Aqueous-phase hydrogenation and hydrodeoxygenation of biomass-derived oxygenates with bimetallic catalysts

Article abstract of DOI:10.1039/c3gc41071d

The reaction rate on a per site basis for aqueous-phase hydrogenation (APH) of propanal, xylose, and furfural was measured over various alumina-supported bimetallic catalysts (Pd-Ni, Pd-Co, Pd-Fe, Ru-Ni, Ru-Co, Ru-Fe, Pt-Ni, Pt-Co, and Pt-Fe) using a high-throughput reactor (HTR). The results in this paper demonstrate that the activity of bimetallic catalysts for hydrogenation of a carbonyl group can be 110 times higher than monometallic catalysts. The addition of Fe to a Pd catalyst increased the activity for hydrogenation of propanal, xylose, and furfural. The Pd₁Fe₃ catalyst had the highest reaction rate for APH of propanal among all catalysts tested in the HTR. The addition of Fe to the Pd catalyst increased the reaction rate for xylose hydrogenation by a factor of 51, compared to the monometallic Pd catalyst. However, no bimetallic catalyst tested in this study was more active than the monometallic Ru catalyst for hydrogenation of xylose. The Pd₁Fe ₃ catalyst had the highest reaction rate for APH of furfural, which was 9 times higher than the rate of the Pd catalyst. The Pd₁Fe ₃/Zr-P, a bimetallic bifunctional catalyst, was 14 times more active on a per site basis than a Pd/Zr-P catalyst for aqueous-phase hydrodeoxygenation (HDO) of sorbitol in a continuous flow reactor. The addition of Fe to the Pd catalyst increased the rate of C-C cleavage reactions and promoted the conversion of sorbitan and isosorbide in HDO of sorbitol. Pd₁Fe ₃/Zr-P also had a higher yield of gasoline-range products than the Pd/Zr-P catalyst.

Full text of DOI:10.1039/c3gc41071d

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Aqueous-phase hydrogenation and
hydrodeoxygenation of biomass-derived
oxygenates with bimetallic catalysts†
Cite this: Green Chem., 2014, 16,
708
Jechan Lee, Yong Tae Kim and George W. Huber*
The reaction rate on a per site basis for aqueous-phase hydrogenation (APH) of propanal, xylose, and fur-
fural was measured over various alumina-supported bimetallic catalysts (Pd–Ni, Pd–Co, Pd–Fe, Ru–Ni,
Ru–Co, Ru–Fe, Pt–Ni, Pt–Co, and Pt–Fe) using a high-throughput reactor (HTR). The results in this paper
demonstrate that the activity of bimetallic catalysts for hydrogenation of a carbonyl group can be 110
times higher than monometallic catalysts. The addition of Fe to a Pd catalyst increased the activity for
hydrogenation of propanal, xylose, and furfural. The Pd1Fe3 catalyst had the highest reaction rate for APH
of propanal among all catalysts tested in the HTR. The addition of Fe to the Pd catalyst increased the reac-
tion rate for xylose hydrogenation by a factor of 51, compared to the monometallic Pd catalyst. However,
no bimetallic catalyst tested in this study was more active than the monometallic Ru catalyst for hydro-
genation of xylose. The Pd1Fe3 catalyst had the highest reaction rate for APH of furfural, which was 9
times higher than the rate of the Pd catalyst. The Pd1Fe3/Zr–P, a bimetallic bifunctional catalyst, was 14
times more active on a per site basis than a Pd/Zr–P catalyst for aqueous-phase hydrodeoxygenation
(HDO) of sorbitol in a continuous flow reactor. The addition of Fe to the Pd catalyst increased the rate of
C–C cleavage reactions and promoted the conversion of sorbitan and isosorbide in HDO of sorbitol.
Pd1Fe3/Zr–P also had a higher yield of gasoline-range products than the Pd/Zr–P catalyst.
Received 5th June 2013,
Accepted 6th January 2014
DOI: 10.1039/c3gc41071d
www.rsc.org/greenchem
of carbon monoxide (CO) and H on the Pt metal surface.16–19
Bimetallic catalysts have also been shown to have a higher
1. Introduction
Aqueous-phase hydrogenation (APH) reactions,
a
critical activity for electrocatalytic reactions.20–24 For example, Pt–Ru
fundamental reaction of aqueous-phase processing (APP), are bimetallic catalyst was more active than monometallic Pt cata-
used in a variety of processes for the conversion of biomass lyst for electrocatalytic oxidation of glycerol.24
into fuels and chemicals.1–7 APH involves the hydrogenation of
Bimetallic catalysts can also modify the reaction selecti-
a
range of functionalities including aldehydes, ketones, vity.11 For instance, Pd–Cu and Ni–Fe bimetallic catalysts had
furans, carbohydrates, and alkenes.8–10
a higher selectivity toward hydrogenation products but lower
Bimetallic catalysts often have a higher activity than either selectivity toward decarbonylation products for furfural conver-
of their parent metals.11 For example, Co and Ni promoted Pt sion than the pure monometallic Pd and Ni catalysts.25,26
catalysts exhibited a higher activity for the hydrogenation of Pt–Ni and Pt–Co bimetallic catalysts showed a higher selecti-
CvO bonds than the corresponding monometallic cata- vity toward deoxygenation products than monometallic Pt cata-
lysts.12,13 A bimetallic Ni–Pd catalyst was more active than lyst for hydrodeoxygenation (HDO) of meta-cresol.27 The
RANEY® Ni for the hydrogenation of HMF and furfural.14 The addition of Sn to Ni catalysts can increase the hydrogen selecti-
addition of Ni, Co, and Fe to a Pt catalyst increased the activity vity from 47% to 93% for APR of ethylene glycol.28 In spite of
for aqueous-phase reforming (APR) of ethylene glycol up to the clear advantages of bimetallic catalysts, they have not been
3 times.15 It was proposed that this increase in catalytic activity widely used for biomass conversion reactions. More infor-
was because the Ni, Co, and Fe decrease the adsorption energy mation about the use of bimetallic catalysts for fundamental
reactions like APH reactions of model biomass compounds
is needed to properly design bimetallic catalytic processes for
realistic biomass conversion processes.
Department of Chemical and Biological Engineering, University of Wisconsin-
Madison, 1415 Engineering Drive, Madison, Wisconsin 53706, USA.
High-throughput techniques have been developed that
E-mail: huber@engr.wisc.edu
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
make possible the rapid screening of large numbers of cata-
lytic materials.10,15,29–33 We have previously used a high-
c3gc41071d
708 | Green Chem., 2014, 16, 708–718
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
throughput reactor (HTR) to measure the initial activity for with deionized water until the pH was 5, dried at 373 K for
APH and hydrogenolysis of model biomass compounds, 12 hours, and then calcined at 673 K (20 K min−1) for 4 hours
including acetaldehyde, propanal, acetone, xylose, furfural, in static air. A 3 wt% Pd/Zr–P catalyst was prepared by incipi-
furfuryl alcohol, tetrahydrofurfuryl alcohol (THFA), and xylitol, ent wetness impregnation with an aqueous solution of
with monometallic catalysts.10 We observed that the reaction Pd(NH3)4(NO3)2 (Strem Chemicals). The Pd/Zr–P catalyst was
rate for hydrogenation of a carbonyl group is dependent on calcined at 523 K (20 K min−1) for 3 hours in static air. A
both the metal surface and the functionality that surrounds Pd–Fe/Zr–P bimetallic catalyst with a Fe to Pd molar ratio of
the carbonyl group.10 Results obtained from the HTR are also 3 : 1 (Pd1Fe3/Zr–P) was prepared by sequentially impregnating
consistent with results obtained from a continuous electrocata- the 3 wt% Pd/Zr–P catalyst dried at 373 K with an aqueous
lytic membrane reactor.34 The objective of this paper is to solution of Fe(NO3)3·9H2O (Sigma-Aldrich). The Pd–Fe/Zr–P
develop relationships between the reaction rate for hydrogen- catalyst was calcined at 523 K (20 K min−1) for 3 hours in
ation of different feed molecules containing a carbonyl group static air.
