Bio-catalytic action of twin-screw extruder enzymatic hydrolysis on the deconstruction of annual plant material: case of sweet corn co-products

O

pen

A

rchive

T

OULOUSE

A

rchive

O

uverte (

OATAO

)

OATAO is an open access repository that collects the work of Toulouse researchers and

makes it freely available over the web where possible.

This is an author-deposited version published in :

http://oatao.univ-toulouse.fr/

Eprints ID : 13605

To link to this article : DOI:10.1016/j.indcrop.2015.01.041

URL :

http://dx.doi.org/10.1016/j.indcrop.2015.01.041

To cite this version :

Vandenbossche, Virginie and Brault, Julien and Vilarem, Gérard

and Rigal, Luc Bio-catalytic action of twin-screw extruder

enzymatic hydrolysis on the deconstruction of annual plant

material: case of sweet corn co-products. (2015) Industrial Crops

and Products, vol. 67 . pp. 239 -248. ISSN 0926-6690

Any correspondance concerning this service should be sent to the repository

administrator:

[email protected]

Bio-catalytic

action

of

twin-screw

extruder

enzymatic

hydrolysis

on

the

deconstruction

of

annual

plant

material:

Case

of

sweet

corn

co-products

Virginie

Vandenbossche

_,

_Julien

_Brault

_,

_Gérard

_Vilarem

_,

_Luc

_Rigal

a_{UniversitédeToulouse,INP,LaboratoiredeChimieAgro-Industrielle,ENSIACET,4AlléeEmileMonso,BP44362,31030ToulouseCedex4,France} b_{INRA,LaboratoiredeChimieAgro-Industrielle,31030ToulouseCedex4,France}

Keywords:

Sweetcornco-product Twin-screwextruder Enzymatichydrolysis Bioextrusion Alkalinepre-treatment

a

b

s

t

r

a

c

t

Acontinuousprocesscombininganalkalinethermo-mechano-chemicalpretreatmentneutralization step,followedbyinjectionofenzymesintothetwin-screwextruder,wasdevelopedusingsweetcorn co-productsasabiomassmodel.Theimplementationofthecontinuousprocessisdescribed.Particular attentionispaidtotheinfluenceofthebio-catalyticactionofenzymatichydrolysisonthedeconstruction ofannualplantmaterialinthetwin-screwextruder(aprocesscalled“bioextrusion”).Theuseofa twin-screwextruderallowsworkingwithhighconsistency(20%),inahighshearenvironment,forashorttime (∼2min).Inthepresentwork,thenatureoftheligno(hemi)cellulosicmaterialtransformations,covering solubilizationandextractionofsaccharidesandmodificationofcellulosicfibers,wereinvestigated.41% ofhemicellulosesand14%ofligninareextractedbythealkalinepretreatment.Hydrolyticenzymesare notdeactivatedduringbioextrusion,whichhasadestructingeffectonthefiber.Itleadstoachangeof rheologicalpropertiesandinducesanincreaseofsugarsreleasedintheformofmonoand polysaccha-rides(upto13%/DMoftotalsugars)withlongerchainsthaninthecaseofabatchreactor.Atthesame time,thedegreeofpolymerizationdecreasesandashorteningofthecellulosechainsoccurs.

1. Introduction

The lignocellulosic materials from agriculture and forest by-products represent an important source of renewable and carbon-neutralenergy.Theseresourcesareclean,cheap,available inlargequantities,andareindependentofgeographicallocation, plustheyarecarbonneutralandrenewable.However,the lignocel-lulosicbiomassisacomplexassemblyofcellulose,hemicelluloses andligninwhoseextractionandconversionrequirestheuseofa refiningprocess(Himmeletal.,2007).

Thesetransformationprocessesareactuallystudiedworldwide and manyreviewsexist onthesubject(VanDykand Pletschke 2012).Theyaretypicallyrealizedintwosteps:oneofpretreatment intendedtoimprovetheaccessibilityofthecelluloseandasecond allowinghydrolysisofthecellulosebyenzymaticaction.Sunand Cheng,2002,Mosieretal.(2005),Zhengetal.(2009),Harmsenetal. (2010)andAlviraetal.(2010)havereviewedpre-treatment

tech-∗ Correspondingauthorat:UniversitédeToulouse,INP,LaboratoiredeChimie Agro-Industrielle,ENSIACET,4AlléeEmileMonso,BP44362,31030ToulouseCedex 4,France.Tel.:+33534323549;fax:+33534323599.

E-mailaddress:[email protected](V.Vandenbossche).

nologiesforabioethanolproductionprocessbasedonenzymatic hydrolysis.Mostcitedpretreatments,amongwhichdiluteacid pre-treatment(Sahaetal.,2005),steamexplosion(Olivaetal.,2003; Caraetal.,2006;Vargaetal.,2004;Ballesterosetal.,2006)and ammoniafiberexplosion(GalbeandZacchi,2007)arethemostwell known,arepenalizedbyimplementationconstraints,technology orreagentcosts,orinhibitorproductionforenzymatichydrolysis orethanolicfermentation.

Soft alkaline pretreatment is one approach that has several potentialadvantagescomparedtootherpretreatmentprocesses, includingloweroperatingcost, reduceddegradationof holocel-lulose, and subsequentformation of inhibitors for downstream processing(Carvalheiroetal.,2008;Kumaretal.,2009;McIntosh andVancov,2011;TaherzadehandKarimi,2008).Themain mech-anismsofsoftalkalinepretreatmentarehydroxylgroupsolvation, leadingtotheruptureofintraorintermolecularhydrogenbonds, andthebreakdownofesterbondswithcleavageoflinkagesinthe lignocellulosiccellwallmatrix,allofwhichleadtothealteration oftheligninstructure,thehydrolysisofthelignin–hemicellulose complex,celluloseswelling,andthepartialdecrystallizationofthe cellulose(Bobleter,1994;SunandCheng,2002).

Awaytoimproveprocesscostsistoincreasethesolid con-centration of biomass in each unit operation. Such high solid

Fig.1.Schematicrepresentationofthesequencesoftheprocesscarriedoutintwosuccessivetwinscrewextruderswithrecirculationofthebioextrudateinthesecondtwin screwextruderwithorwithoutafiltrationmodule.

concentrationsdramaticallyincreaseeconomicviability,reducing operatingcosts duetodecreased volumes. However,increasing solid concentrationincreases theapparentviscosityof biomass slurries,makingmixingandconveyingoperationsmore challeng-ing.Among theprocesses used tocarry out pretreatmentwith a minimum number of steps, twin-screw extrusion technology hasmany advantages andallows workingwithhighsolid con-centrations. It produces a high shear, rapid heat transfer, and effectiveandrapidmixing,inacontinuousoperation,withgood adjustabilityoftreatmentsteps.

Co-penetratingandco-rotatingtwin-screwextrudersaremost common(Dziezak,1989),andaverywidechoiceofscrewelements isavailable.Thescrewprofileisdefinedbythearrangementofthe differentscrewelementsandtheircharacteristics(pitch,length,

numberand shapeofscrew,andanglebetweentwosuccessive elementsincaseofpaddlescrews)indifferentpositionsandspaced differently.

