-
99
9
Research ArticleReceived: 8 July 2012 Revised: 3 September 2012
Accepted: 16 September 2012 Published online in Wiley Online
Library: 14 November 2012
(wileyonlinelibrary.com) DOI 10.1002/jctb.3963
Invasive biomass valorization: environmentallyfriendly processes
for obtaining secondgeneration bioethanol and saccharides fromUlex
europusIria Ana Ares-Peon,a Aloia Roman,a Gil Garrotea and Juan
Carlos Parajob
Abstract
BACKGROUND: Ulex europus (UE) is a widespread invasive shrub
species causing economic problems and environmentalhazards. This
work deals with the valorization of UE by hydrothermal processing
(to obtain hemicelluloses-derived saccharides)followed by
simultaneous saccharification and fermentation of the resulting
solids for manufacturing second generationbioethanol.
RESULTS: Hydrothermal processing of UE resulted in the
solubilization of up to 21.5 wt% of the oven-dry raw material,
leadingto the formation of hemicelluloses-derived saccharides as
major reaction products. Treatments at various severities
resultedin processed solids with enhanced susceptibility to
enzymatic hydrolysis, allowing cellulose to glucose conversions up
to 87%.Simultaneous saccharification and fermentation of solids
pretreated under selected conditions, performed at various
chargesand solid loadings, resulted in bioethanol conversions up to
82% of the stoichiometric amount, with volumetric
concentrationshigher than 30 g ethanol L-1.
CONCLUSION: Hydrothermal processing of UE followed by
simultaneous saccharification and fermentation of pretreated
solidswas suitable for the selective separation of hemicelluloses
as soluble saccharides and for the manufacture of second
generationbioethanol at high yield from the pretreated solids.c
2012 Society of Chemical Industry
Keywords: autohydrolysis; bioethanol; simultaneous
saccharification and fermentation; Ulex europus; valorization
INTRODUCTIONGorse (Ulex europus, UE) is a densely packed,
prickly evergreenshrub less than 4 m tall. The International Union
for Conservationof Nature recognizes gorse as one of the top 100
worst invasivespecies in the world. Gorse is abundant in Galicia
(North West ofSpain), where it is spread over about 30% of the
territory.1 Theproduction of gorse has been estimated at 0.71.9
metric tonsha-1 year-1,2 which corresponds to 6.2105 to 1.7106
metrictons gorse year-1 in our region. UE may grow altering the
soilproperties, increasing erosion and becoming a significant
firehazard,andoutcompeteotherplantswithsubsequent reduction
inproductivity. Finding a profitable way for gorse
valorizationwouldresult in environmental benefit and in improved
development ofsustainable agricultural and silvicultural
exploitation.
Owing to its woody nature, gorse can be included among
thelignocellulosic materials (LCMs), 1,3 whose structural
componentsinclude cellulose (a linear polymer made from glucose
structuralunits), hemicelluloses (branched polymers made up of
sugars andsubstituents) and lignin (a tridimensional polymer made
up fromoxygenated phenylpropane structural units). Additionally,
LCMscontain non-structural components (including extractives,
ashesand protein), which are of minor interest for the objectives
ofthis study. LCMs can be utilized according to the biorefinery
concept46 to obtain various fractions suitable for
specificpurposes such as the manufacture of fuels and/or
chemicals.
LCMs fractionation may be carried out starting with
anautohydrolysis stage, in which the feedstock is treatedwith hot,
compressed water to cause a number of effects,including
hemicelluloses solubilization and structural alterationof the
cellulose remaining in processed solids. Under selectedconditions,
hemicelluloses can be selectively converted intosoluble saccharides
(mainly of oligomeric nature, but alsomonosaccharides). When
hardwoords or agricultural materialsare employed as autohydrolysis
substrates, xylooligomers (XO)
are usually the target products.79 XO can be used for anumber of
applications in the food, pharmaceutical or
chemicalindustries,10,11 or subjected to hydrolysis to yield
solutions
Correspondence to: G. Garrote, Department of Chemical
Engineering, Facultyof Science, University of Vigo (Campus
Ourense), As Lagoas, 32004 Ourense,Spain. E-mail: [email protected]
a Department of Chemical Engineering, Faculty of Science,
University of Vigo(Campus Ourense), As Lagoas, 32004, Ourense,
Spain
b CITI (Centro de Investigacion, Transferencia e Innovacion),
University of Vigo,Tecnopole, San Cibrao das Vinas, Ourense,
Spain
J Chem Technol Biotechnol 2013; 88: 9991006 www.soci.org c 2012
Society of Chemical Industry
-
10
00
www.soci.org IA Ares-Peon et al.
of hemicellulosic sugars12,13 suitable for the manufacture
offermentation media (for example, for lactic acid or
bioethanol
production).1416
The solids from hydrothermal processing of LCMs are mainlymade
up of cellulose and lignin, and are expected to presentincreased
susceptibility to further processing, for example by
delignification17,18 or enzymatic methods.1922
Alternatively, theaqueousprocessingofLCMsmaybeconceivedas a
pretreatment for enzymatic hydrolysis, as it meets a numberof
favourable requirements,23 including savings in chemicals,limited
corrosion, ability to cause structural alteration of LCMs,high
selectivity with respect to cellulose solubilization and
thepossibility of obtaining valuable products from
hemicelluloses.
