1
Op imal condi ions o hemicelluloses ex ac ion om
Eucalyp us globules wood: hyd o he mal ea men in a
semi-con inuous eac o
Gianluca Gallina a, Ál a o Cabeza a, Pie domenico Biasi b and Juan Ga cía-Se na a*
a Depa men o Chemical Enginee ing and En i onmen al Technology, High P essu e P ocesses G oup,
Uni e si y o Valladolid, Valladolid, ES-47011, Spain
b P ocess Chemis y Cen e, Labo a o y o Indus ial Chemis y and Reac ion Enginee ing, Åbo
Akademi, Biskopsga an 8, Tu ku/Åbo, FI-20500, Finland
*To whom co espondence should be add essed. E-mail: [email p o ec ed] (J. Ga cía-Se na)
2
Abs ac
Op imal condi ions o hemicellulose ex ac ion om wooden biomass in a semi-con inuous sys em
ha e been assessed in his wo k. This s udy would cons i u e he i s s age o a p o i able and g een
indus ial p ocess. Eucalyp us globulus was selec ed as aw ma e ial due o i s low wa e consump ion,
high g ow h and i s e iciency in lignocellulose p oduc ion. Mo eo e i s cul i a ion is e y popula in
sou he n Eu ope. Samples o 5.0 g o wood we e ac ioned using a p essu ized ho wa e semi-
con inuous sys em, o p oduce suga s (pen oses and hexoses) and a solid esidue en iched in lignin.
Fi e low a es be ween 2.50 and 20.00 mL/min and ou empe a u es be ween 135.0 and 285.0 ºC
we e es ed in o de o maximize he p oduc ion o suga s, a oiding he o ma ion o deg ada ion
p oduc s.
Op imum condi ions o he ex ac ion o hemicellulose we e iden i ied a 185.0 ºC and 5.00 mL/min,
leading o a pen oses yield o 67.409 w %, wi h 0.702 w % o deg ada ion p oduc s. Almos all he pulp
is ex ac ed a 285.0 ºC.
SEM images show e y well he changes in he wood s uc u e a di e en empe a u es.
A kine ic model was de eloped, desc ibing he ex ac ion and hyd olysis o hemicellulose and cellulose
wi h absolu e a e age de ia ions a ound 10 % o suga ex ac ed mass.
Keywo ds
Eucalyp us, hemicelluloses, ac iona ion, bio e ine y, semi-con inuous.
3
1. In oduc ion
Cellulose and hemicellulose con ained in woody biomass can be hyd olysed o monome ic suga s,
which can be u he e men ed o e hanol, o can be con e ed in highe alue p oduc s [1-3]. Xylose
om hemicellulose, o ins ance, can be con e ed o u u al, which is a p ecu so used in di e en
ields, such as oil e ining, plas ics, pha maceu ical, and ag ochemical indus ies [4]. L-Xylose can be
also hyd ogena ed o enzyma ically ans o med o xyli ol, which is a swee ening agen and is also used
o p e en ing oo h decay [5]. HMF (Hyd oxyme hyl u u al) de i ed om hexose suga s can be
oxidized o ob ain 2,5- u andica boxylic acid, which can subs i u e e eph halic acid (PTA) in he
p oduc ion o polyes e s and o he cu en polyme s con aining an a oma ic moie y [6]. The idea o
ans o ming biomass o ene gy, ma e ials, and chemicals, de ines he concep o a bio e ine y [7-10],
pa icula ly being an in e es ing opic nowadays, conside ing he issues ela ed o ossil combus ibles
and de i a i es.
Di ec e men a ion o biomass o e hanol by enzyma ic diges ion canno succeed wi hou a p e-
ea men o modi y he c oss-linked s uc u e be ween lignin and polysaccha ides and educe he
biomass ecalci ance o enzyma ic hyd olysis [11, 12] .
A p omising, clean and cheap way o ac iona e lignocellulosic ma e ials is he so called
au ohyd olysis, which simply consis s o ea ing biomass wi h ho p essu ised liquid wa e : du ing he
eac ion, mos o he hemicelluloses a e ex ac ed and hyd olysed o monome s, wi h a consequen
elease o ace ic acid o igina ed om he clea age o he ace yl g oups bonded o he oligosaccha ides; a
lowe amoun o cellulose is eleased, due o he c ys alline s uc u e o he polyme , which makes i
mo e di icul o dissol e and hyd olyse [13]. Mo eo e , he s uc u e o he cell walls becomes mo e
accessible o enzyma ic a acks. Hyd olysis and deg ada ion o he ex ac ed p oduc s will be mo e
se e e along wi h empe a u e, esidence ime and low pH.
In compa ison wi h o he p e- ea men s wi h mine al acids [14, 15] o bases [16, 17] added o he
eac ion media, au ohyd olisis has a lowe en i onmen al impac , as he only eagen is wa e and no
u he de oxi ica ion ea men s a e equi ed o neu alize he sludges. Au ohyd olysis is also a cos
4
e ec i e p ocess, as he a ia ion o pH in he liquo s is e y low and he e is no co osion o he
equipmen [18, 19]. In li e a u e he e a e se e al examples o his p ocess, pe o med wi h di e en
biomass c ops and esidues. Ag icul u e was es like whea s aw [20, 21], co n s o e [22, 23] and ice
s aw [24, 25] ha e been in ensi ely explo ed o his kind o p ocess, due o hei abundance and easy
a ailabili y. Also many woody biomass ha e been widely used [26-28], as hey con ain less ino ganic
subs ances han ag icul u al c ops [29] and con ain mo e ace yl g oups ha enhance he ca aly ic ac i i y
o he p ocess (in pa icula ha dwood species) [18].
A p o i able biomass o hemicellulose ex ac ion is needed. This aw ma e ial should ha e a high
g ow h a e, a low wa e consump ion and a high con en o hemicellulose. Eucalyp us globulus has all
hese cha ac e is ics and, in pa icula , i is he wo ld’s mos e icien ee o p oducing pulp. Mo eo e ,
eucalyp us wood has a high densi y enabling he ee o cap u e la ge amoun s o CO2 (0.1359 CO2
/yea / ee) and hus o accumula e mo e ca bon pe uni o olume compa ed o o he o es species [30].
