Online integrated fractionation-hydrolysis of lignocellulosic biomass using sub- and supercritical water

Online In eg a ed F ac iona ion-Hyd olysis o Lignocellulosic 1 Biomass using Sub- and Supe c i ical Wa e 2 3 C is ian M. Pique as1, Ál a o Cabeza2, Gianluca Gallina2, Danilo A. Can e o2, Juan 4 Ga cía-Se na2* and Ma ía J. Coce o2 5 6 1Plan a Pilo o de Ingenie ía Química, PLAPIQUI-Uni e sidad Nacional del Su - 7 CONICET, Camino La Ca indanga km 7-CC 717, (8000) Bahía Blanca, A gen ina. 8 Phone: +542914861700 / Fax:+542914871600 9 2High P essu e P ocesses G oup, Depa men o Chemical Enginee ing and 10 En i onmen al Technology, Uni e si y o Valladolid, Escuela de Ingenie ías 11 Indus iales, (47011) Valladolid, Spain. Phone: +34 983184934 12 *e-mail add ess: jg[email p o ec ed] 13 14 15 Keywo ds: Glucose, Glycolaldehyde, Kine ics, P ocess De elopmen , Xylose, Holm 16 Oak Wood 17 18 Abs ac 19 A no el p ocess coupling he ac iona ion and hyd olysis eac o s is p esen ed. Holm oak was 20 used as eal lignocellulosic biomass o be ea ed. In he ac iona ion eac o , hemicellulose and 21 cellulose we e solubilized and pa ially hyd olyzed in di e en s ages wi h he aim o eeding 22 he hyd olysis eac o wi h high C5 concen a ions o C6 concen a ions. The ac iona ion was 23 pe o med in wo s ages: a 180ºC op imizing he hemicellulose ex ac ion and a 260ºC 24 ex ac ing cellulose and ha d hemicellulose emaining in he biomass s uc u e. Th ee wa e 25 lows we e es ed: 11, 17 and 26 cm3/min. Suga yields om 71 o 75% we e eached, mainly 26 composed o xylose and glucose oligome s and lowe amoun s o o he chemicals, like e o- 27 aldol p oduc s, ace ic acid o 5-HMF. The ou le s eam om he ac iona ion eac o was 28 di ec ly mixed wi h sub o supe c i ical wa e a he inle mixe o a SHR whe e he eac ion 29 ime was p ecisely con olled. The empe a u e, p essu e and eac ion ime we e modi ied o ge 30 an insigh o hei e ec on he yield o e o-aldol condensa ion p oduc s. Yields o 24% o 31 glycolaldehyde, and py u aldehyde we e ound a 8.3 s, 350ºC and 162 ba (hyd olysis eac o 32 condi ions). In o he hand, 25% o lac ic acid was ound a 0.23 s, 396ºC and 245 ba . A 33 discussion based on a known eac ion pa hway is p oposed. Mo eo e , a kine ic model o he 34 hyd olysis eac o was p oposed, being able o ep oduce he expe imen al da a wi h de ia ions 35 lowe han 10 % o suga s and o he p oduc s ex ac ed. This combined p ocess pe o ms a 36 selec i e alo iza ion o eal lignocellulosic biomass, a oiding he cos ly p ocess o ex eme 37 g inding needed o he luidiza ion in a con inuous hyd o he mal p ocess. 38 39 In oduc ion 40 E en i i is easonably assumed ha biomass om plan s will be he main ca bon sou ce in he 41 u u e, he choice o which eac ion medium should be used o depolyme ize and alo ize 42 biomass has no been aken ye . P essu ized luids, especially sub and supe c i ical wa e 43 (Tc=374ºC and Pc= 221 ba ), can be poin ed as a p omising al e na i e o depolyme ize and 44 alo ize biomass [1-5]. Physical and chemical p ope ies o wa e can be modi ied by adjus ing 45 p essu e and empe a u e a ound he c i ical poin , making wa e a eac ion medium able o 46 a o di e en kind o eac ions [1]. Because o his eason, ho p essu ized wa e has been used 47 as eac ion medium o ac iona ion [6-9] , hyd olysis [10-12] and alo iza ion o biomass [13- 48 16]. 49 The composi ion o lignocellulosic biomass is highly dependen on he plan species and 50 g ow h condi ions. Howe e , i can be conside ed ha he a e age composi ion o 51 lignocellulosic biomass is app oxima ely: cellulose (40% w .), hemicellulose (25% w .), lignin 52 (25% w .), ex ac i es and ashes (10% w .) [17]. Al hough biomass is composed by di e se and 53 complex molecules, i can be ac iona ed p incipally in o C6 suga s (mainly glucose), C5 54 suga s (mainly xylose) and lignin [3]. These h ee ac ions can be u he modi ied o p oduce a 55 wide ange o p oduc s like: e hanol, hyd ogen, glycolaldehyde, py u aldehyde, lac ic acid and 56 5-HMF among o he s [3, 18-25]. 