Theperformanceoftwin-screwextrusionandtheinfluenceof theoperatingparametershavebeenstudiedonbiomass fraction-ationfrompoplar (N’Diaye andRigal, 2000), Miscanthussp.(De Vrijeetal.,2002),sugarbeetpulp(Rouillyetal.,2006),sunflower (Evonetal.,2007;Kartikaetal.,2006),soybeanhulls(Yooetal., 2011),ricestraw(Chenetal.,2011), andwheatstraw andbran (Marechaletal.,2004;Zeitounetal.,2010;Vandenbosscheetal., 2014a).

Moreparticularly, twin-screw extrusion canbeused to pre-treatdifferentbiomassesfortheproductionofsugars,andseveral authorshave reportedthistype ofapplicationoverthelast few

Fig.2.Screwconfigurationforthecombinedprocessofpretreatmentandbioextrusionwithorwithoutliquid/solidseparationzone.(T2F=trapezoidaldouble-threadscrew; C2F=conveyingdouble-threadscrew;BB=bilobepaddlescrew;CF2C=reverseddouble-threadscrew.ThenumbersfollowingthetypeofthescrewindicatethepitchofT2F, C2F,andCF2CscrewsortheanglebetweentwosuccessiveelementsincaseofBBscrews).

Table1

Operatingconditions.

Profile Constraintelements SS T NaOH/DM H3PO4/DM pHex L/S FR I Stability (rpm) (◦_{C) (%) (%) (kgh}−1_/kgh−1_{) (A)}

A1 CF2C-25 110 100 32.4 36.6 7.0 10.3 0.06 9 Instabilityofdynamicplug 100 100 10.7 13.1 6.5 3.6 0.20 23 Instability

filtrationmaintainedbut irregular

110 100 9.4 11.0 6.0 7.6 0.08 17 Instability

filtrationmaintainedbut irregular

A2 CF1C-25 110 100 10.5 11.8 6.0 8.2 0.07 17 Stable 110 100 12.1 29.4 5.0 8.4 0.08 21 Stable 110 100 18.6 28.0 5.5 8.1 0.07 14 Stable

ConstraintelementsandoperatingconditionsforAlkalinepre-treatment(CF2C=reverseddouble-threadscrew,CF1C=reversedsimple-threadscrew,DMisthedrymatter ofinletsweetcorn,Ssisthescrewrotationspeed(rpm),pHexisthepHoftheextrudate,L/Sistheliquid/solidratio,FRistheextruderfillingratio,Iisthecurrentfeedingthe motor(A).

Table2

ConstraintelementsandoperatingconditionsforBioextrusionstep.

Profile Constraintelements SS T Enzyme/DM L/S FR Stability (rpm) (◦_{C) (%) (kgh}−1_/kgh−1₎

B1 CF2C-33(25) 280 50 4.0 2.5 0.01 Instabilityofdynamicplug B2 CF2C-33(50) 280 50 4.0 2.5 0.01 Instabilityofdynamicplug

B3 CF2C-16(25) 200 50 4.1 2.5 0.01 Stable

B4 DM-45◦_{(5×5) 200 50 4.6 2.5 0.01 Stable}

CF2C=reverseddouble-threadscrew,DM=monolobepaddle-screw,thelengthofthescrewsisindicatedinbrackets,FR=extruderfillingratio.

years. Zheng and Rehmann (2014) providean overview of the extrusionpretreatmentoflignocellulosicbiomass.

Lamsaletal.(2010)firstshowedthatextrusionledtohigher reducingsugaryieldscomparedtogrinding,andtheystudiedthe process onwheat bran in two steps: impregnationwith water followedbymechanicaltreatment(grindingorextrusion).A com-bination of both extrusion and alkaline pretreatmenthasbeen explored more recently (Zhang et al., 2012; Karunanithy and Muthukumarappan,2011;Duqueetal.,2013;Umetal.,2013;Kang etal.,2013;Hanetal.,2013).KarunanithyandMuthukumarappan, 2013 provide an overview of this type of thermo-mechanical pretreatment,includingthemechanisminfluencingextruderand feedstockparameters,plusevaluationofpretreatmentefficiency.A continuousprocesscombiningalkalinethermo-mechano-chemical pretreatment, followed by injection of enzymes into the twin-screwextruder,called“bioextrusion”wasdevelopedand tested ondifferentbiomasssuchassweetcornresidue(SC),blueagave bagasse,oilpalmemptyfruitbunchasresiduefrompalmoil man-ufacture,and barleystraw(Vandenbosscheetal.,2014b; Duque et al.,2014).This newprocessleadstoexcellentmixing ofthe enzymeswiththepretreatedbiomassathighconcentrations,and allowssaccharificationtobeginduringbioextrusion.

The implementation of the continuous process, and the influence of the bio-catalytic action of hydrolytic enzymes on deconstructionofannualplantmaterialinthetwin-screwextruder during“bioextrusion”,isstudiedhereincaseofaweaklylignified material:thesweetcornco-products.

2. Experimental 2.1. Materials 2.1.1. Feedstocks

Dehydratedsweetcornco-products,providedbySARLSoupro+, camefromanindustrialcorngraincannery.It wasmilledusing ahammermillfittedwitha6mmscreen,andcontained4%ash, 39%cellulose,36%hemicellulosesand4%lignin(allexpressedin %/drymatteranddeterminedaccordingtotheFrenchstandardNF V03-322andADF-NDFmethodcitedinSection2.3).

2.1.2. Enzymaticcocktails

Enzymaticsaccharification wasconductedusing the Saccha-risedC6cocktailfromAdvancedEnzymeTechnologies(India)in acitratebuffer(300mM,pH4.8).Asolutionof1%ofsaccharised C6cocktailischaracterizedby:proteincontentof9mg/ml, cellu-loseactivityof5s5FPU/ml,endoglucanaseactivityof1940nkat/ml, xylanaseacivityof 16,900nkat/ml,andb-glucosidaseactivityof 310nkat/ml. The protein content was measured after acetone precipitation with Bio-Rad DC protein assay kit, cellulose and endoglucanaseactivityaccordingtoGhose(1987),xylanaseactivity accordingtoBaileyetal.(1992)andb-glucosidaseactivity accord-ingtoBaileyandNevalainen(1981).

