REVIEW
1649
Catalytic One-Pot Production of Small Organics from Polysaccharides
Cat lyticOne-PotProdJohan
J.
uctionofSmal Organics
Verendel,a Tamara L. Church,a Pher G. Andersson*a,b
a
Department of Biochemistry and Organic Chemistry, Uppsala University, Box 576, 75123 Uppsala, Sweden
School of Chemistry, University of KwaZulu-Natal, Durban, South Africa
Fax +46(18)4713818; E-mail: [email protected]
Received 17 November 2010; revised 21 January 2011
b
1
1.1
1.2
2
2.1
2.1.1
2.2
2.3
3
3.1
3.2
4
4.1
4.2
5
6
7
Introduction
Scope and Related Topics
A Note on Yield
Acid-Catalyzed Processes
Hydrolysis to Monosaccharides
Decomposition of Sugars in Hot, Dilute Acid
Selective Formation of Furan-2-aldehydes
Selective Formation of Levulinic Acid
Trapping of Monosaccharides by Chemical Modification
Formation of Alkyl Glucosides
Formation of Polyols
Alkaline Hydrolysis
Cellulose Conversion
Product Selectivity
Hydrothermal Degradation
Metal-Catalyzed Oxidative Degradation
Conclusion and Outlook
Key words: catalysis, green chemistry, polysaccharides, renewable
feedstocks, cellulose
other small sugars has long been studied. Most frequently,
dilute or concentrated acid is used to hydrolyze wooden
material to glucose and pentoses. Sugars such as glucose
(the monomer unit of cellulose) and xylose (a common
monomer unit of hemicellulose) can be used as obtained,
or converted into a wide variety of compounds. Historically, ethanol has been the most common target of sugar
degradation, though the obtainable products are quite diverse and have been extensively reviewed.2
Unfortunately, polysaccharides can require harsh conditions to degrade. This is especially true for cellulose,
which is largely crystalline and held together by a network
of hydrogen bonds (Figure 1). Thus the conditions required to hydrolyze polysaccharides to sugars often degrade these same sugars.3 This incompatibility between
high feedstock stability and lower product stability has
been a long-standing challenge for chemists and engineers, who have sought technical and chemical processes
that operate under milder conditions, and thus produce
higher yields of glucose and other sugars. Some progress
toward this goal has been achieved.
β-(1
a)
OH
Introduction
Based upon their abundance, polysaccharides, and especially cellulose and hemicellulose, have the potential to be
major sources of small molecules for a variety of applications. Two polysaccharides, cellulose and hemicellulose,
make up approximately 70–85% of lignocellulose,1 which
in turn makes up the vast majority of woody biomass and
agricultural residues. Additionally, cellulose cannot be digested by humans, so its use as chemical feedstock need
not compete with food resources. Thus the catalytic degradation of natural polysaccharides to obtain glucose and
SYNTHESIS 2011, No. 11, pp 1649–1677xx. 201
Advanced online publication: 19.04.2011
DOI: 10.1055/s-0030-1260008; Art ID: E28910SS
© Georg Thieme Verlag Stuttgart · New York
OH
OH
HO
O
O
O
HO
O
OH
OH
b)
O
HO
HO
n
OH
OH
OH
O
H
O
H
H
O
O
O H
O
OH
OH
OH
O
O
O
O
HO
O H
OH
glucose
cellulose
HO
O
1
4)-glycoside linkage
HO
O
O H
H
O
O
O
O
O
H
H
O
O
OH
O
O
O H
O
HO
O
OH
O H
O
O
OH
OH
Figure 1 (a) Cellulose is a linear homopolymer of anhydroglucose
units bound via b-(1→4)-glycoside linkages. (b) A network of intraand interchain hydrogen bonds causes the high stability and low solubility of cellulose.
Today, biomass samples that contain both hemicellulose
and cellulose are first subjected to a milder hydrolysis,
which degrades hemicellulose, before the more resistant
cellulose is hydrolyzed under harsher conditions. This
procedure improves the yields of pentoses (mainly xylose
and arabinose) and hexoses (mainly mannose and galac-
Downloaded by: the University of Oxford. Copyrighted material.
Abstract: The catalytic techniques that degrade polysaccharides
such as cellulose and elaborate the resulting monomers into organic
feedstocks in situ make up an old area of research that has been reignited of late. One-pot conversions of polysaccharides into small
organic molecules under acidic, basic, oxidative, reductive and hydrothermal conditions have been reported, and a remarkable breadth
of compounds have been produced. 5-Hydroxymethylfurfural
(HMF), levulinic acid and polyols such as sorbitol have been obtained in particularly high yields, whereas serious selectivity struggles remain in reactions that produce organic acids such as lactic or
glycolic acid. This review covers one-pot, catalytic transformations
of polysaccharides into small organic molecules and focuses on
mechanism, selectivity and optimization.
REVIEW
J. J. Verendel et al.
tose) from hemicellulose, which is a branched, inhomogeneous polymer (Figure 2).4
OAc
Additionally, pretreatments that decrease cellulose crystallinity and increase its surface area and porosity, such as
ball-milling5 or steam explosion,6 are often used to permit
lower operation temperatures, thereby minimizing glucose decomposition. Nevertheless, the hydrolysis of cellulose to glucose in high selectivity and conversion
remains difficult.
During the course of efforts to degrade polysaccharides to
monosaccharides for use as renewable sources of organic
chemicals and fuels, a myriad of ‘byproducts’ have been
identified; the character and amount of these byproducts
depend on the conditions used for polysaccharide degradation. Several of these compounds appear frequently in
reviews on useful products that can be derived from
monosaccharides, and efforts to produce them selectively
from biomass have begun.
These compounds offer an opportunity to exploit, rather
than combat, the problem of glucose instability during
cellulose hydrolysis: the selective trapping of decomposition products from the hemiacetal glucose may well be
easier than obtaining quantitative yields of glucose itself.
hemicellulose
HO O
HO
HO O
OH
HO
OH
arabinose
CO2H
HO
HO
HO
OH HO
galactose
OH
O
OH
OH
HO OH
OH
HO
HO
glucuronic acid
HO O
mannose
HO
HO
OH
O
OH
glucose
OH
O
OH
OH
xylose
Figure 2 The term ‘hemicellulose’ describes a family of branched
copolymers containing various saccharide monomers. The structures
of hemicelluloses vary among biomass types.
Biographical Sketches
J. Johan Verendel, born in
1981 in Nynäshamn, Sweden, graduated as a chemical engineer from Uppsala
University in 2007 having
focused his studies on syn-
thetic organic processes and
medicinal compound production. He is currently
working towards a PhD in
the fields of organometallic
synthesis
methodology,
asymmetric catalysis and
sustainable chemistry together with Prof. Pher G.
Andersson.
Tamara L. Church is from
Ottawa, Canada. She studied chemistry at the University of Ottawa, where she
performed research under
the direction of Assoc. Prof.
Susannah L. Scott. She then
investigated catalytic het-
erocycle carbonylation as a
PhD student at Cornell University under the direction
of Prof. Geoffrey W. Coates
before joining the research
group of Prof. Pher G.
Andersson at Uppsala University as a Postdoctoral
Fellow. She is currently
working with Assoc. Prof.
Andrew T. Harris in the
Laboratory of Sustainable
Technology at Sydney University.
Pher G. Andersson was
born 1963 in Växjö, Sweden. He was educated at
Uppsala University where
he received his BSc in 1988
and his PhD in 1991. After
post-doctoral research at
Scripps Research Institute
with Prof. K. B. Sharpless,
he returned to Uppsala
where he became docent in
1994 and full professor in
1999. His main research interests involve organometallic chemistry, stereoselective synthesis, asymmetric catalysis and sustainable processes.
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
1650
REVIEW
these compounds in significant yield. It will also attempt
to cover the recent development of methods to selectively
trap monosaccharides in situ as more stable compounds.
Additionally, this type of strategy could combine the hydrolysis of polysaccharide to sugar with a subsequent
modification in one pot, yielding a more efficient and environmentally benign reaction that requires fewer purification steps and less solvent, time and energy.
The field of polysaccharide degradation is both large and
broad, and will not be covered exhaustively here. In fact,
even the narrower field that concerns the hydrolysis of
polysaccharides to their constituent monomers has seen
almost 200 years research, and has been reviewed.8 Thus
this topic will not be comprehensively addressed. Rather,
we aim to review the direct, catalytic conversion of
polysaccharides, in particular cellulose but also hemicellulose, into small organic molecules in liquid media. Our
focus is directed towards the factors that affect yield and
selectivity in these reactions, and efforts to improve these
quantities. Some of the techniques discussed in this review were discussed in a recent mini-review.9
During the past ten years, there has been significant focus
on developing reusable solid catalysts for biomass degradation and some advances have been made.7 In particular,
the solid-acid-catalyzed hydrolysis of polysaccharides in
various media has been explored and combined with the
in situ metal-catalyzed reduction of the resulting sugars.
Furthermore, the one-pot hydrolysis and dehydration of
solubilized cellulose to 5-hydroxymethylfurfural (HMF)
and levulinic acid has been studied extensively. The oxidative and base-catalyzed degradations of cellulose have
proven more difficult to control, but have also yielded
useful products. By setting the spotlight on the one-pot,
catalytic hydrolysis and further elaboration of polysaccharides, and especially cellulose, we hope to demonstrate
the potential of this strategy to provide a renewable source
of small organic molecules (Figure 3), as well as to identify areas in which further development could offer significant rewards.
A Note on Yield
Discussions of yield and selectivity in cellulose degradation are complicated by the various systems for reporting
yields from these reactions. In the literature, these values
are variously reported as percent by mole, by weight, and
by carbon basis. Though yields in mol% are preferable in
Scope and Related Topics
This review highlights the ‘side reactions’ of sugars that
take place during the catalytic degradation of polysaccharides under various conditions, and the efforts to isolate
O
O
OH
H
OH
formic acid
O
HO
O
O
levulinic acid
OH
HO
hydroxymethyl furfural
O
HO
OH
OH
O
Cl
glucose
OH
O
HO
one pot
chloromethyl furfural
catalyst
OH
OH
OH
OH OH
sorbitol
OH
HO
O
OH
O
O
O
R
alkyl glucopyranoside
OH
OH
O
HO
O
O
HO
OH
OH
HO
OH
n
OH
OH
OH
glucoisosaccharinic acid
cellulose
O
OH
OH
HO
HO
MeOH
OH
OH
OH
OH
OH
HO
OH
OH
HO
OH
HO
OH
O
OH
OH OH OH
mannitol
alcohols
Figure 3
O
OH
O
HO
OH
acids
Small organic compounds obtainable from cellulose by one-pot catalytic techniques
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
Some polysaccharide degradation techniques that are not
covered here have been the subject of other reviews; these
include enzymatic degradation,10 the catalytic production
of hydrogen, carbon monoxide, and syn gas,11 and pyrolysis and gasification techniques.12
1.2
1.1
1651
Catalytic One-Pot Production of Small Organics
REVIEW
J. J. Verendel et al.
even at room temperature when the acid concentration is
≥ 65% w/v and the acid-to-cellulose ratio is >2 mL/g.18 The
hydrolysis of solubilized cellulose is rapid at temperatures
between 50 and 100 °C and results in very little degradation of the formed monosaccharides because of the low
temperature. Despite the high glucose yields available
from this process, problems associated with recovering
the high-boiling sulfuric acid have led to the use of other
strong acids, such as hydrofluoric,19 phosphoric,20 hydrochloric,21 trifluoroacetic22 and formic23 acids.
terms of understanding the reaction, they are not always
possible or reasonable. In the present review, we have
given yields in mol% when the theoretical yield could be
calculated definitively, such as in the case of HMF formation. In more complicated cases, when the initial number
of monomer units is unclear (such as in a biomass sample)
or it is unclear precisely how many equivalents of a product can reasonably be produced from an equivalent of cellulose monomer unit, wt% is used.
2
Acid-Catalyzed Processes (T < 250 °C)
2.1
Hydrolysis to Monosaccharides
To our knowledge, only Hardt and Lamport have isolated
a small organic compound other than the sugar monomer
from concentrated acid hydrolysis. They dissolved wood
chips in liquid hydrofluoric acid and poured the resulting
mixture into a suspension of calcium carbonate in dichloromethane. Aqueous extraction produced a-D-glucopyranosyl fluoride in 20 wt% yield.24
To understand the trapping of glucose-degradation products in aqueous acid, it is necessary to consider the conditions under which glucose can be formed. Braconnot first
demonstrated the solubilization of cellulose with concentrated Brønsted acids in 1819.13 This process works because the highly ionic solvent is able to penetrate and
disrupt the hydrogen bonds between and within the cellulose polymer strands (see Figure 1).14 Additionally, strong
acids can derivatize the hydroxy groups along the cellulose chain, disrupting intermolecular hydrogen bonding
and producing cellulose derivatives that are more reactive
than cellulose itself.15 The fate of cellulose (or its derivatives) dissolved in strong acid depends upon the reaction
conditions. If the solution is diluted immediately after the
cellulose is dissolved and the system is not heated, very
little chain-breaking occurs and cellulose is regenerated.16
On the other hand, diluting the strongly acidic solution
with water and heating to ~100 °C for a few hours yields
glucose, typically in >80% yield.17 This process is commonly denoted ‘concentrated acid hydrolysis’ or ‘homogeneous acid hydrolysis’, and is shown in Scheme 1. The
acid-catalyzed hydrolysis of polysaccharides is a specificacid-catalyzed process, so the particular acid used is irrelevant; the pH alone determines the rate of hydrolysis.
Despite their selective nature, concentrated acid processes
suffer from several serious disadvantages: the acids are
corrosive, and the waste products contain large amounts
of acids that must be neutralized. These issues prevented
the concentrated acid hydrolysis of polysaccharides from
being extensively industrialized, even though it offers access to very high sugar yields.
Dilute acid solutions can also degrade polysaccharides,
though high temperatures must be used to compensate for
the weaker acid solutions. This comes at the cost of selectivity, as monosaccharides are not stable at high temperatures, especially not under acidic conditions.
Nevertheless, commercial plants that generated sugars
from lignocellulose using hot, dilute acid were operative
during the first half of the 20th century. The dilute acid hydrolysis of polysaccharides dates back to an initial report
by Simonsen in 1898.25 Simonsen heated cellulose in
aqueous 0.5 wt% sulfuric acid for four hours under 9 bar
pressure with an approximate temperature of 175 °C and
obtained a 45% yield of glucose, which was subsequently
neutralized and fermented to produce ethanol. This was a
batch process that yielded very dilute sugar solutions, but
it was followed by intense research aimed at improving its
efficiency.26 The dilute acid hydrolysis of cellulose is usu-
Sulfuric acid was traditionally chosen to mediate cellulose
hydrolysis because of its low cost and strong acidity. The
dissolution of cellulose in aqueous sulfuric acid is fast
OH
OH
HO
O
O
O
HO
OH
OH
OH
HO
2n HO
H+
2n H2O
O
n
OH
O
OH
cellulose
glucose
H+
OH
OH
HO
O
O
O+
HO
H
OH
Scheme 1
OH
O
OH
OH
HO
O
O
O+
n
H
OH
The acid-catalyzed hydrolysis of cellulose to yield glucose
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
OH
– H+
O
HO
OH
H2O
n
OH
HO
O
OH O
O
OH
H
O
HO
OH
n
Downloaded by: the University of Oxford. Copyrighted material.
1652
Catalytic One-Pot Production of Small Organics
ally performed between 150 and 250 °C and using ~1 wt%
of a strong, inorganic Brønsted acid. Carboxylic acids
such as acetic and carbonic acids have been used, but
yielded inferior results because they, as weaker acids, are
less suitable for specific-acid-catalyzed processes (such as
cellulose hydrolysis) where the reaction rate is directly
correlated to the pH. Cellulose is not significantly dissolved in dilute acid at temperatures below 250 °C, so the
degradation seen in these reactions is attributed to a combination of the rather fast diffusion of water and protons
into amorphous parts of the cellulose, and slower reactions on the crystal surface that liberate soluble glucose
and oligosaccharides.27 Consequently, the degree of polymerization of cellulose declines in the early stages of dilute acid hydrolysis, but gradually levels off as the fast
hydrolysis of the sensitive amorphous regions is completed and is followed by slow erosion at the surface of the inert crystallites.28 There are also indications that both
chemical modifications of the solid cellulose and the formation of unreactive oligosaccharides occur under these
conditions, further lowering the monomer yield and prohibiting full conversion of poly- and oligosaccharides to
monomers (Scheme 2).29 Hence, in most reports, the decrease in degree of polymerization diminishes quickly at
first, but eventually becomes relatively constant when the
system consists of glucose, soluble oligosaccharides and
insoluble crystalline cellulose.29b,30
easily hydrolyzable cellulose
crystalline cellulose
fast
unreactive
oligosaccharides
slow
oligosaccharides
glucose
unreactive cellulose
dehydrated sugars
degradation products
Scheme 2 Model for the dilute acid hydrolysis of cellulose, adapted
from published models29
Despite problems in achieving complete cellulose degradation from dilute-acid hydrolysis, the most serious drawback of the process is that the desired products are
unstable to the reaction conditions. At 200 °C, for example, the hydrolysis of oligosaccharides is only acid-catalyzed for pH < 4;31 under these conditions, the pentoses
and hexoses formed from the degradation of lignocellulosic material decompose quickly.32 Thus the maximum
yield of glucose from dilute acid cellulose hydrolysis was
long considered to be ca. 60–70%. The quick decomposition of monosaccharides in hot acid and the desire to obtain high yields of monosaccharides, which are essentially
reaction intermediates under these conditions, prompted
the development of several modified processes over the
past century. For example, cellulose pretreatment can be
1653
used to shorten the hydrolysis time and therefore improve
glucose stability.
Dilute acid hydrolysis is often applied in flow reactors,
which minimize the contact time between the formed
monosaccharides and hot acid.33 Additionally, multi-step
hydrolysis processes have been used to improve the total
sugar yields from lignocellulose. In the first step, hemicellulose and easily hydrolyzable cellulose regions are hydrolyzed and the products collected; the more robust
crystalline cellulose is then hydrolyzed under more extreme conditions.34 Hemicellulosic sugars can be produced selectively from lignocellulosic material by dilute
acid hydrolysis. Under carefully controlled conditions,
high yields of xylose and other hemicellulosic monosaccharides can be obtained without significantly altering the
cellulosic fraction.35
During the past ten years, significant research has been
devoted to the acid-catalyzed hydrolysis of polysaccharides by solid acids, which can be recycled. These reactions are generally conducted in water using pretreated
cellulose and temperatures of 100–150 °C; the results are
generally glucose and water-soluble oligosaccharides in
~50% total yield. As in the case of reactions using homogeneous acids, low temperatures are preferred in order to
avoid subsequent reactions of glucose,36 though the formation of decomposition products is significant even under these conditions. This topic has been reviewed.37
Cellulose and other polysaccharides can be solubilized in
ionic liquids38 and in various mixtures of molecular solvents and ionic liquids or salts.39 Li and Zhao reported the
first homogeneous acid-catalyzed degradation of cellulose
to glucose in ionic liquids.40 They obtained 39% glucose
yield from microcrystalline cellulose in a mixture of sulfuric acid and [nBuMIM]+[Cl]– (MIM = 3-methylimidazolium) after 35 minutes at 100 °C. The rest of the
cellulose was also dissolved, and was quantified as oligosaccharidic total reducing sugars (TRS). The same authors later extended their study to other lignocellulosic
materials.41 Processes of this kind42 mainly yield soluble
oligosaccharides, which could be used to produce viscous
aqueous preparations, or hydrolyzed further using other
protocols. In contrast to aqueous heterogenous hydrolysis,
the hydrolysis of cellulose that is dissolved in an ionic liquid or a concentrated acid occurs by random scissions because all of the glycosidic bonds in the polymer are
equally available. Consequently, a short delay in glucose
formation is normally observed after cellulose cleavage
into shorter chains has started.43 Cello-oligomers (DP >
~100) can be precipitated from ionic liquid solution by
adding water.44
As in aqueous media, cellulose hydrolysis in ionic liquids
is faster at lower pH. (Even in ionic liquid systems, a small
amount of water has to be present to promote hydrolysis.)