and different bimetallic catalysts to get insight into which bi-
metallic catalytic systems might be useful for aqueous-phase
2.2. Catalyst characterization
conversion of more complicated biomass molecules. We then Static H2 chemisorption was carried out in a Quantachrome
take one of the most promising bimetallic catalysts obtained Autosorb iQ Automated Gas Sorption system. Before H2 chemi-
from high-throughput studies and test it for HDO of sorbitol.
sorption, catalysts were reduced in situ under H2 flow up to
573 K (1 K min−1), held for 2 hours, purged with He for
2 hours, evacuated for 140 minutes, and cooled down to room
temperature. Hydrogen was dosed on the catalyst until the
equilibrium pressure was 560 mmHg. The hydrogen in the cell
was then evacuated at room temperature, and again hydrogen
2. Experimental
2.1. Catalyst preparation
Gamma-alumina (γ-Al2O3, SBET = 142 m2 g−1) was used as a was dosed on the catalyst to determine the amount of weakly
support of the catalysts tested in the HTR. The γ-Al2O3 was adsorbed hydrogen. The amount of strongly adsorbed hydro-
formed by calcination of Boehmite (Catapal® A Alumina, Sasol gen was determined by subtracting the second isotherm from
North America Inc.) at 873 K (heating rate: 20 K min−1) for the first one.
4 hours in static air. 3 wt% Pd, Pt, and Ru catalysts were pre-
Pulse CO chemisorption was conducted in a Micromeritics
pared by incipient wetness impregnation with aqueous solu- AutoChem II 2920 unit equipped with a thermal conductivity
tions of the following metal precursors: Pd(NH3)4(NO3)2, detector (TCD) to measure the consumption of CO. Before CO
Pt(NH3)4(NO3)2, and Ru(NO)(NO3)3 (Strem Chemicals). 20 wt% chemisorption, catalysts were reduced in situ under H2 flow up
Ni, Co, and Fe catalysts were prepared by incipient wetness to 573 K (1 K min−1), held for 2 hours, and cooled down to
impregnation with aqueous solutions of the following room temperature. The chemisorption was carried out at
metal precursors: Ni(NO3)2·6H2O, Co(NO3)2·6H2O, and 300 K in a 50 mL min−1 stream of He using a pulse chemisorp-
Fe(NO3)3·9H2O (Sigma-Aldrich). After incipient wetness tion technique, in which 500 mL pulses of 10% CO/He were
impregnation, the metal salt solution–support mixture was utilized.
dried at 373 K for 12 hours. The dried catalysts were calcined
Temperature programmed reduction (TPR) was performed
at 523 K (20 K min−1; for Pd and Pt catalysts) and 823 K (20 K in a flow of 10% H2/Ar mixture gas at a flow rate of 50 mL
min−1; for Ni, Co, and Fe catalysts) for 4 hours in static air. min−1 with a temperature ramp of 10 K min−1. A dry ice/
The Ru catalyst reduced at 523 K (1 K min−1) under H2 flow acetone cooling bath removed moisture from TPR effluent
(40 mL min−1, Airgas 99.99%) for 2 hours without calcination, stream at 195 K before the TCD. Before a TPR run, catalysts
followed by purging with helium (He) flow (40 mL min−1
Airgas 99.995%) for 2 hours.
,
were pretreated in He at 423 K for 30 minutes. The consump-
tion of H2 was monitored using the TCD of the Micromeritics
To prepare the γ-Al2O3 supported bimetallic catalysts, the AutoChem II 2920 unit and quantified based on H2 consump-
metal salts of Ni, Co, and Fe (Ni(NO3)2·6H2O, Co(NO3)2·6H2O, tion during TPR of AgO as standard.
and Fe(NO3)3·9H2O (Sigma-Aldrich)) were added to the pre-
Brunauer–Emmett–Teller (BET) surface areas of the Pd/Zr–P
pared 3 wt% Pd/γ-Al2O3, Ru/γ-Al2O3, and Pt/γ-Al2O3 catalysts by and Pd–Fe/Zr–P catalysts were calculated from nitrogen
incipient wetness impregnation. After incipient wetness adsorption data at 77 K obtained using a Micromeritics ASAP
impregnation, the base metal salt solution-noble metal catalyst 2020 system. Before the measurements, the sample was
mixture was dried at 373 K for 12 hours and then calcined at degassed under vacuum at 523 K.
523 K (20 K min−1) for 4 hours in static air.
The bulk crystalline structure of the Pd/Zr–P and Pd–Fe/
Zirconium phosphate (Zr–P, SBET = 361 m2 g−1, total pore Zr–P catalysts was determined by X-ray diffraction (XRD). XRD
volume = 1.05 cm3 g−1) was prepared by co-precipitation patterns were obtained with a Scintag Pad V using Cu Kα radi-
using 1.0 mol L−1 of ZrCl2O·8H2O (Aldrich) and 1.0 mol L−1 ation (k = 0.15406 nm), operated at 45 kV and 40 mA (2.0 kW)
of NH4H2PO4 (Aldrich) aqueous solutions at a molar ratio of at a scan rate of 0.2° (2θ) s−1. Each sample was reduced at
P/Zr = 2, as described by Kamiya et al.35 The precipitate was 573 K for 2 hours in H2 and passivated prior to XRD experi-
aged at room temperature for 30 minutes, filtered, washed ments. The assignment of the crystalline phases was carried
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 708–718 | 709

View Article Online
Paper
Green Chemistry
out using JADE 9.0 software package (Rigaku) for the ICDD
database.
The reaction rate (based on H2 uptake of fresh catalyst) was
calculated by the equation:
ꢀ
Reaction rate hꢀ1
P
ꢁ
¼
ðmoles of productÞ
ðH2 uptakeÞ ꢁ S ꢁ ðg of catalystÞ ꢁ ðreaction timeÞ
2.3. High-throughput studies
APH of propanal, xylose, and furfural over different bimetallic
catalysts were carried out in a HTR manufactured by HEL
Group (Model # CAT24). This HTR reactor consisted of 24 wells
machined into a cylindrical stainless steel high pressure
chamber. Stainless steel cooling tips were inserted into the top
of each glass tube reactor which put in each well in the HTR,
preventing the mixing of vapor between the glass tube reactors.
WinIso E670 system software controlled and monitored the
system temperature, pressure, gas flow rate, and stirring speed.
In this study, we have used the HTR to measure the reaction
rate, not deactivation or regeneration tests.
where S is the number of surface metal atoms adsorbed by
hydrogen. We assumed that hydrogen was dissociatively
adsorbed on all metal surfaces so S = 2. During all experi-
ments, the carbon balance was higher than 90% (see Table S1†).