The production of bioethanol from pretreated solids canbe
carried out by consecutive stages of enzymatic hydrolysisand
fermentation (method known as separate hydrolysis andfermentation,
SHF) or by a single stage with enzymes andmicroorganisms (referred
to as simultaneous saccharification andfermentation, SSF). SSF has
comparative advantages derivedfrom the lower operational costs,
limited contamination risk anddecreased enzyme inhibition caused by
glucose and cellobiose.Because of this, SSF has provided improved
results over those of
SHF.2426
This work deals with the valorization of UE by
aqueousfractionation (to recover hemicelluloses as soluble
saccharides),andwiththefurtherutilizationof
thepretreatedsolidsassubstratesfor bioethanol production by
SSF.
MATERIAL AND METHODSRaw materialUE samples were collected
locally, milled to pass an 8 mm screen,air-dried, homogenized in a
single lot to avoid compositionaldifferences between samples, and
stored in a dark and dry placeuntil use.
Analysis of raw materialUE samples from thehomogenized
lotweremilled to aparticle sizeless than 0.5mmand subjected to the
following assays: extractives(TAPPI T-264-cm-97 method), moisture
(T-264-cm-97 method),ashes (T 211 om-93 method), and quantitative
acid hydrolysis(T-249-cm-85 method). The liquid phase from the
latter assaywas analyzed by high performance liquid chromatography
(HPLC)for sugars and acetic acid (conditions: detector, refractive
index;column, Aminex HPX-87H; mobile phase, 0.01 mol L-1 H2SO4;
flowrate,0.6mLmin-1).
Theconcentrationsofglucose,xylose,arabinoseand acetic acid were
employed to calculate the sample contentsof glucan, xylan,
arabinosyl substituents and acetyl groups. Thesolid phase from the
quantitative acid hydrolysis was consideredas Klason lignin after
correction for ashes. Analyses were carriedout in
quadruplicate.
AutohydrolysisWater andUE samplesweremixedat aproportionof 8
kgwater perkg wood and heated in a stainless steel reactor of 1
gallon volume(Parr InstrumentsCompany,Moline, Illinois) following
thestandardtemperature profile19 to reach the target temperature
(denotedTMAX, in the range 150240
C). Then, the medium was cooled byflowing water through an
internal stainless steel loop. Solids wereseparated by filtration,
washed with water and employed for solid
yield determination (SY, g solid recovered per 100 g raw
material,oven dry basis) and analyzed.
The operational range was selected to cover the
conditionsleading from little to extensive hemicelluloses
solubilization. Theharshness of treatments can be expressed in
terms of severity(S0),27 calculated as:
So = log Ro = log [RoHEATING + RoCOOLING
= log
tMAX0
exp
(T (t) TREF
) dt
+tF
tMAX
exp
(T (t) TREF
) dt
(1)
where Ro is the severity factor, tMAX (min) is the time neededto
achieve the target temperature TMAX (
C), tF (min) is the timeneeded for the whole heatingcooling
period, and T(t) and T (t)represent the temperature profiles in the
heating and coolingstages, respectively. Calculations were made
using the valuesreported usually for and TREF (14.75
C and 100C, respectively).
Analysis of solid phase from autohydrolysisSamples of solids
from autohydrolysis were air dried, milled to aparticle size
-
10
01
Bioethanol and chemicals from Ulex Europus www.soci.org
experiments were performed in duplicate, and average results
arereported.
The results achieved in the enzymatic hydrolysis can
beconveniently compared in terms of the
cellulose-to-glucoseconversion achieved at time t (CGCt),
calculated as:
CGCt = 100 Gt G0Gn100 180162 LSR+1KL/100
(2)
where Gt is the glucose concentration (g L-1) at time t, G0 is
theinitial glucose concentration, Gn is the glucan content of the
solidphase (g glucan per 100 g solid phase, oven dry basis),
180/162 isthe glucose hydration factor, is density (g L-1), LSR is
the liquor tosolid ratio employed in enzymatic hydrolysis (g liquid
g-1 oven-drysolid), and KL is the Klason lignin content of solid
phase (g ligninper 100 g solid phase, oven dry basis).
Yeast cultivation and inoculum preparationThestrainemployed
forbioethanolproductionwasSaccharomycescerevisiaeCECT-1170,obtained
fromtheSpanishCollectionofTypeCultures (Valencia, Spain). Cells
were grown at 32C for 24 h in amedium containing 10 g glucose L-1,
5 g peptone L-1, 3 g maltextract L-1 and 3 g yeast extract L-1.
