Fermentable sugars by chemical hydrolysis of biomass

Article abstract of DOI:10.1073/pnas.0912073107

Abundant plant biomass has the potential to become a sustainable source of fuels and chemicals. Realizing this potential requires the economical conversion of recalcitrant lignocellulose into useful intermediates, such as sugars. We report a high-yielding chemical process for the hydrolysis of biomass into monosaccharides. Adding water gradually to a chloride ionic liquid-containing catalytic acid leads to a nearly 90% yield of glucose from cellulose and 70-80% yield of sugars from untreated corn stover. Ion-exclusion chromatography allows recovery of the ionic liquid and delivers sugar feedstocks that support the vigorous growth of ethanologenic microbes. This simple chemical process, which requires neither an edible plant nor a cellulase, could enable crude biomass to be the sole source of carbon for a scalable biorefinery.

Full text of DOI:10.1073/pnas.0912073107

Fermentable sugars by chemical
hydrolysis of biomass
1
Joseph B. Binder and Ronald T. Raines
Departments of Chemistry and Biochemistry, University of Wisconsin, Madison, WI 53706
Edited by Christopher R. Somerville, University of California, Berkeley, CA, and approved January 5, 2010 (received for review October 19, 2009)
Abundant plant biomass has the potential to become a sustainable
source of fuels and chemicals. Realizing this potential requires the
economical conversion of recalcitrant lignocellulose into useful
intermediates, such as sugars. We report a high-yielding chemical
process for the hydrolysis of biomass into monosaccharides. Add-
ing water gradually to a chloride ionic liquid-containing catalytic
acid leads to a nearly 90% yield of glucose from cellulose and
1948 (20, 21). In the United States, several related processes
using H SO have been developed, typically with 80–90% conver-
2
4
sion of cellulose and hemicellulose into sugars (22–27). Neverthe-
less, the hazards of handling concentrated acids and the
complexities of recycling them have limited the adoption of this
technology.
Less hazardous and more tractable cellulose solvents would
facilitate lignocellulose hydrolysis. Room-temperature ionic
liquids, that is, salts with melting points near or below ambient
temperature, show promise as cellulose solvents for nonwoven
fiber production (28) and chemical derivatization (29–31). Like
concentrated acids, ionic liquids containing chloride, acetate, and
other moderately basic anions disrupt the hydrogen bond network
of cellulose and enable its dissolution (30–34). Although ionic
liquids have been used to pretreat biomass for enzymatic hydro-
lysis (35–37), chemical deconstruction to produce glucose in ionic
liquids has produced only moderate yields (38–43) that contrast
with the nearly quantitative yields of glucose attainable from
cellulose in concentrated acids and other cellulose solvents (44).
Noting that highly efficient hydrolyses employ much higher
water concentrations than those used in ionic-liquid-phase hydro-
lysis, we reasoned that water could play an important role in
delivering high glucose yields. Our investigation of the reactivity
of glucose and cellulose in mixtures of an ionic liquid and water
confirmed this hypothesis and enabled a high-yielding process for
the hydrolysis of cellulose and lignocellulosic biomass. Moreover,
this process generates easily recovered sugars that are superb
feedstocks for microbial growth and biocatalytic ethanol pro-
duction.
7
0–80% yield of sugars from untreated corn stover. Ion-exclusion
chromatography allows recovery of the ionic liquid and delivers sug-
ar feedstocks that support the vigorous growth of ethanologenic
microbes. This simple chemical process, which requires neither an
edible plant nor a cellulase, could enable crude biomass to be the
sole source of carbon for a scalable biorefinery.
biofuel ∣ carbohydrate ∣ ethanol fermentation ∣ ionic liquid ∣ lignocellulose
s the primary component of lignocellulosic biomass, cellulose
is the most abundant organic compound on earth and has the
A
potential to be a renewable source for energy and chemicals. The
1
1
estimated global annual production of biomass is 1 × 10 tons,
2
1
sequestering 2 × 10 J (1, 2). For comparison, annual petroleum
2
0
production amounts to 2 × 10 J, whereas the technically recov-
2
2
erable endowment of conventional crude oil is 2 × 10 J (1).
Hence, in only one decade, Earth’s plants can renew in the form
of cellulose, hemicellulose, and lignin all of the energy stored as
conventional crude oil. The challenge for scientists is to access
these polymers and convert them into fuels and building blocks
for civilization.
Sugars are natural intermediates in the biological and chemical
conversion of lignocellulosic biomass (3–11), but access to sugars
is hindered by the recalcitrance of plant cell walls (4, 12). The
majority of glucose in lignocellulose is locked into highly crystal-
line cellulose polymers. Hemicellulose—a branched polymer of
glucose, xylose, and other sugars—and lignin—a complex aro-
matic polymer—encase the cellulose, fortifying and protecting
the plant. Deriving sugars from this heterogeneous feedstock re-
quires both physical and chemical disruption. Enzymatic methods
of saccharification are the most common, and use physical and
chemical pretreatment processes (13) followed by hydrolysis with
cellulases to produce sugars. The proper combination of pre-
treatment and enzymes for a given feedstock enables high yields
of sugars from both hemicellulose and cellulose components
Results
Cellulose Reactivity in Ionic Liquids. We began by investigating the
intrinsic reactivity of cellulose and sugars under acidic conditions
in ionic liquids. First, we reacted cellulose with H SO and HCl in
2
4
1
-ethyl-3-methylimidazolium chloride ([EMIM]Cl), an ionic
liquid that is an effective solvent for cellulose (39). We observed
the production of 5-hydroxymethylfurfural (HMF), but only mod-
erate yields of glucose (Table 1). HMF inhibits microbial growth
and fermentation (45).