The carbon balance was calculated by the equation:
Carbon balance ð%Þ ¼
moles of carbon out ðdetected by GCÞ
100 ꢁ
moles of carbon in
2.4. Continuous-flow reactor studies
For a run, each catalyst was typically put into two different HDO of sorbitol was conducted in a stainless-steel tubular
random wells to measure the effect of cross-talk between wells. flow reactor (30 cm long, 6.35 mm of outer diameter), heated
We could not detect severe cross-talk between wells in a run. by a furnace (Thermcraft Inc. no. 114-12-1ZH). Reactants
Each catalyst was tested three to five times.
including 20 wt% sorbitol solution and hydrogen gas were
Ten mg of each catalyst was loaded into each well in the flowing upwardly. A uniform temperature profile along the
HTR and then the monometallic Pd, Ru, and Pt catalysts and catalyst bed was achieved using aluminum filler inserted in
all bimetallic catalysts were reduced at 573 K (1 K min−1) and the void between the furnace and the tubular reactor. For cata-
the monometallic Ni, Co, and Fe catalysts were reduced at lytic activity measurements, 3.3 g of the catalyst without dilu-
673 K (1 K min−1) in H2 (50 mL min−1, Airgas 99.99%) for ents was loaded into the reactor with glass wool packed on
2 hours at ambient pressure and purged with He (50 mL both sides. Before the reaction, the catalyst was reduced in situ
min−1, Airgas 99.995%) for 30 minutes. After reduction, the under H2 flow (128 mL min−1) at 573 K (1 K min−1) for
catalysts were cooled to room temperature under He flow, then 2 hours. After the reduction was complete, the furnace was
sealed with closed valves, and transported to the glove box. In cooled to the desired reaction temperature and the reactor
the glove box, the HTR head was removed, magnetic stir bars system was pressurized to 6.21 MPa using a back-pressure regu-
were placed into each glass tube reactor which was loaded into lator. A 20 wt% sorbitol solution was then fed to the reactor
the high pressure chamber, and 2 mL of an aqueous feedstock using an Eldex Laboratories 1SM HPLC pump with co-fed H2
solution was added to each of the glass tube reactors with a (40 mL min−1). A stainless-steel gas–liquid separator (150 mL)
micropipette. The HTR was then put on the heating block and was used to accumulate liquid products at the top of the
pressurized with H2 to 4.38 MPa. After the reactor was pressur- reactor. The gaseous products continued to flow to the back-
ized, the temperature was increased at a rate of 20 K min−1
.
pressure regulator to maintain the reaction pressure. Both an
After the HTR was heated to the reaction temperature and excess flow shut-off valve (Parker Hannifin Corporation, FS190)
then pressurized to 5.41 MPa. A stirring speed of 800 rpm was and a pressure relief valve (Swagelok®, SS-4R3A) were also
used during the reaction. We checked that there was no exter- installed in the reactor system to ensure safe operation. Online
nal mass transfer limitation at 800 rpm in the HTR.10
GC (Agilent 7890A) was used to analyze the reactor effluent
Once a reaction set finished, the HTR was removed from gas. Carbon dioxide (CO2) and CO in the gaseous products
the heater and cooled down to room temperature in an ice were analyzed using a TCD with a GS-Carbon PLOT (Agilent,
bath at a rate of around 14 K min−1. After reaching room temp- Catalog no. 113-3133) capillary column. Alkanes in the
erature, the HTR was depressurized to atmospheric pressure, gaseous products were analyzed using a FID with Rt®-Q-Bond
and samples in each well were taken by a 3 mL syringe and fil- (Restek, Catalog no. 19744) capillary column. The liquid
tered with a syringe filter (0.2 μm). The samples from APH of product accumulated in the gas–liquid separator was drained
propanal and furfural were analyzed by a gas chromatograph periodically into a collecting container in which cyclohexane
(GC; Agilent 7890A) using a flame ionization detector (FID) (Acros Organics, 99%) was preloaded, and passed through a
with Rtx®-VMS (Restek, Catalog no. 19915) capillary column. A 0.2 μm syringe filter. The aqueous phase and organic phase
high-performance liquid chromatography (HPLC; Shimadzu were analyzed using GC-MS (Shimadzu GC-2010), offline
LC-20AT) with ultraviolet-visible (UV-Vis; SPD-20AV) and RID GC-FID (Agilent 7890A) with Rtx®-VMS (Restek, Catalog no.
(RID-10A) detectors was used to analyze the samples from APH 19915) capillary column, and HPLC (Shimadzu LC-20AT) with
of xylose.
UV-Vis (SPD-20AV) and RID (RID-10A) detectors. In HDO of
The average conversion of each catalyst for APH reactions sorbitol, conversion, molar carbon selectivity, yield, and
performed in this study is shown in Table S1.†
710 | Green Chem., 2014, 16, 708–718
turnover frequency (TOF) were calculated according to the
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
following equations:
Mavrikakis demonstrated that hydrogen binds weaker on Fe/
Pd alloy than Pd metal.41
Csorbitol;inletVinlet ꢀ Csorbitol;aqueousVaqueous
Conversion ð%Þ ¼ 100 ꢁ
Csorbitol;inletVinlet
Cproduct;aqueousVaqueous þ Cproduct;organicVorganic þ Cproduct;gaseousVgaseous
Molar carbon selectivity ð%Þ ¼ 100 ꢁ
Csorbitol;inletVinlet ꢀ Csorbitol;aqueousVaqueous
Cproduct;aqueousVaqueous þ Cproduct;organicVorganic þ Cproduct;gaseousVgaseous
Yield ð%Þ ¼ 100 ꢁ
Csorbitol;inletVinlet
ðmolar sorbitol flow rateÞ ꢁ ðconversion=100Þ
Turnover frequency ðhꢀ1Þ ¼
ðH2 uptake after reactionÞ ꢁ 2 ꢁ ðg of catalystÞ
where Csorbitol,inlet and Csorbitol,aqueous are the concentration of
Fig. 1 shows H2-TPR of 3 wt% Pd/γ-Al2O3, 5 wt% Fe/γ-Al2O3,
sorbitol at the inlet and outlet, respectively; Cproduct is the con- and Pd1Fe3/γ-Al2O3 catalysts. Table 2 shows the theoretical
centration of carbon in the product at the corresponding hydrogen consumption and the actual hydrogen consumption
phase; V is the corresponding phase volume. Product carbon during reduction for these three catalysts. The 3 wt% Pd cata-
selectivity was defined as the ratio of carbon moles of product lyst was completely reduced at <363 K.42 The monometallic Fe
divided by the total carbon moles of corresponding categories catalyst started to consume hydrogen at 600 K. Only a small
(light gases, gasoline-range products, and aqueous-phase fraction of the Fe was reduced in the Fe/γ-Al2O3 catalyst with a
products).
hydrogen consumption of 146.1 μmol g−1 at 800 K despite a
Before conducting measurements with different space velo- theoretical hydrogen consumption of 1343 μmol g−1. This indi-
cities, the catalysts had been running at a weight hourly space cates that Fe in the Fe/γ-Al2O3 is a mixture of FeO and Fe2O3 at
velocity (WHSV) of 0.16 h−1 at 518 K for 70 hours. At least a temperature between 600 to 800 K.43 The reduction for the
three liquid samples were collected for each reaction condition bimetallic Pd-Fe catalyst showed several TPR peaks with the
at regular intervals after steady state performance had been largest peak starting at 363 K. The hydrogen consumption
attained with time on stream in the gas phase.
between 363 K and 473 K is 552.1 μmol g−1 which is 36% of
the theoretical hydrogen consumption value. Thus, the exist-
ence of Pd metal helps the reduction of iron oxides. It is well
3. Results and discussion
3.1. High-throughput studies
Table 1 Chemisorption data for γ-Al2O3 supported monometallic and
bimetallic catalysts
3.1.1. Catalyst characterization. Table 1 presents the chemi-
sorption uptake values of the catalysts tested in the HTR.
