

45 and/or IrO₂metal surface, resulting in decreased reducibility of

Re species and the formation of IrꢀReO_xstructure during

reduction.

The surface properties of the bimetallic IrꢀRe materials were

investigated by DRIFT spectra of CO adsorbed on the reduced

⁵⁰materials (Fig. 3). Two CO adsorption bands at 2068 cm^ꢀ1and

1800 cm^ꢀ1are observed for the Ir/KITꢀ6ꢀR, which are assignable

Journal Name, [year], [vol], 00–00

This journal is © The Royal Society of Chemistry [year]

Catalysis Science & Technology

^{Page 5 of 9}Journal Name

Dynamic Article Links

►

DOI: 10.1039/C4CY01285B

Cite this: DOI: 10.1039/c0xx00000x

ARTICLE TYPE

Re are present in the same metal nanoparticle. Moreover, the

STEMꢀEDX spectrum (Fig. 4E) demonstrates the formation of

bulk IrꢀRe alloy for the IrꢀRe/KITꢀ6ꢀR.

Fig. 5 High resolution XPS scans for IrꢀRe/KITꢀ6ꢀR and IrꢀRe/KITꢀ6ꢀCR,

in the Ir 4f region (left panel), and the Re 4f region (right panel).

Based on these results, the influence of thermal treatment on

bimetallic IrꢀRe structure is unravelled. Alloyed IrꢀRe/KITꢀ6ꢀR

can be synthesized by direct reduction of the dried IrꢀRe/KITꢀ6

without calcination treatment. In contrast, IrꢀRe/KITꢀ6ꢀCR with

IrꢀReO_xstructure is synthesized when calcination treatment is

35 retained, which is analogous to the previous reports.^4a,4f,5The

evolution of the two types of IrꢀRe structure is illustrated in

Scheme 1. After drying the impregnated sample, Ir and Re

species are in close contact. The metal precursors existing as Ir⁴⁺

and ReO₄^ꢀspecies are highly reducible and mobile in the hydrated

40 layer on the support surface. Eliminating further calcination

treatment on the impregnated sample renders their reducibility

and mobility. As a result, the formation of IrꢀRe alloy is preferred.

In contrast, further calcination treatment on the impregnated

sample will strengthen the Re⁷⁺ꢀsupport and Re⁷⁺ꢀIrO₂

⁴⁵interactions via the formation of chemical bonds, and decrease

the reducibility and mobility of the Re species. The Re species

can be then partially reduced to ReO_xspecies in the valence of ca.

2.4, which is probably attached on the particle surface in forms of

IrꢀReO_x.

Fig. 4 STEMꢀEDX line scans on a randomly selected individual metal

particle in the IrꢀRe/KITꢀ6ꢀR catalyst: images before (A) and after EDX

line scanning (B); STEMꢀEDX spectra for Ir (C), Re (D); and combined

results for Ir (green) and Re (orange) (E).

The metal oxidation states in the reduced bimetallic IrꢀRe

10 materials were investigated by exꢀsitu XPS (Fig. 5). In both

materials, the presence of Ir 4f doublets at binding energies of

60.4 and 63.3 eV is the characteristic of Ir⁰(4f_7/2= 60.3 ± 0.2 eV,

4f_5/2= 63.4 ± 0.2 eV).^5,19aA small contribution from an Ir 4f

doublet assigned to Ir⁴⁺is also present, likely due to the oxidation

¹⁵of metallic Ir by contact with air during XPS sample preparation.

The Re 4f region indicates the presence of several Re oxidation

states in both materials. Because of the greater oxophilicity of Re

compared with Ir,^3e,4b,8ꢀ10Re metal is more easily oxidized during

XPS sample preparation, leading to the formation of Re⁴⁺, Re⁶⁺

20 and even Re⁷⁺. Nevertheless, metallic Re is observed for Irꢀ

Re/KITꢀ6ꢀR (4f_7/2= 39.7 ± 0.2 eV, 4f_5/2= 42.1 ± 0.2 eV),

suggesting the presence of Re⁰species on the bimetallic

surface.^19bIn contrast, the lowest energy Re 4f_7/2signal for Irꢀ

Re/KITꢀ6ꢀCR is located between the expected binding energies

25 for Re⁰and Re⁴⁺, suggesting the Re oxides with low positive

oxidation states.