Simultaneous saccharification and fermentation (SSF) of
solidphases from autohydrolysisSimultaneous saccharification and
fermentation (SSF) assays werecarried out in duplicate in 250 mL
Erlenmeyer flasks (orbitalshaking at 120 rpm, 35C, pH 5). SSF media
were prepared bymixing the appropriate amounts of solid phases,
water, buffer,nutrients and cells. Suspensions containingwater,
buffer and solidsubstrateswereautoclavedat 121C for 15min
separately fromthenutrients, and thermostated at 35C. At time 0,
enzymes and cellswere added. For 100 mL of media, 10 mL of inocula
(initial yeastconcentration, 1.0 g L-1) and 10 mL of nutrients
(concentrations: 5g peptone L-1, 3 g yeast extract L-1 and 3 g malt
extract L-1) wereadded. At selected fermentation times, samples
were withdrawnfrom themedia, centrifuged at 5000 rpm for 5min, and
aliquots ofsupernatants were assayed for sugars, acetic acid and
ethanol byHPLC using the method described above.
RESULTS AND DISCUSSIONComposition of Ulex europusTable 1 shows
compositional data for UE. Glucan was the mostabundant fraction,
followed by lignin. Hemicelluloses includedxylan, arabinosyl
substituents and acetyl groups.
The glucan content was similar to that of fast growinghardwoods
such as Eucalyptus globulus8 or other invasive speciessuch as
Arundo donax.30 The molar ratio xylose: acetyl groups: arabinose
(10: 6.7 : 0.7) was also similar to that reported forEucalyptus
wood.8
Hydrothermal processingTable 2 shows the results achieved for
SY, NVC content, andsolid phase composition in treatments performed
at selected So.The treatments at TMAX of 205, 215, 230 or 240
C were made intriplicate, and the average values are reported.
Figures 1 and 2show the composition of the liquid phases, measured
by theircontents of oligomers, monosaccharides, acetic acid, HMF
and F.
Table 1. Composition of Ulex europus (data in wt%
standarddeviation; data based on four replicates)
Content
Glucan 41.7 0.2Xylan 17.3 0.4Arabinan 1.2 0.3Acetyl groups 3.8
0.0Klason lignin 29.4 0.1Extracts 5.1 0.1Ashes 0.7 0.1
UE fractionationThe SY determined for experiments performed at
So 3.2 wereclose to 91 g per100 g oven-dry UE. Harsher conditions
leading toSo in the range 3.24.0 resulted in decreased yields (up
to 70 g per100 g oven-dry UE), whereas severe conditions (So >
4) led to SYin the range 6569 g per 100 g oven-dry UE. As this
latter value isbelow the joint contribution of lignin and glucan in
the untreatedfeedstock, it can be inferred that at least one of
these fractionswasdegraded in part under the corresponding
conditions. NVC wasbelow 10 g per 100 g liquor operating at So <
3.2, in the range2025 g per 100 g liquor for treatments performed
at So in therange 3.64.4, and decreased at higher severities owing
to sugardecomposition.
Composition of spent solidsTable 2 also includes data concerning
the composition of solidsfrom autohydrolysis. The most abundant
component was glucan,whose percentage increased with severity up to
57.7 g per 100g solid in the sample obtained at So = 4.38. Higher
severitiesresulted in decreased Gn contents, reaching 55 g per 100
g in theseverest experiment. The glucan recovery (measured as g
glucanprocessed solids per 100 g glucan in UE) presented an
averagevalue of 91.0%, confirming the occurrence of some
solubilization.
The Klason lignin content of processed samples (KL, measuredas g
Klason lignin per 100 g treated solids) increased with
severity,reaching values above 40% in experiments performed at So
4. The average lignin recovery was near 100%, owing to
theparticipation of condensation reactions under harsh
conditions.
The removal of hemicelluloses from solid phase increasedsteadily
with severity. High solubilisation was observed at So> 3.2, and
complete removal was observed under the harshestconditions assayed.
Acetyl groups and xylan showed closelyrelated variation patterns;
whereas arabinosyl groups were splitoff under mild conditions, with
complete removal at So = 3.2(conditions under which about 61% of
the initial xylan and acetylgroups still remained in solid phase).
The selective solubilisationof hemicelluloses occurring under mild
and severe operationalconditions, resulted in the observed increase
in glucan and lignincontents.
The above data confirm that selected autohydrolysis
conditionsmay result in both extensive hemicelluloses removal and
highretentions of cellulose and lignin in solid phase.
Liquid phase compositionThe major solutes in autohydrolysis
liquors corresponded tohemicelluloses-derived compounds, including
oligosaccharides,monosaccharides and sugar degradation
products.
J Chem Technol Biotechnol 2013; 88: 9991006 c 2012 Society of
Chemical Industry wileyonlinelibrary.com/jctb
-
10
02
www.soci.org IA Ares-Peon et al.