The production of HMF at the expense of glucose suggested to
us nascent glucose was being dehydrated to form HMF. To test
this hypothesis, we reacted glucose in [EMIM]Cl containing vary-
ing amounts of water (Fig. 2, Table S1 in the SI Appendix). In the
absence of both acid and water, glucose was recovered intact. On
the other hand, adding H SO led to the rapid decay of glucose
(
14, 15). Nonetheless, the costs of both pretreatment and en-
zymes [one-third of the cost of ethanol production from cellulose
(
16)] and low rates of hydrolysis are potential drawbacks to
2
4
enzymatic hydrolysis.
into HMF and other products in ionic liquid containing little or
no water. Most significantly, we observed that increasing the
Chemistry provides an alternative means to hydrolyze biomass.
As early as 1819, Braconnot demonstrated that linen dissolved in
concentrated H SO , diluted with water, and heated was trans-
2
4
formed into a fermentable sugar (17, 18). The concentrated acid
plays a dual role in biomass hydrolysis. By disrupting its network
of intra- and interchain hydrogen bonds, strong acids decrystallize
cellulose and make it accessible to reagents (19); and by catalyz-
ing the hydrolysis of glycosidic bonds, strong acids cleave cellu-
lose and hemicellulose into sugars (Fig. 1) (4). Bergius took
advantage of these attributes of HCl in the development of a
commercial process that operated in Germany from 1935 to
Author contributions: J.B.B. and R.T.R. designed research; J.B.B. performed research;
J.B.B. and R.T.R. analyzed data; J.B.B. and R.T.R. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Freely available online through the PNAS open access option.
1
To whom correspondence should be addressed. E-mail: rtraines@wisc.edu.
This article contains supporting information online at www.pnas.org/cgi/content/full/
0912073107/DCSupplemental.
4
516–4521 ∣ PNAS ∣ March 9, 2010 ∣ vol. 107 ∣ no. 10
www.pnas.org/cgi/doi/10.1073/pnas.0912073107

Fig. 1. Hydrolysis reactions of cellulose and xylan. Chemical hydrolysis of cellulose and hemicellulose into monomeric sugars proceeds through oligomers and
is accompanied by side reactions that form furans and other degradation products.
water content to 33 wt% in this same acidic solution enabled
nearly 90% of the glucose to remain intact after 1 h.
at 105 °C to allow hydrolysis of the cellulose into shorter, more
soluble segments (Table 1). After a delay, additional water was
added to stabilize the glucose product.
Why does glucose degrade rapidly under nonaqueous condi-
tions in [EMIM]Cl but not in the presence of a high concentra-
tion of water? Le Chatelier’s principle (46) tells us that water
disfavors dehydrative reactions, such as glucose oligomerization
and conversion into HMF. Additionally, the highly nucleophilic
chloride anions of [EMIM]Cl coordinate strongly to carbohy-
drates (47, 48), accelerating acid-catalyzed dehydration reactions
The timing of water addition was critical for high glucose
yields. When the reaction mixture was diluted to 33% water after
5 min, cellulose precipitated, resulting in low yields. Delaying
dilution until after 10 min prevented cellulose precipitation,
and gradually increasing the water content to 43% within
60 min allowed glucose yields of nearly 90% in 2–4 h. These yields
are nearly twice as high as the previous best in ionic liquids and
approach those achieved through enzymatic hydrolysis. We also
examined the effect of the time for dissolution of cellulose in the
ionic liquid prior to hydrolysis, finding that 6 h was optimal
(Table S2 in the SI Appendix). Longer times probably led to in-
creased byproduct formation (50, 51), which was indicated by dis-
coloration of the reaction mixture, whereas shorter times were
insufficient for complete solvation of the cellulose. With this op-
timized procedure, more concentrated cellulose solutions (10 wt
%) could be hydrolyzed in high yields. Tiny cellulose fibers visible
in these reaction mixtures suggest that glucose yields were slightly
lower due to incomplete cellulose breakdown prior to water ad-
dition. It is likely that cellulose is converted into a mixture of glu-
cose and soluble oligomers within the first 30–60 min of reaction,
and that these oligomers hydrolyze subsequently into glucose.
Monitoring the production of glucose and cellobiose (which
is a glucose dimer) during cellulose hydrolysis revealed that
(
49). High concentrations of water solvate chloride and thus
distract it from interacting with carbohydrates.