Scheme 1 Possible evolution pathway of the bimetallic IrꢀRe structures

catalysts upon different thermal treatment procedures.

3.2 Catalytic performance

Glycerol hydrogenolysis was conducted in order to compare the

55 catalytic performance of different structured IrꢀRe/KITꢀ6

catalysts. As can be seen in Table 3, the activities for the two

catalysts are quite different. The glycerol conversion of the Irꢀ

Re/KITꢀ6ꢀR is higher than that of the IrꢀRe/KITꢀ6ꢀCR, i.e., 39.7%

This journal is © The Royal Society of Chemistry [year]

Journal Name, [year], [vol], 00–00 | 5

CREATED USING THE RSC ARTICLE TEMPLATE (VER. 3.0) - SEE WWW.RSC.ORG/ELECTRONICFILES FOR DETAILS

Catalysis Science & Technology

Page 6 of 9

ARTICLE TYPE

DOI: 10.1039/C4CY01285B

vs. 24.4% in 12 h under 8 MPa, 393 K and 20 wt% glycerol 50 PD to 1ꢀPO. Consequently, a higher ratio of 1,3ꢀPD to 1,2ꢀPD is

aqueous solution. The evaluation results at similar level of

glycerol conversion, i.e., 23±1% (Entry 1 and 3) are also

provided in order to compare the selectivity to 1,3ꢀPD over the

two catalysts. The selectivity to 1,3ꢀPD of the IrꢀRe/KITꢀ6ꢀR is

obtained by adding acid promoter.

Interestingly, as shown in Fig. 6, TEM images recorded for the

used IrꢀRe/KITꢀ6ꢀR indicate that the IrꢀRe alloy nanoparticles of

IrꢀRe/KITꢀ6ꢀR are still highly dispersed and stable during

ca. 10% higher than that of the IrꢀRe/KITꢀ6ꢀCR, i.e., 43.9% vs. ⁵⁵reaction, suggesting the high resistance against metal aggregation.

33.5%. This fact suggests that the bimetallic IrꢀRe catalyst with

different structures shows different catalytic performance,

probably due to their different catalytic active sites. The different

In contrast, large aggregates (>10 nm) are observed for the used

IrꢀRe/KITꢀ6ꢀCR. Meanwhile, as studied by ICPꢀAES analysis, no

detectable leaching of active metals occurred during the reaction.

¹⁰active sites are further verified by measuring the reaction order

with respect to glycerol (Fig. S4). Zero and first order with

respect to glycerol are obtained for IrꢀRe/KITꢀ6ꢀR and IrꢀRe/KITꢀ

6ꢀCR, respectively. As a result, the rate of 1,3ꢀPD formation over

the IrꢀRe/KITꢀ6ꢀR is almost twice that of the IrꢀRe/KITꢀ6ꢀCR.

¹⁵Notably, the rate of 1,3ꢀPD formation over the IrꢀRe/KITꢀ6ꢀR is

comparable to that reported for H₂SO₄ꢀpromoted IrꢀReO_x/SiO₂,⁴ⁱ

ꢀ1

i.e., 17.2 vs. 15.4 mol_1,3ꢀPDmol_Irh^ꢀ1, which can be further

enhanced to 25.6 mol_1,3ꢀPDmol_Ir^ꢀ1h^ꢀ1by adding amberlystꢀ15 as

promoter. In 12 h, the glycerol conversion achieved by

²⁰amberlystꢀ15 promoted IrꢀRe/KITꢀ6ꢀR is similar to that reported

for H₂SO₄ꢀpromoted IrꢀReO_x/SiO₂in 24 h, indicating the greatly

improved activity of the alloyed IrꢀRe catalyst compared with Irꢀ

ReO_x. To the best of our knowledge, the IrꢀRe/KITꢀ6ꢀR with

alloyed IrꢀRe structure is so far the most active catalyst for

25 glycerol hydrogenolysis (Table S1).