A high biomass yield and a low wa e consump ion (306 L/kg d y ma e ial agains 400 L/kg d y
ma e ial o oak ees o 1000 and 2000 L/kg d y ma e ial o he baceous species like co n and soya
espec i ely) [30] make eucalyp us e y a ac i e om an indus ial poin o iew, no only o pape
p oduc ion, bu also as a sus ainable and ca bon-neu al sou ce o liquid uels and bio-compounds.
In addi ion, eucalyp us is a ee o conside able impo ance in he Ibe ian Peninsula and in he wo ld,
due o i s wide expansion and i s sp ead use in indus ial applica ions, mainly o he pape indus y.
Ga o e e al. s udied he ac iona ion o Eucalyp us globulus wood [31] , he ex ac ion o
hemicellulose and he p oduc ion o xylose om xylooligosaccha ides [32], a e p e- ea men s in a
ba ch eac o .
While he e a e se e al s udies dealing wi h he au ohyd olysis in ba ch eac o s [33-39], only a smalle
numbe o a icles ega ds low- h ough eac o s [11, 40, 41].
In ou s udy, we in es iga e he au ohyd olysis o Eucalyp us globulus wood in a semi-con inuous
eac o , consis ing in a ubula eac o loaded wi h wood chips, cons an ly h ough by p essu ized ho
wa e . This kind o se -up allows a high solid / liquid a io and a apid emo al o he p oduc s,
5
p e en ing hei deg ada ion. Mo eo e a con inuous supply o esh wa e o he sys em gua an ees a
high concen a ion g adien a he solid-liquid in e ace, hus, enhancing he mass ans e espec o a
ba ch o a semi-ba ch eac o [29].
Recen s udies ha e been comple ed using low- h ough ex ac ions wi h co n s o e biomass. Au ho s
ound ha low- h ough ex ac ion esul ed in highe xylose yield, and g ea e lignin emo al espec o
ba ch eac o s [42]. The emo al o lignin makes he emaining cellulose a e he p e- ea men s mo e
diges ible by he enzymes [43].
Respec o con inuous eac o , whe e biomass and wa e a e con inuously ed in o he eac o , in semi-
con inuous eac o solid pumping and ex eme milling is a oided, educing conside ably he cos s.
All hese cha ac e is ics make, in ou iew, he semi-con inuous eac o he bes o he p e- ea men o
biomass in a u u e indus ial scene y.
Di e en liquid low a es (2.50, 5.00, 10.00, 15.00, 20.00 mL/min) and di e en eac ion empe a u es
(135.0, 185.0, 235.0, 285.0 °C) we e es ed in o de o maximize he yield o polysaccha ides ex ac ion
a oiding he o ma ion o deg ada ion p oduc s ha would inhibi a u he e men a ion s ep [44].
SEM pic u es o he exhaus ed solid bed we e aken o analyse he s uc u e o he wood a e he p e-
ea men s.
In addi ion o op imizing he empe a u e, whose e ec was al eady explo ed in ba ch sys ems [28, 31],
he liquid esidence ime o he ex ac ion o hemicellulose om eucalyp us in a semi-con inuous
eac o was op imized in his wo k. E ec s o he esidence ime we e suddenly checked a empe a u e
no no mally sui able o he ex ac ion o hemicellulose.
Mo eo e an au o ca aly ic kine ic model de eloped by ou g oup, and p e iously alida ed o ano he
aw ma e ial (holm oak), was simpli ied and implemen ed. I can be s a ed ha e en changing he
biomass, he model ep esen e y well he ex ac ion and hyd olysis in a semi-con inuous eac o .

6
2. Ma e ials and Me hods
2.1 Ma e ials
Eucalyp us globulus wood used as he main aw ma e ial o all he expe imen s o igina ed om
Can ab ia (Spain). Wooden b anches we e cu in slices wi h a jigsaw, and hen educed o small pelle s
wi h an a e age Fe e diame e o 0.6 cm. The composi ion o he aw ma e ial (Table 1) in e ms o
s uc u al ca bohyd a es, ex ac i es, ashes, humidi y and lignin we e de e mined acco ding o he
s anda d me hods published by Na ional Renewable Ene gy Labo a o y (NREL) [45].
The column used o he sepa a ion o he compounds was SUGAR SH-1011 Shodex a 50.0 ºC and a
low o 0.80 mL/min, using a solu ion o 0.01N o sulphu ic acid and wa e Milli-Q as mobile phase. A
Wa e s IR de ec o 2414 and Wa e s dual λ abso bance de ec o 2487 (210 nm and 254 nm) was used o
iden i y he suga s and hei de i a i es.
The calib a ion eagen s used o HPLC analysis we e: cellobiose (+98%), glucose (+99%), uc ose
(+99%), glyce aldehyde (95%), py u aldehyde (40%), a abinose (+99%), glycolaldehyde (+98%), 5-
hyd oxyme hyl u u al (99%), lac ic acid (85%), o mic acid (98%), ac ylic acid (99%), mannose
(+99%), xylose (+99%), galac ose(+99%), le ulinic acid (≥ 97%), u u al (+99%), ace ic acid (+99%)
pu chased om Sigma and used wi hou u he modi ica ion.
Fo he analysis o suga s sulphu ic acid (96%) and calcium ca bona e (≥ 99.0%), pu chased om
Pan eac we e used.
2.2 Reac o se -up o he expe imen s
The expe imen s we e ca ied ou in a labo a o y-scale ixed bed eac o (R-01, 38 cm leng h, ½” O.D.
SS316 piping, 0.37” I.D.) (as depic ed in Figu e 1).
The eac o was cha ged wi h 5.00±0.01 g o wood pelle s, wo me allic il e s we e placed a he op
and a he bo om, in o de o a oid he loss o solid pa icles du ing he expe imen s. Deionized wa e
was in oduced con inuously in o he eac o , in up- low, using a PU-2080 HPLC pump.