A preliminary study showed that the mixture of advance enzyme had an optimum temperature at 54◦_{C. Beyond 56}◦_C

enzymeactivitybeginstodecrease,tokeep asafetymarginwe placedat50◦_{Cforthebioextrusiontrials.}

2.2. Twin-screwextruder

Thermo- mechano-chemical alkaline pre-treatment and the neutralizationphase,wereconductedusingaco-penetratingand co-rotating ClextralBC 45 twin-screw extruder. For the bioex-trustionphaseaClextralBC21isused.BothBC45andBC21had sevenmodularbarrels,eachofthesemeasuring200mminlength forBC45 and100mmforBC21.Bothextruderswereequipped withdifferentsegmentalscrewelements.Themetallurgyis, respec-tively,CLX200forBC45andhastelloyforBC21.Botharecorrosion resistant.Moduleswerethermo-regulatedbythermalinduction for BC45and byheater bandfor BC21;modulecoolingwasby watercirculation.Afiltersectionwasusedtoenablethefiltrate tobecollectedandconsistedofsixhemisphericaldisheswith con-icalholes(1mmentry,2mmexit).Theoutputsofextruderswere notequippedwithdie.Schematicrepresentationofthesequence ofstepsfortheprocesscarriedoutintwosuccessivetwin-screw extrudersisshowninFig1.Screwconfigurationforthecombined processofpretreatmentandbioextrusionisshowninFig2.After alkalinepretreatmentandneutralizationintwin-screwextruder BC45,theextrudatewasstoredfrozenanddefrostedjustbefore beingintroducedintoBC21forthebioextrusion.

Table3

Operatingconditionsforalkalinepre-treatmentandbioextrusion. Alkalinepretreatment Bioextrusion

SS(rpm) 110 SS(rpm) 200 T(◦_{C) 100 T(}◦_{C) 50} QS(Kg/h) 8.8 QS(Kg/h) 3.29 DMs(%) 90 DMs(%) 39.6 QLAlk(Kg/h) 20.7 QT+Ez(kg/h) 1.95 QNaOH(Kg/h) 0.8 QEz(g/h) 60 QLAc(Kg/h) 51.6 QH3PO4(Kg/h) 0.9 NaOH/DM(%) 10.5 Enzyme/DM(%) 4.6 H3PO4/DM(%) 11.7 L/S(kg.h−1_/kg.h−1_{) 7.6 L/S(kgh}−1_/kgh−1_{) 2.5} Qext(kg/h) 16.5 Qbioext(kg/h) 5.29 DMext 40.7 DMbioext 29.0 Qfil(kg/h) 62.4 DMfil 4.7 pHext 6 pHfil 5.5 I(A) 17 SME(Wh/kg) 141

WhereSSisthescrewrotationspeed(rpm);Tisthesettemperatureoftheheating modules(◦_{C);QsandDMSarethefeedrate(kg/h)anddrymatter(%)ofsweet} corn;QLAlkandQLAc arerespectivelytheinletflowrateofalkalisolution(kg/h) andacidsolution(kg/h);QNaOHandQH3PO4aretheinletflowofNaOHandH3O4; NaOH/DMandH3PO4/DMarethereagenttosweetcorndrymatterratio(%);L/Sis theliquid/solidratio;Qext,QbioextandQfilare,respectively,theflowrateofextrudate, bioextrudateandfiltrate(kg/h);DMext,DMbioextandDMfilare,respectively,thedry matterofextrudate,bioextrudateandfiltrate.Iisthecurrentfeedingthemotor(A); SMEisthespecificmechanicalenergyconsumedbythemotorperunitweightof solidmatter(Wh/kg).

2.2.1. Alkalinethermo-mechano-chemicalpre-treatmentand neutralizationphase

The alkali used for the pretreatment is sodium hydroxide (NaOH),andtheneutralizationstepusesphosphoricacid(H3PO4).

OperatingconditionsaredescribedinTables1and3.

Feedstockswerefedintotheextruder’sfirstmoduleandthe alkaline solutioninjected usinga piston pump.A first zone of mechanicalpressure,consistingofasuccessionofbilobalpaddles, ensuresgrindingandgoodmixingofthematerialwiththisalkaline solution.Anacidsolutionwastheninjectedusingapistonpump, toneutralizethemediumandreducetheviscosityofthematterto ensuregoodfiltration.Asecondzoneofmechanicalpressure guar-anteesgoodmixingwiththemedium,andfastneutralization.The filtrationzonesituatedinmodule6wascarriedoutbyathirdzone ofmechanicalpressureusingreversedpitchscrewstoensurethe formationofadynamicplug,andthusthepressingofthemixture.

Table4

Compositionandenzymatichydrolysabilityofthepretreatedcoproductofsweet corn.

Composition

DM(%) 40,7

MM(%/DM) 7,3

OM(%/DM) 92,7

Hotwatersoluble(%/DM) 31,4

Cellulose(%/DM) 46,4 Hemicelluloses(%/DM) 28,8 Lignin(%/DM) 4,8 Glucose(%/DM) 54,4 Xylose(%/DM) 22,3 Arabinose(%/DM) 3,0 Galactose(%/DM) 0,7 Mannose(%/DM) 0,1 Enzymatichydrolysability (%/DM) 35

DM=drymatter;MM=mineralmatter;OM=organicmatter.

2.2.2. Bioextrusion

Bioextrusionconsistsofintroducinganenzymecocktailintothe extruder,andthenusingasuccessionofcompressionand expan-sionstepstofacilitatethepenetrationoftheseenzymesintothe matter.OperatingconditionsareshowninTables2and3. 2.3. Analyticalmethods

2.3.1. Drymatterandparietalcompounds

Moisturecontentswere determinedaccordingtotheFrench standardNFV03-903andmineralcontentsaccordingtotheFrench standard NFV 03-322.Anestimationof thethreeparietal con-stituents(cellulose,hemicelluloses,andlignins)containedinthe solids,wasmadeusingtheADF-NDFmethodofVanSoestandWine (1967,1968).Alldeterminationshavebeencarriedoutintriplicate andstandarddeviationwaslessthan1.5%forallmeasurements.

Anestimationofthehotwater-solublecomponentscontained, wasmadebymeasuringthemasslossofthetestsampleafter1hin boilingwater.ThismethodhasbeenadaptedusingstandardTAPPI 204cm-97ontheFibertecTecatorM1017apparatus.All determi-nationswerecarriedoutinduplicate.

2.3.2. Sugars

Reducing sugars were determined using the DNS method (Miller,1959).

Total sugars were determined after acid hydrolysis using a methodadaptedfromNERL(Sluiteretal.,2008).10mloffiltrate wasmixedwith1.25ml72%H2SO4.Hydrolysiskineticswere car-riedoutin sealedtubes between30and 90minat100◦_C.After

cooling,samples wereneutralized with3.6ml NaOH32%(m/v), dilutedandfiltered(0.45mm).Analysiswasthenmadeusing high-performanceliquidchromatography(HPLC).

FreeandtotalsugarsweremeasuredusingHPLIC(DIONEX ICS-3000coupledwithpulsedamperometricdetection(PAD)andfitted withaCarbopacPA1column).

Enzymatic hydrolysability was determined by enzymatic hydrolysisin50mMcitratephosphatebuffer(pH4.6)inthe pres-enceofAdvancedenzymes(2.5%/DMsubstrate)at50◦_Cfor48h.It

hasbeencalculatedasthepercentageofreducingsugarsreleased byenzymatichydrolysis,relativetodrymatter.

Sugarsreleasedafterbioextrusionweremeasuredonwater sol-ublesextractedfrombioextrudates.Thelatter,obtainedafter1,3,5 or7passagesintheBC21twin-screwextruder,werefirstdilutedin waterwithaLiquid/Solidratioof20andthenfilteredontoaglass filter(porositygrade:2).