Table 3 Glycerol hydrogenolysis tests for IrꢀRe/KITꢀ6ꢀR and IrꢀRe/KITꢀ

6ꢀCR. ^a

Entry

Catalyst

Conv. 1,3ꢀPD Carbon

(%) formation balance

Selectivity (%)

1,3ꢀ 1,2ꢀ 1ꢀ 2ꢀ

PD PD PO PO

rate ^b

(%)

60 Fig. 6 TEM images of IrꢀRe/KITꢀ6ꢀR (A and C) and IrꢀRe/KITꢀ6ꢀCR (B

and D), after catalytic use. Reaction conditions: 120 C, 20 wt% aqueous

glycerol, 8 MPa H₂, 12 h.

IrꢀRe/KITꢀ6ꢀCR 24.4

IrꢀRe/KITꢀ6ꢀR 39.7

IrꢀRe/KITꢀ6ꢀR ^c22.4

IrꢀRe/KITꢀ6ꢀR 63.3

+Amberlystꢀ15

9.5

17.2

ꢀ

98.2 33.5 28.7 28.1 9.7

96.8 37.1 25.8 28.0 9.1

97.5 43.9 26.8 20.0 9.3

96.0 34.7 12.6 43.2 9.5

25.6

It has been shown that the monometallic Ir and Re was inactive

in glycerol hydrogenolysis.^4aThe remarkably enhanced activity

65 for Reꢀmodified Ir catalyst is ascribed to the synergy between the

Ir and Re. Chia et al. suggested that the combination of oxophilic

promoters such as Re species with an easily reducible metal, e.g.,

Pt, Rh, leads to the formation of bifunctional catalysts, exhibiting

both metallic and acid sites.^3b,11,25The acid sites, likely M(Pt, Rh,

70 Ir)ꢀReꢀOH groups at the metalꢀsolution interface form by

interaction between oxophilic Re and water molecule,^3b,11,25,28and

can be assessed by NH₃ꢀTPD technique.^3b,11,25Thus, the NH₃ꢀ

TPD technique was employed to assess the acidity generated

from (Ir)ꢀReꢀOH bond for bimetallic IrꢀRe catalyst. Furthermore,

75 the (Ir)ꢀReꢀOH is probably the adsorption site for glycerol,

analogous to the case of glycerol hydrogenolysis over PtꢀRe

based catalyst.^3cAs indicated from NH₃ꢀTPD profiles (Fig. S5),

the IrꢀRe/KITꢀ6ꢀR shows much higher acidity than IrꢀRe/KITꢀ6ꢀ

CR, suggesting the higher number of (Ir)ꢀReꢀOH sites. Therefore,

80 the higher activity for IrꢀRe/KITꢀ6ꢀR could be ascribed to the

higher number of glycerol adsorption site, (Ir)ꢀReꢀOH, which will

facilitate the reaction.

IrꢀReO_x/SiO₂61.1

15.4

43.2 9.2 36.6 9.7

ꢀ

+H₂SO₄

^aReaction conditions: 120 °C, 20 wt% aqueous glycerol, 8 MPa H₂, 12 h,

catalyst (150 mg), additive (50 mg). ^bIn mol_1,3ꢀPDmol_Ir^ꢀ1h^ꢀ1; ^c2 h reaction

30 time; ^d120 °C, 20 wt% aqueous glycerol, 8 MPa H₂, catalyst (150 mg),

H⁺/Ir=1 mol, 24 h reaction time from ref [4i]. PD: propanediol; PO:

propanol.

There are still large amount of byꢀproducts (1,2ꢀPD, 1ꢀPO and

35 2ꢀPO), accompanying with the production of 1,3ꢀPD. The 1,3ꢀPD

and 1,2ꢀPD are the primary products of glycerol hydrogenolysis,

which can be further converted to products of 1ꢀPO and/or 2ꢀPO.