7
The eed low, a oom empe a u e, was p ehea ed by he ou - low o he eac o , h ough a concen ic
ube hea exchange wo king in coun e cu en (E-02, 70 cm leng h, 1/4”-3/8”). A p ehea e (E-01, 200
cm o 1/8” AISI 316 piping) was placed a e he hea exchange and loca ed, oge he wi h he eac o ,
inside a o me ch oma og aphic o en HP5680, which could be se a he desi ed empe a u e.
P essu e was con olled by a Go-backp essu e al e (BPV-01) ins alled a he liquid ou le . The ou -
low pH was measu ed online h ough an elec onic pH-me e (Nahi a model 903).
A hea exchange allowed o eco e be ween he 70% and 85% o he ene gy inpu ; he plan , e en i in
labo a o y scale, was designed o ope a e acco ding o g een concep s o ene gy sa ing.
2.3 Expe imen al p ocedu e and analy ical me hods
A se o 9 expe imen s was ca ied ou : 5 wi h changing wa e low a e (2.50, 5.00, 10.00, 15.00, 20.00
mL/min), a cons an empe a u e (185.0°C), and 4 wi h changing empe a u e (135.0, 185.0, 235.0,
285.0 °C) a cons an low a e (5.00 ml/min). P essu e was kep cons an a 100.0 ba , in o de o
gua an ee he liquid phase o he aqueous eac ion media a all he ope a ed empe a u es. In he ini ial
s age o he expe imen s he eac o was illed wi h a cons an amoun o wood, and hen connec ed o
he sys em. A cold liquid p essu e es was made be o e each expe imen , in o de o check he p esence
o leaks in he sys em, and o ensu e he comple e we ing o he wood. Wa e was hen hea ed in a p e-
hea ing capilla y and when he eac ion empe a u e was eached he o en was u ned on and he pump
was se o he desi ed eed wa e low. Time 0 was de ined as he ime in which he i s d op o liquid
le he sys em, a his poin he measu emen o pH s a ed and a i s sample o liquid was aken.
The o al ime o each expe imen was 90 min, 20.0 mL liquid samples we e aken e e y 10 min, pH
was eco ded online e e y 1 minu e du ing he i s 30 min, and hen e e y 2 minu es un il he end o
he es s. A e 90 min, he pump was swi ched o o ze o low and he o en empe a u e lowe ed o
20.0 ºC; he sys em was hen slowly dep essu ised, he eac o un igh ened and all i s con en was
collec ed in a beake and d ied o 24h a 60.0 °C. Finally he emp y eac o was econnec ed o he
sys em and deionized wa e was lushed o clean all he pipes.
8
To de e mine he amoun o suga s ex ac ed and he deg ada ion p oduc s p oduced a e he
au ohyd olysis, a pos hyd olysis p ocess o he ex ac ed liquo was pe o med o b eak all he
oligome s in monome s, and allowing he accu a e coun o he ex ac ed p oduc s a e a HPLC
sepa a ion.
10.0 mL o each sample we e comple ely hyd olyzed wi h 4.0 mL o sul u ic acid 96 %w . and
consequen ly incuba ed in an o en o 30 min a 30.0 °C. The mix u es we e hen dilu ed wi h 86.0 mL
o deionized wa e and wa med in an o en o 1 hou a 121.0 ºC. A he end o he acid hyd olysis, he
samples we e cooled down o oom empe a u e, calcium ca bona e was added in o de o aise he pH
o a alue be ween 6 and 7, he solu ion was il e ed h ough 0.22 m nylon il e s and he con en o
suga s was de e mined by HPLC.
The solid esul ing om each expe imen was p ocessed as explained by he s anda d me hods published
by Na ional Renewable Ene gy Labo a o y (NREL) [45], esidue o Klason lignin was de e mined as
well as he amoun and quali y o soluble compounds no ex ac ed by he he mohyd olysis.
Unp ocessed eucalyp us wood was cha ac e ized as explained in pa ag aph 2.1, in o de o ela e he
amoun o compounds ex ac ed du ing he expe imen s wi h he aw ma e ial composi ion.
Main peaks and a eas we e iden i ied and calcula ed h ough a band-analysis ia as Fou ie ans o m
(FFT) and band-adjus men by Gaussian unc ions. The adjus men was done by minimizing he
quad a ic e o using a Nelde -Mead algo i hm.
2.4 Unce ain y analysis
In o de o check he eliabili y o he expe imen al da a, he unce ain y o all o hem was calcula ed.
Rega ding he concen a ion ob ained by HPLC Eq. 1 was used o conside he epea abili y, he
expe imen al da a de ia ion and he e ec o he calib a ion. Once his alue was ob ained, a ypical
p opaga ion exp ession was used o ob ain he unce ain y o he alues calcula ed ollowing he NREL
s anda d me hods [45].
9
𝑠x=𝑠y
m·√1
𝐿+1
𝑁+(𝑦x−𝑦)2
𝑚2·𝑆xx
𝑠y=√∑(𝑦i−𝑦i
)2
𝑁−2
𝑠xx=∑(𝑥i−𝑥)
N
𝑖=1
(1)
3. Resul s and discussion
The o al amoun o suga s and ace yl g oups con ained in eucalyp us wood was equal o 64.532 w % o
he o al mass (sum o glucans, xylan, a abinan abd ace yl g oups ep esen ed in Table 1).
Semicon inuous ex ac ion/ eac ion is pa icula in e ms o esidence ime. In he semicon inuous plan
wo di e en eac ion imes we e de ined in he sys em:
 solid esidence ime

sol, which was cons an o all he es s (always 90 min), and co esponded
o he o al du a ion ime o he expe imen s, as he solid was s a ic inside he eac o ;
 liquid esidence ime

liq, which a ied in ela ion wi h he ope a ional condi ions o he
expe imen , in pa icula wi h he liquid low a e and wi h he po osi y ( oid olume o he
bed). An ini ial po osi y

0 was de ined, as he a io be ween he olume o he emp y spaces in
he eac o a he beginning o he expe imen s, and he o al olume o he emp y eac o .