Table 2. Effects achieved by hydrothermal processing: solid
yield (SY , g treated solids per 100 g rawmaterial, oven dry
basis), liquor content of nonvolatile compounds (NVC, g
non-volatile compounds per 100 g raw material, oven dry basis), and
composition of treated solids (expressed as g ofcomponent per 100 g
treated solid, oven dry basis). Standard deviations are based on
three replicates
TMAX (C) So (-) SY NVC Glucan Xylan Arabinan Acetyl groups
Klason lignin
150 2.32 92.3 6.3 43.4 0.6 17.9 0.4 1.8 0.4 3.7 0.1 32.9 2.7160
2.61 91.7 6.2 46.2 0.4 19.0 0.3 1.9 0.1 3.7 0.0 33.2 2.7170 2.91
91.6 7.2 44.7 0.6 18.2 1.4 1.5 0.2 3.5 0.1 35.6 3.5180 3.20 89.3
9.5 44.8 0.6 17.6 0.1 1.4 0.1 3.6 0.0 35.6 1.6190 3.49 81.4 16.5
45.8 0.9 13.0 0.4 < 0.1 2.8 0.0 39.8 1.6195 3.64 75.1 23.1 46.8
0.0 9.6 0.6 < 0.1 1.9 0.0 38.4 0.6200 3.79 71.9 25.0 48.9 0.2
8.3 0.2 < 0.1 1.6 0.1 39.9 0.4205a 3.94 70.8 1.1 24.8 0.1 52.5
0.6 6.4 0.6 < 0.1 1.0 0.1 42.2 1.0210 4.08 66.6 22.5 53.7 0.6
5.3 0.0 < 0.1 0.8 0.0 41.8 0.5215a 4.23 69.2 1.9 22.2 0.4 56.0
0.2 5.1 0.1 < 0.1 0.7 0.0 39.2 2.0220 4.38 67.0 21.1 57.7 1.0
4.7 0.5 < 0.1 0.4 0.2 41.6 3.0225 4.52 65.9 14.7 55.9 0.8 4.1
0.3 < 0.1 0.3 0.0 45.7 1.1230a 4.67 65.6 0.5 13.1 0.2 57.1 1.2
2.5 0.4 < 0.1 0.2 0.0 42.1 3.5235 4.81 67.9 10.9 55.4 0.9 <
0.1 < 0.1 < 0.1 46.1 0.6240a 4.97 66.0 0.9 10.1 0.2 54.5 1.9
< 0.1 < 0.1 < 0.1 39.0 0.8a Average values of three
replicates.
0
2
4
6
8
10
12
14
16
18
20
2 2.5 3 3.5 4 4.5 5So (dimensionless)
Conc
entra
tion
(g/L)
GO
XO
ArO
AcO
Figure 1. Dependence of the concentrations of GO
(glucooligomers, g eq. glucose L-1), XO (xylooligomers, g eq.
xylose L-1), ArO (arabinosyl moieties, geq. arabinose L-1) and AcO
(acetyl groups linked to oligomers, g eq. acetic acid L-1) on So
(severity).
Owing to the analytical method employed for
oligosaccharidequantitation, which is based on total hydrolysis and
furtherdetermination of the structural units, oligomers are
referred to asglucooligomers (GO), xylo-oligomers (XO), arabinosyl
units boundto oligomers (ArO) and acetyl groups esterified to
oligomers(AcO). All of them were measured as monomer
equivalents.Figure 1 shows the dependence of the concentrations of
GO,XO, ArO and AcO on So, and Fig. 2 displays the
interrelationshipbetween the concentrations of monosaccharides
(glucose, xyloseand arabinose) and other compounds present in
themedia (aceticacid, HMF and F) on So.
In general terms, a slight decrease in GO concentration
wasobserved when So increased. Under mild operational conditions(So
< 4.08), the GO concentration averaged 4.0 g L-1, whereasharsher
treatments decreased the concentration to 1 g L-1. XOwere themajor
liquor components,with concentrations increasingwith So in the
range 3.644.08 (to reach a maximum of 19.6 g
L-1 at So = 3.79), and decreasing further up to almost
completeconsumption. The variation pattern observed for AcO was
closelyrelated to the one described for XO, with concentrations
withinthe range 44.6 g L-1 in experiments performed at So
3.644.08(maximum value, 4.56 g AcO/L at So = 3.79), and further
decreaseunder higher severity conditions. Comparatively, ArO
reachedlimited concentrations (0.6 g L-1 as an average), and showed
amaximum at So about 3.5. The behaviour of the overall
oligomerconcentration was governed by the variation pattern
observedfor XO, reaching concentrations in the range 2629 g L-1 at
So of3.644.08 (maximum value, 29.0 g L-1 at So = 3.79,
correspondingto 71% of the initial xylan, which is in agreement
with resultsreported on the autohydrolysis of UE).31
Figure 2 shows that the monosaccharide concentrationsattained in
the reaction media were comparatively limited: theglucose
concentration (average value, 1.8 g L-1) remained fairlyconstant
with So, even if a slight increase in concentration
wileyonlinelibrary.com/jctb c 2012 Society of Chemical Industry
J Chem Technol Biotechnol 2013; 88: 9991006
-
10
03
Bioethanol and chemicals from Ulex Europus www.soci.org
0
1
2
3
4
5
6
7
2 2.5 3 3.5 4 4.5 5So (dimensionless)
Conc
entra
tion
(g/L)
GlucoseXyloseArabinoseAcetic acidHMFF
Figure 2. Dependence of the concentrations of sugars (glucose,
xylose and arabinose, g L-1), acetic acid (g L-1), F (furfural, g
L-1) and HMF(hydroxymethylfurfural, g L-1) on So (severity).