Moreover, as shown in Table 3, with the addition of amberlystꢀ15

to IrꢀRe/KITꢀ6ꢀR as an acidic promoter, the selectivity to 1,2ꢀPD

40 remarkably decreases, while the selectivity to 1,3ꢀPD remains

almost unchanged (Entry 2 and 4). As a result, the ratio of 1,3ꢀPD

to 1,2ꢀPD increases from ca. 1.4 to 3.0 with the addition of

amberlystꢀ15. As shown in Table S2, the activity for IrꢀRe/KITꢀ

6ꢀR catalyzed hydrogenolysis of 1,2ꢀPD is higher than that of

45 glycerol and 1,3ꢀPD, which is ca. 2 times that of glycerol and ca.

6 times that of 1,3ꢀPD. Since the solid acid promoter could

enhance the activities for the hydrogenolysis of both glycerol and

propanediols, the much higher activity for 1,2ꢀPD compared with

glycerol and 1,3ꢀPD remarkably promotes the conversion of 1,2ꢀ

A possible reaction mechanism for the alloyed IrꢀRe/KITꢀ6ꢀR

catalyzed glycerol hydrogenolysis is proposed in Scheme 2. On

85 the surface of an IrꢀRe alloy particle, metallic Re activates the

water molecule to form an (Ir)ꢀReꢀOH absorption site.^11,15The

(Ir)ꢀReꢀOH acts as an absorption site for glycerol molecule.

Meanwhile, the hydride species formed from H₂activated on

Journal Name, [year], [vol], 00–00

Catalysis Science & Technology

^{Page 7 of 9}Journal Name

Dynamic Article Links

►

DOI: 10.1039/C4CY01285B

Cite this: DOI: 10.1039/c0xx00000x

ARTICLE TYPE

35 ^aState Key Laboratory of Chemical Engineering, East China University

of Science and Technology, Shanghai, 200237, P. R. China. E-mail:

jhzhou@ecust.edu.cn

metallic Ir attacks the CꢀO bond of the adsorbed glycerol to

generate unsaturated intermediates.^4f,27Subsequent hydrogenation

of these unsaturated intermediates leads to the formation of the

observed hydrogenolysis products, while further deoxygenation

^bDepartment of Chemical Engineering, University of California, Santa

Barbara, CA 93106-9510, USA. E-mail: sscott@engineering.ucsb.edu

of these unsaturated intermediates results in the formation of 40 † Electronic Supplementary Information (ESI) available: [N₂ꢀadsorption

of KITꢀ6, Table of currently developed catalytic system for glycerol

hydrogenolysis, XRD and NH₃ꢀTPD is available in supporting

information]. See DOI: 10.1039/b000000x/

overꢀhydrogenolysis products (1ꢀPO/2ꢀPO).

a) A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411; b)

Y. Nakagawa and K. Tomishige, Catal. Sci. Technol., 2011, 1, 179;

c) M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi and C. Della

Pina, Angew. Chem. Int. Ed., 2007, 46, 4434.

a) E. S. Vasiliadou, T. M. Eggnenhuisen, P. Munnik, P. E. de Jongh,

K. P. de Jong and A. A. Lemonidou, Appl. Catal. B, 2013, 145, 108;

b) S. Zhu, X. Gao, Y. Zhu, Y. Zhu, H. Zheng and Y. Li, J. Catal.,

2013, 303, 70; c) Y. Nakagawa, M. Tamura and K. Tomishige, J.

Mater. Chem. A, 2014, 2, 6688.

a) L. Ma and D. He, Catal. Today, 2010, 149, 148; b) M. Chia, Y. J.

PaganꢀTorres, 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; (c) C. H. Deng, X. Z. Duan, J. H. Zhou, D. Chen, X. G.

Zhou and W. K.Yuan, Catal. Today, 2014, 234, 208.

Scheme 2 Possible reaction pathway for glycerol hydrogenolysis over

alloyed IrꢀRe/KITꢀ6ꢀR.

Representative reports on glycerol hydrogenolysis to 1,3ꢀpropanediol:

a) Y. Nakagawa, Y. Shinmi, S. Koso and K. Tomishige, J. Catal.,

2010, 272, 191; b) O. M. Daniel, A. Delariva, E. L. Kunkes, A. K.

Datye, J. A. Dumesic and R. J. Davis, ChemCatChem, 2010, 2, 1107;

c) Y. Zhang, X. C. Zhao, Y. Wang, L. K. Zhou, J. Y. Zhang, J.