In ionic liquids, acid-catalyzed cellobiose hydrolysis is
faster than glucose decomposition in a more acidic environment, whereas the opposite is true in less acidic media.43b
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
REVIEW
REVIEW
J. J. Verendel et al.
Rinaldi, Schüth and co-workers have reported the hydrolysis of cellulose to oligosaccharides in ionic liquids using
solid acid catalysts.45 Amberlyst 15 and 35 outperformed
Amberlyst 70, Nafion, g-alumina and zeolites as catalysts
for cellulose depolymerization. The authors suggested
that the pore size and surface area of the solid acid were
pivotal in determining the degree of hydrolysis. In a later
study, the authors concluded that protons were released
from the acidic resin into solution via ion-exchange and
that, consequently, the catalyst was not heterogeneous in
practice.43a The resins could not be reused unless they
were regenerated. In fact, ionic liquids are exceptionally
good at releasing protons from acidic ion-exchange resins
such as Nafion and Amberlyst.46 Many more protons are
released from acidic resins in mixtures of water and
[nBuMIM]+[Cl]– than in pure water, probably because the
high ion concentration in the solution facilitates ionexchange.
2.1.1
Decomposition of Sugars in Hot, Dilute Acid
The degradation of glucose itself in hot, dilute acid has
seen significant research, and occurs primarily by dehydration.29a,47 The main degradation product under these
conditions is HMF, which can be formed both from glucose and from its isomer, fructose (Scheme 3). In fact, it
forms significantly faster, and in higher yield, from fructose than from glucose.48 Thus, because glucose and fructose can be interconverted in aqueous acid by the Lobry de
Bruyn–Alberda van Ekenstein rearrangement,49 HMF can
be formed directly from glucose or via a fructose intermediate. The formation of HMF from fructose and glucose
involves three consecutive dehydration steps and is acidcatalyzed.48,50 Just as HMF is formed from glucose, furfural is a main degradation product of pentoses such as xylose and arabinose in acidic solution.51 Consequently, the
acidic degradation of hemicellulose, a more chemically
diverse compound than cellulose (see Figure 2), generates
xylose, arabinose, and furfural, as well as acetic acid from
the acetyl ester groups that are also present.35a,52 Furfural
formation from xylose also occurs via dehydration,53 and
its rate increases with lower pH and higher temperature.54
Thus, HMF and furfural are often byproducts of polysaccharide hydrolysis in acid.55 Nevertheless, high yields of
hemicellulosic monosaccharides can be obtained from
lignocellulosic material by dilute acid hydrolysis because
hemicellulose can be hydrolyzed under milder conditions,
thus avoiding furfural formation. Alternatively, furfural
can be isolated from hemicellulose under more severe
conditions, as it is more stable in acidic solution than xylose is.56 The production of furfural from hemicellulose
and xylose has been reviewed.2b,57
HMF can also decompose in aqueous media, giving levulinic and formic acids.58 After HMF and levulinic and
formic acids, humic solids are the major decomposition
products from glucose in hot dilute acid, accounting for up
to 20% of the carbon from glucose degradation.56 The formation of insoluble humins and polymers increases at
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
H+
or
HO–
OH
O
HO
HO
HO
OH
O
OH
OH
OH
OH
fructose
glucose
3 H2O
slow
OH
3 H2O
fast
O
HO
O
HMF
Scheme 3
Formation of HMF from glucose and fructose
higher glucose concentration and longer residence time.
In ionic liquids, the yield of solid humins from glucose
and fructose increases linearly with time when acid is
present, whereas solids are not formed under acid-free
conditions.59 In a study conducted in acidic ionic liquid
solution, 16 and 21 wt% humins were produced from glucose and xylose, repectively, after four hours at 120 °C.
Interestingly, although neither HMF nor furfural produced any solid residue under these reaction conditions,
equimolar mixtures of glucose–HMF and xylose–furfural
produced 33 and 41 wt%, respecitvely. These results suggest that humin formation takes place during reactions between furans and other monosaccharide degradation
products, and can be a considerable obstacle to obtaining
high yields of furan-2-aldehydes from monosaccharides
in ionic liquids.59
Several other byproducts of glucose decomposition, including erythrose, glycolic acid, dihydroxyacetone, pyruvaldehyde, glyceraldehyde and lactic acid, can be
detected in trace amounts.50b,54,60 The degradation of glucose during the dilute acid hydrolysis of cellulose is summarized in Scheme 4.
cellulose
oligosaccharides
OH
OH
levoglucosan
OH
OH
O
OH
O
– H2O
HO
HO
– 3 H2O
O
OH
glucose
OH
O
O
HMF
2 H2O
humins,
minor
degradation
products
OH
2
O
O
levulinic acid
+
O
HO
formic acid
Scheme 4 Major pathways for glucose degradation during dilute
acid hydrolysis of cellulose
Downloaded by: the University of Oxford. Copyrighted material.
1654
2.2
Selective Formation of Furan-2-aldehydes
As HMF is the initial product of the acid-catalyzed degradation of glucose, attempts have been made to isolate it,
rather than glucose, in high yield from cellulose degradation, and significant amounts of HMF have been obtained
from dilute aqueous acid systems. However, HMF is also
degraded in acid, so the rate of polysaccharide hydrolysis
must always be balanced against that of HMF degradation. Under the conditions normally employed for the dilute acid hydrolysis of polysaccharides, higher
temperatures and acid concentrations lead to faster HMF
formation,29a and these parameters can be manipulated to
maximize HMF yield.
For example, Agrawal, Jones and co-workers studied the
dilute acid hydrolysis of pine sawdust with various acids
at pH 1.65.61 After heating to 150 °C for two hours, they
found most of the formed monosaccharides intact, but
both pentoses and hexoses decomposed significantly at
higher temperatures. After one hour at 200 °C in aqueous
trifluoroacetic acid, the reaction yielded furfural (57%)
and HMF (18%). When sulfuric acid was used, the furfural yield was 33% but very little HMF was obtained; this
was attributed to further degradation of the latter. The difference between the reactions in trifluoroacetic acid and
sulfuric acid may derive from the decreased dissociation
constants, and hence acidity of the acids in high-temperature water.62 In total, the trifluoroacetic acid solution dissolved 38% of the pine sawdust, whereas sulfuric acid
dissolved 55% of the material. Pure water dissolved 30%
of the biomass and returned 11% HMF at this temperature
(see also section 5).
In addition to Brønsted acids, Lewis acids and electrochemically generated acids have also been applied to the
hydrolysis of cellulose, and have produced HMF. Seri and
co-workers used aqueous 0.01 M lanthanum(III) chloride,
a water-tolerant Lewis acid, to catalyze cellulose hydrolysis at 250 °C. The HMF yield peaked at 19% after three
minutes and then declined rapidly as HMF was being replaced by levulinic acid.63
In 1988, Nobe and co-workers explored cellulose hydrolysis by the acid generated electrochemically from water
and 0.3 M sodium sulfate. The conversion of cellulose
was generally quite low, and depended strongly on temperature. At 160 °C, the degradation of glucose to HMF
began; at 200 °C, levulinic acid and other degradation
products were the main compounds detected.64
Regardless of the temperature and pH used in the aqueous
hydrolysis of cellulose or glucose, high yields of HMF are
not seen due to its slow formation from glucose or cellulose, and its fast rehydration to levulinic and formic acids.
At 190 °C and aqueous 0.1 M sulfuric acid, glucose is
completely converted into HMF and further into levulinic
acid in 75% selectivity after only 30 minutes.65 On the
other hand, high yields of HMF can be obtained starting
from aqueous solutions of fructose or the polyfructan, inulin.66 However, since HMF was named by the US Department of Energy as a top target product from
1655
Catalytic One-Pot Production of Small Organics
biomass,67 the direct conversions of glucose and cellulose
into HMF have attracted great interest. The selective formation of furfural and HMF from mono- and disaccharides has recently been reviewed.2a,b,68 Hydrolysis is the
principal barrier to obtaining HMF in high yields, and efforts toward this end are united in the goal of preventing
reactions between water and HMF.
In one strategy, Mascal and Nikitin successfully trapped
and isolated HMF as its chloro derivative before it could
be hydrolyzed. Using a concentrated acid process, glucose
and even microcrystalline cellulose could be converted
into 5-chloromethylfurfural (CMF) in a single pot.69 They
used 5 wt% lithium chloride in concentrated hydrochloric
acid to dissolve the cellulose, and heated it to 65 °C for 30
hours while continuously extracting with 1,2-dichloroethane to yield 80% conversion of cellulose and a solution of
CMF in 1,2-dichloroethane (Scheme 5, Table 1 entry 9).
After evaporation of the solvent and chromatographic purification, CMF was obtained in 71% yield. The process
was also applied to lignocellulosic material; the authors
obtained 40 wt% yield furfural and 26 wt% CMF from
corn stover.70 Though this method has the advantage of
giving high yields of furan-2-aldehydes from relatively
cheap reagents, the large amounts of chlorinated solvents
and concentrated hydrochloric acid required are likely to
prevent its large-scale implementation.
HCl (concd)
LiCl (5 wt%)
H2O, 65 °C
O
Cl
Cl
extracting
HCl (concd)
LiCl (5 wt%)
H2O, 65 °C
O
Cl
cellulose
71%
O
Cl
corn stover
Cl
Cl
extracting
O
26%
O
+
O
40%
Scheme 5 Direct conversion of polysaccharides into furans using
continuous extraction
Other authors sought to increase the stability of HMF during polysaccharide hydrolysis by minimizing the water
content of the reaction solution. This meant finding a suitable non-aqueous solvent system for the reaction and, as
ionic liquids had already been used as solvents for acidcatalyzed cellulose hydrolysis (vide supra), they were reasonable candidates for HMF formation as well. In 2007,
Zhang and co-workers reported that chromium chloride
salts catalyze glucose–fructose isomerization in ionic liquids, thereby enabling the rapid conversion of glucose
into HMF in the absence of aqueous acid (cf. Scheme 3).
Thus HMF could be isolated in high yield without subsequent hydrolytic fragmentation to levulinic acid and formic acid.71 This work prompted Zhang and several other
research groups to explore the direct conversion of biomass into HMF and furfural in ionic liquids. Zhang and
co-workers dissolved cellulose in [1-EtMIM]+[Cl]–, then
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
REVIEW
Table 1
REVIEW
J. J. Verendel et al.
Conversion of Polysaccharides into Furans
Entry
Substrate
Solvent
Additive, loading
(mol%)a
Tempb
(°C)
Yieldc
(mol%)
Ref.
1
cellulose
[1-EtMIM]+[Cl]–
CrCl2/CuCl2, 6
H2O, 1000
120
480
HMF (57)
72
2
cellulose
[1-EtMIM]+[Cl]–
CrCl2/CuCl2, 6
H2O, 1000
100
240
HMF (16)
72
3
cellulose
[1-BuMIM]+[Cl]–
CrCl2, 6
H2O, 37
MW (400 W)
2
HMF (61)
75
4
xylan
[1-BuMIM]+[Cl]–
CrCl2, 6
H2O, 37
MW (400 W)
2
FUR (63)
76
5
pine sawdust
[1-BuMIM]+[Cl]–
CrCl2, 6
H2O, 37
MW (400 W)
3
HMF (52)d
FUR (31)d
76
6
cellulose
DMAe/LiCl/[1-EtMIM]+[Cl]–
HCl (concd), 6
CrCl2, 25
140
120
HMF (54)
74
7
corn stover
DMAe/LiCl/[1-EtMIM]+[Cl]–
HCl (concd), 10
CrCl2, 10
140
120
HMF (47)f
FUR (37)f
74
8
starch
[1-OctMIM]+[Cl]–
HCl (aq, 0.5 M), 40
CrCl2, 26
120
60
HMF (93)
78
9
cellulose
HCl (concd)/LiCl
–
65
1800
CMF (71)
69
Time
(min)
a
Additive loadings are given in mol% with respect to number of glucose units in polysaccharide starting material.
MW = microwave.
c
HMF = 5-hydroxymethylfurfural; FUR = furfural; CMF = 5-chloromethylfurfural.
d
Molar yields based on total hexose and pentose content in pine wood.
e
DMA = dimethylacetamide.
f
Molar yields based on total cellulose and xylan content in corn stover.
b
heated with copper(II) chloride and/or chromium(II) chloride (total metal loading was 6 mol%; nCu/nCr was varied)
and 10 equivalents of water at 120 °C for eight hours, and
obtained up to 57 mol% of a quite pure HMF (Scheme 6;
Table 1, entry 1), most of which could be extracted with
methyl isobutyl ketone (MIBK).72 The solvent and catalyst could be reused without loss of activity once the accumulated water was removed. The authors also noted
that decreasing the reaction temperature and time gave
more oligo- and monosaccharides at the expense of HMF
yield (Table 1, entries 1 and 2). Binder and Raines used a
combination of two ionic liquids, [1-EtMIM]+[Cl]– and
DMA–LiCl (DMA = dimethylacetamide, which forms an
ion pair with LiCl and has been used as a solvent for cellulose hydrolysis73), along with excess DMA, as a reaction medium to convert glucose, cellulose, and corn stover
into furans (Table 1, entries 6 and 7).74 Their system used
6–10 mol% concentrated hydrochloric acid and 10–25
mol% chromium(II) chloride as catalysts and, after two
hours at 140 °C gave 54% yield of HMF from cellulose,
and 47% HMF and 37% furfural from corn stover. Zhao
and co-workers examined the use of microwave (MW) irradiation for polysaccharide degradation in ionic liquids.
They dissolved 5 wt% cellulose in [1-nBuMIM]+[Cl]– and
added 6 mol% chromium(II) chloride and 37 mol% water,
then heated for two minutes at a microwave power of 400
W (Table 1, entry 3). The result was a 61% yield of HMF,
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
plus 16% reducing sugars as byproducts.75 These authors
later reported that the same process also gives furfural
from xylan and HMF and furfural from pine sawdust in
good yields (Table 1, entries 4 and 5).76
1)
N
N
Cl
100 °C, 1 h
cellulose
2) CuCl2 (1 mol%)
CrCl2 (5 mol%)
H2O (10 equiv)
120 °C, 8 h
O
HO
O
57 mol% yield
95% pure
Scheme 6 One-pot synthesis of HMF from cellulose in ionic liquid–
chromium(II) chloride system
A few less-recalcitrant saccharides have also been decomposed to HMF in ionic liquids. Valente and co-workers
used a similar biphasic system, MIBK/[1-nBuMIM]+[Cl]–
with chromium(II) chloride and 0.3 equivalents of water
at 100 °C for four hours to convert cellobiose (a glucose
dimer) into HMF in 74 wt% yield.77 Neither cellulose nor
starch (which is more readily hydrolyzed in acid than cellulose) could be degraded under these conditions.
Chung and co-workers decomposed starch to HMF in up
to 73% yield using aqueous hydrochloric acid and chromium(II) chloride in [1-nOctMIM]+[Cl]– (Table 1, entry 8).78
Downloaded by: the University of Oxford. Copyrighted material.
1656
Although most reports of polysaccharide hydrolysis in
ionic liquids use water, acid and chromium(II) chloride as
additives (Table 1), the necessity of adding acid (intended
to catalyze polysaccharide hydrolysis and monosaccharide dehydration) in ionic liquids has been questioned.
Qian, Chen and co-workers recently showed that only
trace amounts of water, and no acid, were required for cellulose conversion to HMF in ionic liquids at 120 °C.79 The
simple hydrolysis of cellulose required at least one equivalent of water, but when chromium(II) chloride was added, the dehydration of glucose to HMF produced water
sufficiently rapidly to fuel the hydrolysis. Additional water lowered both the cellulose hydrolysis rate and the
HMF yield. The results were attributed to the amplified
dissociation constant of water (Kw) in ionic liquids. Kw
was highest when the molar ratio of water to ionic liquid
was 0.8; at higher ratios, Kw, and hence [H+], dropped
quickly. The authors compared this reaction to hydrothermal hydrolysis, which takes advantage of the increased Kw
of pure water at temperatures highet than 250 °C (vide infra).
In addition to strategies that trap HMF as its chloro analogue or stabilize it by minimizing contact with water, researchers have also sought to circumvent the
decomposition of furans by continually removing them
from the reaction system. Dumesic and co-workers used
an acidic (pH 1) mixture of dimethyl sulfoxide and water
at 170 °C to degrade mono- and polysaccharides to HMF
and furfural, while continuously extracting with dichloromethane or a mixture of MIBK and 2-butanol. HMF
yields ranged from 29% (from glucose) to 36% (from
starch).80 Although this method provides lower selectivities and yields than the ionic liquid systems, it avoids the
use of expensive ionic liquids and toxic chromium salts.
2.3
Selective Formation of Levulinic Acid
Frost and Kurth prepared levulinic acid from cellulose in
1951.81 In contrast to HMF, levulinic and formic acids
(which are formed together from HMF; see Scheme 4) are
quite stable in hot dilute acid; levulinic acid is stable in
water at pH 1.8 up to at least 200 °C, but decomposes at
higher temperatures.58a The main decomposition products
are the a- and b-angelica lactones.82 The formation of levulinic acid and formic acid from HMF is acid-catalyzed
and they consequently are often encountered in the acidcatalyzed degradation of hexose-containing polysaccharides. Levulinic acid is industrially useful as a platform
chemical for the production of other small organics with
potential applications in organic synthesis, polymer
chemistry and fuel production.83
The selective synthesis of levulinic acid from biomass or
cellulose using aqueous Brønsted acids has been studied
by several researchers.84 The yields from biomass are typically in the range of 30 to 70%, and depend on the reactor
setup and biomass source. Increasing the reaction time or
temperature, or the solution acidity, results in more le-
Catalytic One-Pot Production of Small Organics
1657
vulinic acid formation56,65 and researchers often report
higher yields of levulinic acid under harsher conditions.
Heeres and co-workers studied the kinetics of acid-catalyzed formation of levulinic and formic acid formation
from microcrystalline cellulose in water.58b Despite levulinic acid being stable at 200 °C, it was obtained in
higher yield (>60%) at 150 °C. This is likely because the
activation energy for humin formation from glucose is
greater than that for levulinic acid, so humin formation is
favored at higher temperatures. Reactions run at lower
cellulose concentrations also yielded more levulinic acid.
This is probably also due to a decrease in humin formation, as that reaction is very concentration-dependent.
Heeres and co-workers also studied the formation of levulinic acid and other compounds from the water hyacinth
plant. The highest yield of levulinic acid (53 mol%) was
achieved by combining the substrate with aqueous 1.0 M
sulfuric acid in a batch reactor at 175 °C for 30 minutes.85
Under these conditions, significant amounts of furfural
and acetic acid were also detected; these originated from
the hemicellulose fraction. Interestingly, when the acid
concentration was lowered to 0.1 M, the dominant product
from hemicellulose was propionic acid. Degradation of
xylose to propionic acid is not seen under acidic conditions,54 leading the authors to conclude that the acid was
neutralized by ash components of the biomass. This conclusion was supported by the observations that very low
cellulose conversion was seen under these conditions and
that increasing the substrate concentration significantly
lowered the product yield.