Effect of Water on Celluose Reactivity in Ionic Liquids. We suspected
that increasing the water concentration in an [EMIM]Cl reaction
mixture would enhance glucose yields from cellulose. We were,
however, aware that water precipitates cellulose from ionic
liquids (32). Indeed, we observed that a 5 wt% solution of cellu-
lose in [EMIM]Cl formed an intractable gel when the solution
was diluted to 10 wt% water, making homogeneous hydrolysis
of cellulose in ionic liquid/water solutions impossible. Instead,
we attempted to balance cellulose solubility and glucose stability
by adding water gradually during hydrolysis, expecting that cellu-
lose solubility would increase as the reaction progresses. In these
experiments, we chose to use HCl as the acid so as to match the
anion with that of the ionic liquid. [EMIM]Cl containing 5 wt%
cellulose was first treated with HCl and a small amount of water
Table 1. Hydrolysis of cellulose in [EMIM]Cl
Water content, wt%
Cellulose, wt%
HCl, wt%
0′
5′
10′
20′
30′
60′
Time, h
Glucose yield, %
HMF yield, %
5
5
5
20*
20
20
5
5
5
5
5
33
5
5
33
5
5
33
5
5
33
5
5
33
1
1
1
2
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
3
40
45
14
29
31
64
51
36
40
84
81
77
44
86
83
77
38
85
87
89
71
19
17
ND
ND
ND
19
25
30
ND
7
10
8
ND
7
10
13
ND
5
5
5
5
5
20
20
20
20
10
5
5
5
5
5
5
5
5
5
5
20
20
20
20
20
20
20
25
25
25
20
33
33
33
33
20
33
33
43
43
6
7
ND
1
0
Cellulose was reacted in [EMIM]Cl at 105 °C after its dissolution at 105 °C for 12 h. HCl loading is relative to cellulose mass. Water content is relative to the
total mass of the reaction mixture and is shown at time points (minutes) following the initiation of hydrolysis. Yields are molar yields based on HPLC
analysis, and are relative to the glucose monomers contained in the cellulose. ND, not determined.
*
H2SO4.
Binder and Raines
PNAS
∣
March 9, 2010
∣
vol. 107
∣
no. 10
∣
4517

Table 2. Hydrolysis of cellulose in ionic liquids
Cellulose
concentration,
wt%
Glucose
yield, %
Ionic liquid
[
[
EMIM]OAc
EMIM]OAc
2
5
2
2
0
0
0
0
½EMIMꢁNO
3
1
,3-dimethylimidazolium
dimethylphosphate
,3-dimethylimidazolium
dimethylphosphate
1
5
0
[
EMIM]Br
2
2
2
5
5
5
5
4
0
5
66
73
69
46
½
BMIMꢁBF
4
[
[
EMIM]OTf
BMIM]Cl
-butyl-4-methylpyridinium chloride
-ethylpyridinium chloride
-ethyl-2,3-dimethylimidazolium
chloride
1
1
1
Cellulose was reacted in ionic liquid for 3 h at 105 °C after mixing at 105 °C
for 6 h. HCl loading was 20 wt% relative to cellulose weight. The water
content of the reaction began at 5 wt% and was increased as follows:
Fig. 2. Acid-catalyzed degradation of glucose in [EMIM]Cl. In acidic [EMIM]
Cl, glucose (diamonds) disappears rapidly at 100 °C, forming HMF (squares)
and other degradation products. Increased water content slows glucose loss.
Reaction conditions: Glucose, 10 wt%; H2SO4, 4 wt% relative to glucose; H2O,
2
0% (10 min), 25% (20 min), 33% (30 min), 43% (60 min). Yields are
molar yields based on HPLC analysis and are relative to the glucose
monomers contained in the cellulose.
0
–33 wt% relative to the total mass of the reaction mixture. Product levels
are reproducible to within ꢀ4%.
of glucose based on the xylan and cellulose content of the stover
(
Table 3). Dilution of the reaction mixture to 70% water caused
cellobiose concentrations peaked at 1 h and decayed as glucose
concentrations increased (Fig. 3). Experiments with other ionic
liquids indicate that chloride-based ionic liquids alone are able
to balance cellulose solubility and hydrolytic activity through
strong interactions with cellulose coupled with weak basicity
precipitation of unhydrolyzed polysaccharides and lignin. These
residues were then dissolved in [EMIM]Cl and subjected to an
identical second-stage hydrolysis, which released additional
xylose and glucose, leaving behind lignin-containing solids.
Combined, these two steps resulted in a 70% yield of glucose
and a 79% yield of xylose using only simple chemical reagents.
Given the well-documented versatility of ionic liquids as biomass
solvents, we anticipate that this process will be amenable to other
biomass sources, such as wood and grasses. The brown residue
remaining after two-stage corn stover hydrolysis likely includes
lignin, insoluble polysaccharides, any insoluble hydrolysis
byproducts, protein, and ash. We have not yet characterized
the lignin byproduct, but it is likely to resemble lignin produced
by other acid hydrolysis processes, e.g., Klason lignin (43, 52).
(Table 2).
Hydrolysis of Crude Biomass. Complex and heterogeneous, lignocel-
lulosic biomass presents a more significant challenge for hydro-
lysis than does cellulose. In addition to intractable crystalline
cellulose, lignocellulosic biomass such as corn stover includes
protective hemicellulose and lignin, heterogeneous components
that are major obstacles to many biomass hydrolysis processes
(
4, 12). Nonetheless, chloride ionic liquids are excellent solvents
for lignocellulosic biomass, suggesting that they would easily
disrupt these polymeric barriers. Moreover, preliminary experi-
ments demonstrated that xylan, a hemicellulose, was readily hy-
drolyzed under our conditions, producing xylose in 77% yield.