0 was
de e mined in he lab o be 0.70. A inal po osi y  was calcula ed (Eq. 2) aking in o accoun
he mass a ia ion o he solid pa icles inside he eac o be ween he beginning and he end o
each expe imen ; m0 and m co esponded o he ini ial and he inal mass o he solid in he
eac o . Liquid esidence ime (Eq. 3) was calcula ed conside ing an a e age po osi y

a
be ween he ini ial 𝜀0 and he inal po osi y 𝜀 o he bed in each expe imen . To calcula e he
inal po osi y, i was assumed ha he densi y o he wood pa icles emained cons an du ing
ime, and ha he e we e only a ia ions in he olume o he pa icles.
16
shows ha he model was able o success ully simula e he hemicellulose deace yla ion. In addi ion,
om he adjus men shown in Figu e 5c, i can be concluded ha he model also could ep oduce he
beha io o he hexose suga s, whose maximum dela ed a ound 60 min wi h espec o pen oses due o
he high esis ance o cellulose agains hyd o he mal deg ada ion. The e o e, pen oses a e ex ac ed a
he beginning o he ope a ion and hexoses only a e eco e ed a he inal s age. Howe e , a ound 20 %
o suga deg ada ion was ound a hese ope a ional condi ions (Table 2), which is ela i ely high.
On he o he hand, he A.A.D. o each expe imen is a ayed in Table 3, being he a e age de ia ion
28.57, 39.20 and 5.7 % o pen oses, hexoses and pH espec i ely. These disc epancies we e ela i e
low aken in o accoun he small amoun o sample in oduced in he eac o , he dilu ion o he samples,
he complexi y o he p ocess (which has been simula ed only wi h 8 compounds) and he biodi e si y
o he wood. In addi ion, he de ia ions all up o 10.49 % and 13.17 % o C5 and C6, espec i ely,
when he simula ed and he expe imen al ex ac ed mass a e compa ed (Figu e 2d and 2e). The e o e, i
seems ha he p oposed model was able o ep oduce he biomass ac iona ion by hyd o he mal
ea men s.
3.3.2. Analysis o he model pa ame e s
The physical sense o he pa ame e s lis ed in he Table 4 was checked in his sec ion. Rega ding he
mass ans e coe icien (𝑘𝑗·𝑎), i was seen ha all o hem ollowed a linea unc ion wi h low
( eg ession coe icien s g ea e han 0.91) as i was expec ed. In he same way, i was calcula ed ha he
equilib ium cons an o solubili y (𝐻𝑗) was enhanced wi h empe a u e linea ly (R2 g ea e han 0.91),
which is a common beha iou in a solid dissolu ion p ocess. Finally, i was also ob ained ha he kine ic
cons an s (𝑘𝑖) ollowed he A henius’ law wi h R2 bigge han 0.97.
Fu he mo e,  inc eases wi h empe a u e and dec eases wi h low. Tempe a u e is he main a iable in
his p ocess and i enhances he ex ac ion and jus i ying he inc emen in he accele a ion ac o (). In
con as , he low educes he accele a ion ac o , al hough i also enhances ex ac ion. The eason could

17
be ha i also a ec s he esidence ime o he liquid phase, so a highe low means less deg ada ion
p oduc o ma ion, mo e dilu ion and smo he liquid p o iles. I is also ema kable ha he kine ics o
cellulose ac iona ion a e always lowe han he kine ics o hemicellulose deg ada ion, which ag ees
wi h he ac ha cellulose is s onge han hemicellulose agains hyd o he mal deg ada ion.
3.4. Lignin emo al and s uc u al al e a ion
Table 2 shows ha a 185.0 °C lignin con en in he exhaus solid is almos equal o he one in he aw
ma e ial. A 235.0 °C and 5.00 mL/min a lignin educ ion o app oxima ely 15.00 w % occu ed, while
a educ ion o app ox. 52 w % a 285.0 °C occu ed. A loss in lignin con en in wooden biomass du ing
hyd o he mal p e- ea men s is widely documen ed [12, 55], Leschinsky e al. epo ed ha C he mal
p e ea men o Eucalyp us globulus wood a 170 °C causes a loss o he molecula weigh in lignin
[56].
When lignocellulosic biomass is subjec ed o high empe a u e o mild acidic p e- ea men s, lignin and
lignin-ca bohyd a e complexes coalesce c ea ing some sphe ical o ma ions ha mig a e ou o he wall
cells and deposi in he su ace o he esidual biomass. These d ople s ha e a nega i e e ec on
enzyma ic hyd olysis o cellulose, a ec ing he e iciency o enzyma ic con e sion in a lignocellulosic
bio e ine y [57].
Figu e 6 shows 3 SEM images o he Eucalyp us wood a e ea men a 185.0 °C, 235.0 °C and 285.0
°C, wi h a liquid low a e o 5.00 mL/min. A 185.0°C no e iden changes a e isible in he wood
s uc u e: he h ee-dimensional s uc u e o lignin associa ed wi h linea molecules o cellulose is
isible in igu e 6a.
A modi ica ion in he s uc u e o he wood is e iden a 235.0 °C: lignin-ca bohyd a e d ople s s a o
appea on he wood su ace, as shown in igu e 6b. The numbe o d ople s inc eases signi ican ly a
285.0 °C ( igu e 3c) whe e only spo adic ib es o b oken cellulose a e isible, di ec ly connec ed
h ough hemicellulose uni s (associa ed hemicellulose) as desc ibed in chap e 3.1.
18
While in ba ch eac o s, all he d ople s deposi on he su ace and ha den a e he cooling, c ea ing a
ba ie o enzyma ic a ach, in semi-con inuous eac o his issue is ligh ened, as he liquid low ca ies
hem cons an ly ou o he sys em; his is e iden om he lignin loss obse ed a high empe a u es.
This is ano he big ad an age o using a semicon inuous eac o o he p e ea men o biomass.