2.3.3. Degreeofpolymerization(DP)ofoligosaccharides

The degree of polymerization (DP) of oligosaccharides was determinedinhighperformanceanion-exchangechromatography equippedwithapulsedamperometricdetectorandaneluent gen-eratorEluGen®_{(HPAEC-PAD,ICS-3000,Dionex).Theanalyseswere}

carriedout withananion exchange column(250mm Carbopac PA200) and used a mixtureof two eluents (A:100mM sodium hydroxide and B:100mM sodium hydroxide with 1M sodium acetate)withagradientfrom100%ofAatt=0to75%ofAand 25%ofBatt=35min.Theflow rateofthecolumnwasfixedat 0.35mL/min.

2.3.4. Residencetimedistribution(RTD)

Sweet corn particles colored with Erythrosin were placed quickly(<2s)onthefirstscrewelement.Thisdyewaschosenas atracerforitsneutralcharacteristicswithrespecttotheprocess and becauseof its good colorization of mostvegetable matter, itselfoftenhighlycolored(N’Diaye,1996).Samplesofextrudate aretakenevery5satthesolidoutletsoftheextruder.All sam-plesaredriedat103◦_{Candgroundtohomogenizethecolorand}

0 0,005 0,01 0,015 0,02 0,025 0 20 40 60 80 100 120 140 160 180 200 220 240 Time (s) E (t ) Malaxage Contre filet Retention tim e : 118,2s Retention tim e : 127,9s DM - 45° C2FC - 16

Fig.3.Residencetimedistribution(RTD)ofthematerialintheBC21twin-screwextruderequippedwithprofilesB3andB4.

toeliminatelargesizeparticleswhichaffectmeasurements.The ‘a’coordinateofcoloredoutletsamplesintheCIE(L*a*b)system, isreaddirectlyusingaspectrophotocolorimeter(ModelCM500i, Minolta,Japan,illuminant:D65,observerangle:2◦_).TheRTDdata

areexaminedthroughtypicaldistributionversustimeplots.RTDis definedas:

Ei= _n Ci

X

Ci×Dt

whereCiisthetracerconcentrationineachsampleandDtisthe samplingperiod.Typicalcurvesareshownon(Fig.3).

2.3.5. Dynamicviscosity

Viscosity values were determined in accordance withTAPPI Standard T 230om-94“viscosity ofpulp” (capillary viscometer method).Inourcase,thesamplewassolvatedwith cupriethylene-diamineandpassedthroughacanon-fenskeViscometer,type150 at25◦_{C.Theviscometerreadingswereperformedthreetimesfor}

eachsample.

3. Resultsanddiscussion

Theprocessofdeconstructionoflignocellulosicplantmaterial iscarriedoutusingtwoconsecutiveextrudersBC45andBC21. Itincludedthreedifferentparts:analkalinepre-treatment(inBC 45),aneutralizationphase(inBC45),andanenzymeimpregnation phase(bioextrusion,inBC21)duringwhichthehemicellulosesand cellulosesaccharificationbegan.(Figs.1and2).

3.1. Alkalinepre-treatmentandneutralizationphase

Thefirststepoftheprocessisthedestructurationofcellulosic materialunderalkalineconditions.

Thealkalineextractionconditionsforhemicellulosesandlignin intwin-screwextrudershavebeenwidelystudied,forinstanceon sorghumandpoplarwoodfibers(N’Diayeetal.,1996)oronwheat straw(Magro,1995;Marechal,2001;Zeitounetal.,2010),andthis previousworkhasallowedustodefinetheconfigurationandscrew profilesused inthis study.Theaimofthis firststepis toopen upthecomplexstructureofthebiomass,andfacilitateaccessof thehydrolyticenzymestopolysaccharides,byincreasingthe sur-facearea(KarunanithyandMuthukumarappan2011)andporosity (Zhangetal.,2012;Vandenbosscheetal.,2014b).

Thethermomechanicaleffectoftheflowrestrictingelementsof thescrewprofile,ensuresphysicaldisintegrationofthematerial byseparatingthefiberbundles.Thepresenceofsodium hydrox-ideensuresadditionalchemicaldestructurationbysolubilization

oforganicmatter,especiallyhemicellulosesandlignin,depending ontheamountused.

TheneutralizationphaseisnecessarytoobtainapH compati-blewitheffectiveenzymeactivityduringbioextrusion.Phosphoric acidisusedinthisstepbecauseitistriprotic andsolimitsthe massofacidneeded.Thisneutralizationstepalsodecreasesthe viscosityofthematterviaachangeofpHandadilutioneffect, whichfacilitatesfiltrationandplaysapartindevelopingfriction, shear,andimpactsresidencetime.Theeffectivenessofthefiltration stepdependsontheformationandstabilityofthe“dynamicplug” formedinmodule7,andtothepresenceofconstraintelements (Fig.2).Theselectionoftheseconstraintelementsisimportantfor theformationofastable“dynamicplug”.TheuseofaC2FC-25 reverseddouble-threadscrewfailedtoachieveastablefiltration despitevaryingtheoperatingconditions,includingachangeinthe amountofsodiumhydroxide,intheliquid–solidratio,orinthe degreeoffilling(Table1).Indeed,inthepresenceofinsufficient backpressure,theoperationoftheextruderasaliquid/solid sepa-ratormaybecyclic,withanaccumulationoffluidinthefiltration zoneresultinginalossofdynamicplugstructure.Liquid separa-tionislesseffectiveandthesolidisinsufficientlypressed.Thisis accentuatedinthepresenceofhighlevelsofsodiumorhigh liq-uid/solidratiosandcanexplaintheworstresults,obtainedwitha NaOH/DMratioof32.4%.Achangeinextruderconfiguration, con-sistingofreplacingtheC2FC-25reverseddouble-threadscrewwith areversedsimple-threadscrewC1FC-25,hasachievedasteady stateforaL/Sratiocloseto8,afillingratio(Q/S)of0.07–0.08and forarangeofNaOH/DMratiosbetween10%and19%.Thereversed simple-threadscrewC1FC-25ensuredbetterbackpressurethan theC2FC-25reverseddouble-threadscrew.Forsubsequentworka minimumNaOH/DMratioof10.5%hasbeenretained.Underthese conditions,8%ofthecelluloseisdrivenintothefiltrate,and41% ofthehemicellulosesplus14%oftheligninsareextractedbythe alkalinepretreatmentincludingneutralizationstep(Table4).The pHofthefiltrateandoftheextrudateattheendofthisstepis, respectively,of5.5and6.