Wang, A. Q. Wang and T. Zhang, J. Mater. Chem. A, 2013, 1, 3724;

d) L. Z. Qin, M. J. Song and C. L. Chen, Green Chem., 2010, 12,

1466; e) J. Oh, S. Dash and H. Lee, Green Chem., 2011, 13, 2004; f)

Y. Amada, Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and K.

Tomishige, Appl. Catal. B, 2011, 105, 117; g) J. Ten Dam, K.

Djanashvili, F. Kapteijn and U. Hanefeld, ChemCatChem, 2013, 5,

497; h) R. Arundhathi, T. Mizugaki, T. Mitsudome, K. Jitsukawa

and K. Kaneda, ChemSusChem, 2013, 6, 1345; i) Y. Nakagawa, X.

Ning, Y. Amada and K. Tomishige, Appl. Catal. A., 2012, 433, 128;

j) M. Tamura, Y. Amada, S. Liu, Z. Yuan, Y. Nakagawa and K.

Tomishige, J. Mol. Catal. A: Chem., 2014, 388-389, 177.

10 4. Conclusions

In summary, the correlation between bimetallic IrꢀRe structure

and thermal treatment is investigated, and the resulting different

structured bimetallic IrꢀRe catalysts are applied in glycerol

hydrogenolysis to examine the relationship between bimetallic Irꢀ

¹⁵Re structure and catalytic performance. It is revealed that, direct

reduction of the impregnated sample without prior calcination

renders Re species fully reducible, favoring the formation of Irꢀ

Re alloy. In contrast, when calcination treatment is employed,

reduction of Re species is inhibited, resulting in the formation of

²⁰IrꢀReO_xafter the subsequent reduction of the bimetallic catalyst.

Compared with the catalyst with IrꢀReO_xstructure, the IrꢀRe

alloy catalyst shows remarkably improved activity in glycerol

hydrogenolysis and doubled 1,3ꢀpropanediol formation rate,

which results from the higher density of active center. Besides,

25 the IrꢀRe alloy catalyst also exhibits enhanced resistance against

particle sintering. This readilyꢀprepared alloyed IrꢀRe catalyst

may be useful in other CꢀO bond hydrogenolysis reactions, such

as the disassembly of lignin.

Y. Amada, H. Watanabe, M. Tamura, Y. Nakagawa, K. Okumura and

K. Tomishige, J. Phy. Chem. C, 2012, 116, 23503.

a) D. S. Kim and I. E. Wachs, J. Catal., 1993, 141, 419; b) M. A.

Vuurma, D. J. Stufkens and A. Oskam, J. Mol. Catal., 1992, 76, 263.

X. She, J. Hun Kwak, J. Sun, M. Y. Hu, C. Wang, C. H. F. Peden and

Y. Wang, ACS Catal., 2012, 2, 1020.

Acknowledgements

Z. Wang, S. Skiles, F. Yang and D. W. Goodman, Catal. Today,

2012, 181, 75.

30 This work was supported by the National Natural Science

Foundation of China (21106047). S.L.S. thanks the Center for

Enabling New Technologies through Catalysis (CENTC) for

financial support.

M. J. P. Botman, K.D. Vreugd, H. W. Zandbergen, R. D. Block and

V. Ponec, J. Catal., 1989, 116, 467.

10 E. L. Kunkes, D. A. Simonetti, J. A. Dumesic, W. D. Pyrz, L. E.

Murillo, J. G. G. Chen and D. J. Buttrey, J. Catal., 2008, 260, 164.

Notes and references

Journal Name, [year], [vol], 00–00 | 7

CREATED USING THE RSC ARTICLE TEMPLATE (VER. 3.0) - SEE WWW.RSC.ORG/ELECTRONICFILES FOR DETAILS

Catalysis Science & Technology

Page 8 of 9

ARTICLE TYPE

DOI: 10.1039/C4CY01285B

11 M. Chia, B. J. O’Neill, R. Alamillo, P. J. Dietrich, F. H. Ribeiro, J. T.

Miller and J. A. Dumesic, J. Catal., 2013, 308, 226.