We attempted to extend our process for cellulose hydrolysis to
crude lignocellulosic biomass. Untreated corn stover that had
been mixed with [EMIM]Cl was hydrolyzed with 10 wt% HCl
at 105 °C with the same water-dilution process used for pure cel-
lulose. This process produced a 71% yield of xylose and 42% yield
Sugar and Solvent Recovery. A practical biomass hydrolysis process
requires efficient means for sugar and solvent recovery. We found
that ion-exclusion chromatography enables separation of the
sugars and ionic liquid from the corn stover hydrolysis reaction
mixture. In this technique, a mixture containing electrolyte and
nonelectrolyte solutes is separated by passing it through a
charged resin (53). Charged species, such as the ionic liquid,
are excluded from the resin, while nonelectrolytes, such as sugars,
are retained. Nonpolar species such as HMF and furfural are
adsorbed more strongly than sugars, and elute later. Passing
the corn stover hydrolyzate through a column of [EMIM]-
exchanged Dowex® 50 resin allowed laboratory-scale separation
of the ionic-liquid solvent from the sugars, with >95% recovery of
the ionic liquid, 94% recovery of glucose, and 88% recovery of
xylose. Additionally, this chromatographic step removed inhibi-
tory compounds such as HMF and furfural. We suspect that
ionic-liquid and sugar recovery are limited by the difficulties
intrinsic in small-scale separations and would be improved upon
scale-up.
Microbial Fermentation of Biomass Hydrolyzate. To support biocon-
version, biomass hydrolyzate sugars must be free of contaminants
that inhibit microbial growth and fermentation. We found that
sugars derived from corn stover through our process are excellent
feedstocks for an ethanologenic bacterium and yeast. Whereas
wild-type Escherichia coli ferments a range of sugars into a
Fig. 3. Production of glucose, HMF, and cellobiose production during
cellulose hydrolysis under optimized reaction conditions. Product levels
are reproducible to within ꢀ4%.
4
518
∣
www.pnas.org/cgi/doi/10.1073/pnas.0912073107
Binder and Raines

Table 3. Hydrolysis of untreated corn stover in [EMIM]Cl
Water content, wt%
Stover, wt%
5
Stage
HCl, wt%
0′
5′
10′
20′
30′
60′
Time, h
Glucose yield, %
Xylose yield, %
1
2
20
20
5
5
5
5
20
20
25
25
33
33
43
43
2.5
3.0
42
28
70
19
47
66
17
21
25
27
37
42
44
48
71
8
79
74
1
75
60
73
80
82
5
Overall
1
1
0
0
1
2
10
10
5
5
5
5
20
20
25
25
33
33
43
43
3.5
3.0
Overall
1
1
1
1
2
2
2
2
10
5
5
20
25
33
43
1.0
1.5
2.0
2.5
1.5
2.0
2.5
3.0
10
5
5
20
25
33
43
5
5
5
Untreated corn stover was reacted in [EMIM]Cl at 105 °C after its dissolution at 105 °C for 6 h. HCl loading is relative to stover weight. Water content is
relative to the total mass of the reaction mixture and is shown at time points (minutes) following the initiation of hydrolysis. Yields are molar yields based
on HPLC analysis and are relative to the glucose and xylose monomers contained in the stover.
mixture of ethanol and organic acids, the engineered KO11 strain
produces ethanol selectively (54). Serving as the sole carbon
source, the hydrolyzate sugars from corn stover enabled aerobic
growth of E. coli KO11 at a rate comparable to that of a control
glucose/xylose mixture (Fig. 4A). Moreover, under oxygen-
deficient conditions, E. coli KO11 produced a ð79 ꢀ 4Þ% yield
of ethanol from stover hydrolyzate sugars and a ð76 ꢀ 3Þ% yield
from pure glucose and xylose, demonstrating that sugars from our
hydrolysis process can be converted readily into ethanol.
Engineered bacteria show promise for biofuel production, but
yeast fermentation predominates today (55, 56). Pichia stipitis,
which has an innate ability to ferment xylose, is an especially
viable candidate for bioconversion of lignocellulose-derived
sugars (57–59). Corn stover hydrolyzate sugars are an excellent
carbon source for the growth of this yeast (Fig. 4B), and P. stipitis
efficiently converts them into ethanol. Fermenting xylose and
glucose, the yeasts produced a ð70 ꢀ 2Þ% yield of ethanol from
hydrolyzate and a ð72 ꢀ 1Þ% yield from pure sugars.
Discussion
We have demonstrated an efficient system for polysaccharide hy-
drolysis as well as means to separate and ferment the resulting
sugars. By balancing cellulose solubility and reactivity with water,
we produce sugars from lignocellulosic biomass in yields that are
severalfold greater than those achieved previously in ionic liquids
and approach those of enzymatic hydrolysis. Furthermore, the hy-
drolyzate products are readily converted into ethanol by micro-
organisms. Together, these steps comprise an integrated process
for chemical hydrolysis of biomass for biofuel production (Fig. 5).
First, lignocellulosic biomass such as corn stover is decrystallized
through mixing with [EMIM]Cl. With their defense against
chemical assault breached by the ionic liquid, the hemicellulose
and cellulose are hydrolyzed by treatment with HCl and water.
The residual lignin and cellulose solids are subjected to a second
hydrolysis, while the liquid hydrolyzate is separated through ion-
exclusion chromatography. Ionic liquid recovered in the ion-
exclusion step is stripped of water and recycled, while hydrolyzate
sugars are fermented into fuels and other products.
