View Online / Journal Homepage

Chem Soc Rev Dynamic Article Links

Cite this: DOI: 10.1039/c2cs35188a

www.rsc.org/csr CRITICAL REVIEW

Bimetallic catalysts for upgrading of biomass to fuels and chemicalsw
David Martin Alonso,a Stephanie G. Wettsteinab and James A. Dumesic*ab
Received 24th May 2012
DOI: 10.1039/c2cs35188a
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

Research interest in biomass conversion to fuels and chemicals has increased significantly in the
last decade as the necessity for a renewable source of carbon has become more evident.
Accordingly, many different reactions and processes to convert biomass into high-value products
and fuels have been proposed in the literature. Special attention has been given to the conversion
Downloaded by Cornell University on 09 August 2012

of lignocellulosic biomass, which does not compete with food sources and is widely available as a
low cost feedstock. In this review, we start with a brief introduction on lignocellulose and the
different chemical structures of its components: cellulose, hemicellulose, and lignin. These three
components allow for the production of different chemicals after fractionation. After a brief
overview of the main reactions involved in biomass conversion, we focus on those where
bimetallic catalysts are playing an important role. Although the reactions are similar for cellulose
and hemicellulose, which contain C6 and C5 sugars, respectively, different products are obtained,
and therefore, they have been reviewed separately. The third major fraction of lignocellulose that
we address is lignin, which has significant challenges to overcome, as its structure makes catalytic
processing more challenging. Bimetallic catalysts offer the possibility of enabling lignocellulosic
processing to become a larger part of the biofuels and renewable chemical industry. This review
summarizes recent results published in the literature for biomass upgrading reactions using
bimetallic catalysts.

a
1. Introduction
Department of Chemical and Biological Engineering,
University of Wisconsin-Madison, 1415 Engineering Drive, Madison, Converting lignocellulosic biomass into sustainable chemicals
Wisconsin 53706, USA. E-mail: [email protected] and fuels is of major interest to reduce dependence on fossil
b
DOE Great Lakes Bioenergy Research Center,
University of Wisconsin-Madison, Madison, Wisconsin 53706, USA fuel sources. In 2009, renewable energy accounted for 8% of
w Part of the bimetallic nanocatalysts themed issue. the energy consumption in the United States, with fossil fuels

Dr Martin Alonso earned his Dr Wettstein earned her BSc

Bachelor degree in Chemical in Paper Science at the
Engineering at the University University of Wisconsin –
of Salamanca (Spain) and his Stevens Point and her MSc
PhD at Catalysis and and PhD at the University of
Petroleochemistry Research Colorado – Boulder under the
Institute (Spain) under the supervision of Prof. John
supervision of Dr Rafael Falconer and Prof. Rich
Mariscal working in the Noble. Her thesis research
production of biodiesel using was on the influence of
heterogenous catalysis. In adsorbed species in zeolites as
2009 he moved to UW- well as zeolite membrane
Madison and presently he separations. In 2010 she
is working as a Research became a Research Associate
David Martin Alonso Associate with Prof. James Stephanie G. Wettstein under the guidance of Prof.
A. Dumesic at the University James A. Dumesic at the Uni-
of Wisconsin-Madison. His goal is develop and implement new versity of Wisconsin-Madison studying catalytic lignocellulose
catalytic processes to convert lignocelluloses into valuable conversion and upgrading. She will begin as an Assistant
chemicals and fuels. He is (co)author of 6 patents and 22 Professor at Montana State University in the Fall 2012.
peer-reviewed manuscripts and book chapters.

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

1.3 billion dry tons of biomass annually, which would offset

50% of the carbon demand for gasoline and diesel.4–6
In addition to the production of fuels, it is of interest to
replace petroleum derived chemicals with similar products
derived from biomass, or with new products that have different
properties. Chemicals from renewable sources can reduce the
environmental impact of their manufacture and/or disposal, for
example, biodegradable plastics.7 Combination of the production
of fuel and chemicals into an integrated biomass refinery would
reduce the price of the final products, which is one of the main
impediments to wide-spread biomass use.8
Another major challenge for the effective utilization of
Fig. 1 Sources of energy used in the United States during 2009.1
biomass is catalysis. Biomass is composed of long-chain,
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

oxygen rich molecules, which require different catalysts than

compromising the majority (Fig. 1).1 The Department of
the petroleum industry currently uses for efficient conversion.
Energy forecasted that by 2035, the amount of liquid fuels
Petroleum is underfunctionalized and requires functional
consumed in the U.S. from biofuels should increase 200% due to
groups to be added, typically by oxidation, amination, and
the Federal Renewable Fuels Standard, which requires 36 billion
hydration whereas, biomass is overfunctionalized and removal of
Downloaded by Cornell University on 09 August 2012

gallons of renewable fuel to be blended into transportation fuels

functionality, particularly oxygen containing groups, is required.
by 2022.2 Currently, the demand for renewable fuels is covered by
Many catalysts have been screened for biomass reactions,
the production of starch-derived ethanol and biodiesel; however,
but few are stable with time, recyclable, and selective to the
there remains a lack of inexpensive feedstock for these processes,
desired product. These problems become larger when real
and current feeds compete with food resources.3 Ideally, liquid
feedstock solutions are used, as many impurities and by-products
fuels would be made from second generation feedstocks
are present. Purification and separation methods are widely used
(e.g., crop and forest harvest residues),3 compatible with
to address some of these problems; however, separations increase
current combustion engines, and chemically identical to those
the cost of processing making it less attractive from a commercial
obtained from petroleum. A promising feedstock is lignocellulosic
point of view. Catalyst optimization can be carried out by using
biomass, which contains cellulose, hemicellulose, and lignin, and
combinations of active sites (metals), supports, and promoters.
is abundant in the form of agricultural residues (e.g., corn stover
The use of bimetallic catalysts is a promising option for biomass
and wheat straw), waste streams (e.g., pulp and paper mill and
feedstock upgrading, as the interaction between metals can
food waste), wood, and energy crops (e.g., switch grass and
modify the surface properties of the catalyst. These changes
miscanthus).4 The United States alone could sustainably produce
can significantly improve the catalytic activity, modify the
selectivity to the desired product, and increase the catalyst
stability in presence of biomass-derived impurities and more
Professor James A. Dumesic severe reaction conditions. New bimetallic catalysts need to
earned his BS degree from be optimized specifically for each reaction ensuring high
UW-Madison and his MS and reaction rates, high selectivity, and yields that lead to effective
PhD degrees from Stanford purification steps.
University, under the supervi-
sion of Professor Michel
Boudart. Dumesic joined the 2. Components of lignocellulosic biomass
Department of Chemical Lignocellulosic biomass is the most promising renewable
Engineering in 1976, and he is
source of carbon, as it is widely available around the world
currently the Steenbock Chair
in the College of Engineering at a relatively low cost. It is composed of three main fractions:
and the Michel Boudart cellulose, hemicellulose, and lignin (Fig. 2). Cellulose
Professor of Chemical and Bio- comprises 40–50% of lignocellulosic biomass feedstock9 and
logical Engineering. Through- is a homopolymer consisting of b-D-glucopyranose units linked
James A. Dumesic out his career, Dumesic has via b-glycosidic bonds10 (Fig. 2) that can be deconstructed into
used spectroscopic, microca- glucose monomers. Recent interest in using cellulose as source of
lorimetric, and reaction kinetics techniques to study the surface chemicals and fuels via catalytic processing is reflected in recent
and dynamic properties of heterogeneous catalysts. Dumesic reviews on the topic.11–13 Hemicellulose compromises 15–30% of
pioneered the field of microkinetic analysis, in which diverse lignocellulosic biomass,14 and unlike cellulose, hemicellulose is
information from experimental and theoretical studies is com-
amorphous with a low degree of polymerization (B200) and is a
bined to elucidate the essential surface chemistry that controls
branched polymer (Fig. 2). As such, hemicellulose is easier to
catalyst performance. He has recently studied how aqueous-phase
reforming of biomass-derived carbohydrates can be tailored to hydrolyze than cellulose, which allows hemicellulose to be
selectively produce H2 or directed to produce liquid alkanes. Most removed under milder reaction conditions. Additionally, the
recently, he has been studying the use of levulinic acid and structure of the hemicellulose varies depending on the source
g-valerolactone as biomass-derived platform chemicals for the of the lignocellulose and may consist of pentoses, hexoses, and
production of fuels and chemicals. uronic acids.15 The pentoses (xylose and arabinose) and hexoses

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A
Downloaded by Cornell University on 09 August 2012

Fig. 2 Structure of lignocellulosic biomass with cellulose, hemicellulose, and lignin represented. The building blocks of lignin, p-coumaryl,
coniferyl, and sinapyl alcohol are also shown.

(glucose, galactose, mannose, rhamnose, and fucose) are sources removed either non-selectively, as in the case for thermochemical
of sugars that can be upgraded to biofuels and renewable processes such as pyrolysis, or selectively through dehydration,
chemicals. Most hemicellulose upgrading research has focused hydrogenolysis, decarboxylation, and decarbonylation, as in the
on the main sugar, xylan, which comprises from 8–25% of the case for liquid-phase catalytic processing.21,22 Additionally,
biomass14 and will be the focus of this review as representative of although the functionalized groups allow for many different
the hemicellulose fraction. Studies on xylose conversion typically specialty chemicals to be produced from biomass, a major
focus on the production of furfural and derivatives16–18 due to challenge is controlling the selectivity of the reactions, as many
the value-added chemicals that can be made from furfural. The of the targeted products are intermediates of other products, or
final major component of lignocellulosic biomass is lignin, which, degrade in the reaction medium. For this reason, one of the most
like hemicellulose, is an amorphous polymer that varies in effective ways to process lignocellulosic biomass is in sequential
structure depending on the biomass source.10 Lignin makes up steps. In the first step, oxygen is partially removed, which reduces
15–30% of the biomass weight, is rich in aromatic functionality, the reactivity of the feedstock, and forms molecules, denoted as
and consists of three primary monomers (Fig. 2). A value-added building blocks or platform chemicals.18 In a second step, the
opportunity for processing biomass is converting the lignin remaining functionality is further upgraded to other chemicals or
into chemicals and fuels using the building blocks of lignin: fuels. For example, in the production of fuels, the upgrading
p-coumaryl, coniferyl, and sinapyl alcohols. Lignin contains strategy must involve the formation of carbon–carbon bonds to
approximately 40% of the possible energy of the biomass due increase the molecular weight of the 5 and 6 carbon monomers23
to its high carbon content. A detailed consideration of lignin, to fuel range, 8 (gasoline) to 20 (diesel fuel) carbon atoms.
as well as possible applications, has been recently reviewed by Platform chemicals such as levulinic acid can be upgraded to
Zakzeski et al.19 oxygenated compounds, such as gamma-valerolactone (GVL),
Using chemical, thermal, and microbiological processes, dimethylfuran (DMF), or methyltetrahydrofuran (MTHF), that
each lignocellulosic fraction can be removed from the structure have sufficient energy density to be blended in gasoline for use in
and isolated for separate processing. This takes advantage of gasoline engines.
the different structure and composition of the fractions to As mentioned above, when simplicity of the process is
produce different chemical products and fuels by optimizing preferred versus selective processes, lignocellulosic biomass
the reaction conditions required for each. Alternatively, and can be converted in one step without requirement of fractionation
when simplicity of the process is more important than selectivity (Fig. 3). Biomass can be treated at high temperatures to produce
to target molecules, the entire structure can be converted using synthesis gas by gasification,24–26 and then converted into chemicals
processes as gasification or pyrolysis. by reactions such as Fischer–Tropsch synthesis. The lowest cost
liquid fuel produced from biomass can be obtained by pyrolysis/
liquefaction,27–30 but this bio-oil consists of a mixture of more than
3. Biomass conversion strategies
300 chemicals, which are difficult to separate. Bio-oil is a poor fuel,
Lignocellulosic biomass conversion faces different challenges as it contains high amounts of water (up to 50%) and many highly
than petroleum refining. Since biomass is highly oxygenated oxygenated products. Both gasification31–33 and liquefaction34,35
and highly functionalized, it requires increasing the energy have used bimetallic catalysts for upgrading of syn-gas and bio-oil,
density when fuels are the target product, and reducing the respectively. For gasification, Nishikawa et al. found that the
reactivity to selectively produce high value chemicals at high addition of Pt to Ni based catalysts improved product selectivity
yields.20,21 To increase the energy density, oxygen can be due to alloy formation.32 In the case of pyrolysis, Ardiyanti et al.

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A
Downloaded by Cornell University on 09 August 2012

Fig. 3 Routes for synthesizing renewable fuels and chemicals from lignocellulosic biomass. Opportunities for bimetallic catalysts are shown by
gray arrows [Adapted from ref. 36].

found that the addition of Cu to a Ni/d-Al2O3 significantly As previously mentioned, adding additional metals to
increased the HDO of fast pyrolysis oil compared to the monometallic catalysts can modify the surface properties to
monometallic catalyst.34 improve catalytic activity, can modify the selectivity to the
Liquid-phase processing18,22,24,25,36–40 is a more complex desired product, and can increase the catalyst stability in
process, since multiple steps are required to produce high presence of biomass-derived impurities and under more severe
value chemicals or fuel, but this approach allows for more reaction conditions.
selective processing (Fig. 3). Liquid-phase catalytic processing of Technically viable and economic conversion of ligno-
biomass has been extensively researched because temperatures are cellulosic biomass into chemicals and fuels is an important
relatively low compared to gasification and pyrolysis. While challenge that requires integrated processing and the effective
simultaneous processing of C5 and C6 sugars offers the potential utilization of both hemicellulose and cellulose. The additional
for simplicity of operation, as in gasification and pyrolysis, cost of catalysts is a valid concern. The use of catalysts in the
fractionation of hemicellulose and cellulose allows for separate petroleum industry adds approximately $0.01 per gallon of
processing that can be tailored to exploit the different chemical gasoline.40 Deconstructing the biomass alone costs more than
and physical properties of these fractions, which provides $0.01 per gallon and recent studies estimate the cost of
increased flexibility of operation. For example, it may be desirable enzymes contributes $0.60–$1.60 per gallon of ethanol.42
to employ chemical processing steps to convert the C5 sugars to Additionally, bimetallic catalysts are often used to upgrade
fuels and/or chemicals, while taking advantage of the physical the feedstock, which adds additional costs to the process,43,44
properties of cellulose for pulp and paper applications. and therefore, the additional cost in using catalysts must be
Using lignocellulosic biomass for the production of fuels recovered in increased product value.
and chemicals follows the general strategy outlined in Fig. 3.
The biomass is deconstructed using thermal, enzymatic, and/
4. Upgrading reactions over bimetallic catalyst
or chemical/catalytic methods to produce functional inter-
mediates that are then upgraded to a range of end products.22 Among the several reactions involved in lignocellulosic
An advantage of chemical/catalytic conversion methods is biomass conversion and upgrading, we will focus on those
that the processing conditions are mild relative to thermal that have used bimetallic catalysts, including reforming,
methods, allowing for a more tailored product distribution hydrogenolysis, hydrogenation, and oxidation. Other reactions,
due to the flexible operating conditions. Typically a pretreat- such as sugar hydrolysis and C–C coupling reactions like
ment step removes the hemicellulose and some lignin from the oligomerization, aldol condensation, and ketonization will
biomass.16,41 The hemicellulose stream can then be upgraded to be mentioned when they are essential to meeting the target
furfural, for example, which is a platform chemical that can be products. These reactions typically use acid–base and/or metal
upgraded using bimetallic catalysts to a variety of final products.17 oxide catalysts instead of bimetallic catalysts, which are the
The lignin can be burned for heat, or upgraded using bimetallic main focus of this review. Further details of these reactions
catalysts to aromatic compounds and solvents.19 The remaining can be found in other reviews related to lignocellulosic
solids, the cellulose fraction of the biomass, are then hydrolyzed to biomass conversion.22,45,46 However, we did include some
produce glucose or platform chemicals, such as levulinic acid or metal oxides in this review when they are commonly used, as
5-hydroxymethylfurfural (HMF). in the case of glycerol and lignin upgrading.