The acid-catalyzed production of levulinic acid from biomass has recently been commercialized as the Biofine
process (Scheme 7).86 In this two-stage procedure, lignocellulose is hydrolyzed for less than 30 seconds at
~220 °C with 2–4% mineral acid, giving mainly HMF and
glucose. The mixture is then pumped into a second reactor
and held at ~200 °C for 15–30 minutes. The result is levulinic acid in >60% yield, making this an encouraging
example of how biomass can be directly converted into a
useful small organic molecule via acid-catalyzed hydrolysis.87
lignocellulose
1
H2SO4 (aq)
(CnH2nOn)x
CnH2nOn
HMF
furfural
reactor 1:
T = 220 °C
t = 30 s
furfural +
levulinic acid
2
solid lignin
residue
reactor 2:
T = 200 °C
t = 15–30 min
Scheme 7 The two-step Biofine process for production of levulinic
acid and furfural from biomass
Some alternatives to Brønsted acids have also been explored in the production of levulinic acid from cellulose.
Peng et al. explored the catalytic activity of 0.01 M aqueous solutions of several metal chlorides for the conversion
of microcrystalline cellulose into levulinic acid.88 As expected, after two hours at 180 °C, more acidic metal chloSynthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
REVIEW
REVIEW
J. J. Verendel et al.
rides such as aluminum chloride, copper(II) chloride,
iron(III) chloride and chromium(III) chloride gave the
highest conversions (20–40%) of cellulose into water-soluble compounds. Interestingly aluminum chloride and especially chromium(III) chloride gave considerably more
levulinic acid and less of the intermediates glucose and
HMF than the other metal salts did. In order to correct for
the different Lewis acidities of the metal chlorides, the
cellulose hydrolysis reaction was repeated using solutions
of aluminum chloride, copper(II) chloride, iron(III) chloride and chromium(III) chloride that had pH 3.8 (and
therefore different metal concentrations; these are not given). Under these conditions, aluminum chloride gave the
highest yield, 46 mol% of levulinic acid, whereas iron(III)
chloride gave only a few percent and the copper and chromium salts gave 37 and 32 mol%, respectively. The highest yield of levulinic acid (67%) was obtained by heating
cellulose (2 wt%) in an aqueous chromium(III) chloride
solution ([CrCl3] = 0.02 M) to 200 °C for three hours. Although the actual role of the metal ions in cellulose hydrolysis is uncertain, they may catalyze the isomerization
of glucose to fructose in aqueous solution and the authors
suggested a mechanism (Scheme 8) similar to that reported by Zhang and co-workers for the reaction in ionic liquids.71 In accordance with this proposal, the sulfate and
chloride salts of zinc, chromium and aluminum showed
significant catalytic activity in the conversion of glucose
into HMF, levulinic acid and lactic acid.89
OH
HO
OH HO
MCln⋅xH2O
O
HO
OH
Wang et al. reported the most successful solid-acid-catalyzed synthesis of levulinic acid from a polysaccharide.92
Degrading microcrystalline cellulose with sulfated titanium(IV) oxide, they obtained 27 wt% yield of levulinic
acid after 15 minutes at 240 °C. Not surprisingly, increasing the catalyst loading and sulfate density increased the
yield. They also found that levulinic acid decomposed
slightly under these conditions.
MCln⋅xH2O
OH
HO
OH
Schraufnagel and Rase investigated Amberlyst and
Dowex ion-exchange resins as solid acid catalysts for the
degradation of sucrose and fructose into HMF and levulinic acid in aqueous solution (1.65 £ pH £ 1.95), and
obtained up to 24% yield of the latter using Dowex MSC1H.91 The yields in this case may have been limited by the
relatively low temperature (100 °C) used; unfortunately,
most of the resins could not tolerate higher temperatures.
The authors noted a correlation between selectivity for levulinic acid and the resin pore size (in the range 50–380
Å). The selectivity for each compound varied with time,
but the resins with larger pores generally gave higher selectivity for HMF and lower selectivity for levulinic acid.
The authors suggested that this could be caused by the
slower diffusion of the intermediate, HMF, out of smaller
pores, as this would result in longer residence times in the
pores, and therefore increased contact with the catalyst.
The larger surface area of the resins with smaller pores
may also have produced higher conversion into levulinic
acid. Further experiments would be required to fully elucidate this effect.
OH
3
OH
glucose
MCln⋅xH2O
O
HO
– 3 H2O
O
HMF
HO
OH
O
OH
OH
fructose
OH
Scheme 8 Transition-metal-catalyzed isomerization of glucose into
fructose and subsequent dehydration to HMF
Solid acid catalysts have also been applied to levulinic
acid synthesis. DeBoef, Lucht and co-workers attempted
to synthesize glucose and levulinic acid from cellulose using acidic Nafion SAC 13 (silica-supported Nafion polymer) and iron(III) chloride on silica in water.90
Preliminary results using cellobiose as the substrate were
promising, yielding full conversion and a 1:0 ratio or 1:2
ratio of glucose to levulinic acid with Nafion or iron(III)
chloride on silica, respectively. However, the reaction
fared worse with cellulose: low conversions of cellulose
and low yields of glucose and levulinic acid (<10%) were
obtained after 24 hours at 190 °C. Neither the cellulose
type nor the solution acidity were reported, so it is unclear
whether these catalysts may have fared better under different reaction conditions.
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Trapping of Monosaccharides by Chemical
Modification
Whereas the problem of glucose decomposition during
polysaccharide hydrolysis has inspired some researchers
to focus on isolating the products of sugar decomposition,
others have attempted to selectively produce more stable
compounds via subsequent reactions. The most common
strategy of this type is reductive trapping, though alkylation has also been used.
3.1
Formation of Alkyl Glucosides
In addition to glucose, HMF and levulinic acid, a few other compounds have been isolated directly from acid-catalyzed degradation of cellulose. Zhang, Wang and coworkers isolated methyl glucosides in good yields by performing the dilute-acid-catalyzed hydrolysis of cellulose
in methanol (Scheme 9).93 They heated microcrystalline
cellulose in acidic methanol to 200 °C for 30 minutes, and
obtained 76 wt% conversion and an anomeric mixture of
methyl a- and b-glucopyranoside in 48% molar yield. A
few percent of methyl levulinate was also detected and
this product was favored with increasing acid concentration and prolonged reaction time. The heteropolyacids
tungstosilicic acid (H4SiW12O40) and tungstophosphoric
acid (H3PW12O40), which catalyze aqueous polysaccha-
Downloaded by: the University of Oxford. Copyrighted material.
1658
REVIEW
Catalytic One-Pot Production of Small Organics
OH
O
O
HO
OH
OH
HO
O
O
OH
n
cellulose
MeOH, H+
200 °C, 0.5−1 h
H2O, H+
[1-BuMIM]+[Cl]–
100 oC, 1−1.5 h
OH
HO
OH
O
HO
OH
up to 62% yield
OMe
HO
O
HO
OH
OH
H2O, H+
H+
OH
O
HO
OMe
O+
HO
OH
O
ROH, H+
O
OH
H
OMe
HO
O
HO
OH
R = hexyl or octyl
up to 46% yield
O
R
Scheme 9 The use of O-alkylation by reaction with methanol93 or
longer-chain aliphatic alcohols95 to stabilize glucose during cellulose
hydrolysis
A similar approach was taken by Villandier and Corma
who used ionic liquids to completely dissolve and hydrolyze a-cellulose, then subjected the resulting solution to in
situ Fischer glucosidation (Scheme 9).95 The optimized
process consisted of hydrolysis in [1-nBuMIM]+[Cl]– and
water for 60–90 minutes at 100 °C followed by the addition of excess of 1-hexanol or 1-octanol and reaction at
90 °C under reduced pressure for 24 hours. The reduced
pressure significantly improved the yields, probably by
removing water from the solution and thus driving the alcoholysis equilibria. Homogeneous tungstophosphoric
acid and Amberlyst 15 had similar catalytic activities in
this system, giving up to 46 mol% total yield of alkyl glu-
cosides. Alkyl glucosides are used as non-ionic surfactants in the production of detergents, pharmaceutical
preparations and personal care products.96
3.2
Formation of Polyols
Transition metals can catalyze the hydrogenation of ketones and aldehydes, including monosaccharides,97 and
several researchers have taken advantage of this reaction
to reductively trap glucose as hexitols (Scheme 10),
which are more resilient toward degradation and polymer
formation. Nevertheless, hexitols are not immune to dehydration, hydrogenolysis or other cleavage reactions under
the conditions employed; in fact, the transition-metalcatalyzed degradation of alcohols is used in the aqueous
phase reforming of monosaccharides and polyols to hydrocarbons, hydrogen and carbon dioxide.98
Glucose was traditionally hydrogenated to sorbitol at
120–160 °C and 70–140 bar hydrogen over Raneynickel.99 Low catalyst activity and nickel leaching into solution, however, prompted the search for more robust catalysts, especially for use in the food industry where
sorbitol is used as a sweetener and as a precursor to vitamin C.100 Additionally, D-gluconic acid, a common
byproduct in the metal-catalyzed reduction of glucose,
binds irreversibly to Raney-nickel and thus deactivates
it.101 The platinum group metals (Ru, Rh, Pd, Os, Ir, Pt),
generally show higher activity for the catalytic hydrogenation of monosaccharides than Raney-nickel, and they
are also more resistant to deactivation and leaching.102 Ruthenium in particular shows excellent activity for the hydrogenation of fructose and glucose, and has been used
both as a heterogeneous103 and homogeneous104 catalyst,
usually at temperatures around 100 °C. Even though
monosaccharide reductions of this type are highly selective (typically >95% selectivity), a number of byproducts
have been identified. In addition to D-gluconic acid,
which is formed by the oxidation of glucose, hexitol
isomerization products and shorter polyols have been
quantified.103a,105 At temperatures of about 200 °C, metalcatalyzed hydrogenolysis becomes more pronounced.
With the addition of catalytic amounts of hydroxide ion or
alkaline earth metal oxides, high conversions of hexitols
into short polyols, especially ethylene glycol, propane1,2-diol and glycerol can be achieved.106
Balandin and co-workers exploited the metal-catalyzed
hydrogenation of sugars to isolate polyols directly from
cellulose during the 1950s.107 They reasoned that difficulties in isolating monosaccharides in high yields from the
dilute-acid-catalyzed hydrolysis of polysaccharides could
be circumvented by reductively trapping them as polyols
in situ prior to decomposition (Scheme 10). Although Dsorbitol is the direct product of D-glucose hydrogenation,
the Lobry de Bruyn–Alberda van Ekenstein isomerization,
which is catalyzed by very dilute acid (observed even at
pH 6.69), transforms some of the D-glucose to D-fructose
and D-mannose via enediol intermediates prior to reduction.49 Balandin hydrolyzed hemicellulose and cellulose
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
ride hydrolysis94 and are easier to recover and less corrosive than Brønsted acids, were also tested in the
methanolic reaction, as were various solid acids. The heteropolyacid-catalyzed methanolysis reactions of cellulose
resulted in almost full conversions and yields of methylglucosides comparable to those obtained with sulfuric
acid under similar conditions. Among the solid acids tested in the same reaction, zeolites and sulfated zirconia
showed low activity whereas ion-exchange resins (Nafion
and Amberlyst 15) gave full conversion of cellulose but
relatively low yields of methyl glucosides, which were
further degraded to methyl levulinate when the resins
were present. Use of sulfonated carbon and sulfonated lignin as catalysts also gave high cellulose conversions, and
the latter was also quite selective, giving methyl glucosides in up to 62% yield.
1659
REVIEW
J. J. Verendel et al.
with aqueous 1–2% sulfuric acid at 155–160 °C in the
presence of ruthenium on carbon and 70 bar hydrogen.
After one (for hemicellulose) or two (for cellulose) hours,
they obtained mixtures of polyols in 78–82% total
yield.107 The same authors later showed that aqueous 0.7
wt% phosphoric acid was superior to an equimolar solution of aqueous sulfuric acid for sorbitol production from
cotton cellulose,108 as the stronger sulfuric acid significantly catalyzed the cyclodehydration of sorbitol, resulting in large amounts of sorbitan. On the other hand, up to
90% sorbitol yield could be obtained by treating cellulose
with aqueous 0.7 wt% phosphoric acid and ruthenium on
carbon for 50 minutes at 160 °C under 70 bar hydrogen.
The reaction conditions were crucial to good results. More
sorbitan and other degradation products were formed after
longer times or at higher temperatures or higher concentrations of phosphoric acid, whereas lower temperatures
resulted in slow cellulose hydrolysis. In addition, hydrogenation was slow at hydrogen pressures £60 bar; higher
hydrogen pressures did not affect the reaction outcome.
Both platinum on carbon and palladium on carbon could
also be used as hydrogenation catalysts but gave lower
sorbitol yields than did ruthenium on carbon.108
OH
cellulose
dilute acid
heat
HO
O
HO
OH
glucose
H
OH
H
O
OH
O
O
HO
H+
OH
HO
HO
H+
OH
HO
heat
OH
OH
HO
heat
OH
HO
OH
HO
D-mannose
HO
D-fructose
D-glucose
H2, metal cat.
heat
OH
OH
HO
HO
OH
OH
HO
HO
OH
OH
HO
HO
D-mannitol
D-sorbitol
minor
major
Scheme 10 The combination of acid hydrolysis and metalcatalyzed hydrogenation produces hexitols from cellulose
The same method was later extended to pine sawdust
(acetone-treated). After producing sugar alcohols in
>90% yield, the authors used the same ruthenium on carbon catalyst to isolate phenol in ~35% yield from the hydrogenolysis of the residual lignin under basic conditions
and hydrogen pressure.109 An overview of Balandin’s earSynthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
ly research on the production of polyols from polysaccharides has been authored by Sharkov.110 Robinson and coworkers published a more extensive study using an acid
and ruthenium on carbon catalyst system similar to
Balandin’s, and added a simple method for catalyst reuse
and lignin separation.111
Baudel and co-workers took a two-step hydrolysis–hydrogenation approach to obtain xylitol from a sugarcane bagasse that contained ~65% hemicellulose.112 They
suspended dry bagasse in dilute aqueous acid, heated it
from 20 to 150 °C over 50 minutes and then cooled, neutralized and concentrated the reaction mixture. The crude
pentose hydrolysate was hydrogenated under 20 bar hydrogen over ruthenium on carbon at 80 °C for three hours,
yielding high conversion into polyols (products were not
quantified).
Palkovits and co-workers recently re-investigated
Balandin’s hydrolysis–hydrogenation system in detail.113
They began by comparing the degradation of a-cellulose
in dilute aqueous acid in the presence of platinum, palladium or ruthenium on carbon at 160 °C. Platinum on carbon gave the highest conversion (up to complete cellulose
degradation) and palladium on carbon was almost as effective, but ruthenium on carbon gave significantly less
cellulose conversion (up to 74%). A control reaction without hydrogenation catalyst gave 69% conversion. Thus
the supported metal catalysts clearly affected conversion,
and the authors hypothesized that the metals catalyzed
cellulose breakdown itself, either by producing new acid
sites that catalyze hydrolysis, or by enabling the hydrogenolysis of carbon–carbon or carbon–oxygen bonds in
cellulose. It is also possible that the catalysts favor cellulose hydrolysis because they remove glucose from the
system (via hydrogenation to sorbitol), thus driving the
hydrolysis equilibrium to form more glucose; further
studies are necessary to elucidate the impact of hydrogenation catalysts on cellulose breakdown.
Despite Balandin’s early successes, the hydrolysis–hydrogenation of cellulose is by no means simple. Considerable effort has been made to identify the most common
products of the reaction (summarized in Scheme 11) and
to understand their formation. Palkovits and co-workers
quantified the solution products from cellulose hydrolysis–hydrogenation.113 They found that the product distributions were mainly determined by the catalyst, whereas
the acidity, time and reaction temperature fine-tuned the
distributions. The ruthenium-catalyzed reactions yielded
mainly sorbitol, xylitol (from the hemicellulose contained
in a-cellulose) and their dehydration products; whereas
platinum and palladium yielded mostly glucose and xylose. Thus the palladium- and platinum-catalyzed hydrogenations were clearly slow. The authors also noted that
the solution-phase products did not entirely account for
the weight loss of insoluble cellulose, suggesting that
some gaseous or volatile products were also formed; this
was more pronounced for the palladium- and platinumcatalyzed reactions than for those catalyzed by ruthenium.
Additionally, reactions in aqueous phosphoric acid that
Downloaded by: the University of Oxford. Copyrighted material.
1660
Catalytic One-Pot Production of Small Organics
were catalyzed by ruthenium produced higher fractions of
six-carbon products than those catalyzed by palladium or
platinum under otherwise identical conditions; the situation was less clear in sulfuric acid. Taken together, these
observations led the authors to conclude that palladium
and platinum on activated carbon were more active catalysts for hydrogenolytic cleavage and reforming reactions. It was unclear, however, whether sorbitol was
formed as an intermediate or whether the direct reforming
of glucose was responsible for the formation of shortchain alcohols and gaseous products. Dumesic and coworkers have shown that platinum-catalyzed reforming,
albeit under neutral conditions, degrades sorbitol to gaseous products more rapidly than it does glucose.98a In contrast to these results, the effectiveness of transition-metal
on silica catalysts for aqueous polyol reforming
(T = 210 °C, N2 atmosphere) decreases in the order Pt, Ni
> Rh, Pd > Ir, Ru.114
hydrogenolysis products
OH
HO
H2
metal cat.
O
HO
OH
glucose
OH
OH
OH OH
sorbitol
dehydration
OH
OH
OH
erythritol
H2, metal cat.
heat
hydrogenation
OH
OH
OH
MeOH
H2
metal cat.
methanol
OH
HO
OH
1,2-propane diol
H+
heat
OH
H2
OH metal cat.
HO
OH
HO
ethylene glycol
O
OH
sorbitan
dehydration
HO
HO
H+
heat
H2
metal cat.
O
O
isosorbide
OH
OH
glycerol
gaseous compounds
OH
other small organics
Scheme 11 Products of glucose degradation in the presence of acid,
hydrogen and a metal catalyst; alcohols and polyols with fewer than
six carbon atoms may be generated from glucose, sorbitol, or sorbitol
dehydration products such as sorbitan and isosorbide
94
Heteropolyacids catalyze cellulose hydrolysis, and Sels
and co-workers therefore combined these with ruthenium
on carbon and hydrogen in the hydrolysis–hydrogenation
of cellulose at 190 °C.115 For comparison, they also tested
ruthenium on carbon with sulfuric acid. The reaction using ruthenium on carbon and tungstosilicic acid gave a
combined hexitol yield (including dehydrated hexitols) of
1661
49%, whereas ruthenium on carbon and sulfuric acid gave
only 28%. The authors noted that hexitol selectivity went
through a maximum when sulfuric acid was the hydrolysis
catalyst, but was relatively constant in the presence of the
heteropolyacid. At higher cellulose conversions (>50%),
the authors suggested that sulfuric acid hydrolyzed cellulose to glucose more slowly than the heteropolyacid did
(heteropolyacids have been shown to hydrolyze cellulose
more rapidly than sulfuric acid,94a though these tests were
not conducted at equal pH). Slower cellulose hydrolysis
would cause the glucose concentration in the reaction to
remain low, thus leaving the ruthenium on carbon more
available to catalyze hexitol decomposition by hydrogenolysis. The products of hexitol decomposition were
not examined, however, so this hypothesis remains to be
confirmed in future studies. Interestingly, the tungstosilicic acid system was relatively insensitive to changes in
the cellulose-to-water ratio; the concentration of cellulose
could be increased tenfold without significantly affecting
the yield or conversion as long as the catalyst-to-cellulose
ratio was constant.