3.2. DevelopmentoftheimpregnationprofileonBC21

After thealkaline pretreatment and theneutralization step, enzymesareintroducedintotheconveyingzoneofthe bioextru-sionstep(Fig.2),andimpregnationofthematerialisensuredbythe useofaseriesofmechanicalpressureandrelaxationzones.Four constraintzonesareimplementedinmodules10,12–14.Thefirst threeconsistofaseriesofbilobepaddlescrews,thefirstmounted atanangleof90◦_{(noconveyingeffect),andthelasttwoat−45}◦

(retentioneffectonthematerial).Thefourthzoneisformedeither byareversedscreworbyamonolobepaddlescrewmountedat anangleof−45◦.These constraintelementsprovidegood

mix-0 5 10 15 20 25 30 0 10 20 30 40 50

Time (h)

H

y

d

ro

ly

s

a

b

il

it

y

(

%

/D

M

)

Fig.4.EnzymatichydrolysabilitykineticsoftheextrudateproducedbyprofileB3 andB4.

ingofthematerial.Betweentheseareasofmechanicalpressure, conveyingelementswithdecreasingpitchensurethetransportof themixture.Inordertoincreasetheresidencetimeandoptimize thebiomass–enzymecontact,differentconstraintelementshave beentestedinmodule14(Tables1and2).Thereversed double-thread screws C2FC -33 generated insufficientbackpressure to formastable“dynamicplug”,whateverthelengthtested.Reversed screwswithshorterpitchormonolobepaddle-screwswereused andallowedasteadystatetobeachieved.Thedeterminationof theresidencetimedistributionofthematterintheextruderfor thesetwoprofilesB3andB4,showedthattheuseofmonolobe paddle-screwsresultedinaloweraxialdispersionofresidencetime aswellasaslightshorteningthereof.However,inbothcasesthe retentiontimeisverysimilar,closetotwominutes(Fig.3). Com-parisonoftheyieldsofsugarsreleasedbyenzymatichydrolysisata lowconcentration(2.5%,)withoutaddingfreshenzymes,revealed nosignificantdifferenceafter48h(Fig.4).Atmost,thehydrolysis kineticwasinitiallyslightlyaccelerated.Thisresultalsoshowed thattheimplementationofashearingelementsuchasreversed screws,doesnotcausesignificantdeactivationoftheenzymes. 3.3. Increaseintheresidencetimeoftheenzymeimpregnation step:bioextrusion

AscanbeseenfromFig.3theresidencetimeofthematerialin theBC21forthebioextrusionstepisverycloseto2minwhatever

therestricting elementused. Thisis verylow compared tothe timesrequiredforasignificantincreaseinenzymatichydrolysisfor releaseofthemonosaccharides.Thus,tounderstandandvalidate theactionofenzymesintheextruderitisnecessarytoextendtheir actiontimeduringbioextrusion. Theincreasedextrudercontact timeissimulatedbyrecirculatingthebioextrudateinBC21.The operation is undertaken using the same configuration and the same screwprofileas for the firstpassage (Figs. 1 and 2), but withoutfurtheradditionofenzymesolution.Itisasiftheduration of thecontact time wasmultiplied by thenumber of passages throughtheextruder.

The material is worked on more and more withsuccessive passagesthroughtheextruder,andbecomessticky.Thispasty con-sistenceislessandlesstextured,thematerialisliquefiedgradually, anditsrheologicalpropertieschange.Andthischangeinrheology tendstoprovethatafiberdestructuringeffectisoccurringduring bioextrusion,whichwasalsoobservedbySamaniuketal.(2011), whohaveshownasynergisticrelationshipbetweenmixingand enzymeactivityduringenzymatichydrolysis.Andbyusingatorque rheometer,theydemonstratedthatthetorquerequiredformixing biomasssampleswith20%soliddrymatter,decreasesinthe pres-enceofenzymesasafunctionoftime,andthisisaccompaniedby anincreaseinglucoseconverted.

3.4. Natureofthetransformationsofthelignocellulosicmaterial inthetwin-screwextruderduringenzymeimpregnation 3.4.1. Solubilizationandextractionofsaccharides

Tostudytheimpactofbioextrusionontheevolutionofthe mat-terdestructuration,ananalysisofdissolvedcompoundsduringthis bioextrusionismade.Tothisend,samplesofbioextrudateare col-lectedaftereach passagethroughtheBC21 extruder,and then dilutedinwater,withaliquid/solidratio(L/S)of20.Theyarethen pressedandfiltered.Freesugars,reducingsugarsandtotalsugars arethenmeasuredintherecoveredfiltrates.

Tocomparethethermomechanicaleffectswiththe thermome-chanicaleffectscoupledtoenzymaticaction,atestiscarriedout withouttheadditionofenzyme,maintainingasolidliquidratio of2.5byadditionofwater.Analysisoftheresultantfiltratefrom thetestswithoutenzyme,shows thatvery fewsaccharides are extractedduringextrusionwhateverthenumberofbioextrusion passages(Fig.5).Thethermomechanicaleffectgeneratedby extru-sionundertheseoperatingconditionsdoesnotactontherelease ofsugars,eitherinfreeformorinpolysaccharideform.

0 2 4 6 8 10 12 14 16

TMO TM1 TM3 TM7 TME1 TME3 TME5 TME7 Control

S ug a rs ( % /D M ) Free sugars Reducing sugars Total sugars

Fig.5.Sugarreleasedinbioextrudateafter1,3,5or7passagesthroughthetwinscrewextruderwithonlythermomechanicaleffects(TM)orwiththermomechanicaleffects coupledtoenzymaticaction(TME),comparedwithbiomasstreatedinastirredreactorinthepresenceofadvancedenzymes,4.6%/ms,consistencyof30%,70min,T:50◦_C (control).TMOcorrespondstothealkalinpretreatedcoproductofsweetcorn.

-50 0 50 100 150 200

250 1 - 110502 #22 [modified by Administrateur, 4 peaks manually assigned] ED_1

nC 1 Az 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 -2.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 2 - 110502 #25 m al to 51.7 ED_1 nC m i n 2

Fig.6.HPAE-PADanalysisofoligosaccharidesreleasedintothefiltrateintheseventhcycleofbioextrusioncomparedwiththemaltodextrinestandard.

Duringthebioextrusion,freesugarsappear,whichmeansthat theenzymesareactiveintheextruderandthatmonosaccharides arereleased(1–2%ofdrymatterintroduced).Analysisofthe reduc-ing sugarsand of the total sugarsin the wash filtrates reveals thepresence of sugars releasedby enzymesin the twin-screw extruder.These sugarsarein theformof free sugars, oligosac-charidesandpolysaccharides,andtheirproportionincreaseswith thenumber of passagesthrough theBC21extruder.Looking at thedifferencebetweentheproportionofreducingsugarsandof totalsugarsderivedfromthedifferentdeterminationmethods,the DNSmethodquantifiesonlythereducingendsofmono,oligoand polysaccharides,whereasthemethodbyacidhydrolysiscoversall themonosaccharidesinthesesugars.