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

4.1. Reforming cost of the process. Metal catalyst such as Pd, Pt and Au are
commonly used, and the selectivity depends on the metal used
Catalytic reforming reactions cleave the C–C bonds of the
and if any promoters are present. For example Bi, is typically
carbohydrate backbone to yield a combination of carbon
used as a promoter, which modifies the selectivity to oxidize
monoxide and hydrogen (eqn (A)).47 If water is present,
different functional groups in different positions.
carbon dioxide can be formed via the water-gas-shift reaction
to produce more hydrogen (eqn (B)). Aqueous-phase reforming O2
R  CH2 OH ! R  CHO ðFÞ
(APR) using bimetallic catalysts can produce hydrogen and
carbon dioxide starting with sugars and sugar alcohols.48 The O2
selectivity towards hydrogen can be controlled using bimetallic R  CHO ! R  COOH ðGÞ
catalysts with different metal sites (e.g., Pt) and metal-alloys
(e.g., NiSn).49 4.5. Importance of bimetallic catalysts
CnH2yOn 2 nCO + yH2 (A) The reactions mentioned above are catalyzed by metals. In
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

general, product yields (selectivity and activity) are determined

CO +H2O 2 CO2 + H2 (B)
by the reaction conditions and reactant used. However,
catalyst modification (e.g., addition of supports, promoters,
4.2. Hydrogenolysis
preparation method) can modify the selectivity to a particular
Hydrogenolysis reactions require the cleavage of C–C and product. In this sense, the introduction of a second metal as a
Downloaded by Cornell University on 09 August 2012

C–O bonds. Eqn (C) shows the stoichiometry of hydrogenolysis, promoter has been demonstrated to be successful in maximizing
where X represents an alkyl chain, OH, OR, SH, NH2, yields to desired products. Improving yield is important when
etc.50 In the case of removing oxygen from the molecule, working with biomass due to the high functionality of the initial
hydrogenolysis is typically referred to as hydrodeoxygenation. molecules, which leads to the formation of by-products that
Hydrogenolysis is an important reaction in biomass upgrading reduce final yields and complicate the purification of the
to form polyols, alcohols, and hydrocarbons from glycerol and targeted product. The use of bimetallic catalyst is an interesting
carbohydrates.50 Hydrogenolysis of C–C and C–O bonds in strategy as it offers many possibilities. From a practical point of
polyols usually occurs under basic conditions with supported view, the effect of adding a second metal can be classified in
metal catalysts including Ru, Pd, Pt, Ni, and Cu.50–55 The three main categories: (i) increase catalytic activity, (ii) modify
product selectivity and yields can be controlled by using the selectivity, and (iii) improve catalyst stability. Promotion
bimetallic catalysts.56 may be as simple as the second metal helping to reduce the first
H2 ;cat:
metal by hydrogen spill-over, or promotion may be caused by
R  X ƒƒƒ! R  H þ H  X ðCÞ more complex metal interactions, which can be divided into five
different phenomena:
4.3. Hydrogenation (1) Geometric (or ensemble) effects, wherein the addition of
Hydrogenation reactions saturate the C–C double and triple a metal promoter alters the geometry (or ensemble size) of the
bonds, as well as CQO bonds, increasing the hydrogen active sites comprised of the catalytically active metal.
content (alkene hydrogenation, eqn (D); aldehyde reduction, (2) Electronic (or ligand) effects, wherein the addition of a
eqn (E)) Metal catalysts, such as Pd, Pt, Ni, or Ru, can be metal promoter alters the electron properties of the actives
used, and product selectivity for the hydrogenation reaction sites comprised of the catalytically active metal by electron
depends on factors such as the solvent, partial pressure of transfer between metals (or by acting as a ligand).
hydrogen, temperature, and the nature of catalyst. For example, (3) Stabilizing effects, wherein the addition of a metal
while some metals, as monometallic Pd or Pt are more favorable promoter improves the stability of the catalytically active
for the hydrogenation of CQC bonds, the addition of supports metal (e.g., by inhibiting sintering processes or suppressing
and promoters to form bimetallic catalysts, as RuSn, are more the formation of carbonaceous deposits on the catalyst).
selective to CQO hydrogenation.57–59 Important differences are (4) Synergistic effects, wherein both metal species partici-
found when the CQC is present in ring or in a linear hydro- pate in the chemical bonding to reaction intermediates and
carbon, as the former are typically more easily hydrogenated. transition states.
(5) Bi-functional effects, wherein each metal species provides
H2
R  C ¼ C  R ƒ! R  C  C  R ðDÞ a different function in the reaction mechanism (e.g., one metal
catalyzes hydrogenation–dehydrogenation reactions, while the
H2 second metal catalyzes C–O hydrogenolysis reactions).
R  CHO ƒ! R  CH2 OH ðEÞ
Typically, geometric and electronic effects affect the catalytic
activity and selectivity, while stabilizing effects improve the
4.4. Oxidation
catalyst stability. Synergistic and bi-functional effects normally
Oxidation reactions consist of the reduction of the hydrogen lead to improvement in the reaction rates, but may also lead to
content of an organic molecule (e.g., alcohol oxidation to the creation of new reaction pathways.
aldehyde (primary alcohol, eqn (F)) or ketone (secondary An additional effect of promotion is caused when the second
alcohol)) or an increase in the oxygen content (alcohol/aldehyde metal does not act as a metal, but introduces a second
oxidation to a carboxylic group, eqn (G)). Oxidation is typically functionality in the catalyst; for example, acid or redox sites
carried out using molecular oxygen or air, which reduces the can be introduced by the presence of metal oxides. In these

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

cases, the reaction may take place over a bifunctional catalyst, associated with APR as well as the optimization of process
which may modify the selectivity compared to a pure metal conditions; technical and fundamental details can be found in
catalyst. The difference between a bimetallic catalyst and a reviews by Dumesic et al.36,65 Overall, APR produces H2 and
bifunctional catalyst may be hard to establish, and it will depend CO (synthesis gas) from biomass derived molecules at low
on the degree of reduction of the metals present, as metal oxides temperatures (500–600 K) in the liquid phase. From an
can be formed when one of the metals is not fully reduced. energetic point of view, this is an energy advantage versus
Many bimetallic catalysts have been proposed to upgrade other strategies to produce synthesis gas, as gas phase reactions
biomass to chemicals and fuels, and the advantages of using (e.g., gasification) require high temperatures (800–1000 K). For
bimetallic catalysts are evident; however, there is still a lack of example, temperatures over 675 K are required for steam
knowledge on the mechanisms responsible for these improvements, reforming alkanes, while the steam reforming of oxygenated
and more studies are required to establish correlations between hydrocarbons is thermodynamically favorable at lower
results and catalyst modifications at the molecular and electronic temperatures.66 However, the main disadvantage of APR is
levels. This information will help to better understand the processes that relatively pure molecules are necessary as feedstock, and thus,
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

and design new catalysts in the future for improved performance. it is not possible to use lignocellulose directly as in the case of
Theories and fundamental phenomena established for catalytic gasification. APR is one of the most promising alternatives when
processes in the more-studied petroleum industry can be used as a high purity H2 is required, as in the case of fuel cells. Using pure
starting point for catalysis in biomass conversion, since the catalysts feedstocks reduces the requirement of downstream separation/
used are similar. However, results and conclusions obtained from purifications and using lower temperatures favors the production
Downloaded by Cornell University on 09 August 2012

hydrocarbon reaction studies cannot necessarily be directly applied of CO2 and H2 in the WGS reaction, thus reducing the
to biomass due to the different upgrading strategies for biomass content of CO, which is responsible for deactivation of
(focused on the removal/replacement of functional groups from a electrodes in fuel cells.67–69
highly functionalized molecule for biomass versus the addition of Another advantage of the APR process is that the selectivity
functional groups to a low-functionalized molecule for petroleum.) and reactivity of the catalyst can be modified to produce
Additionally, the reactants and impurities introduced by biomass alkanes instead of H2. Production of alkanes requires a
are new, and their interactions with the catalysts are not well combination of dehydration and hydrogenation reactions,
known, since few studies have been carried out in the presence of and can be accomplished by using a metal catalyst in the
real biomass. presence of a mineral acid, however, the mineral acids typically
Herein, we will focus on the recent advances achieved using deactivates the metal catalysts resulting in low yields. Therefore,
bimetallic catalysts for the conversion of biomass derived the use of heterogeneous and bifunctional catalysts is preferred,
molecules. We will pay attention to the improvements that and the acid functionality is normally introduced in the support
the addition of a second metal have on catalytic activity using acid and amphoteric oxides as Al2O3 or SiO2. Wawrzetz
and selectivity versus monometallic metals at similar reaction et al.70 studied the importance of bifunctional catalysts in the
conditions. More fundamental knowledge on the effect of APR of glycerol.
adding a second metal into a metal catalyst can be found in As listed in Table 1, many metals can catalyze H2 production
other reviews.60–64 from oxygenated compounds using APR. Ru, Rh, Ir and Ni are
active for C–C bond cleavage.65 Other metals such as Co and Fe
are also active for C–C bond cleavage, but less so than the
5. Aqueous phase reforming: production of hydrogen
former; however, Co and Fe catalyze the WGS reaction, which
One reaction that has attracted considerable interest is aqueous is necessary in the process.65 Cu is also active for WGS
phase reforming (APR) for the production of H2 from biomass. reactions, but it is unable to catalyze C–C cleavage.45 Another
Dumesic et al. have studied the thermodynamic and kinetics consideration to maximize the production of H2 is the reaction

Table 1 Typical bimetallic and their respective monometallic catalyst for APR of different feedstocks

Catalyst Feedstock Conversion (%) H2 selectivity (%) Alkane selectivity (%) P (bar) T (K) Ref.
s
RANEY Ni14Sn Sorbitol >90% 46 31 51 538 71
Pt/Al2O3 Sorbitol >90% 46 32 56 538 71
RANEYs Ni14Sn Glycerol >90% 76 17 51 538 71
Pt/Al2O3 Glycerol >90% 51 31 56 538 71
RANEYs Ni14Sn Ethylene glycol >90% 95 7 51 538 71
Pt/Al2O3 Ethylene glycol >90% 88 8 56 538 71
RANEYs Ni Ethylene glycol >90% 28 47 51 538 71
Ni/Al2O3 Ethylene glycol 0.9 30 38 25 483 72
Pt/Al2O3 Ethylene glycol 1.4 92 0 25 483 72
Co/Al2O3 Ethylene glycol 0.1 129 23 25 483 72
Pt1Ni1/Al2O3 Ethylene glycol 3.9 88 0 25 483 72
Pt1Co5/Al2O3 Ethylene glycol 3.8 94 0 25 483 72
1.5 Pt/Al2O3 Ethylene glycol 42 80 7 250 723 73
1.15 Pt0.35Ni/Al2O3 Ethylene glycol 79 83 5 250 723 73
3%Pt/C Glycerol 5.3 56.5 28 498 74
3%Pt/1%Re/C Glycerol 52.4 30.8 28 498 74

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

selectivity, since many of the catalysts that are active in the

conversion of oxygenated compounds also catalyze side reactions,
such as Fischer–Tropsch synthesis and methanation. With all
these considerations, Pt and Pd catalysts are the most frequently
used metals, as they catalyze C–C bond cleavage and the WGS
reaction at moderate rates. The high yields obtained with these
catalysts can be improved significantly by using bimetallic
catalysts65 that promote the WGS reaction or improve C–C
bond cleavage. A high rate of C–C cleavage will produce CO
and H2 and, if followed by WGS, the CO can be converted
into more H2. In contrast, a high rate of C–O cleavage
followed by hydrogenation will produce alkanes (Fig. 4). In
this sequence of reactions, if hydrogenation of the dehydrated
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

products generated after the C–O cleavage is slower than the

rate of C–C cleavage, lighter alkanes are produced generating
H2, which is necessary if no external H2 is supplied, since the
formation of alkanes with equal number of carbon atoms as
the feedstocks requires the addition of H2. Several feedstocks Fig. 5 Overview of aqueous phase reforming for the production of
Downloaded by Cornell University on 09 August 2012

have been studied for APR, and among them the oxygenated synthesis gas and hydrogen.
compounds with a C : O ratio equal to 1 have the highest
yields of H2. The maximum theoretical yields of H2 that can be glycerol conversion at Pt : Ni ratios of 1 : 18, but the glycerol
produced is determined by the amount of carbon in the was converted into 1,2 propylene glycol and ethylene glycol
feedstock (Fig. 5), and depends on the extent of the WGS instead of into H2.75 Huber et al.72 proposed that the increase
reaction, which can account up to 50% of the total yield. in activity over PtNi arose from the formation of alloys
Ethylene glycol was one of the first molecules studied in APR between the metals that cause a reduction in the heats of
due to its simplicity (Fig. 5), and therefore, the number of adsorption of CO and H2,76,77 which increased the availability
by-products that can be generated is small. Davda et al.65 of surface sites for the APR reaction. This effect is important
studied the activity and selectivity of many catalysts and because the addition of Ni increases the number of active sites
established that monometallic Pt was the most active catalyst, present on NiPt alloys, and these sites are more active than
followed by monometallic Ni, Ru, Rh, Pd, and Ir. Using Pt as monometallic Pt ones. An excess of Ni on the surface reduces
the base metal, the APR activity was increased 2–3 times by the interaction with the Pt, leading to the formation of
adding Ni, Co, or Fe. These latter metals are highly active in monometallic Ni, which catalyzes the production of alkanes.
the WGS reaction, and increased H2 production by conversion Finally, the presence of NiPt alloys reduced the methane
of the CO into CO2.36,72 The best TOF using ethylene glycol as formation rate, since Ni is active for the WGS reaction;
feed was observed over a PtNi catalyst with a Pt : Ni atomic therefore, the CO is converted to H2 instead of CH4. Detailed
ratio of 1 : 1. Further increments in Ni addition led to the surface studies were carried out by Skoplyak et al.78 who
formation of alkanes and other by-products due to C–O bond determined that the best results were achieved over Ni–Pt–Pt
cleavage over the Ni, which reduced the production of H2. (111). de Vlieger et al.73 studied a PtNi catalyst for APR of
Iriondo et al. confirmed this change in selectivity by studying ethylene glycol using supercritical water as solvent. They
the effects of increasing the amount of Ni on a Pt catalyst observed that in addition to an increase in conversion using
using glycerol as feedstock. They observed a slight increase in the bimetallic catalyst, the presence of Ni reduced the sintering

Fig. 4 H2 and alkane production by APR.