The patent literature contains reports of the direct hydrolysis–hydrogenation of starch, inulin and polysaccharide
hydrolysates to sugar alcohols by supported metals under
hydrogen without the addition of soluble acids.116 The reported catalysts include nickel, ruthenium, cobalt and other metals or metal combinations deposited on aluminas,
active carbon, zeolites or other materials. For example,
corn starch was converted into hexitols in almost quantitative yield in an aqueous suspension with ruthenium on
HUSY, a protic Y zeolite, under hydrogen at 180 °C.116b
Naturally, these processes, which avoid hazardous acidic
waste and enable simple catalyst reuse, are attractive from
both economic and environmental viewpoints.
In 2006, Fukuoka and Dhepe reported that supported metals could also degrade microcrystalline cellulose.117 They
heated a dilute suspension of cellulose and a solid-supported metal to 190 °C for 24 hours under hydrogen. Platinum on g-alumina or platinum or ruthenium on Y zeolites
gave the highest yields, up to 31%, of hexitols (sorbitol
and mannitol). Both the metal and support material
strongly influenced the product yields, as other transition
metals (Pd, Ir and Ni) and other support materials (ZrO2,
activated carbon and mesoporous silica) gave significantly lower hexitol yields. Fukuoka and Dhepe proposed that
the metal–support match leads hydrogen to spill over from
the metal onto the support material, forming protic sites
on the support surface and consequently an increased pool
of H+ in the system.118
Essayem and co-workers performed a series of control reactions to clarify the functional details of the platinum on
g-alumina catalyst system for cellulose degradation.119
Upon simply stirring an aqueous suspension of microcrystalline cellulose at 150 °C under hydrogen for 24 hours,
they found that 2 wt% of the initial cellulose dissolved; at
190 °C, the conversion was 45%. Conversion is defined in
this context as the weight loss of cellulose over the course
of the reaction and, due to gas formation, is usually ~10%
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
REVIEW
REVIEW
J. J. Verendel et al.
higher than the solubilization (which is based on total organic carbon in the aqueous solution). The conversion
could be increased upon continued heating, with 83%
conversion and 60% dissolution being obtained after 100
hours (Table 2, cf. entries 1 and 2). The degradation of
cellulose at high temperatures and in the absence of catalyst is likely both acid- and base-catalyzed, as the increased ionization constant of water under these
conditions results in higher concentrations of H+ and HO–
(pH = pOH = 5.82 at 150 °C and 5.64 at 200 °C).120 More
details of this kind of polysaccharide degradation are given in section 5.
Comparing the degradation of cellulose in the absence of
catalyst or with g-alumina alone, Essayem and co-workers
noted that a slightly better conversion was achieved when
g-alumina was present (Table 2, entry 3). With platinum
present (entry 4), the conversion of cellulose to solubles
and gases increased to 71% after 24 hours. The amount of
cellulose converted to solubles reached a plateau of 70%
after only 50 hours, as compared to 60% after 100 hours
for the blank reaction. Hence, the metal and support do
improve the rate of cellulose dissolution and may in fact
act as a bifunctional hydrolysis and hydrogenation catalyst. When the reaction was performed in the presence of
platinum on g-alumina but under helium rather than hydrogen, the cellulose conversion was unaffected but a
lower carbon balance (65%, cf. 77% under H2) was observed, providing evidence for the significant formation
of gaseous products.
Table 2
Degradation of Cellulose in the Presence of Hydrogen
P (H2)a
(bar)
Additive
Conv.b
(%)
Ref.
24
50
none
45
119
190
100
50
none
83
119
3
190
24
50
g-Al2O3
57
119
4
190
24
50
Pt/g-Al2O3
71
119
5
230
4
40
none
48
131
6
245
0.5
60
none or ACc
87
128
7
245
0.5
60
Ru/ACc
85
128
Entry
Temp
(°C)
1
190
2
Time
(h)
a
Pressure at room temperature.
Conv. = conversion, loss of cellulose mass.
c
AC = activated carbon.
b
Additionally, the authors re-used cellulose that had been
subjected to the reaction conditions once, and obtained
conversions and yields similar to those from fresh cellulose. The authors speculated that the reaction solution became saturated with soluble oligosaccharides during the
reaction, thereby limiting conversion. Consequently, they
performed the reaction under more dilute conditions but
achieved the same conversion. When a more concentrated
suspension was used, the conversion dropped to 45% but
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
the actual amount of dissolved material was higher, ruling
out the possibility of a pure saturation effect.
Zhang, Chen and co-workers further extended the use of
the platinum on g-alumina catalyst by examining the formation of other polyols during the reaction (see
Scheme 11).121 Under the conditions used by Fukuoka and
Dhepe,117 they found the same combined sorbitol and
mannitol yield, and also noted a 9% yield of ethylene glycol (EG), as well as small amounts of erythritol and 1,2propylene glycol (Table 3, entry 1). Increasing the temperature to 250 °C and decreasing the reaction time to 30
minutes resulted in nearly full conversion of cellulose and
produced more ethylene and propylene glycols (14% and
11% yield respectively) at the expense of the hexitols (entry 2). Changing the catalyst to tungsten carbide on activated carbon (AC) increased the selectivity for ethylene
glycol over hexitols, and adding nickel as well gave a dramatic increase in ethylene glycol yield (entries 4 and 5).
The maximum yield of ethylene glycol, 53 mol% (61
wt%), was obtained using a catalyst, which consisted of
2% nickel and 30% tungsten carbide on activated carbon,
to decompose cellulose at 250 °C under 60 bar hydrogen
for 30 minutes (entry 5). In that reaction, propylene glycol
(included in ROH) was also isolated in 6% yield; only a
few percent hexitols were observed. A subsequent study
demonstrated that the yield of ethylene glycol from cellulose decomposition over tungsten carbide on activated
carbon was relatively independent of temperature in the
range of 225 to 250 °C but decreased at higher temperatures, probably due to decomposition.122 The support type
was very important for the yield of ethylene glycol and the
catalyst nickel–tungsten carbide on activated carbon was
superior to a mixture of nickel on activated carbon and
tungsten carbide on activated carbon combined (entries 4–
6), suggesting a synergistic effect.
Another study tested a variety of metals (Ni, Pd, Pt, Ru, Ir,
W) on activated carbon as catalysts for cellulose degradation under the conditions used for tungsten carbide on activated carbon and found that tungsten was more efficient
in degrading cellulose than any other of the metals tested.123 The choice of metal significantly impacted the specific polyols formed, though all monometallic metal-onactivated-carbon catalysts gave low yields and selectivities. However, catalysts that combined these metals with
tungsten (as M-W/AC) gave yields of ethylene glycol that
were similar to that obtained with tungsten carbide on activated carbon (61 wt%). Hence both the conversion of
cellulose and selectivity for ethylene glycol significantly
increased when bimetallic catalysts were used. This study
also showed that a combination of nickel on activated carbon and tungsten on activated carbon produced more ethylene glycol than either catalyst alone (Table 3, entries 6–
8). In a final study, tungsten carbides were supported on
mesoporous carbon (MC). In this case WCx/MC and 2%
Ni–WCx/MC were equally effective in producing ethylene glycol from cellulose, giving up to 64% yield (74
wt%).124 Moreover, the former catalyst could be reused
several times with only minor reductions in activity. The
Downloaded by: the University of Oxford. Copyrighted material.
1662
REVIEW
Entry
Catalytic Conversion of Microcrystalline Cellulose into Polyols in Water
Temp
P (H2)a Time
(°C)
(bar)
(h)
1
190
60
24
2
250
60
3
250
4
Catalyst
[Metal]b
(mol%)
Conv.c
(wt%)
Yieldd (wt%)
Ref.
Sorbitol Mannitol EG
Other ROH
e
2.5 wt% Pt/g-Al2O3
0.62
46
26
8
9
2
121
0.5
2.5 wt% Pt/g-Al2O3
0.62
98
10
6
14
11
121
60
0.5
2.5 wt% Pt/AC
0.62
66
3
1
8
7
121
250
60
0.5
30% W2C/AC
7.7
97
<1
<1
25
5
121
5
250
60
0.5
[30% W2C + 2 wt% Ni]/AC
Ni: 1.7
W: 7.7
100
3
1
61
9
121
6
245
60
0.5
20 wt% Ni/AC
17
72
10
2
9
4
123
7
245
60
0.5
30 wt% W/AC
7.9
100
0
0
2
0
123
8
245
60
0.5
20 wt% Ni/AC + 30 wt% W/AC Ni: 8.3
W: 4.0
100
3
6
47
11
123
9
245
60
0.5
4 wt% Ru/AC
0.65
86
33
11
6
42
128
10
245
60
0.08
4 wt% Ru/AC
0.65
38
20
5
2
15
128
11
210
60
24
3 wt% Ni/CNFf
4.1
87g
30
5
5
16
131
12
230
60
4
3 wt% Ni/CNFf
4.1
94g
23
5
7
16
131
13
230
40
4
3 wt% Ni/CNFf
4.1
79g
18
4
7
15
131
14
230
40
4
3 wt% Ni/AC
4.1
80g
10
3
9
12
131
15
230
40
4
3 wt% Ni/g-Al2O3
4.1
78g
5
1
12
3
131
a
Pressure of H2 added at room temperature.
100 × nmetal/ncellulose; ncellulose = moles of anhydroglucose repeat units in cellulose.
c
Conv. = conversion, loss of cellulose mass over the course of the reaction.
d
Yield determined by analysis of the crude aqueous solution, see individual publications for details.
e
As various authors have measured different alcohols, these values are only for interest and should not be compared between publications.
f
CNF = carbon nanofibers.
g
Conversion based on total organic carbon in the aqueous phase.
b
loss of activity was attributed to partial oxidation of the
WCx phase, and the lost catalytic activity could be partially recovered by treating the catalyst with hydrogen at
550 °C.
When analyzing and discussing product spectra from
polysaccharides over supported metals, it is important to
consider more than just alcohol formation. Polyols are, in
fact, not very stable over transition metals at high temperatures. In an aqueous phase reforming experiment using
hydrogen and platinum on silica–alumina at 225 °C,
Dumesic and co-workers converted ≥80 mol% of the carbon in a 5 wt% aqueous solution of sorbitol into gaseous
C1–C6 alkanes.125 Completely reduced hydrocarbons can
be formed from sugars or polyols by several routes, including successive dehydration–hydrogenations, carbon–
oxygen and carbon–carbon cleavages and methanation;
they can also be formed via Fischer–Tropsch reactions
from carbon dioxide, which is also formed during aqueous
phase reforming.11d,126 These reactions, in addition to the
various rearrangements and cyclizations that can be catalyzed by metal and/or acid, pose serious selectivity challenges, regardless of whether the desired products are
polyols, alkanes or hydrogen. Overall, reforming and
cleavage reactions are favored at higher temperatures
while the highest hexitol selectivities are obtained at temperatures greater than 200 °C or when using relatively
short reaction times. The most prominent degradation reactions of aqueous sorbitol over transition metals are
shown in Scheme 12.
While Fukuoka and Dhepe investigated cellulose degradation,117 Liu, Kou and co-workers studied the effect of
transition-metal nanoclusters on the cellulose repeat unit
(glucose dimer) cellobiose.127 They heated aqueous solutions containing soluble ruthenium nanoclusters and cellobiose to 120 °C for 12 hours under hydrogen. Under
acidic conditions (pH 2), the sequence of acid-catalyzed
hydrolysis and metal-catalyzed reduction yielded sorbitol
quantitatively, whereas very little cleavage of cellobiose
occurred at neutral pH. Under neutral or basic (pH 10)
conditions, the main product was 4-O-b-D-glucopyranosyl-D-glucitol (the product of cellobiose monoreduction,
Scheme 13).
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
Table 3
1663
Catalytic One-Pot Production of Small Organics
1664
REVIEW
J. J. Verendel et al.
C1–C6 alkanes
H+, H2
metal
Table 4
lystsa
dehydration
hydrogenation
CO, CC cleavage
Entry
Degradation of Cellulose by Supported Ruthenium CataCrystallinityb (%) Catalyst
Yield (mol%)
Hexitols Erythritol Glycerol
cleavages
rearrangements
dehydrations
HO
aldehydes dehydrogenations
acids
dehydration H+/HO–, metal HO
products
OH
OH C–C cleavage
HO–, metal
shorter polyols
OH
1
33
Ru/CNTc
73
5
5
2
33
Ru/Al2O3
25
8
6
3
33
Ru/MgO
0
13
3
4
33
Ru/SiO2
8
0
0
5
33
Ru/CeO2
7
6
7
6
85
Ru/CNTc
13
6
–d
7e
85
Ru/CNTc
40
6
–d
HO
metal reforming
methanation
Fischer–Tropsch
CO2, H2
CH4, H2O
Scheme 12 Sorbitol degradation pathways available at high temperatures and in the presence of a metal catalyst
Liu et al. went on to study the hydrothermal degradation
of cellulose (T = 245 °C) with reductive trapping of glucose by ruthenium-on-carbon-catalyzed hydrogenation.128
In order to assess the effect of the catalyst on cellulose
conversion, blank reactions were compared with reactions
containing various amounts of ruthenium on carbon. The
conversion was, in all cases, 38% and 86% after 5 and 30
minutes, respectively, indicating that the solid catalyst did
not affect the cellulose dissolution (Table 2, entry 6;
Table 3 entries 9 and 10). The product distribution in this
high-temperature system was very time- and temperaturedependent. The selectivity for hexitols, which peaked at
39% yield, decreased at temperatures greater than 245 °C
in favor of shorter-chain alcohols (xylitol, erythritol, propylene glycol, glycerol, ethylene glycol and methanol).
Sorbitan, a cyclic dehydration product of sorbitol
(Scheme 11) was also formed in ~15% yield. The catalyst
loading did not affect product yield or distribution in the
range of 2 to 8 wt%, but the yields of all products decreased dramatically when 1 wt% catalyst was used.
Wang and co-workers studied the conversion of cellulose
samples with varying crystallinity into polyols over supported ruthenium catalysts (Table 4).129 Ruthenium supported on carbon nanotubes was the most effective
catalyst, giving a 73% yield of hexitols and an additional
10% erythritol and glycerol at the optimal catalyst loading
(entry 1); ruthenium on alumina gave 25% yield of hexitols, 8% erythritol and 6% glycerol (entry 2). No information regarding cellulose conversion was included.
Interestingly, ruthenium on magnesia gave no hexitols but
did produce erythritol and glycerol (entry 3); whereas ruthenium on silica produced hexitols but neither erythritol
nor glycerol (entry 4). Ruthenium on ceria produced nearly equal amounts of the three compounds (entry 5). Ethylene glycol formation was not discussed. The origin of the
superior activity observed for the ruthenium-on-carbonSynthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
a
Reaction conditions: 0.16 g cellulose and 0.05 g catalyst (1 wt% Ru)
in 20 mL H2O, 50 bar H2, 185 °C, 24 h.129
b
Determined by X-ray diffraction.
c
CNT = carbon nanotubes.
d
No data given.
e
0.14 g catalyst was used.
nanotube catalyst is not known, though the authors noted
that this catalyst was more acidic and capable of more hydrogen adsorption than the others.
Sorbitol was not obtained when the catalyst was iron, cobalt, nickel, palladium, silver or gold supported on carbon
nanotubes; the corresponding supported platinum, rhodium and iridium catalysts each gave about 30% as much
sorbitol as ruthenium. The ruthenium-on-carbon-nanotube catalyst could also degrade commercial microcrystalline cellulose (crystallinity = 85%) though more
catalyst was required to produce good hexitol yield
(Table 4, entries 6 and 7).
In an attempt to better understand the role of ruthenium on
carbon nanotubes in the degradation of cellulose, Wang,
Zhang and co-workers studied the conversion of cellobiose to polyols by this catalyst.130 Ruthenium-on-carbonnanotube catalysts with more acid sites gave significantly
higher sorbitol yields. The catalyst with the most acid sites
gave almost 90% yield of sorbitol from cellobiose after
treatment for three hours at 185 °C, whereas carbon nanotubes without acid sites gave only 25% yield. Neither cellobiose conversion nor yields of other products were
given. The effect of temperature on product distribution
was examined using the most acidic carbon nanotube support. Below 160 °C, 4-b-O-glucopyranosyl-D-glucitol
was the dominant product, but sorbitol selectively increased at higher temperatures, reaching 85% at 185 °C.
Above 185 °C, the decomposition of sorbitol to shorterchain alcohols became significant. Carbon nanotubes with
various ruthenium cluster sizes were also compared. Ammonia desorption measurements showed that larger ruthenium clusters gave catalysts with higher surface acidity,
and these same catalysts also produced higher conversion
of cellobiose and selectivity for sorbitol. A kinetic study
demonstrated that reduced cellobiose (4-O-b-D-glucopyr-
Downloaded by: the University of Oxford. Copyrighted material.
sorbitol
REVIEW
Catalytic One-Pot Production of Small Organics
OH
O
HO
HO
OH
OH
HO
O
OH
O
OH
D-cellobiose
OH
O
HO
HO
OH
OH
HO
O
O
OH
H2
[Ru]
OH
O
HO
HO
OH
OH
OH
HO
O
OH
OH
OH
4-O-β-D-glucopyranosyl-D-glucitol
H+, H2O
pH 2
pH 7–10
OH
HO
HO
HO
HO
no further reaction
O
OH
D-glucose
+
OH
OH
OH
OH
OH
D-sorbitol
H2
[Ru]
HO
2 HO
OH
OH
OH
OH
2 D-sorbitol
Scheme 13 Reactions of the cellulose repeat unit, cellobiose, at different pH values in the presence of metal catalyst and hydrogen
Sels and co-workers recently reported on polyol production from microcrystalline cellulose using carbon nanofibers (CNF) on nickel on g-alumina.131 The carbon
nanofiber/nickel/g-alumina catalyst was compared with
nickel on g-alumina and nickel on activated carbon and
was more selective towards hexitols, giving up to 30
mol% yield (35 wt%) when heating cellulose in water to
210 °C for 24 hours (Table 3, entry 11). An interesting
pressure effect was observed: increasing the hydrogen
pressure from 20 to 60 bar resulted in a significant increase of hexitol selectivity and almost doubled the yield
of sorbitol and mannitol, despite having essentially no impact on cellulose conversion. Erythritol (12%), glycerol
(4%), propane-1,2-diol (9%) and ethylene glycol (12%)
were also formed under optimized conditions.
4
Alkaline Hydrolysis (T < 250 °C)
The alkaline hydrolysis of cellulose produces C1–C6 carboxylic acids. The production of these compounds via the
alkaline degradation of wood products has a long history;
for a time, oxalic acid was produced commercially from
the fusion of sawdust with solid hydroxides.132 However,
regardless of whether a solid or aqueous base is used, the
reaction is quite complicated and produces a variety of
products whose distribution depends strongly on the conditions. Thus, the aqueous alkaline hydrolysis of cellulose
(or other polysaccharides) has not yet produced carboxylic acids in high yields and selectivities. Nevertheless, the
reaction has several encouraging characteristics. Cellulose swells in concentrated aqueous base ([HO–] > 2 M),
making it more accessible for hydrolysis,133 and alkaline
cellulose hydrolysis has half the activation energy of acidic cellulose degradation.54 The challenge, however, lies in
producing a single (or strongly predominant) compound
from the reaction. Although some work has aimed to manipulate the reaction to improve its selectivity, most studies have sought to identify the products of alkaline
polysaccharide decomposition and to understand their
formation. Some reviews of the topic are available; in particular, Nevell gives a detailed account of the mechanisms
involved,133 and Molton and Demmitt134 and Knill and
Kennedy135 provide extensive lists of compounds that
have been obtained. Alkaline cellulose degradation at low
temperature has been studied in the context of nuclear
waste storage, and many minor products of the reaction
have been identified in these studies. The long reaction
timescales (years) involved in these studies exclude their
use for synthesis, so they are not discussed in detail here,
but an overview of this topic is available.136 Here, we focus on the major reaction products and factors affecting
the reaction yield and selectivity.