4. Conclusions
In his s udy, we ha e ocused in he e ec o liquid low a e on he ex ac ion o hemicellulose om
eucalyp us biomass, mo e speci ically in he esidence ime. Ope a ing in semi-con inuous mode has he
ad an age o sepa a ing he esidence ime o he solid (easy o cha ge and keep inside a ubula eac o )
and he liquid (mo ing h ough he bed c ea ed). Solid esidence imes be ween 20 and 40 min a e
pe ec o ex ac ing hemicelluloses, liquid esidence imes below 2.00 min a oid by-p oduc s wi h a
empe a u e o 185° C. The in luence o he low is no pe cei ed a 135 ºC, while a highe
empe a u es, he inc ease in yield is coun e balanced by he o ma ion o deg ada ion p oduc s. A
simpli ied model conside ing an au o ca aly ic kine ics ep esen s he ope a ion e y well. This wo k
es ablishes he basis o he scale-up o he semi-con inuous hyd o he mal eac ion.
Acknowledgemen s
The au ho s acknowledge he Spanish Economy and Compe i i eness Minis y, P ojec F acBioFuel:
ENE2012-33613 and he egional go e nmen (Jun a de Cas illa y León), P ojec Re e ence:
VA330U13 o unding. MEng. Gianluca Gallina wish o acknowledge he Spanish Economy and
Compe i i eness Minis y o he schola ship/p edoc o al con ac BES-2013-063556. Al a o Cabeza
also wish o acknowledge he Spanish Economy and Compe i i eness Minis y o he
schola ship/p edoc o al con ac FPU2013/01516.
Abb e ia ions and symbols
19
G eek le e s
𝛼𝑖,𝑗: Ini ial eloci y ac o o he compound “j” in he eac ion “i”, dimensionless.
𝛽𝑖,𝑗: Accele a ion ac o o he compound “j” in he eac ion “i”, dimensionless.
𝛽𝐶: Accele a ion ac o o cellulose, dimensionless.
𝛽𝐻𝑐: Accele a ion ac o o hemicellulose, dimensionless.
ℰ: Po osi y o he bed, dimensionless.
ℰ : Po osi y o he bed, calcula ed a he end o he expe imen , dimensionless.
ℰa : A e age po osi y o he bed, be ween he beginning and he end o he expe imen , dimensionless.
ℰ0: Po osi y o he bed, calcula ed a he end o he expe imen , dimensionless.
𝜑: Rela ion ac o be ween po osi y and he o al concen a ion in solid phase, dimensionless.
Ф𝑖,𝑗: S oichiome ic coe icien o he compound “j” o he eac ion “i”, mg.
sol : esidence ime o solid inside he eac o , min
liq : esidence ime o liquid inside he eac o , min
Symbols
𝐶𝑓𝑗: Concen a ion o he compound “j” in he phase “ ”, mg/L
𝐶𝐿𝑗: Concen a ion o he compound “j” in he liquid phase, mg/L
𝐶𝐿𝑗: A e age concen a ion o he compound “j” along he eac o in liquid phase, mg/L
𝐶𝐿𝑗∗: Equilib ium concen a ion o he compound “j” in liquid phase, mg/L
𝐶𝑆𝑗: Cocne a ion o he compound “j” in he solid phase, mg/L
𝐶𝑡: To al concen a ion in he solid, mg/L
Q: Liquid low a e, mL/min
𝐸𝑎/𝑅: Ac i a ion ene gy, K
𝐻𝑗: Equilib ium cons an be ween he solid and he liquid, dimensionless
𝑘: P e-exponen ial ac o o he kine ic cons an , mg-1·min-1
𝑘𝑖: Kine ic cons an , mg-1·min-1
𝑘𝑗·𝑎: Mass ans e coe icien mul iplied by he speci ic exchange a ea, min-1
𝑁: Numbe o compounds, dimensionless
20
𝑛𝑟𝑒𝑐: Numbe o eac ions, dimensionless
𝐿: Leng h o he eac o , m
𝑡: Ope a ing ime, min
𝑟𝑖: Reac ion eloci y “i”, mg/min·L
𝑟𝑗: Reac ion a e o he compound “j”, mg/min·L
𝑅2: Coe icien R2, dimensionless
𝑇: Ope a ing empe a u e, ºC
𝑢: Liquid eloci y in he eac o , m/min
𝑥𝑖𝐸𝑋𝑃: Expe imen al alue o he i ed a iable
𝑥𝑖𝑆𝐼𝑀: Simula ed alue o he i ed a iable
𝑧: Coo dina e along he leng h o he eac o , dimensionless
𝑠𝑥: unce ain y o he expe imen al concen a ion, ppm
𝑠𝑦: s anda d de ia ion o he calib a ion pa ons, ppm
𝑠𝑥𝑥: de ia ion o he HPLC a eas, dimensionless
𝐿: numbe o expe imen epe i ions, dimensionless
𝑁: numbe o calib a ion poin s, dimensionless
𝑚: slope o he calib a ion, ppm
𝑥𝑖: expe imen al HPLC a ea, dimensionless
𝑥: a e age expe imen al HPLC a ea, dimensionless
𝑦𝑖: pa on concen a ion “i”, ppm
𝑦𝑖
: calcula ed pa on concen a ion “i”, ppm
𝑦𝑥: a e age expe imen al concen a ion, ppm
𝑦: a e age pa on concen a ion, ppm
Re e ences
[1] F. Che ubini, The bio e ine y concep : Using biomass ins ead o oil o p oducing ene gy and
chemicals, Ene gy Con e sion and Managemen , 51 (2010) 1412-1421.
[2] I. Egües, M.G. Al iols, Z. He seczki, G. Ma on, J. Labidi, Hemicelluloses ob aining om apeseed
cake esidue gene a ed in he biodiesel p oduc ion p ocess, Jou nal o Indus ial and Enginee ing
Chemis y, 16 (2010) 293-298.
21
[3] S.-H. Yoon, A. an Heiningen, G een liquo ex ac ion o hemicelluloses om sou he n pine in an
In eg a ed Fo es Bio e ine y, Jou nal o Indus ial and Enginee ing Chemis y, 16 (2010) 74-80.
[4] C. Rong, X. Ding, Y. Zhu, Y. Li, L. Wang, Y. Qu, X. Ma, Z. Wang, P oduc ion o u u al om
xylose a a mosphe ic p essu e by dilu e sul u ic acid and ino ganic sal s, Ca bohyd Res, 350 (2012)
77-80.