12 T. Kurosaka, H. Maruyama, I. Naribayashi and Y. Sasaki, Catal.

Commun., 2008, 9, 1360.

13 L. F. Gong, Y. A. Lu, Y. J. Ding, R. H. Lin, J. W. Li, W. D. Dong, T.

Wang and W, M. Chen, Appl. Catal. A, 2010, 390, 119.

14 A. Ciftci, S. Eren, D. A. J. Ligthart and E. J. M. Hensen,

ChemCatChem, 2014, 6, 1260.

15 D. Hibbitts, Q. Tan and M. Neurock, J. Catal., 2014, 315, 48.

10 16 F. Kleitz, S. H. Choi and R. Ryoo, Chem. Commun., 2003, 2136.

17 T. W. Kim, F. Kleitz, B. Paul and R. Ryoo, J. Am. Chem. Soc., 2005,

127, 7601.

18 L. E. S. Rygh and C. J. Nielsen, J. Catal., 2000, 194, 401.

19 a) T. Marzialetti, M. Oportus, D. Ruiz, J. L. G. Fierro and P. Reyes,

Catal. Today, 2008, 133-135, 711; b) W. T. Tysoe, F. Zaera and G.

A. Somorjai, Surf. Sci., 1988, 200, 1.

20 H. I. Karan, K. Sasaki, K. Kuttiyiel, C. A. Farberow, M. Mavrikakis

and R. R. Adzic, ACS Catal., 2012, 2, 817.

21 S. Hirasawa, H. Watanabe, T. Kizuka, Y. Nakagawa and K.

Tomishige, J. Catal., 2013, 300, 205.

22 B. Hammer, L. B. Hansen and J. K. Nørskov, Phys. Rev. B, 1999, 59,

7413.

23 M. Haneda, H. Kudo, Y. Nagao, T. Fujitani and H. Hamada, Catal.

Commun., 2006, 7, 423.

25 24 B. Pholjaroen, N. Li, Y. Huang, L. Li, A. Wang, T. Zhang, Catal.

Today, 2014, DOI: 10.1016/j.cattod.2014.08.011.

25 L. Zhang, A. M. Karim, M. H. Engelhard, Z. H. Wei, D. L. King and

Y. Wang, J. Catal., 2012, 287, 37.

26 G. Borda, H. Rojas, J. Murcia, J. L. G. Fierro, P. Reyes and M.

Oportus, React. Kinet. Catal. Lett., 2007, 92, 369.

27 K. Chen, K. Mori, H. Watanabe, Y. Nakagawa and K. Tomishige, J.

Catal., 2012, 294, 171.

28 Y. Nakagawa, K. Mori, K. Chen, Y. Amada, M. Tamura and K.

Tomishige, Appl. Catal. A., 2013, 468, 418.

35 29 a) Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and K. Tomishige,

Appl. Catal. B, 2010, 94, 318; b) K. Chen, S. Koso, T. Kubota, Y.

Nakagawa and K. Tomishige, ChemCatChem, 2010, 2, 547.

30 T. Ebashi, Y. Ishida, Y. Nakagawa, S. Ito, T. Kubota and K.

Tomishige, J. Phy. Chem. C, 2010, 114, 6518.

40 31 a) S. Liu, M. Tamura, Y. Nakagawa and K. Tomishige, ACS

Sustainable Chem. Eng., 2014, 2, 1819; b) S. Liu, Y. Amada, M.

Tamura, Y. Nakagawa and K. Tomishige, Catal. Sci. Technol., 2014,

4, 2535; c) S. Liu, Y. Amada, M. Tamura, Y. Nakagawa and K.

Tomishige, Green Chem., 2014, 16, 617; d) K. Chen, M. Tamura, Z.

Yuan, Y. Nakagawa and K. Tomishige, ChemSusChem, 2013, 6,

613; e) M. Tamura, K. Tokonami, Y. Nakagawa and K. Tomishige,

Chem. Commun, 2013, 49, 7034.

Journal Name, [year], [vol], 00–00

Page 9 of 9

Catalysis Science & Technology