Below 170 °C and under an inert atmosphere, the aqueous
alkaline hydrolysis of cellulose is dominated by ‘peeling’,
the successive removal of monomer units from the end of
the carbohydrate polymer. This reaction produces one
equivalent of a saccharinic acid, most often D-glucoisosaccharinic acid (4H, Scheme 14) for each glucose monomer that is depolymerized, and occurs even at ambient
temperature, albeit slowly.137 As the saccharinic acid mol-
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
anosyl-D-glucitol) is indeed the initial intermediate in all
cases (Scheme 13), but, when smaller ruthenium clusters
are used, less sorbitol and more mannitol and shortercarbon-chain products are formed. Carbon nanotubes with
smaller ruthenium clusters also converted cellobiose into
4-O-b-glucopyranosyl-D-glucitol much faster than carbon
nanotubes with larger ruthenium clusters; this effect was
attributed to a larger active surface area available for hydrogenation. In contrast, conversion of 4-O-b-glucopyranosyl-D-glucitol was slower over smaller ruthenium
particles, which the authors attributed to their lower acidity. Finally, the degradation of sorbitol to mannitol and
C2–C4 polyols under the reaction conditions was quite
modest using larger ruthenium particles but gave up to
45% sorbitol conversion when using the smallest nanoparticles.
1665
1666
REVIEW
J. J. Verendel et al.
OH
OH
O
H
O
O
iii
HO
O
OH
+
HO
OR1
OR2
HO
HO
ii
O
HO
CH2OH
CH2OH
2
3
HO–
iv
HO–
CH2OH
cellulose
R2 = cellulose polymer chain
(anhydroglucose)n–1
O
H
OH
X–
OR1
i
HO
HO
R3
OH
R
HO
1
HO
OR
CH2OH
HO
CH2OH
CH2OH
1
cellulose
R1 = cellulose polymer chain
(anhydroglucose)n
CH2OH
4
α-4– R3 = OH, R4 = CO2–
β-4– R3 = CO2–, R4 = OH
v HO–
O
O
H
O
O
OH
HO–
OR1
HO
HO
CH2OH
5
Scheme 14
CH2OH
6-
The primary mechanisms of cellulose peeling (reactions i–iv) and chain termination (reactions v, vi) in alkaline solution
ecule can undergo subsequent degradation reactions, the
yield and selectivity of an organic product under these
conditions are determined both by cellulose conversion
(the number of monomers that are peeled from cellulose
to give water-soluble small molecules) and the selectivity
of peeling and subsequent reactions.
Between 170 and 250 °C, the situation is similar, except
that b-(1→4)-glycosidic bonds, which link the anhydroglucose monomers in cellulose, are randomly cleaved as
well. This reaction, caused by hydroxide ion attack at b(1→4)-glycosidic bonds,138 results in a rapid decrease in
the degree of cellulose polymerization,139 and increases
the number of end groups present, thus allowing new
chains to begin peeling. Conversions are therefore higher
for alkaline cellulose hydrolyses carried out above
170 °C.
4.1
OR1
vi
Cellulose Conversion
Cellulose peeling occurs from the reducing end of the cellulose chain (i.e., the aldehyde carbon), and therefore
mechanistic studies on alkaline cellulose hydrolysis are
often carried out using hydrocellulose, which has been
pretreated with acid to give a highly crystalline product in
which most chains terminate with the crucial C1 aldehyde. When the aldehyde groups of hydrocellulose chains
are reduced with sodium borohydride140 or converted into
thiols,141 they are stabilized toward aqueous base, evinc-
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
ing the role of the aldehyde end group in alkaline hydrolysis. In the first step of the reaction, the reducing C1 end
of a cellulose polymer is deprotonated by a Brønsted base
(most often hydroxide, HO–) to form an acyclic enediol
anion (1, i, Scheme 14), the same type of ion that is
formed in the base-catalyzed Lobry de Bruyn–Alberda
van Ekenstein isomerization of glucose to fructose and
mannose.142 In the major peeling pathway (ii), this anion
rearranges to form cellulose with a fructose end group (2).
This rearrangement places the carbonyl group b to the glycosidic bond (i.e., to the cellulose chain). b-Alkoxy elimination from this intermediate (iii) is facile, and produces
the 2,3-diketone 3 [(S)-1,5,6-trihydroxyhexane-2,3-dione,
where the change from an R to an S configuration at C5 reflects a change in substituent priority, not in absolute configuration] as well as a cellulose chain that is shorter by
one monomer unit. The latter can undergo the same reaction again, continuing the peeling process. The released
fragment 3 reacts with hydroxide ion and then undergoes
the benzil–benzilic acid rearrangement143 to produce one
of the two diastereomers of carboxylate 4– (iv), which produces a D-glucoisosaccharinic acid (a- or b-4H) upon protonation. Thus although hydroxide anions catalyze
cellulose hydrolysis, they are also consumed via reaction
with the hydrolysis product 3, causing the reaction pH to
drop over the course of the reaction. Thus a sufficiently
high pH is necessary to maintain the alkaline hydrolysis
reaction. For example, the hydrolysis of the glucose
dimer, cellobiose, in aqueous sodium hydroxide
Downloaded by: the University of Oxford. Copyrighted material.
H
Catalytic One-Pot Production of Small Organics
([NaOH]0 = 0.01 M) at 85 °C comes to a halt when the pH
reaches approximately 9.2.144
Several side reactions preclude the quantitative peeling of
cellulose in alkaline solution via the mechanism shown in
Scheme 14, steps i–iv. In the most important of these, intermediate 1 eliminates a b-hydroxy group to produce the
1,2-dicarbonyl compound 5 (v), which reacts with hydroxide ion to form 6– (vi), a cellulose chain with an alkalistable a-hydroxycarboxylate end group (i.e., a deprotonated metasaccharinic acid group),145 thus terminating the
peeling reaction. Though this is the major termination reaction, more than a dozen other alkali-stable end groups
have been identified,146 indicating that a wide variety of
termination reactions are possible. Termination also occurs when the cellulose chain is peeled back to a (nonalkaline-stable) reducing end group that is inaccessible to
hydroxide ion; this is considered a physical termination,147 and can be a major cause of termination. Once the
alkaline hydrolysis of a cellulose sample ceases, leaving
an alkali-stable cellulose, the alkali-stable end groups can
be removed via acid hydrolysis to produce a material that
can be degraded with alkali again.148 As the alkali degradation of cellulose at temperatures greater than 170 °C occurs only from the chain end and preferentially degrades
the shorter, more soluble chains, the undegraded material
generally has a degree of polymerization that is similar to
or higher than that of the starting material.133 Similarly,
the residual product is more crystalline than the starting
material;31 this is true even when hydrolysis is conducted
Table 5
1667
at temperatures above 170 °C. The relative rates of cellulose peeling (Rpeel) and chemical termination (Rterm) are
important to cellulose conversion, especially for temperatures greater than 170 °C. The factors that affect Rpeel/Rterm
are summarized in Table 5. The major termination reaction (steps v–vi, Scheme 14) has an activation energy of
134 kcal/mol, higher than that of peeling (100 kcal/
mol),147 and can thus be mitigated using mild reaction
temperatures (Table 5, entry 1). At 95 °C and [HO–]0 ≥
1.25 M, for example, the termination reaction is estimated
to be about 100 times slower than peeling,149 though the
ratio can span more than an order of magnitude.147 Nevertheless, termination competes with peeling even at low
temperature.137b Notably, although Rpeel/Rterm is lower at
higher temperature, the weight of cellulose degraded (i.e.,
the reaction conversion) remains relatively constant over
the temperature range 65–132 °C.147 This highlights the
importance of physical chain termination in alkali degradation, and suggests that it is more prevalent at lower temperature. Alkalinity also affects the relative rates of
peeling and termination (entry 2). In the case of cellulose
degradation in aqueous sodium hydroxide at low [HO–]0,
Rpeel/Rterm is approximately constant,149 but conversion increases with [HO–]0,149,150 implicating physical termination as a conversion-limiting factor at low [HO–]0.
Interestingly, when Ba2+ or Sr2+ is used as the counterion,
the conversion decreases with increasing alkalinity in the
range 0.1 M £ [HO–]0 £ 0.4 M.150 In more basic solution,
cellulose conversion reaches a maximum at [NaOH]0 = 6
Effects of Reaction Conditions on the Cellulose Conversion and Rpeel/Rterm During Alkaline Hydrolysis
Entry
Parameter Conditions
Effect on conversiona
Effect on Rpeel/Rtermb
Ref.
1
T
65–132 °C
little effect
↑ T → ↓ Rpeel/Rterm due to greater
termination rate
147
170 °C £ T £ 190 °C
↑ T → ↑ conversion
–c
158
[NaOH]0 < 1 M
↑ [HO–]0 → ↑ conversion
Rpeel/Rterm independent of [HO–]0
149
1 M £ [NaOH]0 £ 6 M
↑ [HO ]0 → ↑ conversion
–
6 M < [NaOH]0
↑ [HO–]0 → ↓ conversion
Rpeel/Rterm small at very high [HO–]0
2
[HO–]
–
4
base
140
140
0.01 M £ [Ba(OH)2]0 £ 0.02 M ↑ [HO ]0 → ↑ conversion
–
c
151
0.1 M £ [E(OH)2]0 £ 0.4 M;
E = Sr2+, Ba2+
↑ [HO–]0 → ↓ conversion
–c
150
HO–, HCO3–, CO32–
similar conversions obtained
using HO– and HCO3–
similar Rpeel/Rterm values
146, 147,
149, 150
base = B4O72– or HPO42–
very low conversion
–c
150
similar Rpeel/Rterm values for alkali
and alkaline earth metals
151
Li+, Na+, K+ > Ca2+, Sr2+, Ba2+
([Ca(OH)2] was not examined at
the same concentrations due to
lower solubility)
150, 151
–
3
c
–
counterion [HO ]0 = 0.02 M
[HO–]0 = 0.1 M or 0.25 M
2+
2+
2+
+
+
Ca , Sr , Ba > Li , Na , K
+
Li+, Na+, K+ > Ca2+, Sr2+, Ba2+
a
Conversion refers to the amount of cellulose solubilized, i.e., the loss of mass of insoluble material.
Rpeel = rate of cellulose peeling; Rterm = rate of chemical termination via the formation of acidic end groups.
c
Data not available.
b
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
REVIEW
REVIEW
J. J. Verendel et al.
M, the concentration at which cellulose swelling is maximized.140 When [NaOH]0 > 6 M, cellulose conversion decreases; the reaction using [NaOH]0 = 18.6 M gave much
less cellulose conversion than that using [NaOH]0 = 6 M
under otherwise identical conditions. This lower conversion was accompanied by an increase in chemical termination (the formation of metasaccharinic acid end groups,
etc.), indicating that these reactions are favored in very
strong alkali.
Conversion is relatively unaffected by changing the basic
anion from hydroxide to bicarbonate or carbonate, but little conversion is observed when the base is tetraborate or
hydrogen phosphate (entry 3).146,150 The counterion, on the
other hand, is quite significant (entry 4). The hydroxides
of alkaline earth metal ions such as Ca2+, Sr2+, and Ba2+
give different ratios of peeling to termination than did
those of alkali metals (Li+, Na+, and K+), though the effect
depends upon alkalinity.151 At [HO–]0 = 0.02 M, greater
cellulose conversion is obtained with alkaline earth metals; the opposite is observed at higher [HO–]0.150,151
4.2
Product Selectivity
Scheme 14 depicts only the most important peeling and
termination reactions of cellulose; as already mentioned,
many minor termination reactions have been observed.
The most important barrier to selectivity in alkaline cellulose degradation, however, is the subsequent decomposition of the saccharinic acids (most often D-glucoisosaccharinic acid), formed from peeling, to give other
acids. The number of acid equivalents produced per unit
glucose lost from a cellulose chain, which would be 1 if
there was no subsequent degradation from the saccharinic
acids, has been measured to be 1.4–1.86 in aqueous sodium hydroxide,138,148,150,152 so further degradation is clearly
common. A great many compounds are produced in varying amounts from the alkaline degradation of cellulose,133–135 meaning that control of selectivity in this type
of reaction is likely to be decisive for its synthetic utility.
In terms of controlling selectivity, the field of alkaline cellulose degradation has not yet progressed as far as that of
acid-catalyzed cellulose hydrolysis, but the impacts of
multiple factors on product selectivity have been examined. Some of the most prominent water-soluble products
of alkaline cellulose hydrolysis (apart from D-glucoisosaccharinic acid) are shown in Figure 4.
sis,153 and has been isolated in 21% molar yield from this
reaction.154 Its formation is favored in the presence of
Ca2+; the reaction in lime water [aqueous Ca(OH)2] produced a much higher selectivity for D-glucoisosaccharinic
acid than did sodium hydroxide.155 This can be understood
by considering intermediate 3 in Scheme 14, which forms
D-glucoisosaccharinic acid (4H) via the benzilic acid rearrangement (step iv in Scheme 14), but can also fragment
to yield formic acid and 3-deoxypentulose (which is not
isolated, but rather converted in situ into 2,5-dihydroxypentanoic acid). Ca2+ ions catalyze the benzilic acid
rearrangement,143 favoring the conversion into D-glucoisosaccharinic acid rather than fragmentation. D-Glucoisosaccharinic acid can be disfavored by hydrolyzing
cellulose using bicarbonate, rather than hydroxide, as the
base. In that case, the fragmentation of intermediate 3 is
apparently favored over its rearrangement to D-glucoisosaccharinic acid, and considerable amounts of formic acid
and 2,5-dihydroxypentanoic acid are produced.146
Most quantitative measurements of the products of alkaline cellulose degradation have been carried out at high
temperature (T ≥ 170 °C), presumably because higher
conversions are available under these conditions. D-Glucoisosaccharinic acid is still obtained, but smaller degradation products make up a greater fraction of the soluble
products. The yields and selectivities of some alkaline cellulose degradations are given in Table 6. Only the most
abundant products from alkaline cellulose degradation are
given. For comparison, results are also given for the products of alkaline starch degradation. Though the alkaline
hydrolysis of starch, in which glucose units are linked by
a-(1→4)-glycosidic bonds, is expected to be slower than
that of the b-(1→4)-linked cellulose,156 it nevertheless
produces many of the same products, albeit with different
distributions.157
Figure 4 Abundant water-soluble products of alkaline cellulose hydrolysis at temperatures at or below 250 °C
Niemelä and Sjöström examined the yield and selectivity
of alkaline cellulose degradation while varying the temperature (170 to 190 °C) and initial concentration of sodium hydroxide (1 or 3 M).158 The main product by weight
was b-glucoisosaccharinic acid (b-4H) in all cases,
though it was produced slightly less selectively when the
initial concentration of sodium hydroxide was raised from
1 M to 3 M (Table 6, entries 1–3 vs. 4–6). The highest
yield of b-4H was obtained at 190 °C and 3 M sodium hydroxide concentration, when 16.8 wt% (15.1 mol%) was
obtained. Selectivities for the other major products were
independent of these two variables in the ranges examined. The yields of almost all products increased with temperature. Niemelä observed similar trends in selectivity
for the alkaline degradation of starch (Table 6, entries 9–
12).157 Notably, the degradation of starch was significantly more selective for b-glucoisosaccharinic acid (b-4H)
formation than was the degradation of cellulose.
At relatively low temperature (100 °C) and alkali concentration ([NaOH]0 = 0.5 M), D-glucoisosaccharinic acid is
the predominant soluble product of cellulose hydroly-
The alkaline hydrolysis of cellulose at 220 °C and
[NaOH]0 = 0.5 M also produced 4H as the major product,
though the a- and b-forms were not quantified separately
in this case. Lactic acid was also a major product.138
O
H
O
OH
formic acid
HO
O
OH
O
oxalic acid
Synthesis 2011, No. 11, 1649–1677
HO
O
OH
glycolic acid
HO
OH
lactic acid
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
1668
REVIEW
Entry
1669
Most Abundant Products from Cellulose and Starch Degradation under Various Alkaline Conditions
Substrate
Temp
Base,
a
Time
Yield of acid by weight (selectivity) (%,%)b
Formic
(°C)
concn (M)
(h)
Glycolic
Ref.
Oxalic
Lactic
b-4H
a-4H
1
cellulose
170
NaOH, 1
4
7.0 (16)
1.9 (4)
–c
5.7 (13)
10.8 (25)
3.9 (9)
158
2
cellulose
180
NaOH, 1
3
7.6 (14)
1.6 (3)
–c
7.3 (13)
12.6 (23)
4.9 (9)
158
3
cellulose
190
NaOH, 1
2
8.8 (14)
1.7 (3)
–c
8.6 (14)
15.0 (24)
5.7 (9)
158
4
cellulose
170
NaOH, 3
4
8.6 (13)
2.6 (4)
–c
7.8 (12)
13.8 (21)
5.2 (8)
158
5
cellulose
180
NaOH, 3
3
9.1 (13)
2.2 (3)
–c
8.8 (13)
14.9 (21)
6.1 (9)
158
2.9 (4)
c
–
9.1 (12)
16.8 (21)
7.0 (9)
158
6
cellulose
190
NaOH, 3
2
10.4 (13)
7
cellulose
240
NaOH, 16
0.42
–c
40.2 (–)c
15.6 (–)c
8.2 (–)c
–c
–c
159
c
c
c
c
c
c
159
8
cellulose
240
KOH, 16
0.42
–
27.1 (–)
9
starch
175
NaOH, 1
1
–c
2.3 (6)
0.2 (<1)
5.9 (15)
12.4 (32)
5.3 (14)
157
c
14.7 (–)
8.0 (–)
–
–
10
starch
190
NaOH, 1
1
–
2.3 (5)
0.2 (<1)
7.0 (14)
14.6 (30)
6.5 (13)
157
11
starch
175
NaOH, 3
1
–c
2.4 (5)
0.1 (<1)
6.8 (15)
14.2 (32)
4.9 (11)
157
c
12
starch
190
NaOH, 3
1
–
2.8 (5)
0.2 (<1)
8.3 (14)
16.0 (27)
7.9 (13)
157
13
starch
240
NaOH, 16
1
–c
45.8 (–)c
24.2 (–)c
6.7 (–)c
–c
–c
159
1
c
c
c
c
c
c
159
14
starch
240
KOH, 16
–
37.5 (–)
18.3 (–)
7.7 (–)
–
–
a
Concn = concentration.
Yields are given in wt% of the starting material. Selectivities are given as wt% of the total soluble acids produced. Refer to Scheme 14 and
Figure 4 for structures.
c
Data not available.
b
In very concentrated alkali ([HO–]0 = 16 M) at 240 °C, the
situation is quite different. Krochta et al. degraded both
cellulose and starch under these conditions, and found
glycolic acid to be the major product (Table 6, entries 7,
8, 13, and 14).159 Oxalic and lactic acids were also produced in significant amounts, indicating considerable
cleavage of sugar acids to two- and three-carbon acids
(4H was not quantified). These products were more abundant when the base was sodium hydroxide rather than potassium hydroxide, and the yields of two-carbon acids
(i.e., glycolic and oxalic acids) were higher from starch
than from cellulose. Lactic acid, on the other hand, was
produced in greater amounts from cellulose. The agricultural product jute stick has been subjected to alkaline hydrolysis under similar conditions; Mathew et al. obtained
52 wt% oxalic acid from jute stick hydrolysis with 50 wt%
aqueous potassium hydroxide at 230 ± 10 °C.160 No other
organic products were quantified.