In comparison to extant enzymatic and chemical processes to
Fig. 4. Aerobic growth of ethanologenic microbes on corn stover hydroly-
zate and pure sugars as their sole carbon source. (A) Escherichia coli strain
KO11: hydrolyzate (⦁), 5 replicates; sugars (○), 20 replicates. (B) Pichia stipitis:
hydrolyzate (⦁), 2 replicates; sugars (○), 6 replicates. Data are mean values
biomass hydrolysis, ours has several attractive features. Like con-
centrated acid processes, it uses inexpensive chemical catalysts
rather than enzymes and avoids an independent pretreatment
step. Working in concert, [EMIM]Cl and HCl produce high sugar
(
ꢀSE) at 0.5 h intervals. Under anaerobic conditions, the cultures produced
yields in hours at just 105 °C, whereas enzymatic hydrolysis can
take days (16) and many pretreatment methods require tempera-
tures of 160–200 °C (13). Also, lignocellulose solubilization by the
ethanol with the following yields based on sugar conversion (three replicates
each). Escherichia coli strain KO11: hydrolyzate, ð79 ꢀ 4Þ%; sugars,
ð76 ꢀ 3Þ%. Pichia stipitis: hydrolyzate, ð70 ꢀ 2Þ%; sugars, ð72 ꢀ 1Þ%.
Binder and Raines
PNAS
∣
March 9, 2010
∣
vol. 107
∣
no. 10
∣
4519

Fig. 5. Integrated process for biofuel production using ionic-liquid biomass hydrolysis. Ionic-liquids solvents enable efficient biomass decrystallization
and hydrolysis. Ion-exclusion chromatography separates the ionic liquids for recycling and the hydrolyzate sugars for fermentation. (Photograph copyright
2
001, Department of Energy/National Renewable Energy Laboratory; credit, Jim Yost.)
ionic liquid allows processing at high concentrations, which can
be a problem in enzymatic hydrolysis. On the other hand, our pro-
cess improves on typical acid hydrolysis methods by avoiding the
use of hazardous concentrated acid. Using catalytic amounts of
dilute acid removes the complexity and danger of recycling large
volumes of concentrated acid. The ionic liquid used in its place is
likely to be far easier to handle. Despite this difference, our
process is similar to commercial processes using concentrated
acid hydrolysis (20, 27) and consequently can exploit proven
engineering and equipment for facile scale-up, particularly for se-
parations and recycling (vide infra).
Still, important challenges remain for implementation of an
ionic-liquid biomass hydrolysis process. Highly viscous bio-
mass-ionic-liquid mixtures might require special handling, and
larger scale fermentation of hydrolyzate sugars might reveal
the presence of inhibitors not detected in our demonstration ex-
periments. Moreover, the sugar concentration resulting from
stover hydrolysis (about 1%) is too low for practical fermentation
and is decreased even further during the chromatographic se-
paration, leading to water-evaporation costs. Methods allowing
a higher starting biomass loading and strategies to concentrate
rather than dilute the sugars during separation from the ionic
liquid would overcome this problem.
Our chemical hydrolysis method offers flexibility for an
integrated biomass conversion process. Because the ionic-liquid
solvent makes biomass polysaccharides readily accessible for
chemical reactions, this process is likely to be compatible with
a broad range of biomass feedstocks. Downstream, the sugars
produced by lignocellulose hydrolysis are flexible feedstocks
for production of a nearly infinite range of fuels and chemicals.
E. coli, which readily use the hydrolyzate sugars, have been
engineered to produce not only fuel ethanol but also 1-butanol,
2
-butanol, branched alcohols, fatty acids, isoprenoids, and even
hydrogen (63–65). Furthermore, the aqueous stream of sugars
can also be converted by catalytic processes into fuels (66) or che-
mical intermediates. In contrast with enzymatic hydrolysis reac-
tions that often require coupling to fermentation (simultaneous
saccharification and fermentation) to prevent product inhibition
(
15, 16), this chemical process can be paired with any downstream
conversion. Finally, the lignin recovered from ionic-liquid bio-
mass hydrolysis could be a valuable coproduct. After washing
to remove ionic liquid, this residue could be burned to generate
process heat and electricity. The lignin residue from similar acid
hydrolysis processes has a heating value similar to that of wood,
making it a useful feed for a solid fuel boiler (62). Alternatively,
the lignin could be an excellent feedstock for high-value products
(
43, 67). As a result, our process, which uses simple chemical re-
Separations and ionic-liquid recycling could pose additional
challenges to commercialization. Our separation method—
ion-exclusion chromatography—is complex, especially with acid,
ionic liquid, multiple sugars, and other soluble species in the
hydrolyzate. Nonetheless, this technique is already used in con-
tinuous carbohydrate processing on a large scale, notably for the
separation of glucose and fructose in corn wet mills (60). The se-
paration of electrolytes such as ionic liquids from sugars is likely
to be easier than this industrial application, and is currently being
scaled up by BlueFire Ethanol, Inc. with sulfuric acid in two
demonstration plants. Lessons from these facilities can aid in
the design of efficient processes based on ionic-liquid hydrolysis.