The addition of Ni and Co to the Ru catalyst and Fe to the Ru
and Pt catalysts increased the H2 uptake. The H2 uptake went
through a maximum at Ni/Pt = 1 or Co/Pt = 1 when Ni and Co
H2 uptake
CO uptake
Catalyst
[μmol g−1
]
[μmol g−1
]
Pd1Ni1
Pd1Ni3
21.6
22.5
21.8
22.9
22.4
10.6
58.1
44.8
81.2
35.5
44.9
36.5
34.5
34.3
56.8
48.7
51.9
46.4
56.0
38.8
37.7
84.7
18.1
67.5
56.9
57.4
50.1
42.9
13.2
118.6
—
—
—
—
—
—
—
—
—
—
—
—
—
were added to the Pt catalyst. However, the H2 uptake of Pd- Pd1Co1
Pd1Co3
based bimetallic catalysts was lower than the monometallic Pd
catalyst by factors of 2 to 6. Hydrogen can adsorb into bulk Pd
Pd1Fe1
Pd1Fe3
metal, causing an overestimation of the number of active Pd
sites.36–38 We performed CO chemisorption for the Pd-based
Ru1Ni1
Ru1Ni3
Ru1Co1
catalysts to check if hydrogen adsorbs into bulk Pd on the cata-
lysts. The metal dispersion based on the H2 uptake and the CO Ru1Co3
Ru1Fe1
Ru1Fe3
Ru
uptake was 41.2% and 42.1% respectively, suggesting that
hydrogen atoms do not go into the Pd bulk for our catalyst
systems. As more Ni and Co were added into the Pd catalyst, Pt1Ni1
the CO uptake decreased, consistent with previous studies.39,40
Pt1Ni3
Pt1Co1
Pt1Co3
Similar to the H2 uptake, the CO uptake also decreased as
more Fe was added to the Pd catalyst. It is possible that Fe Pt1Fe1
Pt1Fe3
Pt
Ni
encapsulates Pd metal sites, allowing less H2 and CO chemi-
sorb on the Pd sites. Shao et al. reported that overlayered Pd
—
—
—
on Fe has a lower d-band center than Pd.23 Greeley and Co
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 708–718 | 711

View Article Online
Paper
Green Chemistry
propanal, 63 times higher than the rate of the 3 wt% Pd
catalyst.
The 3 wt% Ru catalyst was the most active monometallic
catalyst for APH of propanal with a hydrogenation rate of
1810 h−1. The Ru1Co1 catalyst had the highest hydrogenation
rate for propanal conversion among all Ru bimetallic catalysts
tested in this study. Based on a statistical student t-test, the
addition of Fe to the monometallic Ru catalyst did not signifi-
cantly enhance the rate of propanal hydrogenation. Ni addition
at a Ni/Ru = 1 level (Ru1Ni1) also did not statistically show any
significant difference in the rate of propanal hydrogenation
compared to the pure Ru catalyst. Further addition of Ni to
form the Ru1Ni3 catalyst, decreased the catalytic activity for
propanal hydrogenation to levels below that of the monometal-
lic Ru catalyst.
Fig. 1 H2-TPR patterns of γ-Al2O3 supported Pd, Fe, and Pd1Fe3
Addition of Ni, Co, and Fe to the monometallic Pt catalyst
increased the rate of propanal hydrogenation. The maximum
rate for all three of these bimetallic catalysts was observed at a
base metal to Pt molar ratio of 3 to 1. The Pt1Co3 catalyst was
very active for APH of propanal with a rate of 5160 h−1. The
Pt1Fe3 catalyst had a rate of 4700 h−1 for propanal hydrogen-
ation that was 14 times more active than the monometallic
Pt catalyst. The Pt1Ni3 catalyst showed propanal hydrogenation
rate 8 times as high as the 3 wt% Pt catalyst. The results of
Pt–Co bimetallic catalysts being more active for APH of propa-
nal are consistent with the work of Zheng et al. who showed
that a Co–Pt bimetallic catalyst is more active than the
parent monometallic catalysts (Pt and Co) for hydrogenation of
acetaldehyde and acetone.12 They concluded the increase in
activity was because acetaldehyde and acetone adsorbed more
weakly on Pt–Co–Pt(111) bimetallic surface than the parent
catalysts.
Table 2 H2 consumption of γ-Al2O3 supported Pd, Fe, and Pd1Fe3 cata-
lysts determined by H2-TPR measurements
Tmax of H2 consumption
Theoretical H2
consumptiona
[μmol g−1] from TPR
Catalyst
[μmol g−1
]
<363 K
363–473 K
>573 K
3 wt% Pd
5 wt% Fe
Pd1Fe3
281.9
1343.0
1544.3
280.3
0
0
0
0
0
146.1
25.2
552.1
a Theoretical H2 consumption assuming the metal oxides are
completely reduced to their metallic states.
known that the addition of a noble metal into Fe catalyst monometallic surfaces (Pt(111) and Co(0001)).12
increases the reducibility of iron oxides.44–47
The monometallic Ru catalyst was the most active catalyst
3.1.2. High-throughput screening of bimetallic catalysts for APH of xylose among all catalysts tested in the HTR. The
for aqueous-phase hydrogenation of propanal, xylose, and fur- addition of Ni, Co, and Fe to the Ru catalyst decreased the
fural. We previously showed that Ru was the most active activity of the Ru catalyst for xylose hydrogenation.
monometallic catalyst for APH of propanal and xylose.10 The
The addition of Ni, Co, and Fe to the monometallic Pd cata-
activity of monometallic catalysts for APH of furfural decreased lyst increased the activity for APH of xylose where each catalyst
in the order: Pd ∼ Ni > Co > Ru > Pt > Rh.10 We also showed had a maximum rate at a base metal to Pd molar ratio of 3 to
that the rate of propanal hydrogenation is 3 times faster than 1. The order of decreasing activity for xylose hydrogenation
furfural hydrogenation, demonstrating that steric effects are decreased as Pd–Ni > Pd–Fe ∼ Pd–Co. The Pd1Ni3 catalyst
important for APH of carbonyl groups.10 The focus of the HTR was 110 times more active than the monometallic Pd catalyst
screening study is to elucidate activity trends for hydrogen- for APH of xylose. The Pd1Fe3 catalyst had a 50 times higher
ation of different carbonyl compounds with different bi- reaction rate than the monometallic Pd catalyst for xylose
metallic catalysts. Table 3 summarizes the reaction rates for hydrogenation.
APH of propanal, xylose, and furfural over different mono-
metallic and bimetallic catalysts in the HTR.
The addition of Ni, Co, and Fe to the monometallic Pt cata-
lyst increased the activity for APH of xylose with the bimetallic
The Pd-bimetallic catalysts were up to 63 times more active activity decreasing as Pt–Ni > Pt–Co > Pt–Fe. The Pt1Ni3 catalyst
than the monometallic Pd catalyst for propanal hydrogenation. was 16 times more active than the monometallic Pt catalyst.
The bimetallic catalyst activity for propanal hydrogenation
The products observed from APH of furfural were furfuryl
decreased as Pd–Fe > Pd–Ni > Pd–Co. The Pd–Fe and Pd–Ni alcohol and THFA. Addition of Fe and Co to the Pd catalyst
catalysts had the highest activity for propanal hydrogenation at increased the activity of the Pd for APH of furfural. The Pd–Fe
a base metal to Pd molar ratio of 3 : 1 (Pd1Fe3 and Pd1Ni3). In bimetallic catalysts were the most active catalysts for APH
contrast, the Pd–Co catalyst had the highest activity for propa- of furfural. The Pd1Fe3 catalyst was 9 times more active than
nal hydrogenation at a Co to Pd molar ratio of 1 : 1 (Pd1Co1). the monometallic Pd catalyst for APH of furfural. The Pd–Co
The Pd1Fe3 catalyst is the most active catalyst tested for APH of bimetallic catalysts were up to 40% more active than the
712 | Green Chem., 2014, 16, 708–718
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
Table 3 Reaction rate and standard error for APH of propanal, xylose, and furfural over γ-Al2O3 supported monometallic and bimetallic catalysts.