(compatible with GO decomposition) was observed under
harshconditions. Xylose, the most abundant sugar, increased
itsconcentration with So (especially at So > 3.79) up to reach
amaximum (5.0 g L-1 at So = 4.38), and then decreased with
theseverity of treatments up to 1 g L-1. The concentration of
arabinosewas below 1 g L-1, and reached a maximum under
conditionsdefined by So in the range 3.644.23.
The concentrations of acetic acid, F and HMF
increasedsignificantly at So > 4, to reach the following maxima
at So4.55: acetic acid, 6.6 g L-1; HMF, 2.2 g-1; and F, .8 g
L-1.
Enzymatic hydrolysis of Ulex europusThe solids from the various
autohydrolysis treatments (seeTable 2) were assayed for
susceptibility to enzymatichydrolysis. In agreement with literature
reported on the scarceenzymatic digestibility of highly lignified
substrates subjected toautohydrolysis,19,20 and in order to explore
the effects obtainedby drastic hydrothermal processing, solids
autohydrolyzed at So= 5.06 (obtained by processing at 230C for 10
min) were alsoassayed as substrates for enzymatic hydrolysis, but
not achievedimprovements in the results (data not show).
The enzymatic assays were performed at liquid to solid
ratios(LSR) of 6, 12 or 20 g liquid g-1 oven-dry autohydrolyzed
UE,using enzyme to substrate ratios (ESR) of 6.2, 10.3 or 11.5 FPU
g-1
oven-dry autohydrolyzed UE.Autohydrolyzed solids treated at So
< 3.8 presented little
enzymatic digestibility, reaching CGC48 < 20%. This
latterparameter increased to 70% when UE was treated at So =
4.67,and decreased under harsher conditions
The cellulose to glucose conversion was favoured by
increasedvalues of LSR and ESR, the major effects being associated
toLSR within the considered experimental domain. The kinetic
ofenzymatic hydrolysis was interpreted on the basis of the
followingequation:32
CGCt = CGCMAX tt + t1/2 (3)
where CGCMAX and t1/2 are fitting parameters measuring
themaximum glucose conversion achievable at infinite reaction
time,and t1/2 (h) measures the reaction time needed to reach a
glucose
conversion corresponding to 50%of CGCMAX. Figures 3 and 4
showthe dependences of CGCMAX and t1/2 on So.
Solids pretreated undermild conditions (So < 3.8) led to
CGCMAX< 30%, but as expected, 33,34 the susceptibility of
substratesincreased with severity, to reach CGCMAX about 90% in the
case ofthe solid autohydrolyzedat So =4.67.Harsher
conditionsprovidedworse susceptibility of treated solids to
enzymatic hydrolysis.The highest cellulose conversions were
achieved in experimentsperformed at the lowest solid loadings, as
reported in relatedstudies,35 with minor effects associated to
variable ESR within thetested range. The experimental results of
this work compare wellwith data reported for other LCMs of residual
nature. For example,glucose conversions in the range 1576% have
been achievedusing steam-pretreated olive tree trimmings (operating
at ESR =15 FPU g-1 and LSR=20 g g-1); 36 whereas guayule residues
treatedaccording to the AFEX process provided less than 30%
celluloseconversion (operatingat ESR=12mgproteing-1
glucanandLSR=100 g g-1 glucan).37 Sugar cane bagasse subjected to
consecutivestages of steam explosion and NaOH delignification
allowed 72%cellulose conversion (operating at ESR = 20 FPU g-1
cellulose andLSR=11.3 g g-1).38
The kinetic parameter t1/2 decreased with severity from
valueslonger than 40 h to reach a minimum (3.55 h) in the
experimentcarried out at So = 4.97. The limited susceptibility
observed forsolids from too severe treatments were well interpreted
by thevariation pattern of t1/2 (which presented increased values
forthese substrates). In a similar way, the favourable kinetic
effectscaused by increased ESR and decreased LSR were reflected
indecreased values of t1/2.
Bioethanol production by SSFBased on the results obtained in the
above experiments, solidtreated at selected severities were
employed for bioethanolproduction by SSF operating at the best ESR
(in the range 4.1to 11.5 FPU g-1) and intermediate solid loading
(in the range 6 to12 g g-1).
As representative examples, Fig. 5 shows the
ethanolconcentration profiles for selected experiments, which
presenteda typical shape with declining slope.
The ethanol yield (YE, g ethanol per 100 g potential
ethanol),calculated for conversion of glucan into ethanol
without
J Chem Technol Biotechnol 2013; 88: 9991006 c 2012 Society of
Chemical Industry wileyonlinelibrary.com/jctb
-
10
04
www.soci.org IA Ares-Peon et al.