Acontrolispreparedinordertocomparetheenzymaticaction duringbioextrusionwiththeenzymaticactioninastirredreactor. Alkalipretreatedbiomassisplacedincontactwiththeenzymes inastirredreactorunderthesameoperatingconditionsasforthe bioextrusion(T:50◦_{C,enzymerate:4.6%,L/S:2.5)for70min.The}

lattercorrespondstothetimenecessarytoachievetheseven pas-sagesincludingthedeadtimebetweentwosuccessiveones.The sugarreleasedbyenzymesafterthistimeinthestirredreactor,is thesameintermsofreducingsugarsandlessintermsoftotal sug-arsorfreesugars,comparedtothoseobtainedinthetwin-screw extruder.Thisreflects thefact thatin thecase of bioextrusion, thesugarsarereleasedingreateramountsintheformoflonger chainsorintheformofmonosaccharides.Thedifferenceinthe totalamountreleasedandthelengthofsugarchainwouldseem toshowadifferenceofhydrolyticenzymeaction.However,thisis probablyrelatedtothemixingintensityintheextruder,inducing betterpenetrationofendoglucanasesinsidethematerial.Analysis byionchromatographyHPAE-PADwithaDionexCarboPacPA200 column,confirmsthepresenceofpolysaccharidesreleasedduring bioextrusion(Fig.6).Comparedtostandardmaltodextrins,these wouldbeoligomerswithalowdegreeofpolymerization(DP<6–7). Duringthebioextrusion,thesolubilizationofthesaccharidesis accompaniedbyadestructurationofthebiomass,resultinginan increaseininsolublefineparticlescarriedalonginthefiltratewhen washingthebioextrudate(Fig.7).Thisincreaseswithincreasing numberofpassages,andmayreflect theeffectofbreakdownof thebiomassbyhydrolysisofpolysaccharides.Itsproportionnearly

Fig.7.Fineparticlesproducedbyextrusionandbioextrusioninsingleandmultipass mode.

doubledwhenextrusionwasperformedinthepresenceofenzymes for7successivepassages.

Similarly,theamountofhotwatersolublecomponents mea-suredinthebioextrudatesincreases.Itdoublesafter7passagesin thepresenceofenzymescomparedtoextrusionwithoutenzymes (Fig.8),confirmingtheeffectofthesehydrolyticenzymesinthe extruder.

3.4.2. Modificationofligno(hemi)cellulosicfibers

Thestructuralmodificationofthefibersisobservedunderan electronmicroscopethroughoutthepretreatmentand bioextru-sionstepsinthetwin-screwextruder(Fig.9).Afterpretreatment, thefiberstructureappearsmuchwrinkled(Fig.9b),showingthat thefibersurfacehasreleaseditsextractablecompounds,andapart ofthehemicellulosesandlignin.

Fig.8. Hotwatersolublesafterextrusionandbioextrusioninsingleandmultipass mode.

Fig.9. Scanningelectronmicroscopy(SEM)offiberobservedincaseofrawsweet corncoproduct(a),ofalkalinepretreatedsweetcorncoproduct(b),ofbioextruded sweetcorncoproduct(c).

Thisextractionofhemicellulosesandligninisclearingthe cel-lulosemicrofibrilsonthefibersurface.

Followingbioextrusion,theactionofenzymesinthetwin-screw extruderisvisibleonthesurfaceofthefibers(Fig.9c),andthey appeartobedamaged.Theenzymesattackthesurfaceofthefiber givinga“peeling”effectcausedbyapartialdepolymerizationofthe cellulose.

3.4.3. Modificationofthedegreeofpolymerization(DP)of cellulose

Inthepaperindustry,thedegreeofpolymerizationofthe cel-luloseisconventionallyevaluatedbymeasuringtheviscosityafter

0,00 2,00 4,00 6,00 8,00 10,00 12,00

Raw Biomass Pretreated Bioextrudated (1 passage) Bioextrudated (7 passage) Dyna m ic vi s cosi ty ( m P a. s )

Fig.10. Evolutionofthedynamicviscosityofthesweetcornco-productsatdifferent stagesoftreatment.

dissolvingthepaperpulpincupriethylenediamine solution.The greaterthedegreeofpolymerizationthehighertheviscosityofthe resultingsolution.AccordingtoTAPPIT230om94standard,the relationshipbetweentheviscosity(intrinsicanddynamic)andthe DPiscalculatedaccordingtothefollowingequations:

DP=(0.75×V′)1.105 withV′_{=(954×logV )−325}

andV=c×d×t where

c=constantoftheviscometerused;

d=densityoftheworkingsolution,givenbytheTAPPIT230om 94;

t=measurementduration(s); V′_{=intrinsicviscosity(dm3kg}−1_);

V=dynamicviscosity(mPas).

However,this method,designed for pure cellulose fibers, is notintendedtobeapplieddirectlytocomplexmaterialsasisthe case for sweet corncoproducts. To minimize interferencewith the non-cellulosiccompounds, materials obtainedat each step werepurifiedtoisolatetheirholocellulosicfractionandsolubilize thewater-solublesandthelignin.Theprotocolformeasuringthe viscositydirectlyin thecupriethylene diamine,appliedto holo-cellulosicfractionsofrawmaterial,pretreatedextrudatesandthe bioextrudates,revealedthattheviscosityofthesolutionsincrease sharplywiththepre-treatmentandthengraduallydecreasewith thenumberofbioextrusionpassages(Fig.10).Duringpretreatment inthetwin-screwextruder,aproportionofhemicellulosesis dis-solved,whereascelluloseis notaffected,andthis proportionof hemicellulosestocellulosedecreases.Nonetheless,hemicelluloses arecharacterisedbyasignificantlylowerdegreeof polymerisa-tionthancellulose(3–20timeslower)(Klemmetal.,2005;Girio etal.,2010).Thisresultsin ahigher valueofDPfor pretreated extrudatescomparedtorawmatter,andexplainsthehigher vis-cositymeasuredfortheholocellulosefromthepretreatedbiomass. Thereforethedecreaseinmeasuredviscosityobservedwith bioex-trusion,essentiallyreflectsashorteningofthecellulosechainsand confirmstheenzymeactionduringbioextrusion.

4. Conclusion

The implementation of a new process of deconstruction of annualplantmaterialinatwin-screwextruderisdescribedusing sweet cornco-products.The choice of theprofile is explained. Theprocesshasbeeninvestigatedtobetterunderstandenzymatic actionin thebioextrusion step,andtheresultsobtainedin this studyprovethehydrolyticactionoftheenzymesintheextruder.

Thematerialbecomesstickyandless andlesstextured with successivepassagesthroughtheextruder.Itsrheologicalproperties changeanditgraduallyliquefies.Theproportionofsugarreleased increaseswiththenumberofpassages.

Oligosaccharideswithalowdegreeofpolymerization(DP<6–7) werereleasedduringthebioextrusion.Thecontentofhotwater solublecomponentsmeasuredinthebioextrudates,increases.

Partialdepolymerizationofthecelluloseinduceda“peeling”on thefibersurface.

The enzymeaction during bioextrusion was confirmedby a decreaseinmeasuredviscosityofthesolubilizedbioextrudatein cupriethylenediamine,reflectinginfactashorteningofthe cellu-losechains.