Reaction conditions: 373 K, 5.41 MPa, 5 wt% propanal, 5 wt% xylose, and 4.8 wt% furfural solutions as the feed
Reaction rate (h−1
)
Catalyst
Pd1Ni1
Pd1Ni3
Pd1Co1
Pd1Co3
Pd1Fe1
994.7 ( 76.4)
1541.0 ( 135.1)
894.5 ( 69.8)
586.5 ( 38.5)
1234.6 ( 88.6)
6387.3 ( 601.2)
100.6 ( 7.2)
2021.0 ( 118.0)
1293.5 ( 119.7)
3386.8 ( 149.4)
2180.1 ( 126.3)
2026.8 ( 205.7)
2046.3 ( 209.1)
1811.6 ( 122.1)
1969.9 ( 158.1)
2458.2 ( 209.8)
2865.1 ( 224.0)
5162.7 ( 455.4)
2447.9 ( 209.4)
4701.6 ( 442.6)
320.4 ( 13.0)
818.8 ( 78.1)
111.5 ( 7.1)
885.2 ( 69.7)
1552.0 ( 145.4)
101.1 ( 10.7)
724.7 ( 73.9)
109.2 ( 10.5)
701.3 ( 57.0)
13.8 ( 2.4)
1255.4 ( 106.0)
880.2 ( 75.2)
1735.7 ( 165.7)
1160.4 ( 103.7)
899.7 ( 60.5)
705.8 ( 53.7)
2476.9 ( 208.8)
409.1 ( 36.4)
872.7 ( 85.1)
206.0 ( 19.8)
390.1 ( 28.5)
158.9 ( 13.6)
200.2 ( 20.2)
51.3 ( 6.2)
504.5 ( 49.7)
329.7 ( 30.4)
1107.2 ( 105.8)
675.1 ( 63.4)
2147.3 ( 210.1)
4677.4 ( 417.2)
459.9 ( 34.4)
206.6 ( 18.0)
164.7 ( 15.2)
545.9 ( 55.7)
398.3 ( 33.7)
537.2 ( 50.5)
581.4 ( 33.7)
204.3 ( 20.1)
73.2 ( 7.3)
282.9 ( 18.4)
199.1 ( 12.3)
427.2 ( 41.3)
214.4 ( 16.4)
828.1 ( 48.5)
48.8 ( 3.8)
Pd1Fe3
3 wt% Pd
Ru1Ni1
Ru1Ni3
Ru1Co1
Ru1Co3
Ru1Fe1
Ru1Fe3
3 wt% Ru
Pt1Ni1
Pt1Ni3
Pt1Co1
Pt1Co3
Pt1Fe1
Pt1Fe3
3 wt% Pt
20 wt% Ni
20 wt% Co
20 wt% Fe
333.0 ( 30.4)
196.0 ( 18.6)
0
277.0 ( 20.4)
78.0 ( 6.6)
0
0
monometallic Pd catalyst. The Pd1Ni3 bimetallic catalyst was
less active than the 3 wt% Pd catalyst even though Pd and Ni
are both very active metals for APH of furfural.
The addition of Fe and Co to the monometallic Ru catalyst
increased the activity up to 2 times for APH of furfural. Ni
addition to the monometallic Ru catalyst decreased the activity
by 19%. The addition of Ni, Co, and Fe to the Pt catalyst
increased the rate for furfural hydrogenation with the most
active bimetallic catalysts being Pt–Fe. The Pt1Fe3 catalyst was
16 times more active than the 3 wt% Pt catalyst for furfural
hydrogenation. The Pt1Co3 catalyst was 9 times as active as the
3 wt% Pt catalyst for APH of furfural. The Pt1Ni1 catalyst had a
higher hydrogenation rate than the 3 wt% Pt catalyst but a
lower hydrogenation rate than the 20 wt% Ni catalyst.
Fig. 2 Reaction rate for APH of propanal, xylose, and furfural as a func-
tion of Pd molar fraction in Pd–Fe catalysts. Reaction conditions: 373 K,
5.41 MPa, 5 wt% propanal, 5 wt% xylose, and 4.8 wt% furfural solutions
as the feed.
The Pd-based bimetallic catalysts produced both furfuryl
alcohol (88.5 to 93.1%) and THFA (6.9 to 11.5%) from furfural.
This indicates that the Pd-based bimetallic catalysts are also
active in the hydrogenation of the CvC bond in a furan ring.
The Ru-based bimetallic catalysts made THFA with small por-
tions (0.2 to 3.6%). The Pt-based bimetallic catalysts produced surface decreases the binding energy of the carbonyl group,
only furfuryl alcohol.
The results from APH of different carbonyl compounds
performed in the HTR demonstrate that the Fe promoted Pd
resulting in a higher activity for hydrogenation of carbonyl
groups.12,13
On the Pd1Fe3 catalyst the hydrogenation rate for each
catalyst (i.e. Pd1Fe3) is the most active catalyst for propanal molecule decreases in the order: propanal > furfural > xylose.
For xylose hydrogenation, the O atom on the carbonyl group
and furfural hydrogenation and also very active for xylose
hydrogenation (Fig. 2), suggesting that Fe addition to Pd cata- adsorbs on a catalyst surface through O along with an adsorp-
lyst could increase the activity for hydrogenation of carbonyl tion of O atom on the hydroxyl group adjacent to the carbonyl
group on the catalyst surface.48 This adsorbed O atom of the
compounds. One hypothesis is that the Fe in Pd–Fe bimetallic
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 708–718 | 713

View Article Online
Paper
Green Chemistry
hydroxyl group may prevent hydrogenation of CvO bond of
xylose. However, for propanal and furfural, the carbonyl group
adsorbs on Pd metal surface through both C and O.25,49 For
propanal hydrogenation, the adsorbed carbonyl species is
hydrogenated to form a surface alkoxide species via the
addition of adsorbed H to the adsorbed O atom.25,50,51 In con-
trast to propanal, a hydroxyalkyl surface species is the domi-
nant intermediate through H addition to C atom due to the
conjugation with the furan ring of furfural.25,52 Atomic O is
more electronegative than atomic C; hence, the alkoxide inter-
mediate can more easily react with H than the hydroxyalkyl
intermediate resulting in higher activity for hydrogenation of
propanal than of furfural.
Fig. 3 TOF for HDO of sorbitol over Pd/Zr–P, Pd1Fe3/Zr–P, Pt/Zr–P,
and Pt–ReOx/C. The TOF of Pt/Zr–P and Pt–ReOx/C is from the litera-
ture.58 Reaction conditions: 518 K, 6.21 MPa, 20 wt% sorbitol solution as
3.2. Aqueous-phase hydrodeoxygenation of sorbitol over
the feed, and flow rate of H2 is 40 mL min−1
.