0
10
20
30
40
50
60
70
80
90
2 2.5 3 3.5 4 4.5 5So (dimensionless)
CGCM
AX
(%)
(a)(b)(c)(d)
Figure 3. Dependence of the maximum glucose conversion (CGCMAX)
on So (severity) for: (a) experiments carried out at LSR = 12 g g-1
and ESR = 6.2FPU g-1; (b) experiments carried out at LSR = 20 g g-1
and ESR = 10.3 FPU g-1; (c) experiments carried out at LSR = 6 g
g-1 and ESR = 11.5 FPU g-1; (d)experiments carried out at LSR = 12
g g-1 and ESR = 11.5 FPU g-1.
0
10
20
30
40
50
2 2.5 3 3.5 4 4.5 5So (dimensionless)
t 1/2
(h)
(a)(b)(c)(d)
Figure 4. Dependence of the kinetic parameter t1/2 (h) on So
(severity) for: (a) experiments carried out at LSR = 12 g g-1 and
ESR = 6.2 FPU g-1; (b)experiments carried out at LSR = 20 g g-1 and
ESR = 10.3 FPU g-1; (c) experiments carried out at LSR = 6 g g-1
and ESR = 11.5 FPU g-1; (d) experimentscarried out at LSR = 12 g
g-1 and ESR = 11.5 FPU g-1.
degradation as:
YE = 100[
ethanol]
[ethanol
]POT
(4)
where [ethanol] is the highest ethanol concentration (g
L-1)achieved in the experiment and [ethanol]POT is the
potentialethanol concentration, calculated as:
[ethanol
]POT =
Gn
100 92162
LSR + 1 KL100
(5)
where Gn is the glucan content of solids (g glucan per 100
gsolid, oven dry basis), (92/162) is the stoichiometric factor,
isthe density of the reaction medium (average value of 1005 2g
L-1), LSR is the liquid-to-solid ratio in SSF experiments and KL
isthe Klason lignin content of solids (g Klason lignin per 100 g
solid,oven dry basis).
The results determined for the ethanol conversion (EC, definedas
the percentage of ethanol obtained in experiments respect tothe
ethanol that would result from total cellulose saccharificationand
further stoichiometric conversion of glucose into ethanol),
are listed in Table 3. The dependence of EC on the
operationalvariables presented trends closely related to the ones
describedabove for CGC.
The maximum YE (83.3%) was achieved under the
followingconditions: solid pretreated at So = 4.67; LSR and ESR
fixed inthe highest values assayed. The maximum ethanol
concentration(30.2 g L-1) was achieved for the same solid and the
same ESR,but at LSR = 6 g g-1. This result was in the range
reported as athreshold foreconomicprofitability,39
andcompareswellwithdatareported in related studies. For example,
wastes from olive treespretreatedbyacidhydrolysis led to
fermentationmediacontaining20.3 g ethanol L-1,40 whereas olive tree
trimmings pretreated
bysteamexplosionandalkalineperoxidedelignificationwereusedassubstrates
to obtain 29.4 g ethanol L-1 with an ethanol conversionof 65.2%.41
An ethanol concentration of 25 g L-1 was reported fordelignified
sugar cane bagasse.38
CONCLUSIONSThisworkprovidesanexperimental assessmenton
thevalorizationof Ulex europus (an invasive species causing
environmental
wileyonlinelibrary.com/jctb c 2012 Society of Chemical Industry
J Chem Technol Biotechnol 2013; 88: 9991006
-
10
05
Bioethanol and chemicals from Ulex Europus www.soci.org
0
5
10
15
20
25
30
0 20 40 60 80 100 120 140 160Time (h)
Conc
entra
tion
(g/L)
So = 4.03So = 4.38So = 4.67
Figure 5. Ethanol concentration profiles determined in SSF
assays performed at LSR = 6 g g-1 and ESR = 11.5 FPU g-1.
Table 3. Data for SSFexperiments: pretreatment conditions
(definedby severity,So); operational conditionsdefinedbyenzyme to
substrateratio (ESR) and liquor to solid ratio (LSR); and ethanol
yield (YE)
So(dimensionless) ESR(FPU g-1) LSR(g g-1) YE(%)
3.79 11.5 6 6.3 1.24.08 11.5 6 38.8 1.44.38 11.5 6 49.7 0.84.67
11.5 6 60.9 1.14.97 11.5 6 47.5 1.54.23 4.1 10 11.6 0.24.67 4.1 10
23.7 0.54.97 4.1 10 22.0 0.53.94 8.2 10 68.0 3.74.67 8.2 10 82.3
4.23.64 6.2 12 29.7 0.83.79 6.2 12 49.6 2.03.94 6.2 12 50.1 1.0
hazards) by means of an environmentally friendly process,based
on consecutive stages of hydrothermal processing andSSF. The
effects of the pretreatment conditions on the solubleproducts
obtained from hemicelluloses and on the susceptibilityof pretreated
solids to enzymatic hydrolysis were assessedunder conditions
covering the experimental domain of practicalinterest. Under
selected processing conditions, up to 21.5 wt%of the raw material
was recovered as valuable oligomeric ormonomeric saccharides,
whereas the kinetics and yields ofenzymatic hydrolysis presented a
clearly defined optimal range.Based on this latter information, SSF
studies were carried outto establish conditions leading to ethanol
concentrations andconversions that compare well with those in the
literature.