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

of Pt even at high temperatures (723 K), thereby improving the Replacing noble metals with non-precious metals, such
catalyst stability in hydrothermal conditions. as Ni, Co or Cu, is desirable to reduce catalyst costs, but
An increase in the total number of active sites was not non-precious metal catalysts tend to have poor selectivity and
observed when Co was added to Pt/Al2O3, however, an deactivate quickly. By adding promoters to the non-precious
increase in the catalytic conversion of ethylene glycol was metal catalysts, some of these problems can be solved. Mono-
observed. The higher TOF indicates that the sites created upon metallic Ni is a good catalyst for oxygenated hydrocarbons
the addition of Co are more active for the production of H2 conversion since it is able to cleave C–C bonds and has high
than the monometallic Pt.72 At 483 K, the addition of Fe activity for WGS. Unfortunately, it also cleaves C–O bonds
increased the catalytic activity, but not as much as the addition and forms methane, i.e., it is not selective for H2 production.
of Ni and Co. The increase in activity from Fe addition is more Adding Sn to Ni as a promoter improved the H2 selectivity
pronounced at lower temperatures (453 K), where using a 1 : 9 because the formation of methane and alkanes was reduced in
Pt : Fe catalyst, the conversion of ethylene glycol to H2 presence of NiSn alloys49 while the C–C bond cleavage rate
increased 3 times compared with the monometallic Pt/Al2O3 remained high. As more Sn was added, higher selectivities to
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

catalyst.72 Recently, Wang et al.79 also studied the addition of H2 were achieved. At a Ni : Sn atomic ratio of 14 to 1, the H2
Co to Pt supported over carbon nanotubes and observed selectivity was nearly 90% with no production of alkanes.71
increases in H2 selectivity when Co was added, as well as an At low temperatures (498 K), the behavior is similar to that of
increase in catalytic activity. Pt/Al2O3. When the temperature was increased 40 degrees
Another noble metal commonly used in APR due its high (538 K), higher yields of H2 versus alkanes were produced over
Downloaded by Cornell University on 09 August 2012

activity is Pd. In this case, no significant differences were found NiSn using glycerol and ethylene glycol as feeds. The addition
when Ni, Co, and Mn were added to form bimetallic PdNi, of Sn increased the production of H2 and alcohols and reduced
PdCo and PdMn,72 and only Fe showed a promotional effect. acetic acid and alkane production.49 Additionally, RANEYs
Adding Fe to Pd at a ratio of 1 : 9 increased H2 production by NiSn alloys have been demonstrated to be stable for more
up to 46 times compared to the monometallic Pd catalyst; than 340 h at high conversion, which is competitive with
however, this catalyst deactivated more rapidly than the PtFe monometallic noble metals, such as Pt.
catalyst. Also, the intrinsic activity of monometallic Pd is Instead of Sn, Co was added as promoter to Ni/Al2O3
lower than that of monometallic Pt, but Pd is more selective. catalysts and increased the activity for glycerol aqueous phase
The rate limiting step using Pd with ethylene glycol was the reforming due to interaction between the two metals. The
WGS reaction, and the addition of Fe acted as a WGS catalytic system was further improved by the addition of Ce,
promoter, which increased the APR activity to produce H2. which limited catalyst deactivation by suppressing particle
An option to maximize the yields of H2 is to modify the sintering and reducing the formation of methane.86,87 Cheng
reaction system, as was proposed by Kunkes et al. using a dual- et al.88,89 studied the addition of Co to Ni, but at atmospheric
catalytic bed.80 Reaction conditions and catalyst performance pressure and high temperatures (greater than 773 K), such that
were optimized in the first bed to maximize the production of H2 the reforming took place in gas phase instead of liquid phase.
and CO using a catalyst active in C–C bond cleavage. Then, in a They reported less than 10% H2 yield when using both metals
second bed, reaction conditions and catalyst were optimized to individually, but the H2 yields over the bimetallic catalyst were
convert the CO into H2 by using an active WGS catalyst that is up to 40% depending on the amount of Co added. The
not necessarily active in C–C cleavage. For the first bed, they addition of Co increased the H2 selectivity by reducing the
studied several bimetallic Pt/C catalysts (Ru, Re, Os, Sn, Fe, and formation of lower hydrocarbons such as propane and
Cu), and for the second bed they selected Pt/CeO2/ZrO2. PtRe/C ethane,88 and lowered the reduction temperature of the
was the best bimetallic for APR since the addition of Re to Pt catalyst versus the respective monometallic catalysts. Even
decreased the CO coverage, which increased the glycerol though better results were found by adding Co to the catalyst,
conversion 5 and 2 times while co-feeding H2 and in the deactivation of the catalyst by coke formation still occurred.89
absence of external H2 (573 K), respectively.81 According to Chang et al.90 used cellulose as a feedstock for APR with
Ruettinger et al.,82 a weaker interaction of CO over PtRe was PtRu/C and sulfuric acid catalysts for H2 production. H2
observed when small bimetallic particles were present in the yields were low (0.47 mmol H2 per gram of cellulose)
catalyst; however, a stronger CO interaction was observed and further improvements would be required to use biomass
when large particles are present indicating Kunkes et al.80 had feedstock in one reactor. Recently, Zhao et al.91 reported
small PtRe particles present. better yields using a NiCo catalyst on a SBA-15 support
Besides particle size, an important effect to consider when adding (>10 mmol H2 per gram of cellulose in presence of water.).
Re is that the Re must be in its reduced state. In another case Cai et al. used Ir to promote Co/ZnO in the reforming
reported by Zhang et al.,83 the Re was oxidized, which created weak of butanol.92,93 They worked at atmospheric pressure and
acid sites that modified the selectivity of the catalyst to produce temperatures over 773 K (steam reforming instead of APR)
alkanes and liquid product instead of H2 by favoring C–O bond and they observed that the addition of Ir favors the reduction
cleavage as observed by King et al.74 When they added Re to Pt, of Co species leading to the reduction in the formation of
the glycerol conversion increased by one order of magnitude and by-products such as ethylene and coke, which increases
the H2 selectivity was reduced due to the formation of more liquid catalyst stability.
products (alkanes and oxygenated). This effect can be ameliorated if As previously mentioned, if C–O bond cleavage occurs
the presence of ReOx increases the rate of WGS generating more faster than the rate of C–C bond cleavage, APR can be used
H2 by reducing the ReOx with CO84,85 to produce CO2. to produce light alkanes.36,94 The major limitation for this

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A
Downloaded by Cornell University on 09 August 2012

Fig. 6 Conversion of polyols to monofunctional intermediates by PtRe/C.

reaction is that the initial number of carbon atoms in the feed- out the deoxygenation reactions. Instead of alkanes, carboxylic
stock determines the distribution of the alkanes produced. From acid, ketones, secondary alcohols, and heterocyclic compounds
biomass derived molecules, the preferred feedstock for alkane were produced at various conditions. These compounds can be
production is sorbitol, which has 6 carbon atoms and therefore, converted to higher molecular weight hydrocarbons by C–C
can produce alkanes ranging from hexane to methane. External coupling reactions using several processes and combining acid,
H2 can be used in this process; however, if there is no supply of basic, and metal catalysis in successive steps, as can be seen in
external hydrogen, the reaction begins with the production of H2 Fig. 6.95 Using low molecular weight polyols, Ru, Pd, Pt, Re,
and CO2 over a metal catalyst by APR. The next step, over an Ni, and Cu5,45,96,97 were the preferred catalysts, and to date, no
acid site (typically present in the metal catalyst support), consists optimization of the process using bimetallic catalyst have been
of dehydrating the oxygenate molecule and hydrogenating the published.
intermediate generated (CQC) to make the final alkane. With
this pathway and with no external H2 supply, some of the feed 6. Catalytic conversion of glycerol
needs to be converted into CO2 and H2 (eqn (H)–(J)).
As previously mentioned, glycerol can be obtained from
C6H14O6 + 6H2 - C6H14 + 6H2O (H) cellulose and C6 sugars by hydrogenolysis, but currently the
C6H14O6 + 6H2O - 6CO2 + 13H2 (I) most important source of glycerol is as a by-product in the
production of biodiesel. Biodiesel has increased its volume of
19 42 36
C6 H14 O6 ! C6 H14 þ H2 O þ CO2 ðJÞ production, such that glycerin is no considered to be a high-
13 13 13
value chemical, but an abundant and inexpensive feedstock
For example, starting from sorbitol it is not possible to that can be used for the production of biomass-derived
obtain 100% yield of hexane without the addition of H2 and chemicals and fuels. Approximately 0.1 kg of glycerol is
therefore, lower hydrocarbons and CO2 are produced in the produced per kg of biodiesel, and 802 million gallons of
process. Increasing the rate of C–C bond cleavage produced biodiesel have been produced in 2011 in the U.S. alone.98
lighter alkanes due to higher H2 requirements. Unlike APR to Accordingly, many upgrading reaction have been studied in the
produce H2, using APR to produce light alkanes has not been literature that takes advantage of the high functionalization
extensively studied. present in the molecule (1,2,3 propanetriol), increasing the
An advantageous conversion strategy is the partial deoxy- market opportunities for glycerol.99,100 Special attention will
genation of polyols. This pathway combines reforming with be paid here to those products that can be produced using metal
hydrogenolysis, such that by controlling the rate of C–C bond and bimetallic catalyst (Fig. 7) even though other reactions such
cleavage and C–O bond cleavage, it is possible to produce as etherification, amination or dehydration are as important
monofunctional hydrocarbons that retain enough functionality considering the value and interest of the final products obtained
to be upgraded to targeted molecules.47 Addition of Re to Pt by those reactions.
proved to be the most effective catalyst, as it reduced the
6.1. Glycerol hydrogenolysis
C–C bond cleavage on Pt and increased the C–O bond
hydrogenolysis rate. It had sufficient reforming activity to As mentioned in section 5, H2 production by aqueous phase
convert part of the feed into H2, which could be used to carry reforming is one of the most important reactions of glycerol.

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A
Downloaded by Cornell University on 09 August 2012

Fig. 7 Glycerol conversion to high value chemicals and fuels.

If instead of using a catalyst that favors C–C bond cleavage,

catalysts favoring C–O bond cleavage are used, another interesting
alternative becomes possible: hydrogenolysis to produce other
polyols (e.g., 1,2 propanediol, ethylene glycol), alcohols
(e.g., 1-propanol), and hydrocarbons (e.g., propene, propane).
Recently, Palkovits et al. published a detailed review about the
hydrogenolysis of biomass-derived molecules,101 including
more fundamental details about the reaction and catalysts
used. Two mechanisms have been proposed for the hydro-
genolysis of glycerol to produce diols (Fig. 8). The first one,
developed by Montassier et al.,102 proposed the formation of
glyceraldehyde (in equilibrium with the enolic tautomer). This
pathway has been associated with the presence of base and
metal catalysts. The second mechanism, developed by Dasari
et al.,51 proposed the formation of acetol as an intermediate,
due to the presence of acid and metal catalysts.
Typically, hydrogenolysis of glycerol is carried out in
aqueous solution at 373–523 K and at high pressures (up to
200 bar) using metals catalyst such as Pd, Pt, Ru, Ni, Cu
supported on carbon, SiO2, Al2O3, or other solids with low
acidity. The presence of some acidity is advantageous, as a
balance between dehydration and hydrogenation reactions is Fig. 8 Glyerol hydrogenolysis to 1,2 propanediol. (A) In the presence
required for effective hydrogenolysis. In general, Ru has been of a base via glyceraldehyde and (B) in the presence of an acid via
acetol.
reported to be more active than transition metal catalysts and
other noble metals such as Pd and Pt. However, Ru is less
selective for the production of high value products like scale using CuCrO4 catalyst at 473 K and 10 bar. Catalyst
1,2 propanediol (1,2 PDO) and 1,3 propanediol (1,3 PDO), deactivation and Cr leaching are pushing researchers to use other
and leads to fully hydrogenated products such as propane, bifunctional and bimetallic catalysts. Miyazawa et al.54,103,104
which has considerably lower value. Efforts have focused on proposed the addition of acid catalysts, such as ion-exchange
increasing the selectivity of the noble metals and the activity of resins or mineral acids, to Ru/C, Rh/C, Pt/C, and Pd/C to
the transition metals by using promoters to maximize the improve the dehydration reaction. Other authors combined the
production of high value products. Currently, the hydrogenolysis high activity of noble metal catalysts with the high selectivity
of crude glycerol to propylene glycol is carried out at commercial of the transition metal catalyst. For example, Xia et al.105

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

Table 2 Bimetallic catalysts and their respective monometallic catalysts for hydrogenolysis of glycerol

Selectivity (%)
Catalyst T (K) P (bar) Glycerol conversion (%) 1,2 PDO 1,3 PDO EG Other Ref.
Ru/C 473 14 43.7 40 51
Copper–chromite 473 14 54.8 85 51
Ru/TMG-BEN 468 100 90.7 62.3 32.7 5 56
Cu/TMG-BEN 468 100 26.5 83.1 11.6 5.3 56
Ru–Cu/TMG-BEN 468 100 70.9 71.7 23.6 4.7 56
Cu0.4/Mg5.6Al2O8.6 453 20 56.7 97.1 1.1 1.8 105
Pd0.04Cu0.4/Mg5.52Al2O8.52 453 20 76.9 97.2 1.6 1.2 105
Rh0.02Cu0.4/Mg5.6Al1.98O8.57 453 20 74.1 96.1 3.1 0.8 106
Cu/Al2O3 473 36 2 83 0 0 17 108
CuAg/Al2O3 473 36 27 96 0 0 4 108
Cu/SiO2 513 80 4.1 90 3.9 5.1 111
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

Ru–Cu/SiO2 513 80 39.2 85.9 6.9 1.4 111

Ru/ZrO2 453 100 100 48 26 26 112
Cu/ZrO2 453 100 28 100 0 0 112
3Cu–10Ru/ZrO2 453 100 90 90 8 2 112
Ru/C 433 80 29.7 50.9 0.8 14.4 33.9 116
Ru/C + Re2(CO)10 433 80 59.4 56.6 7.2 1.5 34.7 116
Downloaded by Cornell University on 09 August 2012

4 wt% Rh/C 393 34 10.5 17.7 5.3 38.7 118

4 wt% Rh–ReOx/C 393 34 9.8 37.2 24.0 77.0 118

demonstrated that the addition of Pd to Cu supported over to a final PDO yield up to 87% at 503 K and 100 bar H2 at a
layered double hydroxide increased the glycerol conversion 3 : 1 Ru : Cu ratio (Table 2). Further increasing the temperature
from 56% to 76.9% while keeping the selectivity to 1,2 PDO lead to the formation of alcohols, which decreased the selectivity
high at relatively low temperatures and H2 pressures, resulting to 1,2 PDO. Similar results were reported by Vasiliadou and
in yields of over 90% (Table 2). In the case of PdCu, the Lemonidou,111 where addition of Ru to Cu increased the
selectivity was higher than 96% for all the temperatures conversion, with small reductions in the 1,2 PDO selectivity,
studied with a significant increase in conversion (from 10% as the presence of Ru favors the formation of ethylene
to 95%) when temperature was increased from 393 K to glycol. The catalyst support and the preparation method
473 K. The high performance of the catalyst was attributed played important roles, as they modified the metallic surface
to H2 spillover from Pd that assisted the conversion of glycerol area exposed.
over the Cu. The authors found similar results using Rh Liu et al.112 also used CuRu over different supports and
instead of Pd, with a glycerol conversion of 74.1% and a found that the presence of Cu reduced the formation of
selectivity to 1,2 PDO of 96.1%. In both cases, the conversion ethylene glycol. They observed selectivities to ethylene glycol
was increased when the reaction was performed in presence of of 43% (453 K; 100 H2 bar) when using a SiO2 support, and
alcohols, achieving yields of 95.2% while maintaining the the selectivity to ethylene glycol was only 9% with ZrO2 at
selectivity to 1,2 PDO at 98.7%.106 The origin of this behavior similar conditions. When they used zeolite Y (H and Na) as
was suggested to be a weaker interaction of the alcohols with the support, a high selectivity to monoalcohols (62% for HY
the surface that would favor glycerol adsorption, whereas and 87.5% for NaY) was observed, which was attributed to
water is strongly adsorbed and deactivates the catalyst.107 the high activity of this support for dehydration reactions.
Zhou et al.108 used Ag instead of Pd to promote the activity Promotion of Ru can be achieved by using other noble metals
of Cu using g-Al2O3 as the support. The Ag reduced the such as Au and Pt, but even when interactions between Ru and
formation of Cu oxides, which are less active and improved the metals were demonstrated, no changes in the catalytic
dispersion. The optimal amount Cu : Ag was 7 : 3, achieving hydrogenolysis of glycerol were observed.113
100% selectivity to propanediols at low conversion (27%) As the activity of Ru was promoted by a non-precious
working at 473 K and 36 bar of H2. At similar conditions, metal, Cu, Pt was also enhanced by the addition of a non-
the promotion with Zn, Cr, Ni, Co was not observed likely due precious metals, in this case, Sn. Barbelli et al.114 showed that
the formation of the metal oxides, which do not keep the Cu in while no activity was found when using a Pt/SiO2 catalyst,
its metallic form. conversion of glycerol to liquid products up to 54% can be
Jiang et al.56 obtained interesting results adding Cu to Ru. achieved in the presence of N2 and 16% in the presence of H2,
Ru is an active catalyst for the hydrogenolysis of glycerol,53,109 with selectivities of 59% and 84% to 1,2 PDO, respectively
however, it is also active in C–C bond cleavage110 leading to when using a PtSn0.2 catalyst (473 K; 4 bar initial pressure). It
the formation of ethylene glycol instead of propylene glycol. was found that Sn was not only present in metallic form, but it
No changes in the selectivity to 1,2 PDO was found with the is also present in ionic form, thereby, acting as a Lewis acid
pressure, but a drop in the selectivity to liquid products was that promoted C–O bond cleavage to produce acetol that is
observed at lower pressures (25 bar for 483 K and 60 bar for hydrogenated to 1,2 PDO.
503 K). When temperature was increased conversion and Musolino et al.115 co-precipitated Pd with different metals
selectivity to liquid products increased (pressure was 100 bar). (Co, Fe, Zn, Ni) to increase the activity for glycerol conversion
Optimization of the Ru : Cu ratio and reaction conditions leads and selectivity to 1,2 PDO. PdFe, PdZn, and PdNi showed