Pd/Zr–P and Pd–Fe/Zr–P
From the high-throughput screening, we found that the
addition of Fe to Pd increases the activity up to 63 times for
hydrogenation of a carbonyl group. The hydrogenation of car-
bonyl groups is an important fundamental reaction in HDO of
biomass-derived oxygenates.3,53,54 The carbonyl groups are pro-
duced during HDO by several routes including: (1) dehydration
of alcohols or polyols and (2) C–C bond cleavage reactions.53
For instance, ethylene glycol is produced from hydrogenation
of 2-hydroxyacetaldehyde which is produced through dehydro-
genation and retro-aldol condensation of sorbitol.53 Previously
Pd catalyst has been shown to have lower activity but higher
hexane selectivity than Pt for aqueous-phase HDO of sorbitol.3
Pd is also 55% cheaper than Pt.55 Hence, our high-throughput
studies suggest one option to increase the activity of Pd cata-
lysts for HDO of sorbitol would be to add Fe to the Pd catalyst.
HDO reaction occurs through a bifunctional reaction pathway
on metal-acid catalysts. Zr–P is one of the most promising
solid-acids for HDO for this reaction because it has a high
Brønsted to Lewis acid ratio, is hydrothermally stable, and
does not leach during the reaction.56,57 Therefore, we syn-
thesized a Pd1Fe3/Zr–P catalyst and carried out HDO of sorbitol
on the Pd1Fe3/Zr-P catalyst in a continuous flow reactor.
Fig. 4 Molar carbon selectivity of C1–C6 products (normalized without
polymerized humins) over Pd/Zr–P (light gray) and Pd–Fe/Zr–P (dark
gray) for HDO of sorbitol. Reaction conditions: 518 K, 6.21 MPa, WHSV =
0.16 h−1 (no pattern in bars) and 2.92 h−1 (pattern in bars), 20 wt% sorbi-
tol solution as the feed, and flow rate of H2 is 40 mL min−1
.
obtained with Pt/Zr–P.58 However, in spite of the loss of
surface area, both catalysts were stable as shown by the time
on stream data in Fig. S2.†
Table 4 lists the BET surface area and H2 uptake of the
3 wt% Pd/Zr–P and Pd1Fe3/Zr–P catalysts before and after reac-
tion. During the reaction both catalysts underwent phase
transformation from an amorphous material to a crystalline
material as shown by XRD in Fig. S1.† Both catalysts lost BET
and metal surface area after reaction. This is similar to results
The Pd1Fe3/Zr–P was the most active catalyst we have tested
for HDO of sorbitol, as shown in Fig. 3.53,58,59 The TOF of the
Pd1Fe3/Zr–P (based on H2 uptake after reaction) was 14 times
higher than the TOF of the Pd/Zr–P. The Pd1Fe3/Zr–P catalyst
was 13 times more active than a Pt/Zr–P catalyst58 and 18
times more active than a Pt–ReOx/C catalyst.58 This suggests
that the hydrogenation of carbonyl groups is a kinetically slow
step on Pd catalysts and carbonyl intermediates may poison
the Pd catalyst surface.60–62
Fig. 4 shows that the C6 product selectivity decreased and
the C1–C4 products selectivity increased by Fe addition to the
Pd catalyst. Even at the low conversion (WHSV = 2.92 h−1), the
Pd1Fe3/Zr–P catalyst had a higher selectivity toward C1–C4 pro-
ducts than the Pd/Zr–P. The Pd1Fe3/Zr–P catalyst produced
more ethylene glycol, propylene glycol, and glycerol than
the Pd/Zr–P, as shown in Table 5. This indicates that
Table 4 BET surface area and H2 chemisorption data of Pd/Zr–P and
Pd–Fe/Zr–P before and after reaction. Reaction conditions: 518 K,
6.21 MPa, 20 wt% sorbitol solution as the feed, and flow rate of H2 is
40 mL min−1
SBET
H2 uptake
Catalyst
[m2 g−1
]
[μmol g−1
]
3 wt% Pd/Zr–P
3 wt% Pd/Zr–P after reaction
Pd1Fe3/Zr–P
238.0
19.7
102.3
6.7
19.3
1.1
16.4
0.2
Pd1Fe3/Zr–P after reaction
714 | Green Chem., 2014, 16, 708–718
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
Table 5 Molar carbon selectivity of products in gas and liquid phases over Pd/Zr–P and Pd–Fe/Zr–P for HDO of sorbitol. Reaction conditions:
518 K, 6.21 MPa, 20 wt% sorbitol solution as the feed, and flow rate of H2 is 40 mL min−1
Catalyst
3 wt% Pd/Zr–P
Pd1Fe3/Zr–P
WHSV (h−1
)
0.16
97.8
2.92
5.1
0.05
79.1
0.16
100
2.92
15.9
0.14
1200.0
Conversion (%)
Reaction rate (μmolsorbitol gcat−1 s−1
)
TOF (h−1
)
—
—
Alcohol selectivity (%)
Methanol
Ethanol
1-Propanol
2-Propanol
1-Butanol
2-Butanol
1-Pentanol
2-Pentanol
1-Hexanol
2-Hexanol
3-Hexanol
Total alcohols
Aldehyde selectivity (%)
Butanol
0.4
0.5
2.8
0.02
0.4
0
0.1
0.1
1.1
0.02
0.3
5.9
0.5
0
0.3
0
0
0
0
0
0.3
0
0
1.0
3.8
11.3
0.4
1.6
0.3
0.8
0.2
1.6
0.1
0.3
21.3
1.9
1.7
0.4
0
0.02
0
0
0
0.1
0
0
1.1
4.1
0.1
0
0.1
0.4
0
0.4
0.1
0.6
0.7
1.1
0
1.1
Pentanal
Total aldehydes
Ketone selectivity (%)
Acetone
Hydroxyacetone
2-Butanone
2-Pentanone
3-Pentanone
2-Hexanone
0.03
0
0
0
0
0
0.5
0
0
0.7
0.2
1.3
0.4
0.9
0.3
0.6
4.4
0
8.0
0
0
0
0
0
8.0
0.1
0.1
1.0
0.1
0.1
1.4
3-Hexanone
Total ketones
Alkanediol selectivity (%)
Ethylene glycol
Propylene glycol
1,2-Butanediol
1,4-Butanediol
2,3-Butanediol
1,2-Pentanediol
1,4-Pentanediol
1,5-Pentanediol
1,2-Hexanediol
1,6-Hexanediol
2,5-Hexanediol
Total alkanediols
Polyol selectivity (%)
Glycerol
0.5
1.1
0.1
0
0.1
0
0.4
0.04
0.1
0
0.1
0
1.3
0.8
0
0
0
1.1
0.4
0.2
0
1.2
0
0.3
1.1
0.4
0.3
1.0
0.6
0.5
0.1
0
20.0
24.8
2.3
0
1.3
0
0.5
0
0
0.1
0.7
5.1
0
0
49.1
1.9
4.9
0.2
0.1
3.0
3.3
0
0
0
0
0.3
0.02
0
1.8
0.7
0
1,2,6-Hexanetriol
Mannitol
Total polyols
0.3
2.5
Heterocyclic compound selectivity (%)
Tetrahydrofuran
3-Hydroxy-tetrahydrofuran
2-Methyl-tetrahydrofuran
5-Hydroxymethylfurfural
Tetrahydrofurfuryl alcohol
Tetrahydropyran
Tetrahydro-4H-pyran-4-ol
2,5-Dimethyltetrahydrofuran
5-Methyltetrahydrofuran-2-methanol
Tetrahydro-2-methoxymethyl-furan
2-Methyl-tetrahydropyran
Tetrahydropyran-2-methanol
Cyclopentanone
0.4
0.1
0.3
0
0.8
0.7
0.3
2.5
1.4
0.7
3.4
5.0
0
0.2
0
0.3
0
1.0
0
0.7
4.7
2.2
1.0
2.8
0.9
0
0.6
0.8
0.7
0
2.7
1.1
0.9
1.7
3.2
2.2
2.8
9.5
0.1
0
0
0
0
0.1
1.9
0
0.1
1.7
0.7
1.0
2.4
0.2
1.1
0
1,2-Cyclohexanediol
0
0
Isosorbide
Sorbitan
Total heterocyclic compounds
26.8
3.0
45.5
42.7
9.6
65.9
2.0
0.6
29.0
14.2
1.0
24.2
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 708–718 | 715

View Article Online
Paper
Green Chemistry
Table 5 (Contd.)