Ionic liquids are expensive and will likely remain so in compar-
ison to other solvents, even as expanded use reduces their prices
to near those of their raw materials [∼$2.50∕kg for dialkylimida-
zolium chlorides (61)]. Consequently, virtually complete solvent
recovery will be required to make biomass processing with ionic
liquids economical. For instance, one target for the cost of ionic
liquid lost in biomass processing might be $0.13∕L (¼$0.50∕gal)
ethanol, a near-term target for the cost of hydrolytic enzymes. An
optimized process using 1 kg of ionic liquid for each kilogram of
biomass processed would then need 98% recovery of the ionic
liquid in each cycle (see SI Text). This level of solvent recovery
might be feasible in an optimized process, considering that re-
ported sulfuric acid losses in a pilot-scale concentrated acid pro-
cess are only about 38 g∕kg biomass, despite the much lower
incentive to recover sulfuric acid (62). We note, however, that
the accumulation of impurities in an ionic-liquid recycle loop
could be problematic. Future studies will need to address
this issue.
agents to overcome biomass recalcitrance and liberate valuable
sugars, has the potential to underlie a biorefinery.
Materials and Methods
A description of general materials and methods, the hydrolysis of cellulose
and xylan, the recovery of sugars and [EMIM]Cl, the reaction of glucose, ex-
periments with nonchloride ionic liquids, microbial growth and fermenta-
tion, the economics of ionic liquid usage, and Tables S1 and S2 are
provided in SI Text.
Representative Procedure for Hydrolysis of Corn Stover. Corn stover (26.7 mg,
5
4 μmol glucose units, 44 μmol xylose units) and [EMIM]Cl (502 mg) were
mixed at 105 °C for 6 h. To this mixture was added aqueous HCl (1.66 M,
9 μL, equivalent to 5 mg conc. HCl), and the reaction mixture was stirred
2
vigorously at 105 °C. After 10 min, deionized water (100 μL) was added with
stirring, followed by additional aliquots at 20 min (50 μL), 30 min (75 μL), and
60 min (125 μL). After a total reaction time of 2.5 h, the solution was diluted
with water (750 μL). Insoluble materials were removed by centrifugation,
rinsed twice with water (200 μL), and dried. The liquid products (2.046 g) were
analyzed by HPLC (2.0 mg∕g glucose, 42% yield; 2.3 mg∕g xylose, 71% yield).
The brown solids from the first hydrolysis were then heated with [EMIM]Cl
(
1
306 mg) at 105 °C for 4.5 h. To this mixture was added aqueous HCl (1.66 M,
4.5 μL, equivalent to 2.5 mg conc. HCl), and the reaction mixure was vigor-
ously stirred at 105 °C. After 10 min, deionized water (50 μL) was added with
stirring, followed by an additional 25 μL water at 20 min, 67.5 μL water at
3
0 min, and 70 μL water at 60 min. After 3 h total reaction time, the solution
was diluted with water (300 μL) and centrifuged to sediment insoluble
materials. The liquid products (770 mg) were analyzed by HPLC
(
3.56 mg∕g glucose, 28% yield; 0.7 mg∕g xylose, 8% yield). For the two-step
process, the overall yield of glucose was 70% and the overall yield of xylose
was 79%.
In other cases, aliquots of the reaction mixture were removed periodically
for HPLC analysis.
4
520
∣
www.pnas.org/cgi/doi/10.1073/pnas.0912073107
Binder and Raines

ACKNOWLEDGMENTS. We are grateful to J.E. Holladay for helpful conversa-
tions on biomass chemistry in ionic liquids; B.R. Caes for contributive discus-
sions; J.J. Blank and A.V. Cefali for experimental assistance; W.D. Marner,
G.N. Phillips, and T.J. Rutkoski for assistance with bacterial growth studies;
and T.W. Jeffries and T.M. Long for assistance with yeast fermentation. This
work was supported by the Department of Energy through the Great Lakes
Bioenergy Research Center. J.B.B. was supported by an National Science
Foundation Graduate Research Fellowship.
1
. US National Petroleum Council (2007) Facing the Hard Truths about Energy (US Na-
tional Petroleum Council, Washington, DC), pp 110, 156–158.
36. Kilpelainen I, et al. (2007) Dissolution of wood in ionic liquids. J Agric Food Chem
55:9142–9148.
2
. Smeets EMW, Faaij APC, Lewandowski IM, Turkenburg WC (2007) A bottom-up
assessment and review of global bio-energy potentials to 2050. Prog Energ Combust
37. Singh S, Simmons BA, Vogel KP (2009) Visualization of biomass solubilization and
cellulose regeneration during ionic liquid pretreatment of switchgrass. Biotechnol
Bioeng 104:68–75.
38. Fanselow M, Holbrey JD, Seddon KR (2007) Eur Patent Appl 1,860,201.
39. Li C, Zhao ZK (2007) Efficient acid-catalyzed hydrolysis of cellulose in ionic liquid. Adv
Synth Catal 349:1847–1850.
3
3:56–106.
. Hahn-Hägerdal B, Galbe M, Gorwa-Grauslund MF, Lidén G, Zacchi G (2006) Bio-ethanol
The fuel of tomorrow from the residues of today. Trends Biotechnol 24:549–556.
3
4
—
.
(2005) Polysaccharides: Structural Diversity and Functional Versatility, ed Dumitriu S
(
Marcel Dekker, New York), 2nd Ed.
4
4
4
0. Li C, Wang Q, Zhao ZK (2008) Acid in ionic liquid: An efficient system for hydrolysis of
lignocellulose. Green Chem 10:177–182.
1. Rinaldi R, Palkovits R, Schüth F (2008) Depolymerization of cellulose using solid
catalysts in ionic liquids. Angew Chem Int Edit 47:8047–8050.