Xylan and chitin have also been degraded in alkali.157
Though the alkaline hydrolysis of chitin was more selective than those of cellulose or starch, the recalcitrant nature of the feedstock resulted in extremely low yields.
Xylan, on the other hand, could be hydrolyzed in better
yields, and gave mainly xyloisosaccharinic, lactic, and 2hydroxybutanoic acids at temperatures of 175 or 190 °C
and [NaOH]0 of 1 or 3 M. Lactic acid was the most abundant product, and was obtained in 22 wt% yield and 25
wt% selectivity among the identified acids at 190 °C and
3 M initial sodium hydroxide concentration.
To date, the alkaline degradations of cellulose that have
given the highest yields have also used very strong alkali
(for example, see Table 6, entries 7 and 8). This is likely
to limit the utility of ‘low-temperature’ (i.e., T < 250 °C)
alkaline degradations. Alkaline cellulose degradation using [HO–]0 = 1–3 M produces a complex mixture of products that prompted Niemelä and Sjöström to suggest
seeking industrial applications that could use the mixture
directly; they identified detergents and dispersing agents
as candidates.158 However, the situation is different at
At temperatures greater than 200 °C, wood-derived cellulose starts to degrade, albeit slowly, even in pure water.3b
The reaction is rapid above 250 °C; the treatment of pure
cellulose at 265 °C for 40 minutes or 295 °C for just 1
minute in a percolation reactor can produce glucose in
~50% yield.161 At 300 °C, crystalline cellulose is rapidly
converted into amorphous cellulose,162 for which hydrolysis is so facile that its decomposition products give spectra similar to those of glucose decomposition under the
same conditions, though with the addition of soluble oli-
higher temperature (see section 5), and base catalysts may
well be useful in obtaining organics in good yield from the
hydrothermal degradation of cellulose.
5
Hydrothermal Degradation
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
Table 6
Catalytic One-Pot Production of Small Organics
REVIEW
J. J. Verendel et al.
gosaccharides.163 Notably, hydrothermal treatment does
not significantly alter the degree of cellulose crystallinity
at 190119 or 245 °C,128 though it does at 380 °C;164 this is
in contrast to the cases of dilute-acid- and base-catalyzed
hydrolysis, which increase crystallinity by rapidly consuming the amorphous regions of a cellulose sample (vide
supra). Thus hydrothermal degradation (T < 374 °C) does
not selectively consume amorphous cellulose; rather, the
reaction takes place over the entire solid (i.e., cellulose)–
liquid interface. The degradation of polysaccharides in
hot, compressed water (subcritical H2O, T < 374 °C) must
be considered as both an acid- and a base-catalyzed process because the high ionization constant of water under
these conditions (at 300 °C, for example, log K w =
–11.406, so pH = pOH = 5.7)120 results in elevated concentrations of both H+ and HO–, despite the solution remaining neutral. Thus, products of both acid-catalyzed
hydrolysis (glucose, HMF, levulinic acid) and base-catalyzed hydrolysis (lactic and other organic acids) are obtained under these conditions. Over the course of the
reaction, the formed organic acids lower the reaction pH
to 3–4.3b The decomposition of glucose in subcritical water yielded both acid- and base-catalysis products even in
the presence of 0.02 M sulfuric acid or sodium hydroxide.165 Lactic acid is the major product of pyruvaldehyde
decomposition in subcritical water, which suggests that
the benzilic rearrangement can occur in this medium without any added base catalyst,165b and the Lobry de Bruyn–
Alberda van Ekenstein rearrangement also occurs under
these conditions.58c,166 Though quite remarkable, the enTable 7
Entry
hanced ability of subcritical water to act as both an acidic
and a basic catalyst can be a problem for selectivity, as it
increases the number of pathways available for polysaccharide degradation. As the hydrothermal treatments of
cellulose167 and biomass samples31,167a,168 have been reviewed, we focus here on cases in which catalysts have
been added in order to improve the selectivity in this medium. Several such strategies have been reported.
The acid-catalyzed degradation of cellulose can be favored in subcritical water by performing the reaction at
low pH. Using phosphate buffers, Asghari and Yoshida
investigated the hydrothermal decomposition of cellulose
at pH £ 6.5.169 They obtained the best HMF yield, ~32
mol%, after 2–3 minutes at 270 °C and pH 2; less HMF
was obtained at lower or higher temperatures, lower or
higher pH (pH0 ≥6.5), and longer reaction times. Without
phosphate buffer, the HMF yield did not exceed 19 mol%.
The phosphate buffer also improved the yield of HMF
from Japanese red pine wood, producing almost 25 wt%
at pH 2 (cf. ~17 wt% in the absence of buffer). Though
HMF yields higher than 32% have been obtained from
cellulose using ionic liquids (see Table 1), this yield is
high compared to other cellulose decompositions in aqueous acid.
Several authors have applied basic catalysts to the hydrothermal decomposition of polysaccharides. Krochta et
al.170 used sodium hydroxide to bias the reaction toward
the formation of small organic acids, and reported that the
production of these compounds from cellulose at 280 °C
Hydrothermal Decomposition of Polysaccharides Using Basic Metal Salts To Improve Selectivity
Substrate
Temp
Catalyst
(°C)
[Cat]0a
Time
Acid yield (%)b
(M)
(min)
Lactic
Formic
10
18.1
13.3
170b
19.2c
–d
171
Ref.
1
cellulose
280
NaOH
1.5
2
cellulose
300
Ca(OH)2
0.32
1.5
3
cellulose
300
none
0
2
3.3
–d
176
4
cellulose
300
NiSO4
0.0026
2
6.6
–d
176
5
cellulose
300
Cr2(SO4)3
0.0010
2
2.2
–d
176
6
starch
280
NaOH
1.5
10
19.8
13.0
170b
7
starch
300
Ca(OH)2
0.32
1
18.7c
–d
171
8
rice straw
280
NaOH
1.5
10
15.3
9
maize straw
300
none
0
2
4.9
–d
176
d
176
9.7
170b
10
maize straw
300
Cr2(SO4)3
0.0010
2
9.2
–
11
sawdust
300
ZnSO4
0.0025
2
5.5
–d
176
d
176
176
12
rice husk
300
none
0
2
4.5
–
13
rice husk
300
Cr2(SO4)3
0.0010
2
6.7
–d
a
[Cat]0 = initial catalyst concentration.
Unless otherwise noted, yields are given in wt% of the starting material.
c
Yield basis not given.
d
No data given.
b
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
1670
REVIEW
However, the amount of organic acids produced from rice
straw indicated that some must have come from its noncellulose components (most likely hemicellulose), so the
difference in distribution may also be due to the decomposition of other sugar monomers.
Jin and co-workers further investigated the effect of base
on saccharides at high temperature.171 By first investigating the degradation of glucose, they found that a lactic
acid yield of approximately 20% (yield basis not given)
could be achieved using calcium hydroxide as the catalyst
([HO–]0 = 0.64 M) at 300 °C. A higher yield of 27% was
available from sodium hydroxide, but this required more
concentrated base ([HO–]0 = 2.5 M), and resulted in visible reactor corrosion. They thus applied calcium hydroxide as a catalyst in the degradation of polysaccharides
([HO–]0 = 0.64 M; T = 300 °C), producing maximum lactic acid yields of 19.2 and 18.7% from cellulose and
starch, respectively (Table 7, entries 2 and 7). Carbonate
bases (Na2CO3, K2CO3) have been applied to the degradation of cellulose in subcritical water, though the resulting
organic products have not been quantified.172 However,
the presence of these additives increases gas formation
from the reaction.
The hydroxides and carbonates of sodium and potassium
([base]0 = 0.94 M) have been applied to the decomposition of pine sawdust at 280 °C, and increased the selectivity for acetic acid, butyrolactone (isomer not specified),
and 3-hydroxypentan-2-one.173 These catalysts also lowered the amount of char formed, and changed the distribution of phenolic compounds from lignin degradation. The
application of basic catalysts to lignin liquefaction has
been examined by several research groups.174
Li and co-workers turned to transition-metal sulfates,
which catalyze glucose decomposition at 300 °C,175 to improve lactic acid formation from cellulose and biomass
decomposition (Table 7, entries 3–5 and 9–13).176 Nickel(II) sulfate gave the best lactic acid yields from cellulose, with the reaction in the presence of 2.6 mM
nickel(II) sulfate producing twice as much lactic acid as
the uncatalyzed reaction. The yields of lactic acid from
maize straw, sawdust, and rice husk were also increased
by some metal sulfates, though the effect varied depending on both the metal salt and the substrate. All metal salts
tested lowered the yields of lactic acid from wheat bran.
1671
In all cases, lactic acid yields decreased at higher catalyst
concentrations.
Above the critical point of water (374 °C),120 the hydrolysis of cellulose is much faster than mono- and oligosaccharide decomposition, and crystalline cellulose is
partially dissolved and rearranged to a less crystalline
form above 375 °C.164,177 Because cellulose hydrolysis is
very rapid in supercritical water, it can produce glucose
and oligosaccharides selectively if the reaction time is
limited to a few seconds.163,178 At longer reaction times,
organic acids are formed, then subsequently decompose to
gaseous products.179 In fact, catalytic decompositions of
cellulose in supercritical water are often gasification reactions that aim to produce hydrogen.180
The degradation of sugars to acetic acid in supercritical
water can be favored using a base. Calvo and Vallejo reported an acetic acid yield of 36.1 wt% (carbon basis)
from the decomposition of glucose in the presence of
aqueous 0.25 M sodium hydroxide in water at 400 °C and
27.6 MPa,179 suggesting that it may be possible to combine this reaction with cellulose hydrolysis in this medium. Kruse and co-workers also produced acetic acid from
the potassium carbonate catalyzed decomposition of glucose at 400 °C.181 Though the acetic acid yield was not
given, the authors noted much less of this compound in
the reaction mixture when the temperature was 500 °C,
suggesting that it decomposed more rapidly at this temperature.
Sinağ and co-workers applied three dissimilar catalysts,
potassium carbonate, HZSM-5, a protic zeolite, and nickel
on silica, to the decomposition of cellulose in supercritical
water (T = 375 °C).182 The catalysts were used in equal
proportions by mass, and thus in varying amounts by
mole. In all cases, the glucose yield was <1 wt%, and the
major condensed product was acetic acid or acetaldehyde.
The basic catalyst, potassium carbonate, improved the
yield of acetic acid (Table 8, entry 2) compared to the uncatalyzed reaction (entry 1), but increased the yield of
acetaldehyde more. The reaction catalyzed by acidic
HZSM-5 also yielded more acetaldehyde than the uncatalyzed reaction, but produced less acetic acid (entry 3).
Nickel on silica gave the best selectivity for acetic acid
(entry 4), as well as the most organic carbon in the aqueTable 8 Major Products of the Decomposition of Cellulose in
Supercritical Water, in the Absence and in the Presence of Catalysta
Entry
Catalyst
Yield (wt%)
Acetic acid
HMF
Acetaldehyde Phenols
1
none
5.1
0.0
3.6
2.9
2
K2CO3
6.1
0.0
6.2
2.0
3
HZSM-5
4.6
0.0
4.3
2.2
4
Ni/SiO2
6.4
0.1
0.0
0.6
a
Reaction conditions: 2 wt% cellulose, 0.5% (w/w) catalyst, 375 °C,
1 h.182
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
increased more than tenfold when 6 wt% sodium hydroxide (~1.5 M) was added.170a They also found this method
to be applicable to starch and rice straw. The major product from these reactions was lactic acid, which was
formed in 18.1, 19.8 and 15.3 wt% yield from cellulose,
starch and rice straw, respectively (Table 7, entries 1, 6,
and 8).170b Formic, glycolic, acetic, and 2-hydroxybutyric
acids were also significant products. The distributions of
acid products from cellulose and starch were nearly identical. The distribution from rice straw was similar, though
some differences were noted; these may have been caused
by the presence of metal cations in the ash components of
rice straw.
Catalytic One-Pot Production of Small Organics
REVIEW
J. J. Verendel et al.
ous phase (~50%). Nevertheless, the acetic acid yield remained below 10 wt%, and large proportions of the
cellulose were gasified or converted into a carbon-rich
residue. All three catalyzed reactions produced more gas
than the uncatalyzed reaction did, and specifically increased the yields of hydrogen, carbon dioxide, and methane. The nickel on silica catalyst also yielded the most
acetic acid from a real biomass feedstock, sawdust; in that
case, 10.5 wt% acetic acid was produced (cf. 9.1 wt% for
the uncatalyzed reaction).
6
Metal-Catalyzed Oxidative Degradation
Though the oxidative degradation of cellulose in the presence of metals has been well studied, much of this work
has aimed to prevent the reaction. Pulp and paper bleaching processes, which aim to remove the lignin component
of lignocellulosic biomass while leaving the polysaccharide components intact, can oxidatively degrade cellulose
to produce shorter cellulose chains, and this unwanted reaction has been examined in detail.183 The oxidation of
cellulose to give small molecules with carboxylic acid
functionalities has received much less attention. However, mono- and disaccharides are now recognized as
‘green’ sources of these compounds, because they are renewable and can be oxidized with oxygen or hydrogen
peroxide in the presence of noble-metal catalysts,184 and a
few reports have extended the reactions developed for
sugars to the oxidative degradation of cellulose.
In 2000, Patrick and Abraham reported that platinum on
alumina, which catalyzes the oxidation of sugars,184a,b
could be applied to the air oxidation of cellulose.185 Even
in the absence of a transition-metal catalyst, cellulose can
be thermally degraded under oxidation conditions, though
the reaction is nonselective. Thus, after heating cellulose
in the presence of alumina (T = 87–185 °C), the authors
found a mixture of small organic acids in the product.
These were predominantly oxalic and succinic acids,
though a variety of other acids could also be identified.
The platinum-on-alumina catalyst gave oxalic acid as the
dominant organic product, with other products being suppressed. The non-metal-catalyzed oxidation still occurred
in the presence of catalyst, however. Although oxalic acid
was the major organic product from the catalytic oxidation, it was obtained in low yield; the major oxidation
product was carbon dioxide, indicating that over-oxidation was a serious issue.
In an effort to improve the reaction selectivity for small
organic compounds rather than carbon dioxide, Abraham
and co-workers turned to palladium, a weaker oxidizing
agent than platinum, as the catalytic metal.186 The air oxidation of cellulose in aqueous solution over palladium on
alumina at 100 °C produced very little carbon dioxide,
and the major products were malic, acetic, and oxalic
acids (see Figure 4 and Figure 5). The maximum yield of
malic acid, 8 wt%, was produced at 155 °C. Thus palladium produced a higher yield of organic acids, though room
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
for improvement remained. The authors added acetic acid
to the reaction in order to lower its pH, and thus increase
the rate of cellulose conversion. Though cellulose conversion did increase under these conditions, it was not clear
whether this improved the yield of malic (or any other)
acid. In a later publication, Schutt and Abraham performed the same reaction under basic conditions.187 At
pH 11.5 (controlled using NaOH), the palladium-on-alumina-catalyzed hydrolysis–oxidation of cellulose produced acetic acid and glucose as the major products. The
yield of acetic acid was ~20 wt% and the yield of glucose
~15%. Malic, succinic, and oxalic acids were also formed,
though in much smaller amounts. When the pH was held
at 9.0 using sodium tetraborate, a Lewis base, the oxidation of glucose to acids was greatly accelerated, though
the effect of this change on the product yields and selectivities was not described.
O
HO
O
OH
O
succinic acid
HO
OH
O HO
malic acid
OH
O
HO
OH
HO
glyceric acid
Figure 5
cellulose
O
HO
HO
O
OH
OH
glucuronic acid
Products of palladium-on-alumina-catalyzed oxidation of
Though the system of Abraham and co-workers still requires development to improve the conversion of cellulose and yield of organic products, it stands as a rare
example of a direct, fairly selective oxidation route from
cellulose to small organic acids. The particular design of
their reactor, called a ‘monolith froth reactor’ offered a
specific advantage in the noble-metal-catalyzed oxidation
of glucose to acids. The reactor essentially operated as a
four-phase system (solid catalyst, solid cellulose, aqueous
solution of cellulose oligomers, sugars, and oxidation
products, and gaseous air), so the catalyst had only intermittent contact with oxygen, which is known to over-oxidize palladium- and platinum-on-alumina catalysts during
glucose oxidation.184a,b Minimizing over-oxidation of the
catalyst extended its lifetime.
Although the oxidative degradations of cellulose by palladium on alumina and platinum on alumina have not been
subject to mechanistic study, Abraham and co-workers
have produced kinetic models suggesting that oxidation
products can be formed both directly from the polysaccharide and in two steps, via hydrolysis to sugars and subsequent sugar hydrolysis or oxidation.185–187 The initial
oxidation of aldoses occurs via hydride extraction
(Scheme 15),184b and the oxidation of polysaccharides is
likely to follow a similar mechanism. At elevated temperatures, such as those used by Abraham and co-workers,
platinum also oxidizes the C6 alcohol of glucose.184c
Downloaded by: the University of Oxford. Copyrighted material.
1672
Catalytic One-Pot Production of Small Organics
O
R
OH
HPt
½ O2
Pt0
HO
HO
O
O OH
HO
R
H
HO
R
H
HO
Scheme 15 The platinum-catalyzed oxidation of aldoses occurs via
hydride transfer184b
Recently, gold catalysts on various supports have been applied to cellobiose oxidation,188 and have proven more selective than supported platinum and palladium catalysts.
To our knowledge, these have not yet been used in combination with cellulose degradation.
Finally, there is a method of catalytic cellobiose oxidation
whose application to cellulose has thus far been restricted
to bleaching studies. Modifications of these systems could
be investigated for cellulose breakdown. Heteropolyanions have been used in pulp and paper bleaching,189 and
Neto and co-workers have studied cellulose degradation
by oxygen in the presence of [PMo(12–n)VnO40](3+n)–.190 As
part of this study, they examined the oxygen oxidations of
cellulose and cellobiose by partially reduced
[PMo7V5O40]8– in acidic solution, and attributed the oxidation reactions to both the free radical hydroxy and hydroperoxy species formed upon catalyst oxidation, and to
the VO2+ ion, which is released from molybdovanadophosphates in acidic solution. The major product from the
degradation of cellobiose was glucuronic acid (Figure 5);
together with its lactone, this compound comprised 53%
of the reaction mixture. Thus oxidation at C6 was dominant in this system. Other identified products included
glyceric, malic, glycolic, and oxalic acids (Figures 4
and 5).
7
Conclusion and Outlook
A large number of compounds are available directly from
polysaccharides through one-pot catalytic transformations. Acid-catalyzed hydrolysis coupled with dehydrative or reductive transformations have yielded HMF,
levulinic acid and polyols such as sorbitol in good yields.
Alkaline, oxidative and hydrothermal degradations of
polysaccharides have also yielded useful compounds, albeit less selectively. Metal catalysis has proven useful in
all of the aforementioned approaches, but considerable
work remains to develop truly effective and selective metal catalysts. Significant interest has recently been directed
towards the use and design of solid catalysts for polysaccharide, and especially cellulose, degradation; however, a
deeper understanding of how factors such as a pore size,
acidic sites, and the addition of metals affect hydrolysis
and subsequent steps is required.
1673
Solid acids are not yet very effective catalysts for cellulose hydrolysis and the conversion of cellulose, at least in
aqueous systems, is mainly determined by temperature.