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

good selectivities to 1,2 PDO, 71%, 59%, and 84% respectively at combined yield of 1-propanol and 2-propanol of 87% in
conversions over 90% (453 K and 5 bar H2). PdCo was the most aqueous solution at 393 K and 80 bar using ReRu/SiO2. As
active catalyst, but the selectivity to 1,2 PDO was low (10%). mentioned above, ReOx adds an acid functionality, which
Adding a precious metal, Re, increased the activity of the improves dehydration pathways and increases the yields to
Ru by a factor of 3 and increased the 1,2 PDO selectivity by a monoalcohols as reported previously.123,124 Co-precipitated PdCo
factor of 2, using different supports and Re2(CO)10 as precursor has promising potential as a catalyst to produce 1-propanol, as
at 433 K and 80 H2 bar (Table 2).116 The presence of Re reduced selectivities of 81% were found.115
the production of degradation products, such as ethylene glycol, When the target is the production of ethylene glycol, the
from 14% when using Ru/C to less than 2% when using Ru/C rate of C–C bond cleavage needs to be improved. Addition of
promoted with Re2(CO)10. The Re also increased the glycerol Pt, Ir and Rh to Ni/Al2O3 increased the selectivity of glycerol
conversion from 30% to 59% and selectivity to 1,2 PDO from hydrogenolysis to ethylene glycol, while other metals such as
50% to 56%. The main by-product using Ru promoted by Re Ru and Pd did not show significant effects.125 The most active
are propanols, indicating that reaction conditions can likely be promoter was Pt. Adding small amounts of Pt (0.1 : 5 Pt : Ni
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

optimized to produce monoalcohols. ratio) increased the conversion from 12 to 16% and the
Shinmi et al.117 studied the addition of oxophilic metals selectivity to ethylene glycol from 17% to 48%. At the best
such as Re, Mo and W to Rh/SiO2. This combination conditions, Ueda et al. reported yields of ethylene glycol
increased the selectivity of glycerol to 1,3 PDO, which is more around 32% (80% conversion with 40% selectivity to ethylene
valuable than 1,2 PDO. The best catalyst was RuRe at a glycol) with methane being the main by-product. Promotion of
Downloaded by Cornell University on 09 August 2012

0.5 atomic ratio. The addition of Re reduced the rate of C–C Ru with other noble metals such as Au and Pt to increase the
bond cleavage compared to the monometallic catalyst at 393 K production of 1,2 PG did not improve the catalytic hydrogenolysis
and 80 bar H2.117 In this case, Re added an acid functionality to of glycerol.113 However, Ag and Pt increased the selectivity slightly
the catalyst due to the presence of Re oxide,118,119 and the catalyst to ethylene glycol, making PtRu/C and AuRu/C potential
acted as a bifunctional catalyst. Similar to Cu, the addition of Re catalysts for the production of ethylene glycol from glycerol.
promotes the formation of monoalcohols, and therefore, diols
yields are low. Better results were reported using Ir promoted
6.2. Glycerol oxidation
by Re instead of Rh, at the same reaction conditions120 with
60% selectivity to 1,3 PDO at 50% glycerol conversion, with Another route to increase the value of glycerol is through
1-propanol being the main by-product. This was attributed to oxidation to glyceraldehyde, dihydroxyacetone, and acids
the OH adjacent to a primary OH being more active when such as glycolic and glyceric (Fig. 7). In these oxidation
using the IrRe catalyst. reactions, bimetallic catalysts have proven to be important
Daniel et al.,121 promoted Pt/C with Re and observed that by increasing the catalytic activity, selectivity, and improving
glycerol conversion increased up to 5 times upon addition of catalyst stability. There are two main pathways for the oxidation
Re. Additionally the selectivity of the Pt/C catalyst was of glycerol, as can be seen in Fig. 7. One pathway is to oxidize the
modified (74% to 1,2 PDO, 23% ethylene glycol) to produce secondary alcohol leading to the production of dihydroxyacetone
up to 36% 1,3 PDO and 33% monoalcohols at 443 K and and hydroxypiruvic acid, and another route is to oxidize the
40 bar H2. The addition of Re was suggested to create Re–OH primary alcohols leading to the formation of glyceraldehyde and
species at the surface, which activate the C–OH bond and glyceric acid. AuPd, which produces glyceric acid126–128 in basic
facilitate the hydrogenolysis of glycerol. Better results with medium, and PtBi, which produced dihydroxyacetone129–131 in
sintered PtRe/C catalysts were explained due to better atomic acidic medium, are the most common bimetallic catalysts used
mixing of the Pt and Re after treatment at high temperatures for glycerol oxidation. Pd was highly selective for the production
(723 K). of glyceric acid, but has low activity, while Au was more
Similar results were observed by Chia et al.,118 using a RhRe active, but less selective, producing glycolic acid in addition to
catalyst. Addition of Re to Rh increased the activity and glyceric acid.
modified the selectivity of Rh/C, which increased the production Bianchi et al.132 reported yields up to 77% for glyceric acid
of diols from 18% 1,2 PDO and 5.3% 1,3 PDO to 37.2% 1,2 using AuPd, and the main by-product was tartronic acid
PDO and 24% 1,3 PDO. Also, the formation of monoalcohols coming from further oxidation of glyceric acid; the selectivity
decreased from 77% to 40% at 393 K and 34 bar H2 (Table 2). for which increased with conversion. The formation of alloys
The change in selectivity upon addition of Re was again suggested increased the catalytic activity, but there were no differences in
to be the formation of acid sites that make the catalyst bifunctional the selectivities compared with the monometallic catalyst.
with metal and acid functionality. The highest selectivities were found with monometallic Pd
In general, when the goal is to obtain monoalcohols such as (88% with a TOF 541 h1 versus 77% with a TOF 1077 h1
1-propanol and 2 propanol, monometallic catalysts are not for AuPd/C).
effective, having either low activity (RANEYs Ni, Co, Cu) or Promoting Au with Pt was not effective. Using AuPt/C
low selectivity (Rh/SiO2, Rh/C, Ru/C), with degradation catalysts resulted in higher activity compared with the mono-
products such as propane.122 The addition of Re, Mo, or W metallic catalyst, but lower selectivities were achieved, produ-
increased the activity for production of 1,2 PDO and decreased the cing a mixture of glyceric acid and glycolic acid. The higher
production of over-hydrogenated products, such as propane and activity observed in the bimetallic catalyst was associated with
degradation products (ethanol, ethane, methane). At optimum slower deactivation as compared to the monometallic catalyst.
conditions, 1-propanol yields of 68% were achieved with a total Pd promotes tartronic acid and Pt, glycolic acid. According to

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

Dimitratos et al.,133 the particle size in the bimetallic catalyst is the degree of functionalization by oxidation to produce aldonic,
important, as conversion increases for smaller particles, while aldaric, and keto aldonic/aldaric acids,46 which are used as
selectivity decreases. Using a non-precious metal promoter, specialty chemicals and food additives.45,140 Normally, the oxygen
Cu, showed improved conversion and selectivity compared required for the oxidation comes from molecular oxygen or air,
with the monometallic Pt/C catalyst since the addition of Cu which reduces the cost, but chemicals such as hydrogen peroxide
reduced C–C bond cleavage over Pt, which reduced the can be used.
formation of by-products.134 Glucose has 3 different types of carbons: aldehyde, primary
Kimura et al.130,131 studied the oxidation of glycerol to alcohol, and secondary alcohol. All three positions can be
dihydroxacetone, using noble metals such as Pt, Pd, Ru and oxidized leading to different products, and the reactivity of
Re. In acidic media, only Pd was demonstrated to be active. these positions was studied by Heyns and Paulsen.141
Addition of promoters, like Bi, to Pd slightly decreased The oxidation of the aldehyde will produce a carboxylic acid
glycerol conversion, but increased dihydroxacetone selectivity (e.g., gluconic acid) while the oxidation of the primary and
from 10% to 80% (at 25% conversion). While Pt carried out secondary alcohol can produce glucuronic acid, keto-glucose,
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

the oxidation, Bi promoted oxidation of the secondary alcohol and keto-acids, as presented in Fig. 9.
by blocking Pt (111) sites. Worz et al.129 identified the The most studied catalysts are based on Pt and Pd, which
formation of glyceric acid as being responsible for catalyst are active and selective for glucose oxidation to produce
deactivation, and the presence of Bi may reduce the formation gluconic acid. Addition of Bi increased the activity and
of this compound, leading to higher activity of the PtBi selectivity to gluconic acid, which is produced by oxidizing
Downloaded by Cornell University on 09 August 2012

catalyst versus the monometallic. Using nanoparticles of PtBi, the C1 (anomeric) carbon in glucose. Besson et al.142 observed
Liang et al. reported selectivities to dihydroxyacetone up to that the addition of Bi increased the catalytic activity, such
95%; however, the selectivity decreased with conversion due to that the reaction became limited by oxygen diffusion instead of
further oxidation to oxalic acid, which is another possible catalyst reactivity with yields over 99% to gluconic acid from
product from glycerol.134 glucose. This increase in activity is related to an improvement
Lactic acid can be produced by combining oxidation with in catalyst stability compared to the monometallic catalyst that
dehydration. Using a AuPt catalyst to carry out the oxidation deactivates at high conversions. The origin of deactivation was
and NaOH for the dehydration, Shen et al.135 reported yields that the Pd surface is covered by oxygen and acid production,
of lactic acid up to 86%, which are similar to yields observed making the active center inaccessible. The presence of Bi prevents
with monometallic Pt, but in this case at lower TOF. The the over-oxidation of Pd since oxygen is adsorbed on the Bi,
higher activity of the bimetallic catalyst was associated with which is oxidized to Bi2O3. Additionally, the high reaction rate
the formation of alloys and electron transfer from Au to Pt. for glucose oxidation using PdBi catalysts limits the formation of
These interactions improve not only the activity, but the by-products, such as the oxidation of the alcohol groups. The
stability of the catalyst. The alloys also prevent the sintering PdBi catalyst can also be used for the oxidation of dissacharides,
of the Au particles and the poisoning of the Pt sites by O2. for example, lactose.143
The promotion of Pd with other metals such as Te, Sn and
7. Cellulose conversion by bimetallic catalysts Co was not as satisfactory as PdBi.144 Addition of Te
increased the activity of 5% Pd/SiO2 from 60% to 100% when
Cellulose can be converted by hydrolysis to its monomer, 1% Te was added, while conversion of glucose was only 63%
glucose (C6H12O6), a six-carbon sugar containing an aldehyde and 82% when 1% and 8% Bi were added, respectively. Also,
moiety (an aldohexose).136 Biological conversion of glucose by the selectivity to gluconic acid was 100% using 5%Pd–1%Te/
fermentation has been extensively studied for the production of SiO2 versus 73% when using monometallic Pd and 92%
many products, one of the most important being bioethanol.137,138 when using 5%Pd–8%Bi/SiO2. This increase in activity and
Here, we focus on the catalytic conversion of glucose, and more selectivity was related to the formation of intermetallic PdTe
particularly, on those reactions that involve the use of bimetallic
catalysts. The hydrolysis of cellulose can be accomplished
enzymatically or chemically, mainly by acid catalysis. Higher
selectivities and yields have been achieved by enzymatic
methods, which are preferred when glucose is to be upgraded
by metal catalysis (oxidation, hydrogenation). However, when
glucose is to be used to produce platform molecules such as
levulinic acid or 5-hydroxymethylfurfural (HMF) by dehydration
reactions, acid hydrolysis is preferred, as it is less expensive, faster,
and cellulose hydrolysis and glucose dehydration can be carried
out at similar conditions in the same reactor. Many methods for
cellulose hydrolysis can be found in the literature.12,13,139

7.1. Glucose oxidation

Glucose upgrading can be achieved by decreasing its degree of
functionalization by APR to produce H2, alkanes, or mono-
functional intermediates, or it can be achieved by increasing Fig. 9 Possible oxidation products obtained from glucose.

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

and PdTe2 phases after activation of the catalyst by a

reduction treatment. Higher additions of Te resulted in lower
conversion and selectivities.145
Pt is one order of magnitude more active than Pd in the
oxidation of glucose; however, Pt is able to oxidize the alcohol
groups, which led to the formation of glucuronic acid and
glucaric acid, and oxidize the aldehyde group, which forms
gluconic acid. Pt is not selective146 to any of these species, as
protection of the anomeric carbon would be necessary to
produce selectively glucuronic acid. The higher activity is
related to a lower coverage of the surface by oxygen, which
facilitates the interaction of the surface with the glucose.
Addition of Bi to Pt increased the catalytic activity up to
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

4 times and modified the selectivity for the production of

2-ketogluconic acid using gluconic acid as feedstock (60%
selectivity at 68% gluconic acid conversion).146 Similar
changes in selectivity were observed using a PbPt catalyst.147 Fig. 10 Hydrogenation products derived from glucose and cellulose.
Abbadi et al. reported selectivities up to 98% when the pH was
Downloaded by Cornell University on 09 August 2012

lower than 6.147,148 The presence of Bi or Pb modified how the adsorb glucose and favor the nucleophilic attack on the
gluconic acid adsorbed on the catalyst surface by forming carbon atom by H2 dissociated on the Ni. The presence of
a chelate with the carboxyl and a-hydroxyl group. This both species, Ni and Cr, is necessary in this mechanism, and
promoted the dehydrogenation of the a-hydroxyl group, while explains the decrease in activity observed when larger additions
Pt adsorbs the hydrogen atom. Addition of Au to Pt increased of Cr to Ni are used, as it reduced the amount of Ni exposed.
the yields even at low levels of Au (Au : Pt ratio 0.03).146 Addition of Fe increased the activity up to 8.5 times, and in the
According to the authors, Au prevented coverage of the surface case of Cr, Fe promotes activation of the CQO bond, but
by degradation products, increasing the catalytic activity of Pt. Fe leaches into the medium leading to catalyst deactivation.
Several other authors have used Au/C to carry out Addition of Sn and Mo increased the activity up to 5 and
the oxidation reaction, showing that Au can have a direct 4 times, respectively. Similar results were reported by Li et al.
participation in the reaction.149–158 Comotti et al.159 studied using NiB/SiO2 promoted by Cr, Mo and W.163 Hoffer et al.164
the promotion of Au with Pt, Pd, and Rh in acidic conditions. used Mo and Cr/Fe as promoters of RANEYs Ni and
They observed that AuPt was the most active bimetallic (best observed that glucose hydrogenation increased when adding
Au : Pt ratio was 2), followed by AuPd, and both were more Mo or Cr/Fe while keeping selectivity to sorbitol over 98%;
active than the respective monometallic catalyst with 100% however, the most active catalyst, Cr/Fe, promoted catalyst
selectivity to gluconic acid; AuRh showed a lower conversion deactivation by leaching, but no leaching was detected for
than monometallic Au. High activity of AuPt was observed NiMo catalyst. Another possible option is replace the Ni by
only at low pH and no differences between the AuPt and Au Co adding B/Cr or W.165,166
catalyst were observed at basic pH. Recently Zhang et al. Ru-based catalysts are more expensive than Ni, but have
reported the promotion of Au by adding Ag using PVP (poly- demonstrated to be more stable.167 Guo et al. prepared a RuB
vinylpyrrolidone) as protective agent, resulting in a catalyst catalyst by reducing the Ru with KBH4 and then promoting it
3 times more active than the monometallic Au.160,161 with Cr.168,169 They observed that while Ru was completely
reduced, only part of the B formed an alloy, with the remainder in
form of salt or oxide. The amorphous RuB catalyst was 9 times
7.2. Glucose hydrogenation
more active than monometallic Ru and stable for several
Cellulose, and its monomer, glucose, can be converted into cycles of reaction. The effect of B is not clear, but the higher
several polyalcohols by hydrogenation (Fig. 10). For example, dispersion observed in the RuB catalyst, formation of alloys,
glucose can be converted to sorbitol by hydrogenation of the and amorphous structure were identified as possible reasons.
carbonyl group. Sorbitol is an extensively studied polyol, and Adding Cr to the catalyst resulted in higher activity, but as
is used in the production of vitamin C and as a low-caloric observed before, the activity decreased upon addition of larger
sweetener.18,45,46,140 Many systems have been studied for amounts of Cr.
glucose hydrogenation using both batch and continuous A mixture of hexitols can be produced directly from cellulose by
reactors45,140 with the main catalyst utilized in the process hydrogenolysis of glucose (hydrogenation plus dehydration), using
based on Ni. Promotion by other metals increased the activity Ni-promoted catalyst supported over mesoporous carbon.170
and stability, with catalyst leaching being one of the main Using monometallic Ni, 42% hexitols yield can be achieved at
issues in glucose hydrogenation. RANEYs Ni is active and 84.5% conversion of cellulose. Noble metals such as Ru, Pt and
selective (approximately 98% to sorbitol); however, it deactivates Pd are more active than Ni,171,172 and have been used to promote
and leaches. Gallezot et al.162 studied the effect of adding the activity of monometallic Ni. Adding Pt to Ni increased the
promoters such as Sn, Mo, Cr and Fe to RANEYs Ni and yield to 47.4% at 100% conversion of the cellulose. The effect of
found that the addition of Cr increased the rate of glucose Pd and Ru was even more significant, achieving yields of 53.2%
hydrogenation up to 5 times. Cr(III) acted as a Lewis acid to and 54.2%, respectively at 100% conversion. The addition of