[5] P. Mäki-A ela, B. Holmbom, T. Salmi, D.Y. Mu zin, Recen P og ess in Syn hesis o Fine and
Special y Chemicals om Wood and O he Biomass by He e ogeneous Ca aly ic P ocesses, Ca alysis
Re iews: Science and Enginee ing, 49 (2007) 197 - 340.
[6] P.N.N.L. (PNNL), N.R.E.L. (NREL), Top Value Added Chemicals F om Biomass in: U.S.D.o.
Ene gy (Ed.), 2004.
[7] J.H. Cla k, V. Buda in, F.E.I. Deswa e, J.J.E. Ha dy, F.M. Ke on, A.J. Hun , R. Luque, D.J.
Macqua ie, K. Milkowski, A. Rod iguez, O. Samuel, S.J. Ta ene , R.J. Whi e, A.J. Wilson, G een
chemis y and he bio e ine y: A pa ne ship o a sus ainable u u e, G een Chemis y, 8 (2006) 853-
860.
[8] M. González, A. Ga cía, A. Toledano, R. Llano-Pon e, M.A. De And és, J. Labidi, Lignocellulosic
eeds ock bio e ine y p ocesses. Analysis and design, Chemical Enginee ing T ansac ions, 17 (2009)
1107-1112.
[9] D. Mu zin, Chemical Enginee ing o Renewables Con e sion, Fi s ed., 2013.
[10] J.J. Bozell, G.R. Pe e sen, Technology de elopmen o he p oduc ion o biobased p oduc s om
bio e ine y ca bohyd a es- he US Depa men o Ene gy's "Top 10" e isi ed, G een Chemis y, 12
(2010) 539-554.
[11] T. Rogalinski, T. Ing am, G. B unne , Hyd olysis o lignocellulosic biomass in wa e unde
ele a ed empe a u es and p essu es, The Jou nal o Supe c i ical Fluids, 47 (2008) 54-63.
[12] Y. Pu, F. Hu, F. Huang, B. Da ison, A. Ragauskas, Assessing he molecula s uc u e basis o
biomass ecalci ance du ing dilu e acid and hyd o he mal p e ea men s, Bio echnology o Bio uels, 6
(2013).
[13] P. Mäki-A ela, T. Salmi, B. Holmbom, S. Will ö , D.Y. Mu zin, Syn hesis o suga s by hyd olysis
o hemicelluloses- A e iew, Chemical Re iews, 111 (2011) 5638-5666.
[14] F. Ca asco, C. Roy, Kine ic s udy o dilu e-acid p ehyd olysis o xylan-con aining biomass, Wood
Science and Technology, 26 189-208.
[15] J. A eola-Va gas, V. Ojeda-Cas illo, R. Snell-Cas o, R.I. Co ona-González, F. Ala is e-
Mond agón, H.O. Méndez-Acos a, Me hane p oduc ion om acid hyd olysa es o Aga e equilana
bagasse: E alua ion o hyd olysis condi ions and me hane yield, Bio esou ce Technology, 181 (2015)
191-199.
[16] S.M. Hosseini, H.A. Aziz, Sya alni, M.V. Kiamahalleh, E alua ion o signi ican pa ame e s on
alkaline p e ea men p ocess o ice s aw, KSCE Jou nal o Ci il Enginee ing, 17 (2013) 921-928.
[17] Y. He, Y. Pang, X. Li, Y. Liu, R. Li, M. Zheng, In es iga ion on he Changes o Main
Composi ions and Ex ac i es o Rice S aw P e ea ed wi h Sodium Hyd oxide o Biogas P oduc ion,
Ene gy & Fuels, 23 (2009) 2220-2224.
[18] G. Ga o e, H. Domínguez, C.J. Pa ajó, Hyd o he mal p ocessing o lignocellulosic ma e ials, Holz
als Roh- und We ks o , 57 191-202.
[19] H. Ami i, K. Ka imi, Au ohyd olysis: A p omising p e ea men o he imp o emen o ace one,
bu anol, and e hanol p oduc ion om woody ma e ials, Chemical Enginee ing Science, 137 (2015) 722-
729.
[20] F. Ca alhei o, T. Sil a-Fe nandes, L.C. Dua e, F.M. Gí io, Whea s aw au ohyd olysis: P ocess
op imiza ion and p oduc s cha ac e iza ion, Applied Biochemis y and Bio echnology, 153 (2009) 84-
93.
[21] D. Naba la z, A. Eb inge o á, D. Mon ané, Au ohyd olysis o ag icul u al by-p oduc s o he
p oduc ion o xylo-oligosaccha ides, Ca bohyd a e Polyme s, 69 (2007) 20-28.
[22] K.L. Kadam, C.Y. Chin, L.W. B own, Op imiza ion o a biomass ac iona ion p ocess and
e alua ion o i s economic easibili y, in: 2007 AIChE Annual Mee ing, 2007.

22
[23] C.T. Bu uiana, C. Vizi eanu, G. Ga o e, J.C. Pa ajó, Op imiza ion o co n s o e bio e ine y o
cop oduc ion o oligome s and second gene a ion bioe hanol using non-iso he mal au ohyd olysis,
Indus ial C ops and P oduc s, 54 (2014) 32-39.
[24] S. Imman, J. A n hong, V. Bu apa ana, N. Laosi ipojana, V. Champ eda, Au ohyd olysis o
opical ag icul u al esidues by comp essed liquid ho wa e p e ea men , Applied Biochemis y and
Bio echnology, 170 (2013) 1982-1995.
[25] P. Moniz, J. Lino, L.C. Dua e, L.B. Rosei o, C.G. Boe iu, H. Pe ei a, F. Ca alhei o, F ac iona ion
o hemicelluloses and lignin om ice s aw by combining au ohyd olysis and op imised mild
o ganosol deligni ica ion, BioResou ces, 10 (2015) 2626-2641.