ACKNOWLEDGEMENTSThe authors are grateful to Xunta de Galicia for
financial supportof this work, in the framework of the Research
Project Use of forestresidues for biofuels production (reference
08REM002383PR).
REFERENCES1 Vega A and Bao M, Fractionation of lignocellulose
materials with
phenol and dilute HCl. Wood Sci Technol 25:459466 (1991).
2 Carballas T, Los incendios forestales en Galicia, in
Reflexiones sobre elmedio ambiente en Galicia, ed by Casares Long
JJ. Xunta de Galicia,Santiago de Compostela, 363415 (2003).
3 Vega A and Bao M, Organosolv fractionation of Ulex europus
withdilute hydrochloric acid and phenol. Two simple kinetic
modelsfor pre-hydrolysis and delignification. Wood Sci Technol
27:6168(1992).
4 Gullon P, Roman A, Vila C, Garrote G and Parajo JC,
Potentialof hydrothermal treatments in lignocellulose
biorefineries. BiofuelBioprod Bioref 6:219232 (2012).
5 Yanez R, Roman A, Garrote G, Alonso JL and Parajo JC,
Experimentalevaluation of alkaline treatment as a method for
enhancing theenzymatic digestibility of autohydrolysed Acacia
dealbata. J ChemTechnol Biotechnol 84:10701077 (2009).
6 Gosselink RJA, Teunissen W, van Dam JEG, de Jong E,
Gellerstedt G,Scott EL and Sanders JPM, Lignin depolymerisation in
supercriticalcarbon dioxide/acetone/water fluid for the production
of aromaticchemicals. Bioresource Technol 106:173177 (2012).
7 Garrote G, Domnguez H and Parajo JC, Mild autohydrolysis:an
environmentally friendly technology for
xylooligosaccharideproduction from wood. J Chem Technol Biotechnol
74:11011109(1999).
8 Garrote G, Kabel MA, Schols HA, Falque E, Domnguez H and
Parajo JC,Effects of eucalyptus globulus wood autohydrolysis
conditions onthe reaction products. J Agric Food Chem 55:90069013
(2007).
9 Carvalheiro F, Garrote G, Parajo JC, Pereira H and Girio FM,
Kineticmodeling of brewerys spent grain autohydrolysis. Biotechnol
Prog21:233243 (2005).
10 Parajo JC, Garrote G, Cruz JM and Domnguez H, Production
ofxylooligosaccharides by autohydrolysis of lignocellulosic
materials.Trends Food Sci Technol 15:115120 (2004).
11 Moure A, Gullon P, Domnguez H and Parajo JC, Advances in
themanufacture, purification and applications of
xylo-oligosaccharidesas food additives and nutraceuticals. Proc
Biochem 41:19131923(2006).
12 Garrote G, Domnguez H and Parajo JC, Manufacture of
xylose-based fermentation media from corncobs by posthydrolysis
ofautohydrolysis liquors. Appl Biochem Biotechnol 95:195207
(2001).
13 Garrote G, Domnguez H and Parajo JC, Generation of xylose
solutionsfrom Eucalyptus globulus wood by
autohydrolysis-posthydrolysisprocesses: posthydrolysis kinetics.
Bioresource Technol 79:155164(2001).
14 TamakawaH, Ikushima S and Yoshida S, Efficient production of
L-lacticacid from xylose by a recombinant Candida utilis strain. J
BiosciBioeng 37:174179 (2012).
15 Pereira SR, Ivanusa S, Evtuguin DV, Serafim LS and Xavier
ARMB,Biological treatment of eucalypt spent sulphite liquors: a way
toboost the production of second generation bioethanol.
BioresourceTechnol 103:131135 (2012).
16 Alvira P, Tomas-Pejo E, Negro MJ and Ballesteros M,
Strategies ofXylanase supplementation for an efficient
saccharification andcofermentation process from pretreated wheat
straw. BiotechnolProg 27:944950 (2011).
J Chem Technol Biotechnol 2013; 88: 9991006 c 2012 Society of
Chemical Industry wileyonlinelibrary.com/jctb
-
10
06
www.soci.org IA Ares-Peon et al.
17 Ruiz HA, Ruzene DS, Silva DP, da Silva FFM, Vicente AA and
Teixeira JA,Development and characterization of an environmentally
friendlyprocess sequence (autohydrolysis and organosolv) for wheat
strawdelignification. Appl Biochem Biotechnol 164:629641
(2011).
18 Caparros S, Ariza J, Garrote G, Lopez F and Daz MJ,
Optimization ofPaulownia
fortuneiL.autohydrolysis-organosolvpulpingasasourceof xylooligomers
and cellulose pulp. Ind Eng Chem Res 46:623631(2007).
19 Roman A,GarroteG,AlonsoJLandParajo JC,
Experimentalassessmenton the enzymatic hydrolysis of hydrothermally
pretreatedeucalyptus globulus wood. Ind Eng Chem Res 49:45634663
(2010).