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

Rh and Ir showed greater increases, with 59.8% and 57.5%

yield respectively at 100% conversion. An improvement of
catalyst stability was found upon addition of promoters, such
as Ir. Monometallic Ni catalyst deactivates rapidly, and 20%
yield to hexitols was achieved after 5 cycles, while 4%Ir/
4%Ni/MC retains its activity 4 cycles with yields around
60%.170 The authors found that the support played an important
role to achieve these promotional effects in activity and stability,
and good results were found with the use of mesoporous carbon.
Ji et al.173 promoted W2C supported on activated carbon
with Ni to convert cellulose into ethylene glycol. The addition
of small amounts of Ni (2 wt%) increased the selectivity to
ethylene glycol from 25.4% to 61% at 518 K and 60 bar H2.174
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

The Ni promoted the formation of tungsten carbides. For the

same reaction, Wang et al.175 used Ni to promote the hydro-
genolysis of cellulose using Cu/ZnO. The addition of Ni
increased the hydrogenation activity, and the selectivity to
ethylene glycol increased from 8.4% to 22.8% with a total
Downloaded by Cornell University on 09 August 2012

increment in 1,2 alkanediol from 46.4% to 72.5% at 518 K

and 40 bar H2 using a 2 : 3 Ni : Cu ratio compared to pure
Cu/ZnO. Larger amounts of Ni resulted in lower selectivities
due to further hydrogenation of the alcohols to produce
alkanes and degradation products. The catalyst was used to
process biomass, cornstalk and corncob, but lower selectivities
to ethylene glycol (15.9% and 18.4%, respectively) and to 1,2 Fig. 11 Production of building blocks from glucose dehydration.
alkanediols (48.5% and 55.5%, respectively) were obtained
due to the presence of hemicellulose and other impurities. functionality. Further hydrogenation would produce 2,5
Glucose and other C6 sugars can be converted by dehydration dihydroxymethyltetrahydrofuran; 2,5 bishydroxymethyltetra-
into attractive building blocks such as 5-hydroxymethylfur- hydrofuran).193 As in the oxidation reaction, reduction of
fural (HMF),176–179 levulinic acid,22,180,181 or gamma- HMF is typically done with monometallic catalysts, such as
valerolactone182,183 using acid catalyst (Fig. 11). Acid hydrolysis Ru.194 Dumesic et al. proposed the hydrogenolysis of C–O
does not involve the use of bimetallic catalysts, and therefore bonds to produce 2,5 dimethylfuran (DMF) using CuRu/C.
is not considered in this review. HMF is a highly reactive Addition of Cu modifies the selectivity to produce DMF since
platform molecule that can be upgraded into various Cu is active in the hydrogenolysis reaction; a 3 : 1 CuRu/C
chemicals by rehydration, aldol condensation, oxidation, catalyst produced DMF at 71% yield.195
hydrogenation, hydrolysis, etherification, and many other Levulinic acid, another important building block, can be
reactions.184 The main products obtained by these reactions produced from HMF in acid media or directly from C6 sugars
can be observed in Fig. 12. Here we will focus in the oxidation (Fig. 11). Reduction of levulinic acid to gamma valerolactone
and hydrogenation, as these are the reactions where bimetallic (GVL) has attracted significant interest recently, as GVL has
catalyst can be used. demonstrated to be a good solvent, can be used directly as fuel,
HMF can be converted into 5-hydroxymethylfuranoic acid or be converted into liquid fuels.183,196 Reduction of levulinic
(HFCA) by oxidation of the formyl group, 2,5-diformylfuran acid is normally done using noble metal catalysts, such as Pt,
(DFF) by oxidation of the hydroxy group, or 2,5-furandicar- Pd, Au and Ru,197–201 which present high selectivity to GVL,
boxylic acid (FDCA) by oxidation of both groups. Catalysts but deactivate with time-on-stream. This deactivation is more severe
used for the oxidation reaction are mainly noble metals, such when levulinic acid was produced by cellulose deconstruction using
as Au, Ru, Pd and Pt, usually promoted with other metals, mineral acids. Braden et al.202 showed a 2 order of magnitude
such as Pb and metal oxides.45,184,185 Verdeguer et al. studied decrease in activity when using Ru/C in presence of sulfuric acid,
the use of Pt to oxidize HMF to FDCA, which is a possible which is commonly used in the production of levulinic acid.
replacement for terephthalic acid in the production of polymers. Addition of Re improved the catalyst stability and the activity in
They observed that the catalytic activity was improved considerably presence of sulfuric acid by the formation of ReRu alloys; however,
by adding Pb to Pt, achieving yields close to 100%.186 Similar in absence of sulfuric acid the bimetallic catalyst was less active than
systems were used by Kokoh and Belgsir187 to produce 2,5 the monometallic Ru.
diformylfuran, but using electrocatalytic systems. Kroger et al. Another option would be the use of solvents to remove
used PtBi to produce FDCA directly from fructose, but yields the sulfuric acid. Wettstein et al. used sec-butylphenol to
were only 25%.188 Other oxidation reactions have been carried selectively extract the levulinic acid thus, separating it from
out using monometallic catalyst as Au, Pd and Pt.189–192 the sulfuric acid. While monometallic Ru/C deactivated
Another option to upgrade HMF is the reduction of rapidly in presence of the alkylphenol solvent, the addition
its functional groups to form 2,5 dihydroxymethylfuran of Sn led to a catalyst that was stable for more than 300 h on
(2,5 bishydroxymethylfuran; BHMF) by reducing the aldehyde stream.59,203

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A
Downloaded by Cornell University on 09 August 2012

Fig. 12 Oxidation and hydrogenation HMF derivatives products.

8. Hemicellulose conversion by bimetallics the Quaker oats process.207 From furfural, many chemicals,
catalysts including biofuels, can be produced (Fig. 13). Lange et al.
recently wrote a thorough review on the conversion of furfural
Historically, hemicellulose has been less studied than cellulose. to biofuels.17
Although its composition depends on the source of the Furfural can be oxidized to produce furonic acid using
lignocellulosic biomass, hemicellulose is typically a polymer noble metal catalysts and basic media, which is necessary
of C5 sugars, mainly xylose. A major difference compared to due to catalyst deactivation by acids formed along the process.
cellulose (C6 sugars), is that xylose is not processed well by A monometallic Pt catalyst produced furonic acid with
biological treatments. This, together with the fact that hemi- selectivities around 50% at 40% conversion (308 K). Addition
cellulose is more reactive than cellulose, makes it attractive for of Pb and Bi as promoters increased the selectivity to 92% and
heterogeneous catalytic processing. Many catalytic strategies 72%, respectively. However, while addition of Pb increased
have been developed using xylose as a feedstock, and many of the conversion of furfural to 88%, addition of Bi reduced the
the reactions to upgrade hemicellulose are similar to those activity to 29%.208 Furfural hydrogenation to furfuryl alcohol
used for cellulose conversion; however, the final products are is an important reaction due to the wide number of uses for
substantially different. Of the many reactions to upgrade furfuryl alcohol. Reduction of furfural can also lead to
hemicellulose reported in the literature, we focus here on those hydrogenation of the CQC bonds (tetrahydrofurfural
using bimetallic catalysts, oxidation and hydrogenation. (THFF)) in addition to hydrogenation of the CQO bond
(furfuryl alcohol).209 The CQC bonds of furfural are less
8.1. Xylose to xylitol
reactive than the CQO bond, probably due to steric effects,210
Xylitol, a common sweetener used in the food industry, can be making the production of THFF difficult.
produced by hydrogenation of xylose. Noble metals, such as Pt, Furfural hydrogenation is typically limited to Cu-based
Pd, Ru, and non-precious metals, such as Cu and Ni, have been catalysts with Cu/chromite being the most widely used
used for xylose hydrogenation, with Ru and RANEYs Ni commercially.211,212 Several other supported copper catalysts
often being used to produce xylitol. Although high selectivity to have been successfully applied in vapor phase hydrogenation
xylitol is achieved, typically xylose hydrogenation catalysts of furfural, for example Cu/C213 and Cu/MgO,214 as well as
deactivate due to structural degradation, organic species depos- bimetallics, CuLa/MCM-41,215 NiCu/SiO2, CoNi/SiO2,
iting in the pores, and leaching.204 Using a Ru/C catalyst CuCo/SiO2216 and CuCa/SiO2.217 Vapor phase hydrogenation
(373 K, 5.5 MPa H2), over 80% of the xylose was converted at 473 K resulted in almost 98% yield over Cu/MgO catalysts,214
to xylitol after 100 min.205 Using RANEYs Cu, 4.5 mmol of as well as the bimetallic CuCo/SiO2 catalyst.216 The addition of
xylitol can be produced per hour and gram of catalyst at 373 K Ca did not have a significant effect on the reaction,217 but the
and 40 bar of hydrogen. The catalytic activity was increased 3.7 addition of La resulted in a higher conversion at a given time
times by adding Ru to the catalyst.206 when compared to a Cu/MCM-41 catalyst. The addition of La
reduced the required reduction temperature of the CuO, which
8.2. Fufural to furfuryl alcohol
may be the reason for the increased activity. Additionally, support
Dehydration of xylose using acid catalysts results in the effects were seen with the CuLa catalyst. Using mordenite as a
formation of furfural, which is produced commercially using support resulted in 24–76% selectivity to 2-methylfuran, while a

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

PtSn catalyst for furfural hydrogenation and achieved 80%

selectivity to furfuryl alcohol.222 Merlo et al.209 used a PtSn
catalyst in a liquid-phase system that was more active than the
monometallic Pt catalyst. At a given reaction time, PtSn led
to higher conversions while achieving high furfuryl alcohol
selectivities (greater than 96%), which was also seen by Vetere
et al.219 As with the xylose hydrogenation catalysts, the PtSn
catalyst deactivated, losing 25% of its activity upon 3 reuses,
but the selectivity to furfuryl alcohol remained unchanged.
The authors attributed the increase in reaction rate of the
bimetallic catalyst to the modification of the electronic nature
of the active site with the addition of Sn.209 Sitthisa et al.
hypothesized similar reasoning for the increase in hydrogena-
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

tion yield of PdCu/SiO2 compared with monometallic Pd and

Cu on SiO2. On Pd/SiO2, the reaction pathway favored
decarbonylation resulting in production of furan. When Cu
was added to form PdCu/SiO2, the percent conversion was cut
in half, but hydrogenation was the preferred reaction pathway,
Downloaded by Cornell University on 09 August 2012

resulting in furfuryl alcohol production and the hydrogenation

yield almost doubled.223
Chia et al.,118 studied the hydrogenolysis of tetrahydrofurfuryl
alcohol to produce 1,5 pentanediol. Using a Rh/C catalyst,
Fig. 13 Hydrogenation products of furfural and its derivatives.
selectivities to 1,5 pentanediol of 59% at 8.5% conversion
(393 K 34 bar) were observed. The selectivity to 1,5 pentanediol
sodium beta zeolite catalyst resulted in 80% 2-methylfuran at increased to 91%, while the conversion increased up to 6 times at
low conversions (9.5%), but greater than 99% furfuryl alcohol similar conditions when Mo was added as a promoter; and,
selectivity at high conversions (98%). The high selectivity of selectivities of 97% at 47% conversion were achieved by using Re
the MCM-41 catalyst, greater than 99%, to the desired as promoter. The increase in selectivity was suggested to be due
product, furfuryl alcohol, was attributed to the mesoporous to the presence of acid sites on the bimetallic catalyst from the
structure, which deters pore blocking and possibly forms presence of ReOx species. The acid was responsible for the ring
different active species.215 opening at the most substituted carbon. DFT calculations
Recently, Ni and NiFe/SiO2 catalysts have been used to showed that the secondary carbenium ion, stabilized by an
convert furfural. The addition of Fe to the Ni/SiO2 catalyst at adjacent –OH group, was more stable than the primary
50% conversion resulted in a significant selectivity change. carbenium ion, leading to the formation of 1,5 pentanediol
The major products from the Ni/SiO2 catalyst were furan and instead of 1,2 pentanediol.118
C4 products (butanal, butanol, and butane), while the addition
of 2 wt% Fe produced C4 products and 2-methylfuran at
9. Lignin upgrading
0.025 h W/F and 523 K. However, the conversion was only
28% for the bimetallic catalyst compared to 51% for the Lignin is the most complex fraction present in lignocellulosic
Ni/SiO2 catalyst. The C4 products result from the ring-opening biomass and is harder to process compared with the other
of the furan ring via C–O hydrogenolysis, which is common fractions. This complexity has led lignin to be treated as a
for Ni catalysts. The 2-methylfurfan is also a secondary waste stream, and typically burned to produce energy. However,
product, derived from C–O hydrogenolysis of furfuryl alcohol. lignin is a significant portion of the total carbon in biomass, and
Sitthisa et al. determined that the reaction pathways, decarbony- better use of this fraction is a requirement to improve the economic
lation versus hydrogenation, are not strongly dependent on balance of any biorefinery. For this reason, research on upgrading
conversion, but strongly depend on the active sites of the catalyst lignin has become of recent interest, as many interesting products,
and temperature.218 The addition of Sn to Ni resulted in a mainly aromatics, can potentially be produced from lignin. As
selectivity of 97% for furfuryl alcohol at 8 h reaction time and a feedstock rich in aromatic functionality,224 lignin can be
a Sn : Ni atomic ratio of 0.8. At an atomic ratio of 0.2, the leveraged to produce chemicals such as benzene,225 cyclohexane,226
furfuryl alcohol selectivity dropped to 78% with 2-isopropoxy- and phenols225–227 upon cracking, hydrogenation, and
methylfuran as the remaining compound stemming from the hydrodeoxygenation (HDO).
isopropanol solvent, as well as some light compounds from Due to the complex structure of lignin, most studies use
secondary reactions.219 model compounds, such as phenol, cresol, guaiacol, and anisole
Monometallic Pt catalyst typically form the fully hydro- (Fig. 14) instead of real lignin. These aromatic compounds
genated species in the vapor phase, although Pt/C is selective for contain functional groups found in lignin, and are used to
furfuryl alcohol production in the liquid phase.220 Bimetallic determine the products and selectivity of the more complex
catalysts formed by coupling Pt with electropositive promoters, species that would be expected if lignocellulosic lignin were
such as Sn, Fe, and Ga, demonstrated improved selectivity used. Typically, sulfided CoMo and NiMo bimetallic catalyst
toward furfuryl alcohol.221 For example, Marinelli et al. used a systems are used for lignin hydrogenation and HDO,225,226,228