Theseresultsprovethatenzymesareactiveintheextruderand thattheprocessofdeconstructionofannualplantmaterialusing thisapparatusishighlyadvantageousandcouldbeaveryattractive pre-treatmentfortheproductionofbioethanol.Itenables work-ingwithahighconsistency,inahighshearenvironment,overa shorttime,andinitiatingenzymatichydrolysiswithoutexcessive dilution.Inasubsequentstudy,theprocesswillbeadaptedand optimizedinonecontinuousstep.

Acknowledgements

Thiswork hasbeen partiallyfunded bythe European Com-munity(SeventhFrameworkProgrammeunderGrantagreement referencen◦_{227498(BABETHANOLPROJECT)).Theauthorswould}

liketothankthestafffromVTTforthehydrolysabilityanalyses.

References

Alvira,P.,Tomás-Pejó,E.,Ballesteros,M.,Negro,M.J.,2010.Pretreatment

technologiesforanefficientbioethanolproductionprocessbasedon

enzymatichydrolysis:areview.Bioresour.Technol.101(13),4851–4861.

Bailey,M.J.,Biely,P.,Poutanen,K.,1992.Inter-laboratorytestingofmethodsfor

assayofxylanaseactivity.J.Biotechnol.23,257–270.

Bailey,M.,Nevalainen,K.M.H.,1981.Induction,isolationandtestingofstable

Trichodermareeseimutantwithimprovedproductionofsolubilizingcellulase.

EnzymeMicrob.Technol.3,153–157.

Ballesteros,I.,Negro,M.,Oliva,J.M.,Caba ˜nas,A.,Manzanares,P.,Ballesteros,M., 2006.Ethanolproductionfromsteam-explosionpretreatedwheatstraw.Appl.

Biochem.Biotechnol.130,496–508.

Bobleter,O.,1994.Hydrothermaldegradationofpolymersderivedfromplants.

Prog.Polym.Sci.19,797–841.

Cara,C.,Ruiz,E.,Ballesteros,I.,Negro,M.J.,Castro,E.,2006.Enhancedenzymatic

hydrolysisofolivetreewoodbysteamexplosionandalkalineperoxide

delignification.ProcessBiochem.41,423–429.

Carvalheiro,F.,Duarte,L.C.,Gírio,F.M.,2008.Hemicellulosebiorefineries:areview

onbiomasspretreatments.J.Sci.Ind.Res.67,849–864.

Chen,W.H.,Xu,Y.Y.,Hwang,W.S.,Wang,J.B.,2011.Pretreatmentofricestraw

usinganextrusion/extractionprocessatbench-scaleforproducingcellulosic

ethanol.Bioresour.Technol.102,10451–10458.

DeVrije,T.,deHaas,G.G.,Tan,G.B.,Keijsers,E.R.P.,Claassen,P.A.M.,2002.

Pre-treatmentofMiscanthusforhydrogenproductionbyThermotogaelfii.Int.J.

HydrogenEnergy27(11),1381–1390.

Duque,A.,Manzanares,P.,Ballesteros,I.,Negro,M.J.,Oliva,J.M.,González,A., Ballesteros,M.,2014.Sugarproductionfrombarleystrawbiomasspretreated

bycombinedalkaliandenzymaticextrusion.Bioresour.Technol.158,262–268.

Duque,A.,Manzanares,P.,Ballesteros,I.,Negro,M.J.,Oliva,J.M.,Sáez,F., Balles-teros,M.,2013.Optimizationofintegratedalkaline-extrusion

pretreatmentofbarleystrawforsugarproductionbyenzymatichydrolysis.

ProcessBiochem.48,775–781.

Dziezak,J.D.,1989.Singleandtwin-screwextrudersinfoodprocessing.Food

Technol.43,164–174.

Evon,Ph.,Vandenbossche,V.,Pontalier,P.Y.,Rigal,L.,2007.Directextractionofoil

fromsunflowerseedsbytwin-screwextruderaccordingtoanaqueous

extractionprocess:feasibilitystudyandinfluenceofoperatingconditions.Ind.

CropsProd.26(3),351–359.

Ghose,T.K.,1987.Measurementofcellulaseactivities.PureAppl.Chem.59, 257–268.

Girio1,F.M.,Fonseca,C.,Carvalheiro,F.,Duarte,L.C.,Marques,S.,Bogel-Łukasik,R., 2010.Hemicellulosesforfuelethanol:areview.BioresourceTechnol.101(13), 4775–4800.

Harmsen,P.F.H.,Huijgen,W.J.J.,BermúdezLópez,L.M.,Bakker,R.R.C.,2010. LiteratureReviewofPhysicalandChemicalPretreatmentProcessesfor

LignocellulosicBiomass.Wageningen:WageningenURFoodBiobased Research,(Rapport1184)54p.

Himmel,M.E.,Ding,S.-Y.,Johnson,D.K.,Adney,W.S.,Nimlos,M.R.,Brady,J.W., Foust,T.D.,2007.Biomassrecalcitrance.Engineeringplantsandenzymesfor

biofuelsproduction.Science315,804–807.

Galbe,M.,Zacchi,G.,2007.Pretreatmentoflignocellulosicmaterialsforefficient

bioethanolproduction.Adv.Biochem.Eng.108,41–65.

Han,M.,Kang,K.E.,Kim,Y.,Choi,G.W.,2013.Highefficiencybioethanolproduction

frombarleystrawusingacontinuouspretreatmentreactor.ProcessBiochem.

48,488–495.

Kang,K.E.,Han,M.,Moon,S.K.,Kang,H.W.,Kim,Y.,Cha,Y.L.,etal.,2013.

Optimizationofalkali-extrusionpretreatmentwithtwin-screwforbioethanol

productionfromMiscanthus.Fuel109,520–526.

Kartika,I.A.,Pontalier,P.Y.,Rigal,L.,2006.Extractionofsunfloweroilby

twin-screwextruder:screwconfigurationandoperatingconditioneffects.

Bioresour.Technol.97(18),2302–2310.

Karunanithy,C.,Muthukumarappan,K.,2011.Optimizationofalkaliconcentration,

switchgrassparticlesizeandextruderparametersformaximumsugar

recoveryusingreponsesurfacemethodology.Chem.Eng.Technol.34(9),

1413–1426.

Karunanithy,C.,Muthukumarappan,K.,2013.Thermo-mechanicalpre-treatment

offeedstocks.In:Gu,T.(Ed.),GreenBiomassPretreatmentforBiofuels

Production.

Klemm,D.,Heublein,B.,Fink,H.P.,Bohn,A.,2005.Cellulose:fascinating

biopolymerandsustainablerawmaterial.Angew.Chem.Int.Ed.Engl.44(22),

3358–3393.

Kumar,R.,Mago,G.,Balan,V.,Wyman,C.E.,2009.Physicalandchemical

characterizationsofcornstoverandpoplarsolidsresultingfromleading

pretreatmenttechnologies.Bioresour.Technol.100,3948–3962.

Lamsal,B.,Yoo,J.,Brijwani,K.,Alavi,S.,2010.Extrusionasathermo-mechanical

pre-treatmentforlignocellulosicethanol.BiomassBioenergy34,

1703–1710.