Carboxylic acid selectivity (%)
Propanoic acid
Butylic acid
Pentanoic acid
Hexanoic acid
Total carboxylic acids
Alkane selectivity (%)
Methane
Ethane
Propane
Butane
Pentane
0.3
0.1
0.2
0.2
0.8
0
0
0
0.6
0.6
0.5
0.2
0.3
0.6
1.6
0
0
0.04
1.2
1.3
0.05
0.2
0.5
0.5
0.2
1.0
2.5
0.6
0.7
36.5
0.02
0.1
0.1
0.2
0.1
0.8
1.3
0.2
1.0
1.2
0.6
1.4
0.8
0.4
0.5
0.5
4.2
1.5
5.4
24.1
0.02
0.01
0
0
0.01
0.01
0.04
0.3
5.0
0.3
Hexane
Total alkanes
CO selectivity (%)
CO2 selectivity (%)
Humins (%)
the addition of Fe to the Pd catalyst promotes C–C bond
cleavage (dehydrogenation/retro-aldol condensation) compared
4. Conclusions
to the monometallic Pd and Pt58 catalysts. This might be The addition of Ni, Co, or Fe to the monometallic Pd and Pt
due to interactions between Fe-containing bimetallic sites catalysts usually increased the activity for APH of propanal and
and FeO species existing on the periphery of the bimetallic furfural. In contrast, all bimetallic catalysts tested in this study
sites.63
had a lower activity than the monometallic Ru catalyst for APH
Table 5 shows the detailed product selectivity for each cata- of xylose. The reaction rate for APH of propanal increased by
lyst on a molar carbon basis grouped by product functionality. the addition of Ni, Co, and Fe to the monometallic Pd catalyst
Only trace amounts of aldehydes were detected for both cata- with the bimetallic Pd1Fe3 catalyst being most active for APH
lysts, indicating that aldehydes rapidly converted to their of propanal. The addition of Fe to the monometallic Pd cata-
corresponding alcohols on the Pd and Pd–Fe catalysts. There lyst increased the reaction rate for APH of xylose by a factor of
were up to 7 times more ketones than aldehydes in the liquid 51. The Pd1Fe3 bimetallic catalyst had the highest activity for
phase. We observed that aldehyde hydrogenation was up to APH of furfural. The Pd-based bimetallic catalysts produced
58% faster than ketone hydrogenation on Pd catalysts.10 The not only furfuryl alcohol but also THFA for APH of furfural,
Pd1Fe3/Zr–P also had a lower selectivity toward sorbitan and meaning that the Pd-based bimetallic catalysts are also active
isosorbide that are made through dehydration of sorbitol at not only for CvO bond hydrogenation but also for CvC bond
acid sites of Zr–P than the Pd/Zr–P (Table 5). By adding Fe to hydrogenation. The Pd1Fe3 bimetallic catalyst was used for
the Pd catalyst, the C6 alcohols, C6 diols, C6 polyols, and HDO of sorbitol with Zr–P as a support. The Pd1Fe3/Zr–P cata-
C6 cyclic ethers selectivity increased from 16% to 24.3% lyst was the most active catalyst we have tested for HDO of sor-
at WHSV = 0.16 h−1. This suggests that the addition of Fe bitol. The addition of Fe to the Pd enhanced C–C bond
to the Pd catalyst promotes the conversion of sorbitan and iso- cleavage and C–O hydrogenolysis reactions. Moreover, Fe
sorbide through a series of dehydration/hydrogenation (or addition to the Pd increased the conversion of sorbitan and
hydrogenolysis) reactions. At the high conversion (WHSV = isosorbide to C6 alcohols, C6 diols, C6 polyols, and C6 cyclic
0.16 h−1) for both catalysts we observed a white insoluble ethers. The Pd1Fe3/Zr–P was also more selective to gasoline-
product in the liquid phase that we report as humins in range products than the Pd/Zr–P.
Table 5. The humins were probably formed via polymerization
of isosorbide.64
Acknowledgements
In Table S2,† we grouped the products into three cat-
egories: (1) light gases such as C1–C4 alkanes and CO2, (2)
This material is based upon work supported as part of the
gasoline-range products including C5–C6 alkanes, C2–C6
Institute for Atom-efficient Chemical Transformations (IACT),
monofunctional compounds (including alcohols, alde-
an Energy Frontier Research Center funded by the U.S. Depart-
hydes, cyclic ethers, carboxylic acids, and ketones), and (3)
ment of Energy, Office of Science, Office of Basic Energy
aqueous-phase oxygenates including methanol, polyols,
Sciences.
mannitol, sorbitan, and isosorbide.59,65 The Pd1Fe3/Zr–P had
a higher yield (55.1%) of gasoline-range products than the
3
wt% Pd/Zr–P (24.1%). The estimated research octane
References
number
also higher than the RON of the Pd/Zr–P (92.2) at WHSV =
0.16 h−1
(RON)
of
the
Pd1Fe3/Zr–P
(101.3)
was
1 J. W. Shabaker, G. W. Huber, R. R. Davda, R. D. Cortright
and J. A. Dumesic, Catal. Lett., 2003, 88, 1–8.
.
716 | Green Chem., 2014, 16, 708–718
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
2 R. R. Davda, J. W. Shabaker, G. W. Huber, R. D. Cortright 28 G. W. Huber, J. W. Shabaker and J. A. Dumesic, Science,
and J. A. Dumesic, Appl. Catal., B, 2003, 43, 13–26. 2003, 300, 2075–2077.
3 G. W. Huber, R. D. Cortright and J. A. Dumesic, Angew. 29 S. Senkan, K. Krantz, S. Ozturk, V. Zengin and I. Onal,
Chem., Int. Ed., 2004, 43, 1549–1551. Angew. Chem., Int. Ed., 1999, 38, 2794–2799.
4 R. R. Davda, J. W. Shabaker, G. W. Huber, R. D. Cortright 30 R. J. Hendershot, S. S. Lasko, M. F. Fellmann,
and J. A. Dumesic, Appl. Catal., B, 2005, 56, 171–186.
5 G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic,
Science, 2005, 308, 1446–1450.
6 G. W. Huber and J. A. Dumesic, Catal. Today, 2006, 111,
119–132.
7 J. N. Chheda and J. A. Dumesic, Catal. Today, 2007, 123,
59–70.
8 J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew.
Chem., Int. Ed., 2007, 46, 7164–7183.
9 H. Olcay, L. J. Xu, Y. Xu and G. W. Huber, ChemCatChem,
2010, 2, 1420–1424.
10 J. Lee, Y. Xu and G. W. Huber, Appl. Catal., B, 2013,
140–141, 98–107.
G. Oskarsdottir, W. N. Delgass, C. M. Snively and
J. Lauterbach, Appl. Catal., A, 2003, 254, 107–120.
31 P. Serna, L. A. Baumes, M. Moliner and A. Corma, J. Catal.,
2008, 258, 25–34.
32 L. A. Baumes, P. Serna and A. Corma, Appl. Catal., A, 2010,
381, 197–208.