20 Ares-Peon IA, Vila C, Garrote G and Parajo JC, Enzymatic
hydrolysis ofautohydrolyzedbarleyhusks. JChemTechnolBiotechnol
86:251260(2011).
21 Ruiz HA, Ruzene DS, Silva DP, Quintas MAC, Vicente AA and
TeixeiraJA, Evaluation of a hydrothermal process for pretreatment
of wheatstraw - effect of particle size and process conditions. J
Chem TechnolBiotechnol 86:8894 (2011).
22 Daz MJ, Cara C, Ruiz E, Perez-Bonilla M and Castro E,
Hydrothermalpre-treatment and enzymatic hydrolysis of sunflower
stalks. Fuel90:32253229 (2011).
23 Roman A, Garrote G and Parajo JC, Bioethanol productionfrom
autohydrolyzed Eucalyptus globulus by simultaneoussaccharification
and fermentation operating at high solids loading.Fuel 94:305312
(2012).
24 Sassner P, Galbe G and Zacchi G, Techno-economic evaluation
ofbioethanolproductionfromthreedifferent
lignocellulosicmaterials.Biomass Bioenerg 32:422430 (2008).
25 Ohgren K, Bura R, Lesnicki G, Saddler J and Zacchi G, A
comparisonbetween simultaneous saccharification and fermentation
andseparate hydrolysis and fermentation using steam-pretreated
cornstover. Proc Biochem 42:834839 (2007).
26 Manzanares P, Negro MJ, Oliva JM, Saez F, Ballesteros I,
BallesterosM, Cara C, Castro E and Ruiz E, Different process
configurations forbioethanol production from pretreated olive
pruning biomass. JChem Technol Biotechnol 86:881887 (2011).
27 Lavoie JM, Capek-Menard E, Gauvin H and Chornet E, Production
ofpulp from Salix vinimalis energy crops using the FIRSST
process.Bioresource Technol 101:49404946 (2010).
28 Ghose TK, Measurement of cellulose activities. Pure Appl
Chem59:257268 (1987).
29 Paquot M and Thonart P, Hydrolyse enzymatique de la
celluloseregeneree. Holzforschung 36:177181 (1982).
30 Caparros S, Garrote G, Ariza J and Lopez F, Autohydrolysis of
Arundodonax L., a kinetic assessment. Ind Eng Chem Res
45:89098920(2006).
31 Ligero P, de Vega A, van der Kolk JC and van Dam JEG, Gorse
(Ulexeuropus) as apossible sourceof xylansbyhydrothermal
treatment.Ind Crops Prod 33:205210 (2011).
32 Holtzapple MT, Caram HS and Humphrey AE, A comparison of
twoempirical models for the enzymic hydrolysis of pretreated
poplarwood. Biotechnol Bioeng 26:936941 (1984).
33 Cara C, Romero I, Oliva JM, Saez F and Castro E, Liquid hot
waterpretreatmentofolive treepruning residues.Appl Biochem
Biotechnol136140:379394 (2007).
34 Roman A, Garrote G, Alonso JL and Parajo JC,
Bioethanolproduction from hydrothermally pretreated Eucalyptus
globuluswood. Bioresource Technol 101:87068712 (2010).
35 Cara C, Moya M, Ballesteros I, Negro MJ, Gonzalez A and
RuizE, Influence of solid loading on enzymatic hydrolysis of
steamexploded or liquid hot water pretreated olive tree biomass.
ProcBiochem 42:10031009 (2007).
36 Cara C, Ruz E, Oliva JM, Saez F and Castro E, Conversion of
olivetree biomass into fermentable sugars by dilute acid
pretreatmentand enzymatic saccharification. Bioresource Technol
99:18691876(2008).
37 Chundawat SPS, Chang L, Gunawan C, Balan V, McMahan C andDale
BE, Guayule as a feedstock for lignocellulosic biorefineriesusing
ammonia fiber expansion (AFEX) pretreatment. Ind CropsProd
37:496492 (2012).
38 Santos JRA, Lucena MS, Gusmao NB and Gouveia ER,
Optimizationof ethanol production by Saccharomyces cerevisiae
UFPEDA 1238in simultaneous saccharification and fermentation of
delignifiedsugarcane bagasse. Ind Crops Prod 36:584588 (2012).
39 Zhang J and Lynd LR, Ethanol production from paper sludgeby
simultaneous saccharification and co-fermentation usingrecombinant
xylose-fermentingmicroorganisms. Biotechnol Bioeng107:235244
(2010).
40 Cuevas M, Sanchez S, Bravo V, Garca JF, Baeza J, Parra C
andFreer J, Determination of optimal pre-treatment conditions
forethanol production from olive-pruning debris by
simultaneoussaccharification and fermentation. Fuel 89:28912896
(2010).
41 Ruiz E, Cara C, Ballesteros M, Manzanares P, Ballesteros I
and Castro E,Ethanol production from pretreated olive tree wood and
sunflowerstalksbyanSSFprocess.ApplBiochemBiotechnol
129132:631643(2006).
wileyonlinelibrary.com/jctb c 2012 Society of Chemical Industry
J Chem Technol Biotechnol 2013; 88: 9991006