Magro,C.,1995.ValorisationdesPaillesdebléparFractionnement Thermo-Mécano-ChimiquedansunRéacteurbivis.INPT,Toulouse

http://www.theses.fr/1995INPT057G

Marechal,P.,2001.AnalysedesPrincipauxFacteursImpliquésdansle Fraction-NementCombinédePaillesetdeSonsdebléenExtrudeurbi-vis: Obtentiond’agro-Matériaux.INPT,Toulouse

http://www.theses.fr/2001INPT031G

Marechal,P.,Jorda,J.,Pontalier,P.Y.,Rigal,L.,2004.Twin-screwextrusionand

filtrationforxylanproductionfromwheatstrawandbran.PaulGatenholm

andMaijaTenkanenWashington(USA)In:Hemicelluloses:Sciencesand

technology–ACSSymposiumSeries864;ED,864,pp.38–50.

McIntosh,S.,Vancov,T.,2011.Optimisationofdilutealkalinepretreatmentfor

enzymaticsaccharificationofwheatstraw.BiomassBioenergy35(7),

3094–3103.

Miller,G.L.,1959.Useofdinitrosalicylicacidreagentfordeterminationofreducing

sugar.Anal.Chem.31,426–428.

Mosier,N.,Wyman,C.,Dale,B.,Elander,R.,Lee,Y.Y.,Holtzapple,M.,Ladisch,M., 2005.Featuresofpromisingtechnologiesforpre-treatmentoflignocellulosic

biomass.Bioresour.Technol.96(6),673–686.

N’Diaye,S.,1996.FractionnementdelaMatiereVe´ıge´ıtale:MiseauPointd’un

Proce´ıdeThermo-Me´ıcano-ChimiqueetMode´ılisationduRe´ıacteurbi-vis

Thesis.INPT,ToulouseFrance.

N’Diaye,S.,Rigal,L.,2000.Factorsinfluencingthealkalineextractionofpoplar

hemicellulosesinatwin-screwreactor:correlationwithspecificmechanical

energyandresidencetimedistributionoftheliquidphaseoriginalresearch

article.Bioresour.Technol.75(1),13–18.

Oliva,J.M.,Sáez,F.,Ballesteros,I.,Gónzalez,A.,Negro,M.J.,Manzanares,P., Ballesteros,M.,2003.Effectoflignocellulosicdegradationcompoundsfrom

steamexplosionpretreatmentonethanolfermentationbythermotolerant

yeastKluyveromycesmarxianus.Appl.Microbiol.Biotechnol.105,

141–154.

Rouilly,A.,Jorda,J.,Rigal,L.,2006.Thermo-mechanicalprocessingofsugarbeet

pulp.I.Twin-screwextrusionprocess.Carbohydr.Polym.66(1),

81–87.

Saha,B.C.,Iten,L.B.,Cotta,M.A.,Wu,Y.V.,2005.Diluteacidpretreatment,

enzymaticsaccharificationandfermentationofwheatstrawtoethanol.

ProcessBiochem.40,3693–3700.

Samaniuk,J.R.,Scott,C.T.,Root,T.W.,Klingenberg,D.J.,2011.Theeffectofhigh

intensitymixingontheenzymatichydrolysisofconcentratedcellulosefiber

suspensions.Bioresour.Technol.102(6),4489–4494.

Sluiter,A.,Hames,B.,Ruiz,R.,Scarlata,C.,Sluiter,J.,Templeton,D.,Crocker,D., 2008.Determinationofstructuralcarbohydratesandlignininbiomass.In:

LaboratoryAnalyticalProcedure.

Sun,Y.,Cheng,J.,2002.Hydrolysisoflignocellulosicmaterialsforethanol

production:areview.Bioresour.Technol.83,1–11.

Taherzadeh,M.J.,Karimi,K.,2008.Pretreatmentoflignocellulosicwastesto

improveethanolandbiogasproduction:areview.Int.J.Mol.Sci.9,

1621–1651.

Um,B.H.,Choi,C.H.,Oh,K.K.,2013.Chemicalseffectontheenzymaticdigestibility

ofrapestrawoverthethermo-mechanicalpretreatmentusingacontinuous

twinscrew-drivenreactor(CTSR).Bioresour.Technol.130,38–44.

Varga,E.,Reczey,K.,Zacchi,G.,2004.Optimizationofsteampretreatmentofcorn

stovertoenhanceenzymaticdigestibility.Appl.Biochem.Biotechnol.113–116,

Vandenbossche,V.,Doumeng,C.,Rigal,L.,2014a.Thermomechanicaland

thermo-mechano-chemicalpretreatmentofwheatstrawusingatwin-screw

extruder.BioResources9(1),1519–1538.

Vandenbossche,V.,Brault,J.,Vilarem,G.,Hernández-Meléndezc,O.,Vivaldo-Lima, E.,Hernández-Luna,M.,Barzana,E.,Duque,A.,Manzanares,P.,Ballesteros,M., Mata,J.,Castellón,E.,Rigal,L.,2014b.Anewlignocellulosicbiomass

deconstructionprocesscombiningthermo-mechanochemicalactionand

bio-catalyticenzymatichydrolysisinatwin-screwextruder.Ind.CropsProd.

55(2014),258–266.

VanDyk,J.S.,Pletschke,B.I.,2012.Areviewoflignocellulosebioconversionusing

enzymatichydrolysisandsynergisticcooperationbetweenenzymes-factors

affectingenzymes,conversionandsynergy.Biotechnol.Adv.30,1458–1480.

VanSoest,P.J.,Wine,R.H.,1967.Useofdetergentsintheanalysisoffibriousfeeds.

IV.Determinationofplantcellwallconstituents.J.AOACInt.50,50–55.

VanSoest,P.J.,Wine,R.H.,1968.Determinationofligninandcelluloseinacid

detergentfiberwithpermanganate.J.AOACInt.51,780–784.

Yoo,J.,Alavi,S.,Vadlani,P.,Amanor-Boadu,V.,2011.Thermo-mechanical

extrusionpre-treatmentforconversionofsoybeanhullstofermentablesugars.

Bioresour.Technol.102,7583–7590.

Zhang,S.,Keshwani,D.R.,Xu,Y.,Hanna,M.A.,2012.Alkalicombinedextrusion

pretreatmentofcornstovertoenhanceenzymesaccharification.Ind.Crops

Prod.37,352–357.

Zeitoun,R.,Pontalier,P.-Y.,Marechal,P.,Rigal,L.,2010.Twin-screwextrusionfor

hemicelluloserecovery:influenceonextractpurityandpurification

performance.Bioresour.Technol.101,9348–9354.

Zheng,Y.,Pan,Z.,Zhang,R.,2009.Overviewofbiomasspretreatmentforcellulosic

ethanolproduction.Int.J.Agric.Biol.Eng.2(3),51–68.

Zheng,J.,Rehmann,L.,2014.Extrusionpretreatmentoflignocellulosicbiomass:a review.Int.J.Mol.Sci.15(18),967–18984,