33 I. Onal, D. Duzenli, A. Seubsai, M. Kahn, E. Seker and
S. Senkan, Top. Catal., 2010, 53, 92–99.
34 S. K. Green, J. Lee, H. J. Kim, G. A. Tompsett, W. B. Kim
and G. W. Huber, Green Chem., 2013, 15, 1869–1879.
35 Y. Kamiya, S. Sakata, Y. Yoshinaga, R. Ohnishi and
T. Okuhara, Catal. Lett., 2004, 94, 45–47.
36 F. A. Lewis, Platinum Metals Rev., 1960, 4, 132–137.
11 D. M. Alonso, S. G. Wettstein and J. A. Dumesic, Chem. Soc. 37 F. D. Manchester, A. San-Martin and J. M. Pitre, J. Phase
Rev., 2012, 41, 8075–8098. Equilib., 1994, 15, 62–83.
12 R. Y. Zheng, Y. X. Zhu and J. G. G. Chen, ChemCatChem, 38 G. Fagherazzi, A. Benedetti, S. Polizzi, A. Dimario, F. Pinna
2011, 3, 578–581. and M. Signoretto, Catal. Lett., 1995, 32, 293–303.
13 R. Y. Zheng, M. P. Humbert, Y. X. Zhu and J. G. Chen, 39 A. Sarkany, Z. Zsoldos, G. Stefler, J. W. Hightower and
Catal. Sci. Technol., 2011, 1, 638–643. L. Guczi, J. Catal., 1995, 157, 179–189.
14 Y. Nakagawa and K. Tomishige, Catal. Commun., 2010, 12, 40 N. S. Babu, N. Lingaiah and P. S. S. Prasad, Appl. Catal., B,
154–156. 2012, 111, 309–316.
15 G. W. Huber, J. W. Shabaker, S. T. Evans and J. A. Dumesic, 41 J. Greeley and M. Mavrikakis, J. Phys. Chem. B, 2005, 109,
Appl. Catal., B, 2006, 62, 226–235. 3460–3471.
16 J. C. Bertolini, B. Tardy, M. Abon, J. Billy, P. Delichere and 42 L. L. Cai, G. Z. Lu, W. C. Zhan, Y. Guo, Y. L. Guo,
J. Massardier, Surf. Sci., 1983, 135, 117–127.
17 H. Igarashi, T. Fujino, Y. M. Zhu, H. Uchida and
Q. S. Yang and Z. G. Zhang, J. Mater. Sci., 2011, 46,
5639–5644.
M. Watanabe, Phys. Chem. Chem. Phys., 2001, 3, 306– 43 H.-J. Wan, B.-S. Wu, C.-H. Zhang, H.-W. Xiang,
314.
Y.-W. Li, B.-F. Xu and F. Yi, Catal. Commun., 2007, 8,
1538–1545.
44 L. Guczi, Catal. Rev. Sci. Eng., 1981, 23, 329–376.
18 E. Christoffersen, P. Liu, A. Ruban, H. L. Skriver and
J. K. Norskov, J. Catal., 2001, 199, 123–131.
19 J. Greeley and M. Mavrikakis, Nat. Mater., 2004, 3, 810– 45 D. Richard, J. Ockelford, A. Giroirfendler and P. Gallezot,
815. Catal. Lett., 1989, 3, 53–58.
20 A. U. Nilekar, Y. Xu, J. L. Zhang, M. B. Vukmirovic, 46 F. J. Berry, C. H. Xu and S. Jobson, J. Chem. Soc., Faraday
K. Sasaki, R. R. Adzic and M. Mavrikakis, Top. Catal., 2007,
46, 276–284.
21 W. P. Zhou, X. F. Yang, M. B. Vukmirovic, B. E. Koel, J. Jiao,
Trans., 1990, 86, 165–169.
47 J. Xu, C. H. Bartholomew, J. Sudweeks and D. L. Eggett,
Top. Catal., 2003, 26, 55–71.
G. W. Peng, M. Mavrikakis and R. R. Adzic, J. Am. Chem. 48 M. Yadav, D. K. Mishra and J.-S. Hwang, Appl. Catal., A,
Soc., 2009, 131, 12755–12762. 2012, 425–426, 110–116.
22 V. R. Stamenkovic, B. S. Mun, M. Arenz, K. J. J. Mayrhofer, 49 M. Mavrikakis and M. A. Barteau, J. Mol. Catal. A: Chem.,
C. A. Lucas, G. F. Wang, P. N. Ross and N. M. Markovic,
Nat. Mater., 2007, 6, 241–247.
23 M. H. Shao, P. Liu, J. L. Zhang and R. Adzic, J. Phys. Chem.
B, 2007, 111, 6772–6775.
24 H. J. Kim, S. M. Choi, M. H. Seo, S. Green, G. W. Huber and
W. B. Kim, Electrochem. Commun., 2011, 13, 890–893.
25 S. Sitthisa, T. Pham, T. Prasomsri, T. Sooknoi, R. G. Mallinson
and D. E. Resasco, J. Catal., 2011, 280, 17–27.
1998, 131, 135–147.
50 R. Alcala, J. Greeley, M. Mavrikakis and J. A. Dumesic,
J. Chem. Phys., 2002, 116, 8973–8980.
51 G. M. R. van Druten and V. Ponec, Appl. Catal., A, 2000,
191, 163–176.
52 S. Sitthisa, T. Sooknoi, Y. G. Ma, P. B. Balbuena and
D. E. Resasco, J. Catal., 2011, 277, 1–13.
53 N. Li and G. W. Huber, J. Catal., 2010, 270, 48–59.
26 S. Sitthisa, W. An and D. E. Resasco, J. Catal., 2011, 284, 54 B. M. Moreno, N. Li, J. Lee, G. W. Huber and M. T. Klein,
90–101. RSC Adv., 2013, 3, 23769–23784.
27 P. T. M. Do, A. J. Foster, J. G. Chen and R. F. Lobo, Green 55 Mineral commodity summaries 2012, U.S. Geological
Chem., 2012, 14, 1388–1397.
Survey, 2012.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 708–718 | 717

View Article Online
Paper
Green Chemistry
56 R. Weingarten, G. A. Tompsett, W. C. Conner and 61 M. Burgener, R. Wirz, T. Mallat and A. Baiker, J. Catal.,
G. W. Huber, J. Catal., 2011, 279, 174–182. 2004, 228, 152–161.
57 R. Weingarten, Y. T. Kim, G. A. Tompsett, A. Fernández, 62 P. Maki-Arvela, J. Hajek, T. Salmi and D. Y. Murzin, Appl.
K. S. Han, E. W. Hagaman, W. C. Conner Jr., J. A. Dumesic
and G. W. Huber, J. Catal., 2013, 304, 123–134.
58 Y. T. Kim, J. A. Dumesic and G. W. Huber, J. Catal., 2013,
304, 72–85.
59 N. Li, G. A. Tompsett and G. W. Huber, ChemSusChem,
2010, 3, 1154–1157.
Catal., A, 2005, 292, 1–49.
63 B. D. Li, J. Wang, Y. Z. Yuan, H. Ariga, S. Takakusagi and
K. Asakura, ACS Catal., 2011, 1, 1521–1528.
64 A. Yamaguchi, N. Hiyoshi, O. Sato and M. Shirai, Green
Chem., 2011, 13, 873–881.
65 N. Li, G. A. Tompsett, T. Y. Zhang, J. A. Shi, C. E. Wyman
and G. W. Huber, Green Chem., 2011, 13, 91–101.
60 F. Delbecq and P. Sautet, J. Catal., 1995, 152, 217–236.
718 | Green Chem., 2014, 16, 708–718
This journal is © The Royal Society of Chemistry 2014