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

below 673 K, all Rh-based catalysts were significantly more

active than sulfided CoMo and NiMo catalysts. The Rh-based
catalysts resulted in the initial hydrogenation of the aromatic
ring (Fig. 16), followed by two pathways including demethoxy-
lation and dehydroxylation.226 The main products are listed in
Table 3, entries 4 and 5, and at 673 K, the major products were
2-methoxycyclohexanol, 2-methoxycyclohexanone, and cyclo-
hexane. The sulfided CoMo and NiMo catalysts had different
selectivities, which, similar to the PtSn/CNF/Inconel catalyst,
followed a reaction mechanism similar to Fig. 15a. The guaiacol
is first demethylated, demethoxylated, and deoxygenated, and
then the benzene ring is hydrogenated to form cyclohexane. The
Mo-based catalysts produced significant amounts of coke,
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

which was attributed to catechol formation even though no

catechol was observed. At higher temperatures and longer
reaction times, the coke production decreased and the major
Fig. 14 Compounds found in lignin conversion reactions and used as
model compounds. Guaiacol (1), phenol (2), anisole (3), catechol (4),
products listed in Table 3 were formed (entries 6 and 7), mainly
4-methylguaiacol (5), p-cresol (6), 4-methylanisole (7), and o-cresol (8). mono-oxygen compounds and hydrocarbons. Additionally,
Downloaded by Cornell University on 09 August 2012

sulfur stripping occurred due to the lack of sulfur in the feed,

and increased with increasing severity of treatment.226
although other metallic catalysts have been used, such as Other research using guaiacol and a sulfided CoMo catalyst
catalysts containing noble metals, e.g., PtSn,227 and PtRh.226 showed a significant improvement in HDO rate compared to a
Details about lignin conversion to chemicals and fuels can be sulfided Mo catalyst.231,232 The addition of cobalt enhanced
found in recent reviews by Zakzeski et al.19 and Kobayashi the demethoxylation and deoxygenation pathways leading to a
et al.229 higher fraction of HDO compounds.231 The HDO reaction
At high temperatures (673 K), Pt/Inconel catalysts were rate was an order of magnitude higher over ZrO2, suggesting a
used for HDO of guaiacol, but the conversion was low (less similar reaction pathway as shown in Fig. 15a, excluding the
than 30%) and the catalyst had poor stability with time- catechol intermediate, and the major oxygen-free product was
on-stream. A bimetallic PtSn catalyst led to the formation of benzene (Table 3, entry 8), and a large amount of phenol also
phenol with over 80% selectivity (Table 3, entry 1). The appeared. Different reaction pathways and product distributions
bimetallic catalyst deactivated at a much lower rate than the were found when the support was g-Al2O3 or TiO2. On the
monometallic catalyst, and the guaiacol conversion was greater g-Al2O3 support, the major products were methylated products,
than 80% after 3 h time-on-stream. Phenol remained the major and on TiO2, phenol and cyclohexane were the major products
product with benzene and o-cresol as minor products. Using a (Table 3, entries 9 and 10).232 Methylated compounds can
carbon nanofiber (CNF)/Inconel support, the guaiacol conversion include o-cresol, p-cresol, and di- and trimethylated phenols.225
remained high over PtSn, but benzene became a larger portion of The authors hypothesized that the ZrO2 support suppressed
the product distribution (Table 3 entry 2), which was attributed to coke formation due to the close interaction of the CoMo and
the higher surface area of the catalyst support leading to higher support.232
catalyst activity. Fig. 15a shows the simplified reaction mechanism In other recent work for biomass-derived feedstock conversion,
proposed. The same reaction using anisole as the feed (Fig. 15b) Jongerius et al. found that HDO, demethylation, and hydrogena-
resulted in lower conversion than with guaiacol, and benzene was tion reactions take place simultaneously at 573 K and 5 MPa over
formed as the main product along with phenol, o-cresol, and a sulfided CoMo/Al2O3 catalyst (Table 3, entries 11–18). In the
toluene as minor products (Table 3, entry 3).227 case of o-cresol, p-cresol, and phenol, non-oxygen containing
Monometallic catalysts using noble metals result in a high extent species (toluene and benzene) were the major products. For
of hydrogenation of the lignin aromatic ring (e.g., cyclohexanol, anisole, catechol, guaiacol, 4-methylanisole, and 4-methylguaiacol,
2-methoxycyclohexanol, cyclohexane, etc.);227,230 however, the major products were phenol and p-cresol. Additionally,
the support can have a significant effect on the product although hydrogenation products were formed, no oxygenated
distribution.227,230–233 Lee et al. reported that for a given cyclohexanes formed, which indicates that demethylation and
support, Pt, Rh, Pd, and Ru monometallic catalysts had HDO take place prior to hydrogenation, and therefore, the
similar product distributions, which were all hydrogenated reaction can be stopped prior to hydrogenation leading to
compounds due to the low reaction temperatures. At higher more aromatic compounds. They determined that oxygen-
reaction temperatures (higher than 523 K), HDO is preferred. demethylation was the fastest reaction pathway and HDO
For the monometallic catalysts, the support had a more significant occurs at a slower rate for the mono-aromatics evaluated.225
effect than the noble metal on product distribution.230 Although CoMo catalysts are preferred over NiMo catalysts
Lin et al.226 used Rh-based catalysts for guaiacol reactions due to their lower hydrogenation activity, many studies have
and determined that the addition Pt or Pd did not improve the been published for lignin conversion using NiMo.19 Using
activity or selectivity of Rh/ZrO2 catalysts. For temperatures an unsupported sulfided NiMo catalyst with atomic ratios
ranging from 573 to 673 K, the PdRh catalyst had significantly (Ni/(Mo + Ni)) varying from 0.1 to 0.6, Yoosuk et al.234
lower conversion than the PtRh and Rh catalysts; however, reported that the most common HDO products from phenol

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

Table 3 Recent bimetallic catalysts used for lignin conversion

Reaction conditions
Lignin (model)
Entry Catalyst Support T (K) P (MPa) t (h) compound Top 3 major products+ Conv. (%) Ref.
1 PtSn Inconel 673 0.1 W/F = 3.2 gcat/gr h Guaiacol Phenol >80 227
2 PtSn CNF/Inconel 673 0.1 W/F = 3.2 gcat/gr h Guaiacol Phenol >90 227
Benzene
3 PtSn CNF/Inconel 673 0.1 W/F = 2.5 gcat/gr h d Anisole Benzene B80 227
Phenol
4 PtRh ZrO2 573–673 5 0.3–1 Guaiacol 2-Methoxycyclohexanol B100 226
2-Methoxycyclohexanone
Cyclohexane
5 PdRh ZrO2 673 0.3–1 Guaiacol 2-Methoxycyclohexanol B100 226
2-Methoxycyclohexanone
Cyclohexane
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

6 CoMo 673 0.3–1 Guaiacol Phenol o15 226

Methoxybenzene
Cyclohexene
7 NiMo 673 0.3–1 Guaiacol Cyclohexane o15 226
Phenol
Methoxybenzene
a
B80
Downloaded by Cornell University on 09 August 2012

8 CoMo ZrO2 573 4 Guaiacol Phenol 231

Light compounds
Benzene
9 g-Al2O3 Guaiacol Phenol B80 232
Methyl phenols
Catechol
10 TiO2 Guaiacol Phenol B80 232
Light compounds
HDO products
11 CoMo Al2O3 573 5 4 o-Cresol Toluene 23 225
12 p-Cresol Toluene 22 225
13 CoMo Al2O3 573 5 4 Phenol Benzene 27 225
14 Anisole Phenol 58 225
Methylated products
15 Catechol Phenol 80 225
16 Guaiacol Phenol 84 225
Catechol
Methylated products
17 4-Methylanisole p-Cresol 85 225
Toluene
Methylated products
18 4-Methylguaiacol p-Cresol 89 225
Methylated products
4-Methylbenzene-1,2-diol
19 NiMo 623 2.8 1 Phenol Cyclohexane 96 234
Benzene
Cyclohexene
20 NiCu Al2O3 593 17 1 Guaiacol Cyclohexane 80 236
1-Methylcyclohexane-1,2-diol
Cyclohexanone
21 CeO2–ZrO2 Guaiacol Cyclohexanone 94 236
1-Methylcyclohexane-1,2-diol
22 SiO2 Guaiacol Cyclohexane 87 236
23 SiO2–ZrO2–La2O3 Guaiacol Cyclohexane 85 236
Benzene
a
Between 0.3 and 8 s contact t.

are cylcohexane, benzene, and cyclohexene, which are similar for the hydrogenation pathway is much higher when Ni is
products to those previously reported.226 The NiMo catalyst added for phenol HDO.234
with a ratio of 0.3 (Table 3, entry 19) removed over 95 mol% Ni is highly active for hydrogenation reactions, but deacti-
of the oxygen in the feed, which is an improvement over the vates due to sintering and coke formation. Recently, it was
non-Ni containing catalyst, Mo, that removed 70 mol%. The found that Cu can be added as a modifier to Ni catalysts to
authors attributed the increase in HDO activity to more active facilitate reduction of nickel oxide at lower temperatures.235
catalyst particles to the addition of Ni; however, the selectivity Bykova et al.236 evaluated several NiCu catalysts on various
of the hydrogenation pathway was much higher when Ni supports for guaiacol HDO, and they found the main products
was added. were cyclohexane, cyclohexanone, and 1-methylcycloheane-
The benzene selectivity for Mo was 65.6 mol%, whereas the 1,2-diol (Table 3, entries 20–23). Although the monometallic
selectivity for NiMo-0.3 was only 30.4 mol%. The selectivity Ni/SiO2 catalyst had the highest extent of HDO (97.3%), the

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

the bio-oil produced is a low quality fuel, due to the water

content and oxygenated compounds, which reduce its heating
value, as well as many acids, which make it difficult to use in
current engines. Even though bio-oil consists of a mixture of
more than 300 chemicals, isolation of those compounds to
upgrade them to targeted products is not promising in view of
the complex separation/purification processes required.
Due to the composition of lignocellulosic biomass (containing
hemicellulose, cellulose and lignin), a promising option is to
separate each fraction and process these fractions separately to
take advantage of the different properties of each component
to produce specialty chemicals and fuels at high conversions and
selectivities. These fractions of biomass consist of highly function-
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

alized molecules, and it is hard to control the selectivity to

Fig. 15 Simplified HDO reaction pathways for (a) guaiacol and (b)
produce target chemicals in a single processing step. Accordingly,
anisole [Adapted from ref. 227].
a successful strategy has been to first convert biomass into
building blocks (i.e., platform molecules), which are more stable,
but retain sufficient functionality to be upgraded to chemicals
Downloaded by Cornell University on 09 August 2012

and fuels in a second step.

10.2. Catalytic upgrading of lignocellulose to fuels and

chemicals
Selection of the proper catalyst (together with reaction conditions)
is critical to maximizing the yields of the selected processes. In this
review, we show how the addition of promoters to metal catalysts
can significantly increase the conversions observed at the same
reaction conditions, and more importantly, how addition of
promoters can modify the selectivity leading to the formation of
Fig. 16 Simplified HDO reaction pathways for guaiacol by Rh-based
completely different products. Although similar trends are
catalysts [Adapted from ref. 226].
observed for the majority of the systems, the results obtained
can be different depending on the properties of the starting
NiCu/ZrO2–SiO2–La2O3 catalyst (96.1% HDO) was identified
materials, the presence of impurities, the feed composition,
as the most promising system in terms of guaiacol conversion,
and the reaction studied. For example, in APR, carbohydrates
deoxygenation extent, low gaseous product formation, and
and polyols can be converted to H2 and CO2, or converted to
coke deposition. On this catalyst, as the temperature was
alkanes of different lengths, depending on the catalyst used. In
increased from 553 K to 633 K, the extent of deoxygenation
addition, with careful catalyst selection, the feedstocks can be
increased, but the guaiacol conversion decreased due to coke
converted into a mixture of mono-functional compounds
formation.
taking advantage of H2 produced in situ combined with partial
conversion to alkanes.
10. Concluding remarks and future prospects Not only is the type of the promoter important, but it is also
important how the promoter interacts with the catalytic metal.
10.1. Lignocellulosic biomass as source of carbon
For example, the addition of Ni to Pt can change the exposed
Concerns about diminishing petroleum reserves, increasing bimetallic surface72 creating synergistic effects the make both
fuel prices, and minimizing environmental impact make it metals more active. In other cases, the addition of a promoter
necessary to identify alternatives to petroleum carbon. One modifies the geometry of the surface (ensemble effect), changing
possibility is the use of lignocellulosic biomass as a source of the selectivity. For example, the addition of Sn to Ru suppress
carbon to produce renewable liquid fuels and chemicals. This CQC hydrogenation, making the catalyst more selective to
approach can be advanced through the use of heterogeneous CQO hydrogenation.59 The interaction can be even more simple,
catalysts to maximize yields, minimize residues, decrease such that the addition of the second metal helps to reduce the
environmental impact of chemical processes, and produce catalytic metal at lower temperature, as is the case for the
products at low prices, such that these products can be addition of Co or Cu to Ni.89,235 Special consideration is given
competitive in the current market. when the second catalytic component can introduce or enhance a
Lignocellulosic biomass can be processed in a single step to new pathway, such as the addition of water-gas shift catalysts
produce synthesis gas via gasification, or bio-oil via pyrolysis. (Ni, Fe, Co) to a metal catalyst in the APR process,65 or the
In the first case, the biomass is completely deconstructed into addition of Re to Rh which introduces an acid functionality to
H2 and CO/CO2, which can be upgraded to produce fuels the catalyst by the creation of acid sites.118
and chemicals. However, the presence of biomass-derived One of the fields were bimetallic catalyst can play a significant
impurities complicates the process. In the case of pyrolysis, role is the improvement of catalyst stability. This effect has

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

proven to be successful in many systems, such as modifying the Notes and references
selectivity to suppress formation of by-products. Examples of
1 Annual Energy Review 2009. Report No. DOE/EIS-0384, 2010, U.S.
improving stability include the addition of Bi to Pt129 in Department of Energy: Washington, DC, Available electronically at
glycerol oxidation, or making the catalyst tolerant to real http://www.eia.gov.
feedstock/upstream processes, as caused by the addition of 2 U.S. Environmental Protection Agency, Renewable Fuel Standard
(RFS), 2011. Available electronically at http://www.epa.gov/otaq/
Re to Ru to hydrogenate levulinic acid in presence of sulfuric
fuels/renewablefuels/index.htm.
acid.202 3 Annual Energy Outlook 2011 with Projections to 2035, Report No.
Due to the wide range of effects that can be caused by DOE/EIA-0383, 2011, U.S. Department of Energy: Washington,
formation of bimetallic systems, each system needs to be DC, Available electronically at http://www.eia.gov.
4 R. D. Perlack, L. L. Wright, A. F. Turhollow, R. L. Graham,
studied carefully, and fundamental studies are required to better B. J. Stokes and D. C. Erbach, Biomass as a feedstock for a
understand the effects of the promoter and its interaction with bioenergy and bioproducts industry: the technical feasibility of a
the primary metal. To help in this task, the principles established billion-ton annual supply.
by studies performed with petroleum-derived chemicals can be 5 J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem.,
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

Int. Ed., 2007, 46, 7164–7183.

initially applied to the catalysts used in biomass conversion. 6 G. W. Huber and B. E. Dale, Sci. Am., 2009, 301, 52–59.
However, the effects observed and results obtained can be 7 R. A. Gross and B. Kalra, Science, 2002, 297, 803–807.
different for petroleum versus biomass catalysis, as the reaction 8 D. J. Hayes, S. Fitzpatrick, M. H. B. Hayes and J. R. H. Ross,
Biorefineries-Industrial Processes and Products. Status Quo and
pathways, by-products generated, and the presence of impurities
Future Directions, WILEY-VCH, Weinheim, 2006, vol. 1,
are different when working with biomass (for example, addition pp. 139–164.
Downloaded by Cornell University on 09 August 2012