[26] L. Senila, M. Senila, C. Va a iceanu, C. Roman, L. Silaghi-Dumi escu, Au ohyd olysis
P e ea men and Deligni ica ion o Sil e Fi Wood o Ob ain Fe men able Suga s o Bioe hanol
P oduc ion, Ene gy Sou ces, Pa A: Reco e y, U iliza ion and En i onmen al E ec s, 37 (2015) 1890-
1895.
[27] Z.M. Zhao, H.Q. Shi, G. Liu, N. Li, J. Zhang, M.H. Niu, Q.W. Ping, Dissolu ion and
depolyme iza ion o hemicelluloses o acacia wood du ing au ohyd olysis, Zhongguo Zaozhi
Xuebao/T ansac ions o China Pulp and Pape , 30 (2015) 6-11.
[28] T. Sil a-Fe nandes, L.C. Dua e, F. Ca alhei o, M.C. Lou ei o-Dias, C. Fonseca, F. Gí io,
Hyd o he mal p e ea men o se e al lignocellulosic mix u es con aining whea s aw and wo
ha dwood esidues a ailable in Sou he n Eu ope, Bio esou ce Technology, 183 (2015) 213-220.
[29] P. Kilpeläinen, V. Ki unen, J. Hemming, A. P ano ich, H. Il esniemi, S. Will ö , P essu ized ho
wa e low- h ough ex ac ion o bi ch sawdus – E ec s o sawdus densi y and sawdus size, No dic
Pulp & Pape Resea ch Jou nal, 29 (2014) 547-556.
[30] ENCE, Sus ainable o es managemen and Eucalyp us, in: ENCE Ca alog 2009, 2009.
[31] A. Romaní, G. Ga o e, F. López, J.C. Pa ajó, Eucalyp us globulus wood ac iona ion by
au ohyd olysis and o ganosol deligni ica ion, Bio esou ce Technology, 102 (2011) 5896-5904.
[32] G. Ga o e, H. Domınguez, J.C. Pa ajó, Gene a ion o xylose solu ions om Eucalyp us globulus
wood by au ohyd olysis–pos hyd olysis p ocesses: pos hyd olysis kine ics, Bio esou ce Technology, 79
(2001) 155-164.
[33] Q. Hou, Y. Wang, W. Liu, L. Liu, N. Xu, Y. Li, An applica ion s udy o au ohyd olysis
p e ea men p io o popla chemi- he momechanical pulping, Bio esou Technol, 169 (2014) 155-161.
[34] G. Ga o e, H. Domınguez, J.C. Pa ajó, Kine ic modelling o co ncob au ohyd olysis, P ocess
Biochemis y, 36 (2001) 571-578.
[35] D. Sidi as, F. Ba zias, R. Ranjan, M. Tsapa sis, Simula ion and op imiza ion o ba ch
au ohyd olysis o whea s aw o monosaccha ides and oligosaccha ides, Bio esou ce Technology, 102
(2011) 10486-10492.
[36] T. Ojumu, B.a. A ahDaniel, E. Be iku, B. Solomon, Au o-hyd olysis o lignocellulosics unde
ex emely low sulphu ic acid and high empe a u e condi ions in ba ch eac o , Bio echnology and
Biop ocess Enginee ing, 8 (2003) 291-293.
[37] M.H. Sipponen, V. Pihlajaniemi, S. Sipponen, O. Pas inen, S. Laakso, Au ohyd olysis and aqueous
ammonia ex ac ion o whea s aw: e ec o ea men se e i y on yield and s uc u e o hemicellulose
and lignin, RSC Ad ances, 4 (2014) 23177-23184.
[38] J.V. Rissanen, H. G énman, S. Will ö , D.Y. Mu zin, T. Salmi, Sp uce Hemicellulose o
Chemicals Using Aqueous Ex ac ion: Kine ics, Mass T ans e , and Modeling, Indus ial & Enginee ing
Chemis y Resea ch, 53 (2014) 6341-6350.
[39] L. Cheng, X.P. Ye, R. He, S. Liu, In es iga ion o apid con e sion o swi chg ass in subc i ical
wa e , Fuel P ocessing Technology, 90 (2009) 301-311.
[40] X. Chen, M. Lawoko, A.V. Heiningen, Kine ics and mechanism o au ohyd olysis o ha dwoods,
Bio esou ce Technology, 101 (2010) 7812-7819.
[41] T. Ing am, T. Rogalinski, V. Bockemühl, G. An anikian, G. B unne , Semi-con inuous liquid ho
wa e p e ea men o ye s aw, The Jou nal o Supe c i ical Fluids, 48 (2009) 238-246.
23
[42] C. Liu, C.E. Wyman, Impac o Fluid Veloci y on Ho Wa e Only P e ea men o Co n S o e in a
Flow h ough Reac o , in: M. Finkels ein, J.D. McMillan, B.H. Da ison, B. E ans (Eds.) P oceedings o
he Twen y-Fi h Symposium on Bio echnology o Fuels and Chemicals Held May 4–7, 2003, in
B ecken idge, CO, Humana P ess, To owa, NJ, 2004, pp. 977-987.
[43] H. Pei, L. Liu, X. Zhang, J. Sun, Flow- h ough p e ea men wi h s ongly acidic elec olyzed wa e
o hemicellulose emo al and enzyma ic hyd olysis o co n s o e , Bio esou ce Technology, 110
(2012) 292-296.
[44] M.S. Mo eno, F.E. Ande sen, M.S. Díaz, Dynamic Modeling and Pa ame e Es ima ion o Uni
Ope a ions in Lignocellulosic Bioe hanol P oduc ion, Indus ial & Enginee ing Chemis y Resea ch, 52
(2013) 4146-4160.
[45] N.R.E. Labo a o y, De e mina ion o S uc u al Ca bohyd a es and Lignin in Biomass, in, Golden,
CO, 2011.
[46] C.E. Wyman, S.R. Decke , M.E. Himmel, J.W. B ady, C.E. Skopec, L. Viika i, Hyd olysis o
cellulose and hemicellulose, in: S. Dumi iu (Ed.) Polysaccha ides: S uc u al Di e si y and Func ional
Ve sa ili y, CRC P ess, 2005, pp. 1023-1062.