2. Vanoye L, Fanselow M, Holbrey JD, Atkins MP, Seddon KR (2009) Kinetic model for the
hydrolysis of lignocellulosic biomass in the ionic liquid, 1-ethyl-3-methyl-imidazolium
chloride. Green Chem 11:390–396.
5
6
. Peters D (2007) Raw materials. Adv Biochem Eng Biotechnol 105:1–30.
. Lange J-P (2007) Lignocellulose conversion: An introduction to chemistry, process and
economics. Biofuel Bioprod Bior 1:39–48.
. Stephanopoulos G (2007) Challenges in engineering microbes for biofuels production.
Science 315:801–804.
. Chheda JN, Huber GW, Dumesic JA (2007) Liquid-phase catalytic processing of
biomass-derived oxygenated hydrocarbons to fuels and chemicals. Angew Chem Int
Edit 46:7164–7183.
. Van Haveren J, Scott EL, Sanders J (2008) Bulk chemicals from biomass. Biofuel Bioprod
Bior 2:41–57.
7
8
4
4
4
3. Sievers C, et al. (2009) Ionic-liquid-phase hydrolysis of pine wood. Ind Eng Chem Res
4
8:1277–1286.
9
4. Cao NJ, Xu Q, Chen LF (1995) Acid hydrolysis of cellulose in zinc chloride solution. Appl
Biochem Biotechnol 51–52:21–28.
5. Almeida JRM, Betilsson M, Gorwa-Grauslund MF, Gorsich S, Liden G (2009) Metabolic
effects of furaldehydes and impacts on biotechnological processes. Appl Microbiol
Biotechnol 82:625–638.
1
1
1
1
1
0. Christensen CH, Rass-Hansen J, Marsden CC, Taarning E, Egeblad K (2008) The renew-
able chemicals industry. ChemSusChem 1:283–289.
1. Binder JB, Raines RT (2009) Simple chemical transformation of lignocellulosic biomass
into furans for fuels and chemicals. J Am Chem Soc 131:1979–1985.
2. Himmel ME, et al. (2007) Biomass recalcitrance: Engineering plants and enzymes for
biofuels production. Science 315:804–807.
3. Mosier N, et al. (2005) Features of promising technologies for pretreatment of
lignocellulosic biomass. Bioresource Technol 96:673–686.
4
4
6. Le Chatelier HL (1884) On a general expression of the laws of chemical equilibrium
(Translated from French). CR Hebd Seances Acad Sci 99:786–789.
7. Remsing RC, Swatloski RP, Rogers RD, Moyna G (2006) Mechanism of cellulose dissolu-
tion in the ionic liquid 1-n-butyl-3-methylimidazolium chloride: A 13C and 35∕37Cl NMR
relaxation study on model systems. Chem Commun 1271–1273.
4. Wyman CE, et al. (2005) Comparative sugar recovery data from laboratory scale
application of leading pretreatment technologies to corn stover. Bioresource Technol
0
4
4
8. Remsing RC, et al. (2008) Solvation of carbohydrates in N -dialkylimidazolium ionic
liquids: A multinuclear NMR spectroscopy study. J Phys Chem B 112:11071–11078.
9. Mascal M, Nikitin EB (2008) Direct, high-yield conversion of cellulose into biofuel.
Angew Chem Int Edit 47:7924–7926.
9
6:2026–2032.
1
1
1
5. Lau MW, Dale BE (2009) Cellulosic ethanol production from AFEX-treated corn
stover using Saccharomyces cerevisiae 424A(LNH-ST). Proc Natl Acad Sci USA
1
06:1368–1373.
50. Ebner G, Schiehser S, Potthast A, Rosenau T (2008) Side reaction of cellulose with com-
mon 1-alkyl-3-methylimidazolium-based ionic liquids. Tetrahedron Lett 49:7322–7324.
51. Vitz J, Erdmenger T, Hänsch C, Schubert US (2009) Extended dissolution studies of cel-
lulose in imidazolium based ionic liquids. Green Chem 11:417–424.
52. Hatfield RD, Jung HJG, Ralph J, Buxton DR, Weimer PJ (1994) A comparison of the
insoluble residues produced by the Klason lignin and acid detergent lignin procedures.
J Sci Food Agr 65:51–58.
53. Asher DR (1956) Sugar purification by ion exclusion. Ind Eng Chem 48:1465–1466.
54. Dien BS, Cotta MA, Jeffries TW (2003) Bacteria engineered for fuel ethanol produc-
tion: Current status. Appl Microbiol Biotechnol 63:258–266.
55. Ho NWY, Chen Z, Brainard AP, Sedlak M (1999) Successful design and development of
genetically engineered Saccharomyces yeasts for effective cofermentation of glucose
and xylose from cellulosic biomass to fuel ethanol. Adv Biochem Eng Biotechnol
6. Aden A, et al. (2002) Lignocellulosic Biomass to Ethanol Process Design and Economics
Utilizing Co-Current Dilute Acid Prehydrolysis and Enzymatic Hydrolysis for Corn
Stover (National Renewable Energy Laboratory, Department of Energy, Golden, CO).
7. Braconnot H (1819) Memoir on the conversion of wood particles in rubber, in sugar,
and in a special natural acid, by means of sulfuric acid; conversion of the same woody
substance in ulmin by potash (Translated from French). Ann Chim Phys 12:172–195.