This is hardly surprising considering the slow dissolution
of cellulose at temperatures below 200 °C. Once dissolution has occurred, however, the hydrolytic cleavage and
chemical modification of mono- and oligosaccharides are
more dependent on the solid catalyst, with different metal–support combinations giving very different product
spectra. Solids bearing both Brønsted and Lewis acid sites
have been used and re-used, and have given promising results. The major barrier to effective cellulose conversion
in the liquid phase is the dissolution of the very stable
crystalline polymer. Two main approaches have been
used to dissolve cellulose: low-temperature dissolution in
highly ionic solvents such as concentrated acids or ionic
liquids, and high-temperature dissolution in sub- or supercritical water. Each of these options has important advantages and disadvantages. Although low-temperature
dissolution gives better selectivity, the use of ionic liquids
for the bulk processing of wooden material is not optimal.
Water, on the other hand, is cheap and a truly green solvent, but the high-temperature processes in which it is
useful are energetically less favorable. Regardless of
which approach is taken, one is left with the challenge of
extracting the often very hydrophilic products from the reaction medium. In this respect, one-pot transformations of
cellulose to HMF, CMF, levulinic acid and syn-gas with
direct extraction into another phase are inspiring examples.
Clearly, processed cellulose, even microcrystalline cellulose that has a crystallinity of ~95%, is transformed into
small organic molecules more easily and with higher selectivity than is crude lignocellulose. Although the
polysaccharides in lignocellulose, or any type of biomass,
are protected by lignin and other stabilizing components,
they are, on average, only 70% crystalline. However, they
generally have a much higher degree of polymerization.
More work is required to adapt the candidate processes
discussed here to lignocellulosic material, which is much
more complex than model systems.
Acknowledgment
The authors are grateful to Prof. Thomas Norberg and Dr. Henrik
Ottosson for useful discussions. Energimyndigheten (The Swedish
Energy Agency), Nordic Energy Research (N-INNER II), The Swedish Research Council (VR), The Knut and Alice Wallenberg Foundation and VR/SIDA supported this work.
References
(1) McKendry, P. Bioresour. Technol. 2002, 83, 37.
(2) (a) Chheda, J. N.; Huber, G. W.; Dumesic, J. A. Angew.
Chem. Int. Ed. 2007, 46, 7164. (b) Corma, A.; Iborra, S.;
Velty, A. Chem. Rev. 2007, 107, 2411. (c) Kennedy, J. F.;
Knill, C. J.; Taylor, D. W. Conversion of cellulosic
feedstocks into useful products, In Polysaccharides:
Structural Diversity and Functional Versatility; Dumitriu,
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
REVIEW
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
REVIEW
J. J. Verendel et al.
S., Ed.; CRC Press: New York, 1998, 835.
(d) Lichtenthaler, F. W.; Peters, S. C. R. Chim. 2004, 7, 65.
(a) Abatzoglou, N.; Chornet, E. Acid hydrolysis of
hemicelluloses and cellulose: theory and applications, In
Polysaccharides: Structural Diversity and Functional
Versatility; Dumitriu, S., Ed.; CRC Press: New York, 1998,
1007. (b) Bobleter, O. Hydrothermal degradation and
fractionation of saccharides and polysaccharides, In
Polysaccharides: Structural Diversity and Functional
Versatility, 2nd ed.; Dumitriu, S., Ed.; CRC Press: New
York, 2005, 893.
Hamelinck, C. N. van Hooijdonk, G.; Faaij, A. P. C.
Biomass Bioenergy 2005, 28, 384.
Zhao, H.; Kwak, J. H.; Wang, Y.; Franz, J. A.; White, J. M.;
Holladay, J. E. Energy Fuels 2006, 20, 807.
Sun, Y.; Cheng, J. Bioresour. Technol. 2002, 83, 1.
(a) Lin, Y.-C.; Huber, G. W. Energy Environ. Sci. 2009, 2,
68. (b) Rinaldi, R.; Schuth, F. Energy Environ. Sci. 2009, 2,
610.
Wyman, C. E.; Decker, S. R.; Himmel, M. E.; Brady, J. W.;
Skopec, C. E.; Viikari, L. Hydrolysis of cellulose and
hemicellulose, In Polysaccharides: Structural Diversity and
Functional Versatility, 2nd ed.; Dumitriu, S., Ed.; CRC
Press: New York, 2005, 995.
Van de Vyver, S. Geboers, J.; Jacobs, P. A.; Sels, B. F.
ChemCatChem 2010, 3, 82.
(a) Margeot, A.; Hahn-Hagerdal, B.; Edlund, M.; Slade, R.;
Monot, F. Curr. Opin. Biotechnol. 2009, 20, 372.
(b) Wilson, D. B. Curr. Opin. Biotechnol. 2009, 20, 295.
(c) Coughlan, M. P. Anim. Feed Sci. Technol. 1991, 32, 77.
(a) Stöcker, M. Angew. Chem. Int. Ed. 2008, 47, 9200.
(b) Kamm, B. Angew. Chem. Int. Ed. 2007, 46, 5056.
(c) Huber, G. W.; Corma, A. Angew. Chem. Int. Ed. 2007,
46, 7184. (d) Huber, G. W.; Dumesic, J. A. Catal. Today
2006, 111, 119.
(a) Lange, J.-P. Lignocellulose conversion: an introduction
to chemistry, process and economics, In Catalysis for
Renewables: From Feedstock to Energy Production ; Centi,
G.; van Santen, R. A., Eds.; Wiley-VCH: Weinheim, 2007,
21. (b) Florin, N. H.; Harris, A. T. Chem. Eng. Sci. 2008, 63,
287.
Braconnot, H. Ann. Phys. 1819, 63, 347.
Xiang, Q.; Lee, Y. Y.; Pettersson, P. O.; Torget, R. W. Appl.
Biochem. Biotechnol. 2003, 107, 505.
(a) Roman, M.; Winter, W. T. Biomacromolecules 2004, 5,
1671. (b) Fredenhagen, K.; Cadenbach, G. Angew. Chem.
1933, 46, 113.
(a) Zhang, Y.-H. P.; Ding, S.-Y.; Mielenz, J. R.; Cui, J.-B.;
Elander, R. T.; Laser, M.; Himmel, M. E.; McMillan, J. R.;
Lynd, L. R. Biotechnol. Bioeng. 2007, 97, 214. (b) Zhang,
Y. H. P.; Cui, J.; Lynd, L. R.; Kuang, L. R.
Biomacromolecules 2006, 7, 644.
Ritter, G. J.; Mitchell, R. L.; Seborg, R. M. J. Am. Chem.
Soc. 1933, 55, 2989.
Camacho, F.; González-Tello, P.; Jurado, E.; Robles, A.
J. Chem. Technol. Biotechnol. 1996, 67, 350.
Selke, S. M.; Hawley, M. C.; Hardt, H.; Lamport, D. T. A.;
Smith, G.; Smith, J. Ind. Eng. Chem. Prod. Res. Dev. 1982,
21, 11.
Garves, K. Rapid hydrolysis of celluloses in homogeneous
solution, In Hydrolysis of cellulose: Mechanisms of
enzymatic and acid catalysis; Brown, R. D. Jr.; Jurasek, L.,
Eds.; American Chemical Society: Washington DC, 1979,
159.
Higgins, F. J.; Ho, G. E. Agric. Wastes 1982, 4, 97.
Fengel, V. D.; Wegener, G.; Heizmann, A.; Przyklenk, M.
Holzforschung 1977, 40, 65.
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
(23) Sun, Y.; Lin, L.; Pang, C.; Deng, H.; Peng, H.; Li, J.; He, B.;
Liu, S. Energy Fuels 2007, 21, 2386.
(24) Hardt, H.; Lamport, D. T. A. Biotechnol. Bioeng. 1982, 24,
903.
(25) Simonsen, E. Angew. Chem. 1898, 11, 195.
(26) Harris, E. E. Adv. Carbohydr. Chem. 1949, 4, 153.
(27) Krässig, H. A. Cellulose: Structure, Accessibility and
Reactivity; Gordon and Breach Publ.: Singapore, 1993.
(28) Philipp, B. Pure Appl. Chem. 1984, 56, 391.
(29) (a) Mok, W. S.; Antal, M. J. Jr.; Varhegyi, G. Ind. Eng.
Chem. Res. 1992, 31, 94. (b) Bouchard, J.; Abatzoglou, N.;
Chornet, E.; Overend, R. P. Wood Sci. Technol. 1989, 23,
343. (c) Bouchard, J.; Overend, R. P.; Chornet, E.;
Van Calsteren, M.-R. J. Wood Chem. Technol. 1992, 12,
335.
(30) Feller, R. L.; Lee, S. B.; Bogaard, J. Advances in Chemistry
Series, Vol. 212: The kinetics of cellulose deterioration;
Needles, H. L.; Zeronian, S. H., Eds.; American Chemical
Society: Washington DC, 1986.
(31) Bobleter, O. Prog. Polym. Sci. 1994, 19, 797.
(32) Saeman, J. F. Ind. Eng. Chem. 1945, 37, 43.
(33) Taherzadeh, M. J.; Karimi, K. Bioresources 2007, 2, 472.
(34) Harris, J. F.; Baker, A. J.; Connor, A. H.; Jeffries, T. W.;
Minor, J. L.; Pettersen, R. C.; Scott, R. C.; Springer, E. L.;
Wegner, T. H.; Zerbe, J. L. Two-stage, dilute sulfuric acid
hydrolysis of wood: An investigation of the fundamentals,
Report FPL-45; U.S. Department of Agriculture Forest
Products Laboratory: Madison WI, 1985.
(35) (a) Herrera, A.; Téllez-Luis, S. J.; Ramírez, J. A.; Vázquez,
M. J. Cereal Sci. 2003, 37, 267. (b) Nguyen, Q.; Tucker,
M.; Keller, F.; Eddy, F. Appl. Biochem. Biotechnol. 2000,
84-86, 561.
(36) (a) Suganuma, S.; Nakajima, K.; Kitano, M.; Yamaguchi,
D.; Kato, H.; Hayashi, S.; Hara, M. J. Am. Chem. Soc. 2008,
130, 12787. (b) Onda, A.; Ochi, T.; Yanagisawa, K. Green
Chem. 2008, 10, 1033. (c) Bootsma, J. A.; Entorf, M.; Eder,
J.; Shanks, B. H. Bioresour. Technol. 2008, 99, 5226.
(37) Rinaldi, R.; Schüth, F. ChemSusChem 2009, 2, 1096.
(38) (a) Pinkert, A.; Marsh, K. N.; Pang, S.; Staiger, M. P. Chem.
Rev. 2009, 109, 6712. (b) Rosatella, A. A.; Branco, L. C.;
Afonso, C. A. M. Green Chem. 2009, 11, 1406.
(c) Remsing, R. C.; Swatloski, R. P.; Rogers, R. D.; Moyna,
G. Chem. Commun. 2006, 1271. (d) Swatloski, R. P.; Spear,
S. K.; Holbrey, J. D.; Rogers, R. D. J. Am. Chem. Soc. 2002,
124, 4974.
(39) (a) Rinaldi, R. Chem. Commun. 2011, 47, 511. (b) Kuo, C.H.; Lee, C.-K. Bioresour. Technol. 2009, 100, 866.
(c) Saalwächter, K.; Burchard, W.; Klüfers, P.; Kettenbach,
G.; Mayer, P.; Klemm, D.; Dugarmaa, S. Macromolecules
2000, 33, 4094.
(40) Li, C.; Zhao, Z. K. Adv. Synth. Catal. 2007, 349, 1847.
(41) Li, C.; Wang, Q.; Zhao, Z. K. Green Chem. 2008, 10, 177.
(42) Amarasekara, A. S.; Owereh, O. S. Ind. Eng. Chem. Res.
2009, 48, 10152.
(43) (a) Rinaldi, R.; Meine, N.; Stein, J. v.; Palkovits, R.; Schüth,
F. ChemSusChem 2010, 3, 266. (b) Vanoye, L.; Fanselow,
M.; Holbrey, J. D.; Atkins, M. P.; Seddon, K. R. Green
Chem. 2009, 11, 390.
(44) Rinaldi, R.; Engel, P.; Büchs, J.; Spiess, A. C.; Schüth, F.
ChemSusChem 2010, 3, 1151.
(45) Rinaldi, R.; Palkovits, R.; Schüth, F. Angew. Chem. Int. Ed.
2008, 47, 8047.
(46) Dwiatmoko, A. A.; Choi, J. W.; Suh, D. J.; Suh, Y.-W.;
Kung, H. H. Appl. Catal., A 2010, 387, 209.
(47) Conner, A. H.; Wood, B. F.; Hill, C. G.; Harris, J. F. J. Wood
Chem. Technol. 1985, 5, 461.
Downloaded by: the University of Oxford. Copyrighted material.
1674
(48) Kuster, B. F. M.; van der Baan, H. S. Carbohydr. Res. 1977,
54, 165.
(49) Speck, J. C. Jr. Adv. Carbohydr. Chem. 1958, 13, 63.
(50) (a) Moreau, C.; Durand, R.; Razigade, S.; Duhamet, J.;
Faugeras, P.; Rivalier, P.; Ros, P.; Avignon, G. Appl. Catal.,
A 1996, 145, 211. (b) Antal, M. J.; Mok, W. S. L.; Richards,
G. N. Carbohydr. Res. 1990, 199, 91.
(51) (a) Hurd, C. D.; Isenhour, L. L. J. Am. Chem. Soc. 1932, 54,
317. (b) Dunlop, A. P. Ind. Eng. Chem. 1948, 40, 204.
(52) (a) Karimi, K.; Kheradmandinia, S.; Taherzadeh, M. J.
Biomass Bioenergy 2006, 30, 247. (b) Rahman, S. H. A.;
Choudhury, J. P.; Ahmad, A. L. Biochem. Eng. J. 2006, 30,
97.
(53) Antal, M. J.; Leesomboon, T.; Mok, W. S.; Richards, G. N.
Carbohydr. Res. 1991, 217, 71.
(54) Oefner, P. J.; Lanziner, A. H.; Bonn, G.; Bobleter, O.
Monatsh. Chem. 1992, 123, 547.
(55) (a) Kim, S. B.; Lee, Y. Y. Biotechnol. Bioeng. Symp. 1986,
15, 81. (b) Kobayashi, H.; Komanoya, T.; Hara, K.;
Fukuoka, A. ChemSusChem 2010, 3, 440. (c) Bootsma, J.
A.; Shanks, B. H. Appl. Catal., A 2007, 327, 44.
(d) Moreau, C.; Durand, R.; Duhamet, J.; Rivalier, P.
J. Carbohydr. Chem. 1997, 16, 709. (e) Yamaguchi, D.;
Kitano, M.; Suganuma, S.; Nakajima, K.; Kato, H.; Hara, M.
J. Phys. Chem. C 2009, 113, 3181. (f) Onda, A.; Ochi, T.;
Yanagisawa, K. Top. Catal. 2009, 52, 801.
(56) Baugh, K. D.; McCarty, P. L. Biotechnol. Bioeng. 1988, 31,
50.
(57) Mamman, A. S.; Lee, J.-M.; Kim, Y.-C.; Hwang, I. T.; Park,
N.-J.; Hwang, Y. K.; Chang, J.-S.; Hwang, J.-S. Biofuels,
Bioprod. Biorefin. 2008, 2, 438.
(58) (a) Asghari, F. S.; Yoshida, H. Ind. Eng. Chem. Res. 2007,
46, 7703. (b) Girisuta, B.; Janssen, L. P. B. M.; Heeres, H. J.
Ind. Eng. Chem. Res. 2007, 46, 1696. (c) Kabyemela, B.
M.; Adschiri, T.; Malaluan, R. M.; Arai, K. Ind. Eng. Chem.
Res. 1999, 38, 2888.
(59) Sievers, C.; Musin, I.; Marzialetti, T.; Valenzuela Olarte, M.
B.; Agrawal, P. K.; Jones, C. W. ChemSusChem 2009, 2,
665.
(60) Knežević, D.; van Swaaij, W. P. M.; Kersten, S. R. A. Ind.
Eng. Chem. Res. 2009, 48, 4731.
(61) Marzialetti, T.; Valenzuela Olarte, M. B.; Sievers, C.;
Hoskins, T. J. C.; Agrawal, P. K.; Jones, C. W. Ind. Eng.
Chem. Res. 2008, 47, 7131.
(62) (a) Sue, K.; Ouchi, F.; Minami, K.; Arai, K. J. Chem. Eng.
Data 2004, 49, 1359. (b) Xiang, T.; Johnston, K. P.;
Wofford, W. T.; Gloyna, E. F. Ind. Eng. Chem. Res. 1996,
35, 4788.
(63) Seri, K.-i.; Sakaki, T.; Shibata, M.; Inoue, Y.; Ishida, H.
Bioresour. Technol. 2002, 81, 257.
(64) Yu, J. C.; Baizer, M. M.; Nobe, K. J. Electrochem. Soc.
1988, 135, 83.
(65) Chang, C.; Ma, X.; Cen, P. Chin. J. Chem. Eng. 2006, 14,
708.
(66) (a) Kuster, B. F. M. Starch – Stärke 1990, 42, 314.
(b) Carlini, C.; Giuttari, M.; Maria Raspolli Galletti, A.;
Sbrana, G.; Armaroli, T.; Busca, G. Appl. Catal., A 1999,
183, 295.
(67) Werpy T., Petersen G.; Top value added chemicals from
biomass, Volume I: Results of screening for potential
candidates from sugars and synthesis gas PNNL-14808;
U.S. Department of Energy Pacific Northwest National
Laboratory: Richland WA, 2004.
(68) Moreau, C.; Belgacem, M. N.; Gandini, A. Top. Catal. 2004,
27, 11.
(69) Mascal, M.; Nikitin, E. B. Angew. Chem. Int. Ed. 2008, 47,
7924.
Catalytic One-Pot Production of Small Organics
1675
(70) Mascal, M.; Nikitin, E. B. ChemSusChem 2009, 2, 423.
(71) Zhao, H.; Holladay, J. E.; Brown, H.; Zhang, Z. C. Science
2007, 316, 1597.
(72) Su, Y.; Brown, H. M.; Huang, X.; Zhou, X.-D.; Amonette, J.
E.; Zhang, Z. C. Appl. Catal., A 2009, 361, 117.
(73) Potthast, A.; Rosenau, T.; Sartori, J.; Sixta, H.; Kosma, P.
Polymer 2003, 44, 7.
(74) Binder, J. B.; Raines, R. T. J. Am. Chem. Soc. 2009, 131,
1979.
(75) Li, C.; Zhang, Z.; Zhao, Z. K. Tetrahedron Lett. 2009, 50,
5403.
(76) Zhang, Z.; Zhao, Z. K. Bioresour. Technol. 2010, 101, 1111.
(77) Lima, S.; Neves, P.; Antunes, M. M.; Pillinger, M.; Ignatyev,
N.; Valente, A. A. Appl. Catal., A 2009, 363, 93.
(78) Chun, J.-A.; Lee, J.-W.; Yi, Y.-B.; Hong, S.-S.; Chung, C.H. Starch – Stärke 2010, 62, 326.
(79) Zhang, Y.; Du, H.; Qian, X.; Chen, E. Y. X. Energy Fuels
2010, 24, 2410.
(80) Chheda, J. N.; Roman-Leshkov, Y.; Dumesic, J. A. Green
Chem. 2007, 9, 342.
(81) Frost, T. R.; Kurth, E. F. Tappi 1951, 34, 80.
(82) Leonard, R. Ind. Eng. Chem. 1956, 48, 1330.
(83) (a) Bozell, J. J.; Moens, L.; Elliott, D. C.; Wang, Y.;
Neuenscwander, G. G.; Fitzpatrick, S. W.; Bilski, R. J.;
Jarnefeld, J. L. Resour. Conserv. Recycl. 2000, 28, 227.