of Sn can have different effects depending on the primary metal 9 C. E. Wyman, B. E. Dale, R. T. Elander, M. Holtzapple,
considered, Pt, Ni, Ru, and the reaction conditions).59,71,209 To M. R. Ladisch and Y. Y. Lee, Bioresour. Technol., 2005, 96,
1959 –1966.
achieve the level of understanding that currently exists for 10 Analytical methods in wood chemistry, pulping, and papermaking,
petroleum catalysis, more fundamental and characterization ed. E. Sjöström and R. Alen, Springer-Verlag, Berlin Heidelburg,
studies of in situ reactions and post-reaction catalysts are 1999.
necessary for biomass conversion processes. Research needs 11 P. L. Dhepe and A. Fukuoka, ChemSusChem, 2008, 1, 969–975.
12 R. Rinaldi and F. Schueth, ChemSusChem, 2009, 2, 1096–1107.
to be carried out on reaction mechanisms involved in biomass 13 S. Van de Vyver, J. Geboers, P. A. Jacobs and B. F. Sels,
conversion, since the high functionality of these molecules ChemCatChem, 2011, 3, 82–94.
leads to the formation of many by-products. These by-products 14 N. Mosier, C. Wyman, B. Dale, R. Elander, Y. Y. Lee,
M. Holtzapple and M. Ladisch, Bioresour. Technol., 2005, 96,
interact differently with the surfaces of the bimetallic catalysts due
673–686.
to ensemble and electronic effects, which modify the selectivities 15 Biomass to renewable energy processes, ed. J. Cheng, CRC Press,
and final products. For example, adding Re to Pt leads to Boca Raton, 2010.
production of monofunctional oxygenated hydrocarbons, 16 A. S. Mamman, J.-M. Lee, Y.-C. Kim, I. T. Hwang, N.-J. Park,
Y. K. Hwang, J.-S. Chang and J.-S. Hwang, Biofuels, Bioprod.
or adding Re to Rh leads to selective hydrogenolysis of Biorefin., 2008, 2, 438–454.
oxygenated compounds. Another important consideration 17 J. P. Lange, E. van der Heide, J. van Buijtenen and R. Price,
for biomass conversion is the presence of water, which can ChemSusChem, 2012, 5, 150–166.
lead to hydrothermal deactivation of catalysts and can be 18 J. J. Bozell and G. R. Petersen, Green Chem., 2010, 12, 539–554.
19 J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius and
alleviated by the use of bimetallic catalysts, for example, to B. M. Weckhuysen, Chem. Rev., 2010, 110, 3552–3599.
prevent sintering of particles. 20 D. Vlachos, J. Chen, R. Gorte, G. Huber and M. Tsapatsis,
The use of bimetallic catalysts will be crucial to developing a Catal. Lett., 2010, 140, 77–84.
21 J. C. Serrano-Ruiz and J. A. Dumesic, Energy Environ. Sci., 2011,
biomass-based economy. Researchers have made significant 4, 83–99.
advances to discover new pathways/reactions and to demonstrate 22 D. M. Alonso, J. Q. Bond and J. A. Dumesic, Green Chem., 2010,
that many valuable products can be produced at high yields 12, 1493–1513.
using model compounds. A key step will be to correlate reaction 23 D. A. Simonetti and J. A. Dumesic, ChemSusChem, 2008, 1,
725–733.
kinetics results with catalyst structures to better understand the 24 J. Kopyscinski, T. J. Schildhauer and S. M. A. Biollaz, Fuel, 2010,
effects that lead to catalytic improvement. This knowledge can be 89, 1763–1783.
used first to prepare better catalysts and second to implement the 25 R. W. Breault, Energies (Basel, Switz.), 2010, 3, 216–240.
reaction pathways proposed using real biomass as feedstock. 26 M. Balat, M. Balat, E. Kirtay and H. Balat, Energy Convers.
Manage., 2009, 50, 3158–3168.
27 D. C. Elliott, D. Beckman, A. V. Bridgwater, J. P. Diebold,
S. B. Gevert and Y. Solantausta, Energy Fuels, 1991, 5, 399–410.
Acknowledgements 28 A. V. Bridgwater and M. L. Cottam, Energy Fuels, 1992, 6,
113–120.
This work was funded in part by the DOE Great Lakes 29 A. V. Bridgwater and G. V. C. Peacocke, Renewable Sustainable
Bioenergy Research Center (DOE Office of Science BER Energy. Rev., 2000, 4, 1–73.
DE-FC02-07ER64494). In addition, this work was supported 30 M. Balat, M. Balat, E. Kirtay and H. Balat, Energy Convers.
through funding from the Defense Advanced Research Projects Manage., 2009, 50, 3147–3157.
31 S. Albertazzi, F. Basile, J. Brandin, J. Einvall, G. Fornasari,
Agency (Surf-cat: Catalysts for Production of JP-8 range mole- C. Hulteberg, A. Sanati, F. Trifiro and A. Vaccari, Energy Fuels,
cules from Lignocellulosic Biomass). The views, opinions, and/or 2009, 23, 573–579.
findings contained in this article are those of the author and 32 J. Nishikawa, T. Miyazawa, K. Nakamura, M. Asadullah,
K. Kunimori and K. Tomishige, Catal. Commun., 2008, 9,
should not be interpreted as representing the official views or
195–201.
policies, either expressed or implied, of the Defense Advanced 33 M. M. Yung, W. S. Jablonski and K. A. Magrini-Bair, Energy
Research Projects Agency or the Department of Defense. Fuels, 2009, 23, 1874–1887.

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

34 A. R. Ardiyanti, S. A. Khromova, R. H. Venderbosch, 72 G. W. Huber, J. W. Shabaker, S. T. Evans and J. A. Dumesic,

V. A. Yakovlev and H. J. Heeres, Appl. Catal., B, 2012, 117, Appl. Catal., B, 2006, 62, 226–235.
105–117. 73 D. J. M. de Vlieger, A. G. Chakinala, L. Lefferts, S. R. A.
35 A. Gutierrez, R. K. Kaila, M. L. Honkela, R. Slioor and A. O. Kersten, K. Seshan and D. W. F. Brilman, Appl. Catal., B,
I. Krause, Catal. Today, 2009, 147, 239–246. 2012, 111, 536–544.
36 G. W. Huber and J. A. Dumesic, Catal. Today, 2006, 111, 74 D. L. King, L. Zhang, G. Xia, A. M. Karim, D. J. Heldebrant,
119–132. X. Wang, T. Peterson and Y. Wang, Appl. Catal., B, 2010, 99,
37 D. M. Alonso, J. Q. Bond, J. C. Serrano-Ruiz and J. A. Dumesic, 206–213.
Green Chem., 2010, 12, 992–999. 75 A. Iriondo, J. F. Cambra, V. L. Barrio, M. B. Guemez,
38 G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic, P. L. Arias, M. C. Sanchez-Sanchez, R. M. Navarro and J. L. G.
Science, 2005, 308, 1446–1450. Fierro, Appl. Catal., B, 2011, 106, 83–93.
39 G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106, 76 J. Greeley and M. Mavrikakis, Nat. Mater., 2004, 3, 810–815.
4044–4098. 77 E. Christoffersen, P. Liu, A. Ruban, H. L. Skriver and
40 G. W. Huber, NSF, DOE, and American Chemical Society Work- J. K. Norskov, J. Catal., 2001, 199, 123–131.
shop, Washington, DC, 2007. 78 O. Skoplyak, M. A. Barteau and J. G. Chen, ChemSusChem,
41 C. E. Wyman, L. R. Lynd and T. U. Gerngross, Biotechnol. Prog., 2008, 1, 524–526.
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

1999, 15, 777–793. 79 X. M. Wang, N. Li, L. D. Pfefferle and G. L. Haller, J. Phys.

42 D. Klein-Marcuschamer, P. Oleskowicz-Popiel, B. A. Simmons Chem. C, 2010, 114, 16996–17002.
and H. W. Blanch, Biotechnol. Bioeng., 2012, 109, 1083–1087. 80 E. L. Kunkes, R. R. Soares, D. A. Simonetti and J. A. Dumesic,
43 F. K. Kazi, A. D. Patel, J. C. Serrano-Ruiz, J. A. Dumesic and Appl. Catal., B, 2009, 90, 693–698.
R. P. Anex, Chem. Eng. J. (Lausanne), 2011, 169, 329–338. 81 D. A. Simonetti, E. L. Kunkes and J. A. Dumesic, J. Catal., 2007,
44 S. M. Sen, C. A. Henao, D. J. Braden, J. A. Dumesic and 247, 298–306.
Downloaded by Cornell University on 09 August 2012

C. T. Maravelias, Chem. Eng. Sci., 2012, 67, 57–67. 82 W. Ruettinger, X. Liu, X. Xu and R. J. Farrauto, Top. Catal.,
45 A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2008, 51, 60–67.
2411–2502. 83 L. Zhang, A. M. Karim, M. H. Engelhard, Z. H. Wei, D. L. King
46 P. Gallezot, Catal. Today, 2007, 121, 76–91. and Y. Wang, J. Catal., 2012, 287, 37–43.
47 E. L. Kunkes, D. A. Simonetti, R. M. West, J. C. Serrano-Ruiz, 84 K. G. Azzam, I. V. Babich, K. Seshan and L. Lefferts, J. Catal.,
C. A. Gartner and J. A. Dumesic, Science, 2008, 322, 417–421. 2007, 251, 163–171.
48 R. D. Cortright, R. R. Davda and J. A. Dumesic, Nature, 2002, 85 K. G. Azzam, I. V. Babich, K. Seshan and L. Lefferts, Appl.
418, 964–967. Catal., B, 2008, 80, 129–140.
49 G. W. Huber, J. W. Shabaker and J. A. Dumesic, Science, 2003, 86 N. J. Luo, K. Ouyang, F. H. Cao and T. C. Xiao, Biomass
300, 2075–2077. Bioenergy, 2010, 34, 489–495.
50 J. Chaminand, L. a. Djakovitch, P. Gallezot, P. Marion, C. Pinel 87 S. Y. Foo, C. K. Cheng, T.-H. Nguyen and A. A. Adesina, Catal.
and C. Rosier, Green Chem., 2004, 6, 359–361. Today, 2011, 164, 221–226.
51 M. A. Dasari, P. P. Kiatsimkul, W. R. Sutterlin and G. J. Suppes, 88 C. K. Cheng, S. Y. Foo and A. A. Adesina, Ind. Eng. Chem. Res.,
Appl. Catal., A, 2005, 281, 225–231. 2010, 49, 10804–10817.
52 D. G. Lahr and B. H. Shanks, Ind. Eng. Chem. Res., 2003, 89 C. K. Cheng, S. Y. Foo and A. A. Adesina, Catal. Today, 2011,
42, 5467. 164, 268–274.
53 D. G. Lahr and B. H. Shanks, J. Catal., 2005, 232, 386–394. 90 A. C. C. Chang, R. F. Louh, D. Wong, J. Tseng and Y. S. Lee,
54 T. Miyazawa, Y. Kusunoki, K. Kunimori and K. Tomishige, Int. J. Hydrogen Energy, 2011, 36, 8794–8799.
J. Catal., 2006, 240, 213–221. 91 M. Zhao, T. L. Church and A. T. Harris, Appl. Catal., B, 2011,
55 U. Saxena, N. Dwivedi and S. R. Vidyarthi, Ind. Eng. Chem. Res., 101, 522–530.
2005, 44, 1466. 92 W. Cai, N. Homs and P. Ramirez de la Piscina, Green Chem.,
56 T. Jiang, Y. Zhou, S. Liang, H. Liu and B. Han, Green Chem., 2012, 14, 1035–1043.
2009, 11, 1000–1006. 93 W. Cai, P. Ramirez de la Piscina and N. Homs, Bioresour.
57 P. Maki-Arvela, J. Hajek, T. Salmi and D. Y. Murzin, Appl. Technol., 2012, 107, 482–486.
Catal., A, 2005, 292, 1–49. 94 G. W. Huber, R. D. Cortright and J. A. Dumesic, Angew. Chem.,
58 V. Ponec, Appl. Catal., A, 1997, 149, 27–48. Int. Ed., 2004, 43, 1549–1551.
59 S. G. Wettstein, J. Q. Bond, D. M. Alonso, H. N. Pham, 95 R. M. West, E. L. Kunkes, D. A. Simonetti and J. A. Dumesic,
A. K. Datye and J. A. Dumesic, Appl. Catal., B, 2012, 117–118, Catal. Today, 2009, 147, 115–125.
321–329. 96 Y. Shen, S. Wang, C. Luo and H. Liu, Prog. Chem., 2007, 19,
60 B. Coq and F. Figueras, J. Mol. Catal. A: Chem., 2001, 173, 431–436.
117–134. 97 S. J. You, I. G. Baek, Y. T. Kim, K.-E. Jeong, H.-J. Chae,
61 B. Coq and F. Figueras, Coord. Chem. Rev., 1998, 178, T.-W. Kim, C.-U. Kim, S.-Y. Jeong, T. J. Kim, Y.-M. Chung,
1753–1783. S.-H. Oh and E. D. Park, Korean J. Chem. Eng., 2011, 28,
62 C. T. Campbell, Annu. Rev. Phys. Chem., 1990, 41, 775–837. 744–750.
63 V. Ponec, Adv. Catal., 1983, 32, 149–214. 98 The National Biodiesel Board (NBB). www.nbb.org.
64 J. K. A. Clarke and A. C. M. Creaner, Ind. Eng. Chem. Prod. Res. 99 A. Brandner, K. Lehnert, A. Bienholz, M. Lucas and P. Claus,
Dev., 1981, 20, 574–593. Top. Catal., 2009, 52, 278–287.
65 R. R. Davda, J. W. Shabaker, G. W. Huber, R. D. Cortright and 100 M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi and C. Della
J. A. Dumesic, Appl. Catal., B, 2005, 56, 171–186. Pina, Angew. Chem., Int. Ed., 2007, 46, 4434–4440.
66 R. R. Soares, D. A. Simonetti and J. A. Dumesic, Angew. Chem., 101 A. M. Ruppert, K. Weinberg and R. Palkovits, Angew. Chem.,
Int. Ed., 2006, 45, 3982–3985. Int. Ed., 2012, 51, 2564–2601.
67 X. Cheng, Z. Shi, N. Glass, L. Zhang, J. J. Zhang, D. T. Song, 102 C. Montassier, D. Giraud and J. Barbier, in Stud. Surf. Sci.
Z. S. Liu, H. J. Wang and J. Shen, J. Power Sources, 2007, 165, Catal., ed. M. Guisnet, J. Barrault, C. Bouchoule, D. Duprez,
739–756. C. Montassier and G. Pérot, Elsevier, 1988, vol. 41, pp. 165–170.
68 S. Gottesfeld and J. Pafford, J. Electrochem. Soc., 1988, 135, 103 T. Miyazawa, S. Koso, K. Kunimori and K. Tomishige, Appl.
2651–2652. Catal., A, 2007, 329, 30–35.
69 Q. F. Li, R. H. He, J. A. Gao, J. O. Jensen and N. J. Bjerrum, 104 T. Miyazawa, S. Koso, K. Kunimori and K. Tomishige, Appl.
J. Electrochem. Soc., 2003, 150, A1599–A1605. Catal., A, 2007, 318, 244–251.
70 A. Wawrzetz, B. Peng, A. Hrabar, A. Jentys, A. A. Lemonidou 105 S. X. Xia, Z. L. Yuan, L. N. Wang, P. Chen and Z. Y. Hou, Appl.
and J. A. Lercher, J. Catal., 2010, 269, 411–420. Catal., A, 2011, 403, 173–182.
71 J. W. Shabaker, G. W. Huber and J. A. Dumesic, J. Catal., 2004, 106 S. Xia, Z. Yuan, L. Wang, P. Chen and Z. Hou, Bioresour.
222, 180–191. Technol., 2012, 104, 814–817.

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012
 View Online

107 A. Bienholz, F. Schwab and P. Claus, Green Chem., 2010, 12, 145 I. Witonska, M. Frajtak and S. Karski, Appl. Catal., A, 2011, 401,
290–295. 73–82.
108 J. X. Zhou, L. Y. Guo, X. W. Guo, J. B. Mao and S. G. Zhang, 146 M. Besson, G. Fleche, P. Fuertes, P. Gallezot and F. Lahmer,
Green Chem., 2010, 12, 1835–1843. Recueil Des Travaux Chimiques Des Pays-Bas-Journal of the
109 J. Feng, H. Fu, J. Wang, R. Li, H. Chen and X. Li, Catal. Royal Netherlands Chemical Society, 1996, 115, 217–221.
Commun., 2008, 9, 1458–1464. 147 P. C. C. Smits, B. F. M. Kuster, K. Vanderwiele and
110 C. Montassier, J. C. Ménézo, L. C. Hoang, C. Renaud and H. S. Vanderbaan, Carbohydr. Res., 1986, 153, 227–235.
J. Barbier, J. Mol. Catal., 1991, 70, 99–110. 148 A. Abbadi and H. Vanbekkum, J. Mol. Catal. A: Chem., 1995, 97,
111 E. S. Vasiliadou and A. A. Lemonidou, Appl. Catal., A, 2011, 396, 111–118.
177–185. 149 A. Corma and M. E. Domine, Chem. Commun., 2005, 4042–4044.
112 H. Z. Liu, S. G. Liang, T. Jiang, B. X. Han and Y. X. Zhou, 150 S. Biella, L. Prati and M. Rossi, J. Catal., 2002, 206, 242–247.
Clean: Soil, Air, Water, 2012, 40, 318–324. 151 S. Biella, L. Prati and M. Rossi, J. Mol. Catal. A: Chem., 2003,
113 E. P. Maris, W. C. Ketchie, M. Murayama and R. J. Davis, 197, 207–212.
J. Catal., 2007, 251, 281–294. 152 Y. Önal, S. Schimpf and P. Claus, J. Catal., 2004, 223, 122–133.
114 M. L. Barbelli, G. F. Santori and N. N. Nichio, Bioresour. 153 P. Beltrame, M. Comotti, C. Della Pina and M. Rossi, Appl.
Technol., 2012, 111, 500–503. Catal., A, 2006, 297, 1–7.
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