[47] G.-J. L , S.-B. Wu, R. Lou, KINETIC STUDY FOR THE THERMAL DECOMPOSITION OF
HEMICELLULOSE ISOLATED FROM CORN STALK, 2010.
[48] X.F. Sun, F. Xu, R.C. Sun, Z.C. Geng, P. Fowle , M.S. Bai d, Cha ac e is ics o deg aded
hemicellulosic polyme s ob ained om s eam exploded whea s aw, Ca bohyd a e Polyme s, 60 (2005)
15-26.
[49] G. Ga o e, H. Domínguez, J.C. Pa ajó, S udy on he deace yla ion o hemicelluloses du ing he
hyd o he mal p ocessing o Eucalyp us wood, Holz als Roh- und We ks o , 59 (2001) 53-59.
[50] G. Ga o e, M.E. Eugenio, M.J. Diaz, J. A iza, F. López, Hyd o he mal and pulp p ocessing o
Eucalyp us, Bio esou ce Technology, 88 (2003) 61-68.
[51] C. Liu, C. Wyman, The e ec o low a e o comp essed ho wa e on xylan, lignin, and o al mass
emo al om co n s o e , Ind Eng Chem Res, 42 (2003) 5409 - 5416.
[52] A. Cabeza, F. Sob ón, F.M. Yed o, J. Ga cía-Se na, Two-phase modelling and simula ion o he
hyd o he mal ac iona ion o holm oak in a packed bed eac o wi h ho p essu ized wa e , Chemical
Enginee ing Science, 138 (2015) 59-70.
[53] R. Capa , L. Khezami, A.K. Bu nham, Assessmen o a ious kine ic models o he py olysis o a
mic og anula cellulose, The mochimica Ac a, 417 (2004) 79-89.
[54] W. P ess, S. Teukolsky, W. Ve e ling, B. Flanne y, Nume ical ecipes 3 d edi ion: The a o
scien i ic compu ing, (2007).
[55] H. T ajano, N. Engle, M. Fos on, A. Ragauskas, T. Tschaplinski, C. Wyman, The a e o lignin
du ing hyd o he mal p e ea men , Bio echnology o Bio uels, 6 (2013) 110.
[56] M. Leschinsky, G. Zucke s ä e , K. Webe Hedda, R. Pa , H. Six a, E ec o au ohyd olysis o
Eucalyp us globulus wood on lignin s uc u e. Pa 1: Compa ison o di e en lignin ac ions o med
du ing wa e p ehyd olysis, in: Holz o schung, 2008, pp. 645.
[57] J.D. DeMa ini, S. Pa a hil, U. A ci, K. Szekalski, K. Mazumde , M.G. Hahn, C.E. Wyman,
Applica ion o monoclonal an ibodies o in es iga e plan cell wall decons uc ion o bio uels
p oduc ion, Ene gy & En i onmen al Science, 4 (2011) 4332-4339.
24
Table and Figu e Cap ions
Table 1. Composi ion o he aw ma e ial
Table 2. Expe imen al able o he s udy o eucalyp us wood au ohyd olysis in a semicon inuous
eac o .
Table 3. A.A.D. o he i ed expe imen s.
Table 4. Mass ans e and kine ic pa ame e s ob ained om he i ing o he model o expe imen al
da a.
Figu e 1. Schema ic low diag am o he expe imen al sys em. equipmen : D-01 eede , P-01 pump,
E-01 eed wa e p ehea e , R-01 hyd o he mal eac o , F-01 eac o ai con ec ion o en, E-02 p ehea
capilla y, BPV-01 p oduc dep essu iza ion Go-backp essu e al e, E-03 p oduc coole , D-02 liquid
p oduc essel.
Figu e 2. F ac ion o soluble compounds in liquid as a unc ion o solid esidence ime a a cons an
liquid low a e 5 mL/min and cons an empe a u e 185.0 ºC.
Figu e 3. Yield o soluble compounds as a unc ion o empe a u e a a cons an liquid low a e 5
mL/min (a). pH in unc ion o sol a di e en empe a u es and cons an low a e 5 mL/min (b). Yield
o soluble compounds as a unc ion o he liquid eac ion ime a 185.0°C (c). pH in unc ion o sol a
di e en liquid low a es and cons an empe a u e 185.0°C (d).
Figu e 4. Simpli ied eac ion pa hway o eucalyp us hyd o he mal ac iona ion.
Figu e 5. Adjus men o he expe imen a 285.0 ºC and 5 ml/min. (a) Suga C5, (b) pH and (c) Suga
C6 and compa ison be ween he expe imen al and simula ed ex ac ed mass o C5 (d) and C6 (e) . C5-
25
SIM: Simula ed concen a ion o suga s C5. C6-SIM: Simula ed concen a ion o suga s C6. pH-SIM:
Simula ed pH. C5-EXP: Expe imen al concen a ion o suga s C5. C6-EXP: Expe imen al concen a ion
o suga s C6. pH-EXP: Expe imen al pH.
Figu e 6. SEM images o Eucalyp us exhaus ed wood a e p e ea men a a) 185.0°C, b) 235.0 °C, and
c) 285.0 °C wi h a liquid low a e o 5 mL/min.
Tables and igu es
Table 1.
w % w % w % w % w % w % w % w %
Umidi y Ex ac i es Ashes Klason Lignin Glucans Xylan A abinian Ace il G oups
6.501 3.088 0.138 25.741 39.742 18.796 2.593 3.401
32
Figu e 3.
(a)
(b)
(c)
(d)

33
Figu e 4.
2
CELLULOSE SUGARS C6
1
HEMICELLULOSE SUGARS C5
3
HEMICELLULOSE SUGARS C5
4
CELLULOSE SUGARS C6
DISSOLUTION DISSOLUTION
DISSOLUTION DISSOLUTION
DEACETYLATED
FIBER
5
ACETIC
ACID
DISSOLUTION
ACETIC
ACID
6H+
DISSOLUTION
34
Figu e 5.
(b)
(c)
(a)
(d)
(e)
35
Figu e 6.
a)
b)
c)