(b) Bozell, J. J.; Petersen, G. R. Green Chem. 2010, 12, 539.
(84) (a) Efremov, A. A.; Pervyshina, G. G.; Kuznetsov, B. N.
Chem. Nat. Compd. 1998, 34, 182. (b) Chang, C.; Cen, P.;
Ma, X. Bioresour. Technol. 2007, 98, 1448. (c) Fang, Q.;
Hanna, M. A. Bioresour. Technol. 2002, 81, 187.
(85) Girisuta, B.; Danon, B.; Manurung, R.; Janssen, L. P. B. M.;
Heeres, H. J. Bioresour. Technol. 2008, 99, 8367.
(86) Fitzpatrick, S. W. US5608105, 1997; Chem. Abstr. 1997,
126, 117739.
(87) Ritter, S. Chem. Eng. News 2006, 84 (43), 47.
(88) Peng, L.; Lin, L.; Zhang, J.; Zhuang, J.; Zhang, B.; Gong, Y.
Molecules 2010, 15, 5258.
(89) Rasrendra, C.; Makertihartha, I.; Adisasmito, S.; Heeres, H.
Top. Catal. 2010, 53, 1241.
(90) Hegner, J.; Pereira, K. C.; DeBoef, B.; Lucht, B. L.
Tetrahedron Lett. 2010, 51, 2356.
(91) Schraufnagel, R. A.; Rase, H. F. Ind. Eng. Chem. Prod. Res.
Dev. 1975, 14, 40.
(92) Wang, P.; Zhan, S.; Yu, H. Adv. Mater. Res. 2010, 96, 183.
(93) Deng, W.; Liu, M.; Zhang, Q.; Tan, X.; Wang, Y. Chem.
Commun. 2010, 46, 2668.
(94) (a) Shimizu, K.-i.; Furukawa, H.; Kobayashi, N.; Itaya, Y.;
Satsuma, A. Green Chem. 2009, 11, 1627. (b) Arai, K.;
Ogiwara, Y. J. Appl. Polym. Sci. 1985, 30, 3051.
(95) Villandier, N.; Corma, A. Chem. Commun. 2010, 46, 4408.
(96) von Rybinski, W.; Hill, K. Angew. Chem. Int. Ed. 1998, 37,
1328.
(97) Mäki-Arvela, P.; Holmbom, B.; Salmi, T.; Murzin, D. Y.
Catal. Rev. 2007, 49, 197.
(98) (a) Cortright, R. D.; Davda, R. R.; Dumesic, J. A. Nature
2002, 418, 964. (b) Huber, G. W.; Shabaker, J. W.;
Dumesic, J. A. Science 2003, 300, 2075.
(99) Wisnlak, J.; Simon, R. Ind. Eng. Chem. Prod. Res. Dev.
1979, 18, 50.
(100) Crawford, T. C.; Crawford, S. A. Synthesis of L-Ascorbic
Acid, In Advances in Carbohydrate Chemistry and
Biochemistry; Tipson, R. S.; Derek, H., Eds.; Academic
Press: San Diego, 1980, 79.
(101) Hoffer, B. W.; Crezee, E.; Devred, F.; Mooijman, P. R. M.;
Sloof, W. G.; Kooyman, P. J.; van Langeveld, A. D.;
Kapteijn, F.; Moulijn, J. A. Appl. Catal., A 2003, 253, 437.
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
REVIEW
REVIEW
J. J. Verendel et al.
(102) Makkee, M.; Kieboom, A. P. G.; van Bekkum, H.
Carbohydr. Res. 1985, 138, 225.
(103) (a) Broekhuis, R. R.; Budhlall, B. M.; Nordquist, A. F. Ind.
Eng. Chem. Res. 2004, 43, 5146. (b) Heinen, A. W.; Peters,
J. A.; van Bekkum, H. Carbohydr. Res. 2000, 328, 449.
(104) (a) Kolarić, S.; Šunjić, V. J. Mol. Catal. A: Chem. 1996, 110,
189. (b) Joó, F.; Tóth, Z.; Beck, M. T. Inorg. Chim. Acta
1977, 25, L61.
(105) Andrews, M. A.; Klaeren, S. A. J. Am. Chem. Soc. 1989,
111, 4131.
(106) (a) Clark, I. Ind. Eng. Chem. 1958, 50, 1125. (b) van Ling,
G.; Ruijterman, C.; Vlugter, J. C. Carbohydr. Res. 1967, 4,
380. (c) Dubeck, M.; Knapp, G. G. US4476331, 1984;
Chem. Abstr. 1984, 100, 138589. (d) Arena, B. J.
US4496780, 1985; Chem. Abstr. 1985, 102, 167115.
(107) Balandin, A. A.; Vasyunina, N. A.; Barysheva, G. S.;
Chepigo, S. V. Bull. Acad. Sci. USSR 1957, 6, 403.
(108) Balandin, A. A.; Vasyunina, N. A.; Chepigo, S. V.;
Barysheva, G. S. Proc. Acad. Sci. USSR 1959, 128, 839.
(109) Vasyunina, N. A.; Balandin, A. A.; Chepigo, S. V.;
Barysheva, G. S. Bull. Acad. Sci. USSR 1960, 9, 1419.
(110) Sharkov, V. I. Angew. Chem. Int. Ed. 1963, 2, 405.
(111) Robinson, J. M.; Burgess, C. E.; Bently, M. A.; Brasher, C.
D.; Horne, B. O.; Lillard, D. M.; Macias, J. M.; Mandal, H.
D.; Mills, S. C.; O’Hara, K. D.; Pon, J. T.; Raigoza, A. F.;
Sanchez, E. H.; Villarreal, J. S. Biomass Bioenergy 2004, 26,
473.
(112) Baudel, H. M.; de Abreu, C. A. M.; Zaror, C. Z. J. Chem.
Technol. Biotechnol. 2005, 80, 230.
(113) Palkovits, R.; Tajvidi, K.; Procelewska, J.; Rinaldi, R.;
Ruppert, A. Green Chem. 2010, 12, 972.
(114) Davda, R. R.; Shabaker, J. W.; Huber, G. W.; Cortright, R.
D.; Dumesic, J. A. Appl. Catal., B 2003, 43, 13.
(115) Geboers, J.; Van de Vyver, S.; Carpentier, K.; Blochouse, K.
d.; Jacobs, P.; Sels, B. Chem. Commun. 2010, 46, 3577.
(116) (a) Capik, R. J.; Wright, L. W. US3670035 (DE1911093A),
1972; Chem. Abstr. 1970, 72, 25576. (b) Jacobs, P.;
Hinnekens, H. EP0329923, 1989; Chem. Abstr. 1989, 111,
233477.
(117) Fukuoka, A.; Dhepe, P. L. Angew. Chem. Int. Ed. 2006, 45,
5161.
(118) (a) Hattori, H.; Yamada, T.; Shishido, T. Res. Chem.
Intermed. 1998, 24, 439. (b) Fukuoka, A.; Dhepe, P. L.
Chem. Rec. 2009, 9, 224.
(119) Jollet, V.; Chambon, F.; Rataboul, F.; Cabiac, A.; Pinel, C.;
Guillon, E.; Essayem, N. Green Chem. 2009, 11, 2052.
(120) Marshall, W. L.; Franck, E. U. J. Phys. Chem. Ref. Data
1981, 10, 295.
(121) Ji, N.; Zhang, T.; Zheng, M.; Wang, A.; Wang, H.; Wang,
X.; Chen, J. G. Angew. Chem. Int. Ed. 2008, 47, 8510.
(122) Ji, N.; Zhang, T.; Zheng, M.; Wang, A.; Wang, H.; Wang,
X.; Shu, Y.; Stottlemyer, A. L.; Chen, J. G. Catal. Today
2009, 147, 77.
(123) Zheng, M. Y.; Wang, A. Q.; Ji, N.; Pang, J. F.; Wang, X. D.;
Zhang, T. ChemSusChem 2010, 3, 63.
(124) Zhang, Y.; Wang, A.; Zhang, T. Chem. Commun. 2010, 46,
862.
(125) Huber, G. W.; Cortright, R. D.; Dumesic, J. A. Angew.
Chem. Int. Ed. 2004, 43, 1549.
(126) Tanksale, A.; Beltramini, J. N.; Lu, G. M. Renewable
Sustainable Energy Rev. 2010, 14, 166.
(127) Yan, N.; Zhao, C.; Luo, C.; Dyson, P. J.; Liu, H.; Kou, Y.
J. Am. Chem. Soc. 2006, 128, 8714.
(128) Luo, C.; Wang, S.; Liu, H. Angew. Chem. Int. Ed. 2007, 46,
7636.
(129) Deng, W.; Tan, X.; Fang, W.; Zhang, Q.; Wang, Y. Catal.
Lett. 2009, 133, 167.
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
(130) Deng, W.; Liu, M.; Tan, X.; Zhang, Q.; Wang, Y. J. Catal.
2010, 271, 22.
(131) Van de Vyver, S.; Geboers, J.; Dusselier, M.; Schepers, H.;
Vosch, T.; Zhang, L.; Van Tendeloo, G.; Jacobs, P.; Sels, B.
ChemSusChem 2010, 3, 698.
(132) Othmer, D. F.; Gamer, C. H.; Jacobs, J. J. Jr. Ind. Eng. Chem.
1942, 34, 262.
(133) Nevell, T. P. Degradation of cellulose by acids, alkalis, and
mechanical means, In Cellulose Chemistry and its
Applications; Nevell, T. P.; Zeronian, S. H., Eds.; Ellis
Horwood: Chichester, 1985, 223.
(134) Molton, P. M.; Demmitt, T. F. Polym. Plast. Technol. Eng.
1978, 11, 127.
(135) Knill, C. J.; Kennedy, J. F. Carbohydr. Polym. 2003, 51,
281.
(136) Van Loon, L. R.; Glaus, M. A. J. Polym. Environ. 1997, 5,
97.
(137) (a) Pavasars, I.; Hagberg, J.; Borén, H.; Allard, B. J. Polym.
Environ. 2003, 11, 39. (b) Glaus, M. A.; Van Loon, L. R.
Environ. Sci. Technol. 2008, 42, 2906.
(138) Corbett, W. M.; Richards, G. N. Svensk Papperstidn. 1957,
60, 791.
(139) Samuelson, O.; Grangård, G.; Jönsson, K.; Schramm, K.
Svensk Papperstidn. 1953, 56, 779.
(140) Lai, Y.-Z.; Ontto, D. E. J. Appl. Polym. Sci. 1979, 23, 3219.
(141) Procter, A. R.; Wiekenkamp, R. H. J. Polym. Sci., Part C:
Polym. Symp. 1969, 28, 1.
(142) Lobry de Bruyn, C. A.; Alberda van Ekenstein, W. Recl.
Trav. Chim. Pays-Bas 1895, 14, 203.
(143) Selman, S.; Eastham, J. F. Q. Rev. Chem. Soc. 1960, 14, 221.
(144) Bonn, G.; Binder, H.; Leonhard, H.; Bobleter, O. Monatsh.
Chem. 1985, 116, 961.
(145) Machell, G.; Richards, G. N. J. Chem. Soc. 1957, 4500.
(146) Johansson, M. H.; Samuelson, O. J. Appl. Polym. Sci. 1978,
22, 615.
(147) Haas, D. W.; Hrutfiord, B. F.; Sarkanen, K. V. J. Appl.
Polym. Sci. 1967, 11, 587.
(148) Richtzenhain, H.; Lindgren, B. O.; Abrahamsson, B.;
Holmberg, K. Svensk Papperstidn. 1954, 57, 363.
(149) Ziderman, I.; Bel-Ayche, J. J. Appl. Polym. Sci. 1978, 22,
711.
(150) Colbran, R. L.; Davidson, G. F. J. Text. Inst. 1961, 52, T73.
(151) Ziderman, I. Cellul. Chem. Technol. 1980, 14, 703.
(152) Samuelson, O.; Wennerblom, A. Svensk Papperstidn. 1954,
57, 827.
(153) Richards, G. N.; Sephton, H. H. J. Chem. Soc. 1957, 4492.
(154) Alfredsson, B.; Samuelson, O. Svensk Papperstidn. 1968,
71, 679.
(155) Machell, G.; Richards, G. N. J. Chem. Soc. 1960, 1924.
(156) Dryselius, E.; Lindberg, B.; Theander, O. Acta Chem. Scand.
1958, 12, 340.
(157) Niemelä, K. Carbohydr. Res. 1990, 204, 37.
(158) Niemelä, K.; Sjöström, E. Biomass 1986, 11, 215.
(159) Krochta, J. M.; Tillin, S. J.; Hudson, J. S. Appl. Biochem.
Biotechnol. 1988, 17, 23.
(160) Mathew, M. D.; Gopal, M.; Banerjee, S. K. Agric. Wastes
1984, 11, 47.
(161) Bonn, G.; Concin, R.; Bobleter, O. Wood Sci. Technol. 1983,
17, 195.
(162) Deguchi, S.; Tsujii, K.; Horikoshi, K. Green Chem. 2008,
10, 191.
(163) Sasaki, M.; Fang, Z.; Fukushima, Y.; Adschiri, T.; Arai, K.
Ind. Eng. Chem. Res. 2000, 39, 2883.
(164) Sasaki, M.; Adschiri, T.; Arai, K. J. Agric. Food Chem.
2003, 51, 5376.
Downloaded by: the University of Oxford. Copyrighted material.
1676
(165) (a) Jin, F.; Zhou, Z.; Kishita, A.; Enomoto, H. J. Mater. Sci.
2006, 41, 1495. (b) Jin, F.; Zhou, Z.; Enomoto, H.; Moriya,
T.; Higashijima, H. Chem. Lett. 2004, 33, 126.
(166) Kabyemela, B. M.; Adschiri, T.; Malaluan, R. M.; Arai, K.
Ind. Eng. Chem. Res. 1997, 36, 1552.
(167) (a) Jin, F.; Enomoto, H. Bioresources 2009, 4, 704.
(b) Matsumura, Y.; Sasaki, M.; Okuda, K.; Takami, S.;
Ohara, S.; Umetsu, M.; Adschiri, T. Combust. Sci. Technol.
2006, 178, 509.
(168) Garrote, G.; Domínguez, H.; Parajó, J. C. Eur. J. Wood
Wood Prod. 1999, 57, 191.
(169) Asghari, F. S.; Yoshida, H. Carbohydr. Res. 2010, 345, 124.
(170) (a) Krochta, J. M.; Hudson, J. S. Agric. Wastes 1985, 14,
243. (b) Krochta, J. M.; Tillin, S. J.; Hudson, J. S. J. Appl.
Polym. Sci. 1987, 33, 1413.
(171) Yan, X.; Jin, F.; Tohji, K.; Kishita, A.; Enomoto, H. AIChE
J. 2010, 56, 2727.
(172) (a) Fang, Z.; Minowa, T.; Smith, R. L.; Ogi, T.; Koziński, J.
A. Ind. Eng. Chem. Res. 2004, 43, 2454. (b) Kumar, S.;
Gupta, R. B. Ind. Eng. Chem. Res. 2008, 47, 9321.
(c) Minowa, T.; Zhen, F.; Ogi, T. J. Supercrit. Fluids 1998,
13, 253.
(173) Karagöz, S.; Bhaskar, T.; Muto, A.; Sakata, Y.; Oshiki, T.;
Kishimoto, T. Chem. Eng. J. 2005, 108, 127.
(174) (a) Ogi, T.; Yokoyama, S.; Koguchi, K. Chem. Lett. 1985,
14, 1199. (b) Miller, J. E.; Evans, L.; Littlewolf, A.; Trudell,
D. E. Fuel 1999, 78, 1363. (c) Karagöz, S.; Bhaskar, T.;
Muto, A.; Sakata, Y. Fuel 2004, 83, 2293. (d) Karagöz, S.;
Bhaskar, T.; Muto, A.; Sakata, Y. J. Chem. Technol.
Biotechnol. 2005, 80, 1097. (e) Liu, A.; Park, Y.; Huang, Z.;
Wang, B.; Ankumah, R. O.; Biswas, P. K. Energy Fuels
2006, 20, 446. (f) Ross, A. B.; Biller, P.; Kubacki, M. L.; Li,
H.; Lea-Langton, A.; Jones, J. M. Fuel 2010, 89, 2234.
(175) Bicker, M.; Endres, S.; Ott, L.; Vogel, H. J. Mol. Catal. A:
Chem. 2005, 239, 151.
(176) Kong, L.; Li, G.; Wang, H.; He, W.; Ling, F. J. Chem.
Technol. Biotechnol. 2008, 83, 383.
Catalytic One-Pot Production of Small Organics
1677
(177) Sasaki, M.; Adschiri, T.; Arai, K. AIChE J. 2004, 50, 192.
(178) Sasaki, M.; Kabyemela, B.; Malaluan, R.; Hirose, S.;
Takeda, N.; Adschiri, T.; Arai, K. J. Supercrit. Fluids 1998,
13, 261.
(179) Calvo, L.; Vallejo, D. Ind. Eng. Chem. Res. 2002, 41, 6503.
(180) (a) Resende, F. L. P.; Savage, P. E. Ind. Eng. Chem. Res.
2010, 49, 2694. (b) Peterson, A. A.; Vogel, F.; Lachance, R.
P.; Froling, M.; Antal, M. J. Jr.; Tester, J. W. Energy
Environ. Sci. 2008, 1, 32. (c) Yamamura, T.; Mori, T.; Park,
K. C.; Fujii, Y.; Tomiyasu, H. J. Supercrit. Fluids 2009, 51,
43. (d) Lu, Y. J.; Guo, L. J.; Ji, C. M.; Zhang, X. M.; Hao, X.
H.; Yan, Q. H. Int. J. Hydrogen Energy 2006, 31, 822.
(e) Watanabe, M.; Inomata, H.; Arai, K. Biomass Bioenergy
2002, 22, 405.
(181) Sinağ, A.; Kruse, A.; Schwarzkopf, V. Ind. Eng. Chem. Res.
2003, 42, 3516.
(182) Sinağ, A.; Gülbay, S.; Uskan, B.; Güllü, M. J. Supercrit.
Fluids 2009, 50, 121.
(183) Gavrilescu, D. Cellul. Chem. Technol. 2005, 39, 531.
(184) (a) Besson, M.; Gallezot, P. Catal. Today 2000, 57, 127.
(b) Arts, S. J. H. F.; Mombarg, E. J. M.; Van B, H.; Sheldon,
R. A. Synthesis 1997, 597. (c) Heyns, K.; Paulsen, H. Adv.
Carbohydr. Chem. Biochem. 1962, 17, 169.
(185) Patrick, T. A.; Abraham, M. A. Environ. Sci. Technol. 2000,
34, 3480.
(186) Schutt, B. D.; Serrano, B.; Cerro, R. L.; Abraham, M. A.
Biomass Bioenergy 2002, 22, 365.
(187) Schutt, B. D.; Abraham, M. A. Chem. Eng. J. 2004, 103, 77.
(188) (a) Tan, X.; Deng, W.; Liu, M.; Zhang, Q.; Wang, Y. Chem.
Commun. 2009, 7179. (b) Mirescu, A.; Prüße, U. Appl.
Catal., B 2007, 70, 644.
(189) Gaspar, A. R.; Gamelas, J. A. F.; Evtuguin, D. V.;
Pascoal Neto, C. Green Chem. 2007, 9, 717.
(190) Shatalov, A. A.; Evtuguin, D. V.; Pascoal Neto, C.
Carbohydr. Polym. 2000, 43, 23.
Synthesis 2011, No. 11, 1649–1677
© Thieme Stuttgart · New York
Downloaded by: the University of Oxford. Copyrighted material.
REVIEW