8. Stern AL (1895) X. Contributions to the chemistry of cellulose. I. Celluose-sulphuric
acid, and the products of its hydrolysis. J Chem Soc, Trans 67:74–90.
9. Xiang Q, Lee YY, Pettersson PO, Torget RW (2003) Heterogeneous aspects of acid
hydrolysis of α-cellulose. Appl Biochem Biotechnol 105:505–514.
1
1
2
2
0. Bergius F (1937) Conversion of wood to carbohydrates. Ind Eng Chem 29:247–253.
1. Schoenemann K (1958) The perfecting of wood hydrolysis in the Rheinau process.
Chim Ind (Paris) 80:140–150.
6
5:163–192.
2
2
2
2. Dunning JW, Lathrop EC (1945) The saccharification of agricultural residues: A contin-
uous process. Ind Eng Chem 37:24–29.
3. Tsao GT, Ladisch MR, Voloch M, Bienkowski PR (1982) Production of ethanol and
chemicals from cellulosic materials. Process Biochem 17:34–38.
4. Wright JD, D’Agincourt CG (1984) Evaluation of Sulfuric Acid Hydrolysis Processes for
Alcohol Fuel Production (Solar Energy Research Institute, Department of Energy, Gold-
en, CO).
5
6. Van Maris AJA, et al. (2007) Development of efficient xylose fermentation in Sacchar-
omyces cerevisiae: Xylose isomerase as a key component. Adv Biochem Eng Biotechnol
1
08:179–204.
7. Jeffries TW (1985) Emerging technology for fermenting D-xylose. Trends Biotechnol
:208–212.
8. Agbogbo FK, Haagensen FD, Milam D, Wenger KS (2008) Fermentation of acid-
pretreated corn stover to ethanol without detoxification using Pichia stipitis. Appl Bio-
chem Biotech 145:53–58.
5
5
3
2
5. Wright JD, Power AJ (1987) Comparative technical evaluation of acid hydrolysis
processes for conversion of cellulose to alcohol. Energy from Biomass and Wastes,
5
9. Agbogbo FK, Coward-Kelly G (2008) Cellulosic ethanol production using the naturally
occurring xylose-fermenting yeast, Pichia stipitis. Biotech Lett 30:1515–1524.
0. Colonna WJ, Clarke MA, Godshall MA, Cleary M, White JS (2004) Kirk-Othmer Ency-
clopedia of Chemical Technology, ed Kroschwitz JI (Wiley-Interscience, New York)
p 485.
(
Institute of Gas Technology, Chicago, IL), 10, pp 949–971.
2
6. Farina GE, Barrier JW, Forsythe ML (1988) Fuel alcohol production from agricultural
lignocellulosic feedstocks. Energ Source Part A 10:231–237.
6
2
2
7. Farone WA, Cuzens JE (1998) US Patent 5,726,046.
8. Hermanutz F, Meister F, Uerdingen E (2006) New developments in the manufacture of
cellulose fibers with ionic liquids. Chem Fibers Int 56:342–344.
9. Graenacher C (1934) US Patent 1,943,176.
0. Zhu S, et al. (2006) Dissolution of cellulose with ionic liquids and its application: A mini-
review. Green Chem 8:325–327.
6
6
6
1. Reddy RG (2006) Ionic liquids: How well do we know them? J Phase Equilib Diff
2
7:210–211.
2
3
2. Cuzens JC, Miller JR (1997) Acid hydrolysis of bagasse for ethanol production. Renew
Energ 10:285–290.
3. Maeda T, Sanchez-Torres V, Wood TK (2007) Enhanced hydrogen production from
glucose by metabolically engineered Escherichia coli. Appl Microbiol Biotechnol
3
1. El Seoud OA, Koschella A, Fidale LC, Dorn S, Heinse T (2007) Applications of ionic
liquids in carbohydrate chemistry: A window of opportunities. Biomacromolecules
8
2. Swatloski RP, Spear SK, Holbrey JD, Rogers RD (2002) Dissolution of cellose with ionic
liquids. J Am Chem Soc 124:4974–4975.
7
7:879–890.
:2629–2647.
6
4. Atsumi S, Hanai T, Liao JC (2008) Non-fermentative pathways for synthesis of
3
3
3
3
branched-chain higher alcohols as biofuels. Nature 451:86–89.
65. Fortman JL, et al. (2008) Biofuel alternatives to ethanol: Pumping the microbial well.
Trends Biotechnol 26:375–381.
66. Kunkes EL, et al. (2008) Catalytic conversion of biomass to monofunctional hydrocar-
bons and targeted liquid-fuel classes. Science 322:417–421.
67. Lora JH, Glasser WG (2002) Recent industrial applications of lignin: A sustainable
alternative to nonrenewable materials. J Polym Environ 10:39–48.
3. Mikkola J-P, et al. (2007) Ultrasound enhancement of cellulose processing in ionic
liquids: From dissolution towards functionalization. Green Chem 9:1229–1237.
4. Pinkert A, Marsh KN, Pang S, Staiger MP (2009) Ionic liquids and their interaction with
cellulose. Chem Rev 109:6712–6728.
5. Dadi AP, Varanasi S, Schall CA (2006) Enhancement of cellulose saccharification kinetics
using an ionic liquid pretreatment step. Biotechnol Bioeng 95:904–910.
Binder and Raines
PNAS
∣
March 9, 2010
∣
vol. 107
∣
no. 10
∣
4521