115 M. G. Musolino, L. A. Scarpino, F. Mauriello and 154 M. Comotti, C. Della Pina, E. Falletta and M. Rossi, J. Catal.,
R. Pietropaolo, ChemSusChem, 2011, 4, 1143–1150. 2006, 244, 122–125.
116 L. Ma, D. H. He and Z. P. Li, Catal. Commun., 2008, 9, 155 A. Mirescu and U. Prüße, Appl. Catal., B, 2007, 70, 644–652.
2489–2495. 156 C. Baatz and U. Prüße, J. Catal., 2007, 249, 34–40.
117 Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and K. Tomishige, 157 C. Baatz and U. Prüße, Catal. Today, 2007, 122, 325–329.
Appl. Catal., B, 2010, 94, 318–326. 158 C. Baatz, N. Decker and U. Prüße, J. Catal., 2008, 258, 165–169.
118 M. Chia, Y. J. Pagan-Torres, D. Hibbitts, Q. H. Tan, 159 M. Comotti, C. Della Pina and M. Rossi, J. Mol. Catal. A:
Downloaded by Cornell University on 09 August 2012

H. N. Pham, A. K. Datye, M. Neurock, R. J. Davis and Chem., 2006, 251, 89–92.

J. A. Dumesic, J. Am. Chem. Soc., 2011, 133, 12675–12689. 160 H. Zhang, M. Okumura and N. Toshima, J. Phys. Chem. C, 2011,
119 Y. Amada, Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and 115, 14883–14891.
K. Tomishige, Appl. Catal., B, 2011, 105, 117–127. 161 H. Zhang, J. Okuni and N. Toshima, J. Colloid Interface Sci.,
120 Y. Nakagawa, Y. Shinmi, S. Koso and K. Tomishige, J. Catal., 2011, 354, 131–138.
2010, 272, 191–194. 162 P. Gallezot, P. J. Cerino, B. Blanc, G. Fleche and P. Fuertes,
121 O. M. Daniel, A. DeLaRiva, E. L. Kunkes, A. K. Datye, J. Catal., 1994, 146, 93–102.
J. A. Dumesic and R. J. Davis, ChemCatChem, 2010, 2, 163 H. Li, H. X. Li and J. F. Deng, Catal. Today, 2002, 74, 53–63.
1107–1114. 164 B. W. Hoffer, E. Crezee, P. R. M. Mooijman, A. D. van Lagneveld,
122 Y. Amada, S. Koso, Y. Nakagawa and K. Tomishige, Chem- F. Kapteijn and J. A. Moulijn, Catal. Today, 2003, 79, 35–41.
SusChem, 2010, 3, 728–736. 165 H. Li and J. F. Deng, J. Chem. Technol. Biotechnol., 2001, 76,
123 M. Schlaf, P. Ghosh, P. J. Fagan, E. Hauptman and 985–990.
R. M. Bullock, Angew. Chem., Int. Ed., 2001, 40, 3887–3890. 166 H. X. Li, H. Li and M. H. Wang, Appl. Catal., A, 2001, 207,
124 M. Schlaf, P. Ghosh, P. J. Fagan, E. Hauptman and 129–137.
R. M. Bullock, Adv. Synth. Catal., 2009, 351, 789–800. 167 P. Gallezot, N. Nicolaus, G. Fleche, P. Fuertes and A. Perrard,
125 N. Ueda, Y. Nakagawa and K. Tomishige, Chem. Lett., 2010, 39, J. Catal., 1998, 180, 51–55.
506–507. 168 Y. D. Wu, H. B. Guo, Y. Wan, M. H. Wang and H. X. Li, Chin.
126 N. Dimitratos, F. Porta and L. Prati, Appl. Catal., A, 2005, 291, J. Catal., 2004, 25, 533–536.
210–214. 169 H. B. Guo, H. X. Li, Y. P. Xu and M. H. Wang, Mater. Lett.,
127 A. Villa, C. Campione and L. Prati, Catal. Lett., 2007, 115, 2002, 57, 392–398.
133–136. 170 J. Pang, A. Wang, M. Zheng, Y. Zhang, Y. Huang, X. Chen and
128 D. Wang, A. Villa, F. Porta, L. Prati and D. S. Su, J. Phys. Chem. T. Zhang, Green Chem., 2012, 14, 614–617.
C, 2008, 112, 8617–8622. 171 A. Fukuoka and P. L. Dhepe, Angew. Chem., Int. Ed., 2006, 45,
129 N. Worz, A. Brandner and P. Claus, J. Phys. Chem. C, 2010, 114, 5161–5163.
1164–1172. 172 C. Luo, S. Wang and H. Liu, Angew. Chem., Int. Ed., 2007, 46,
130 H. Kimura, K. Tsuto, T. Wakisaka, Y. Kazumi and Y. Inaya, 7636–7639.
Appl. Catal., A, 1993, 96, 217–228. 173 N. Ji, T. Zhang, M. Y. Zheng, A. Q. Wang, H. Wang,
131 H. Kimura, K. Tsuto, T. Wakisawa, Y. Kazumi and Y. Inaya, X. D. Wang, Y. Y. Shu, A. L. Stottlemyer and J. G. G. Chen,
Appl. Catal., A, 1995, 123, 323. Catal. Today, 2009, 147, 77–85.
132 C. L. Bianchi, P. Canton, N. Dimitratos, F. Porta and L. Prati, 174 N. Ji, T. Zhang, M. Y. Zheng, A. Q. Wang, H. Wang,
Catal. Today, 2005, 102, 203–212. X. D. Wang and J. G. G. Chen, Angew. Chem., Int. Ed., 2008,
133 N. Dimitratos, J. A. Lopez-Sanchez, D. Lennon, F. Porta, 47, 8510–8513.
L. Prati and A. Villa, Catal. Lett., 2006, 108, 147–153. 175 X. C. Wang, F. Wu, S. X. Yao, Y. J. Jiang, J. Guan and
134 D. Liang, J. Gao, J. H. Wang, P. Chen, Y. F. Wei and Z. Y. Hou, X. D. Mui, Chem. Lett., 2012, 41, 476–478.
Catal. Commun., 2011, 12, 1059–1062. 176 M. S. Feather, J. F. Harris and R. S. T. a. D. Horton, in Adv.
135 Y. H. Shen, S. H. Zhang, H. J. Li, Y. Ren and H. C. Liu, Carbohydr. Chem. Biochem., Academic Press, 1973, vol. 28,
Chem.–Eur. J., 2010, 16, 7368–7371. pp. 161–224.
136 C. J. Biermann, Essentials of Pulping and Papermaking, Academic 177 B. F. M. Kuster, Starch-Starke, 1990, 42, 314–321.
Press, Inc., San Diego, 1993. 178 J. Lewkowski, Arkivoc, 2001, 2, 17–54.
137 A. K. Agarwal, Prog. Energy Combust. Sci., 2007, 33, 233–271. 179 X. Tong, Y. Ma and Y. Li, Appl. Catal., A, 2010, 385, 1–13.
138 A. Demirbas, Prog. Energy Combust. Sci., 2007, 33, 1–18. 180 D. W. Rackemann and W. O. S. Doherty, Biofuels, Bioprod.
139 Y. Sun and J. Y. Cheng, Bioresour. Technol., 2002, 83, 1–11. Biorefin., 2011, 5, 198–214.
140 P. Maki-Arvela, B. Holmbom, T. Salmi and D. Y. Murzin, Catal. 181 J. C. Serrano-Ruiz, R. M. West and J. A. Durnesic, Annu. Rev.
Rev. Sci. Eng., 2007, 49, 197–340. Chem. Biomol. Eng., 2010, 1, 79–100.
141 K. Heyns and H. Paulsen, Adv. Carbohydr. Chem., 1962, 17, 182 D. Fegyverneki, L. Orha, G. Lang and I. T. Horvath, Tetrahedron,
169–221. 2010, 66, 1078–1081.
142 M. Besson, F. Lahmer, P. Gallezot, P. Fuertes and G. Fleche, 183 I. T. Horvath, H. Mehdi, V. Fabos, L. Boda and L. T. Mika,
J. Catal., 1995, 152, 116–121. Green Chem., 2008, 10, 238–242.
143 H. E. J. Hendriks, B. F. M. Kuster and G. B. Marin, Carbohydr. 184 A. A. Rosatella, S. P. Simeonov, R. F. M. Frade and C. A. M.
Res., 1990, 204, 121–129. Afonso, Green Chem., 2011, 13, 754–793.
144 S. Karski, T. Paryjczak and I. Witonska, Kinet. Catal., 2003, 44, 185 C. Carlini, P. Patrono, A. M. R. Galletti, G. Sbrana and V. Zima,
618–622. Appl. Catal., A, 2005, 289, 197–204.

This journal is c The Royal Society of Chemistry 2012 Chem. Soc. Rev.
 View Online

186 P. Verdeguer, N. Merat and A. Gaset, J. Mol. Catal., 1993, 85, 212 G. Seo and H. Chon, J. Catal., 1981, 67, 424–429.
327–344. 213 R. S. Rao, R. T. K. Baker and M. A. Vannice, Catal. Lett., 1999,
187 K. B. Kokoh and E. T. Belgsir, Tetrahedron Lett., 2002, 43, 60, 51–57.
229–231. 214 B. M. Nagaraja, V. S. Kumar, V. Shasikala, A. H. Padmasri,
188 M. Kroger, U. Prusse and K. D. Vorlop, Top. Catal., 2000, 13, B. Sreedhar, B. D. Raju and K. S. R. Rao, Catal. Commun., 2003,
237–242. 4, 287–293.
189 B. Saha, S. Dutta and M. M. Abu-Omar, Catal. Sci. Technol., 215 X. Y. Hao, W. Zhou, J. W. Wang, Y. Q. Zhang and S. X. Liu,
2012, 2, 79–81. Chem. Lett., 2005, 1000–1001.
190 M. A. Lilga, R. T. Hallen and M. Gray, Top. Catal., 2010, 53, 216 B. M. Reddy, G. K. Reddy, K. N. Rao, A. Khan and I. Ganesh,
1264–1269. J. Mol. Catal. A: Chem., 2007, 265, 276–282.
191 S. E. Davis, B. N. Zope and R. J. Davis, Green Chem., 2012, 14, 217 J. Wu, Y. M. Shen, C. H. Liu, H. B. Wang, C. Geng and
143–147. Z. X. Zhang, Catal. Commun., 2005, 6, 633–637.
192 S. E. Davis, L. R. Houk, E. C. Tamargo, A. K. Datye and 218 S. Sitthisa, W. An and D. E. Resasco, J. Catal., 2011, 284, 90–101.
R. J. Davis, Catal. Today, 2011, 160, 55–60. 219 V. Vetere, A. B. Merlo, J. F. Ruggera and M. L. Casella, J. Braz.
193 V. Schiavo, G. Descotes and J. Mentech, Bull. Soc. Chim. Fr., Chem. Soc., 2010, 21, 914–920.
1991, 704–711. 220 P. D. Vaidya and V. V. Mahajani, Ind. Eng. Chem. Res., 2003, 42,
Published on 08 August 2012 on http://pubs.rsc.org | doi:10.1039/C2CS35188A

194 R. Alamillo, M. Tucker, M. Chia, Y. Pagan-Torres and 3881–3885.

J. Dumesic, Green Chem., 2012, 14, 1413–1419. 221 T. Marinelli, L. Ponec, C. G. Raab and J. A. Lercher, in
195 Y. Roman-Leshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic, Heterogeneous Catalysis and Fine Chemicals Iii, eds.
Nature, 2007, 447, U982–U985. M. Guisnet, J. Barbier, J. Barrault, C. Bouchoule, D. Duprez,
196 J. Q. Bond, D. M. Alonso, D. Wang, R. M. West and G. Perot and C. Montassier, Elsevier Science Publ B V, Amster-
J. A. Dumesic, Science, 2010, 327, 1110–1114. dam, 1993, vol. 78, pp. 195–202.
197 L. Deng, Y. Zhao, J. Li, Y. Fu, B. Liao and Q.-X. Guo,
Downloaded by Cornell University on 09 August 2012

222 T. B. L. W. Marinelli, S. Nabuurs and V. Ponec, J. Catal., 1995,

ChemSusChem, 2010, 3, 1172–1175. 151, 431–438.
198 H. Heeres, R. Handana, D. Chunai, C. B. Rasrendra, B. Girisuta 223 S. Sitthisa, T. Pham, T. Prasomsri, T. Sooknoi, R. G. Mallinson
and H. J. Heeres, Green Chem., 2009, 11, 1247–1255. and D. E. Resasco, J. Catal., 2011, 280, 17–27.
199 P. P. Upare, J.-M. Lee, D. W. Hwang, S. B. Halligudi, 224 J. H. Lora and W. G. Glasser, J. Polym. Environ., 2002, 10, 39–48.
Y. K. Hwang and J.-S. Chang, J. Ind. Eng. Chem., 2011, 17, 225 A. L. Jongerius, R. Jastrzebski, P. C. A. Bruijnincx and
287–292. B. M. Weckhuysen, J. Catal., 2012, 285, 315–323.
200 Z.-p. Yan, L. Lin and S. Liu, Energy Fuels, 2009, 23, 3853–3858. 226 Y. C. Lin, C. L. Li, H. P. Wan, H. T. Lee and C. F. Liu, Energy
201 X. L. Du, Q. Y. Bi, Y. M. Liu, Y. Cao and K. N. Fan, Fuels, 2011, 25, 890–896.
ChemSusChem, 2011, 4, 1838–1843. 227 M. A. Gonzalez-Borja and D. E. Resasco, Energy Fuels, 2011, 25,
202 D. J. Braden, C. A. Henao, J. Heltzel, C. T. Maravelias and
4155–4162.
J. A. Dumesic, Green Chem., 2011, 13, 1755–1765.
228 M. J. Girgis and B. C. Gates, Ind. Eng. Chem. Res., 1991, 30,
203 D. M. Alonso, S. G. Wettstein, J. Q. Bond, T. W. Root and
2021–2058.
J. A. Dumesic, ChemSusChem, 2011, 4, 1078–1081.
229 H. Kobayashi, H. Ohta and A. Fukuoka, Catal. Sci. Technol.,
204 J.-P. Mikkola, H. Vainio, T. Salmi, R. Sjöholm, T. Ollonqvist and
2012, 2, 869–883.
J. Väyrynen, Appl. Catal., A, 2000, 196, 143–155.
230 C. R. Lee, J. S. Yoon, Y. W. Suh, J. W. Choi, J. M. Ha, D. J. Suh
205 J. Wisniak, M. Hershkowitz and S. Stein, Ind. Eng. Chem. Prod.
Res. Dev., 1974, 13, 232–236. and Y. K. Park, Catal. Commun., 2012, 17, 54–58.
206 C. Montassier, J. C. Menezo, J. Moukolo, J. Naja, L. C. Hoang, 231 V. N. Bui, D. Laurenti, P. Afanasiev and C. Geantet, Appl.
J. Barbier and J. P. Boitiaux, J. Mol. Catal., 1991, 70, 65–84. Catal., B, 2011, 101, 239–245.
207 K. J. Zeitsch, The chemistry and technology of furfural and its 232 V. N. Bui, D. Laurenti, P. Delichere and C. Geantet, Appl. Catal.,
many by-products, Elsevier, Amsterdam, Sugar Series, 1st edn, B, 2011, 101, 246–255.
2000, vol. 13, pp. 34–69. 233 A. Centeno, E. Laurent and B. Delmon, J. Catal., 1995, 154,
208 P. Verdeguer, N. Merat and A. Gaset, Appl. Catal., A, 1994, 112, 288–298.
1–11. 234 B. Yoosuk, D. Tumnantong and P. Prasassarakich, Fuel, 2012,
209 A. B. Merlo, V. Vetere, J. F. Ruggera and M. L. Casella, Catal. 91, 246–252.
Commun., 2009, 10, 1665–1669. 235 L. De Rogatis, T. Montini, A. Cognigni, L. Olivi and
210 T. B. L. W. Marinelli, J. H. Vleeming and V. Ponec, Stud. Surf. P. Fornasiero, Catal. Today, 2009, 145, 176–185.
Sci. Catal., 1993, 75, 1211–1222. 236 M. V. Bykova, D. Y. Ermakov, V. V. Kaichev, O. A. Bulavchenko,
211 R. Rao, A. Dandekar, R. T. K. Baker and M. A. Vannice, A. A. Saraev, M. Y. Lebedev and V. A. Yakovlev, Appl. Catal., B,
J. Catal., 1997, 171, 406–419. 2012, 113, 296–307.

Chem. Soc. Rev. This journal is c The Royal Society of Chemistry 2012