57 The ac iona ion o biomass can be de ined as he selec i e sepa a ion o C5 suga s, C6 suga s 58 and lignin om he o iginal biomass ma ix. This p ocess was s udied unde hyd o he mal 59 condi ions in di e en ways o ope a ion: ba ch, semi ba ch and con inuous [3, 26]. Semi ba ch 60 and con inuous p ocesses allow ob aining highe yields o suga s and chemical compounds han 61 ba ch eac o s, because i is possible o con ol he empe a u e (T) and he esidence ime ( ) 62 mo e accu a ely han in ba ch p ocesses [27]. Con inuous p ocesses a e he mos app op ia e o 63 con ol he eac ion condi ions (T and ), howe e , in mos cases i is necessa y o apply 64 expensi e p e ea men s o he aw ma e ial be o e he ac iona ion+hyd olysis p ocess, o 65 example: exhaus i e size educ ion [28]. On he o he hand, he con inuous p ocess can be 66 pe o med a di e en ope a ing condi ions in o de o sepa a e he C5 suga s om he C6 67 suga s. 68 The ex ac ion o hemicellulose om woody biomass can be ca ied ou a empe a u es 69 be ween 130ºC and 260ºC, solid eac ion imes be ween 20 and 60 min and liquid esidence 70 imes inside he eac o be ween 0.1 min and 1 min. A hose condi ions, hemicellulose can be 71 bo h ex ac ed and hyd olyzed [29, 30]. A e he ex ac ion a 180ºC, wo p oduc s a e usually 72 ob ained: a liquid composed mainly o C5 suga s and a solid composed o C6 suga s and lignin. 73 These wo p oduc s can be sepa a ed by il a ion. Then, he cellulose in he solid can be 74 hyd olyzed a supe c i ical condi ions o ob ain a wa e solu ion o C6 suga s and a solid 75 en iched in lignin. These p ocesses can be ca ied ou in wo eac o s wi h a il a ion ope a ion 76 be ween hem. Ano he op ion which allows he in ensi ica ion o he p ocess is using one ixed 77 bed eac o . In such a case, he biomass is loaded in he eac o and he hyd olysis empe a u e 78 is changed in o de o hyd olyze C5 o C6 suga s [31]. The semi ba ch p ocess allows high 79 pe o mances on he yields o C5 suga s hyd olysis. Howe e , when he eac ion empe a u e is 80 inc eased o hyd olyze he ecalci an cellulose and hemicellulose, he yield o eco e ed suga s 81 dec eases because o he inc emen o he suga s u he eac ions [10, 11, 32]. 82 The con inuous eac o s ha e been employed in many applica ions o he alo iza ion o suga 83 s eams allowing a p ecise con ol o e he eac ions [19-21]. These eac ions can be managed 84 using p essu ized wa e and choosing he adequa e eac ion condi ions. Fo example, a 85 empe a u es be ween 200ºC and 300ºC (250 ba ) he wa e molecules a e highly dissocia ed 86 a o ing he ionic eac ions, like he p oduc ion o 5-HMF om uc ose and glucose [1]. On he 87 o he hand, a 400ºC (250 ba ) he wa e molecules a e highly associa ed a o ing he non-ionic 88 eac ions, like he e o aldol condensa ion eac ions [1]. 89 In his a icle, a no el in eg a ed ac iona ion- alo iza ion p ocess was designed and buil using 90 wooden biomass as aw ma e ial and wa e (subc i ical and supe c i ical) as eac ion medium. 91 The wooden biomass was ac iona ed in a ixed bed eac o a di e en empe a u es. The 92 solubilized p oduc s we e di ec ly injec ed o a con inuous nea c i ical wa e eac o o 93 e icien ly con e C5 and C6 suga s in o aluable p oduc s, like glycolaldehyde, 94 py u aldehyde and lac ic acid a oiding a u he hyd olysis o o ganic acids. In addi ion, a 95 kine ic analysis o he biomass hyd olysis was done in o de o s udy he di e ences in he 96 p ocess when subc i ical and supe c i ical condi ions we e used. 97 The objec i e o his esea ch pape was o design a no el p ocess capable o con e ing 98 lignocellulosic biomass in o aluable p oduc s eluding he excessi e milling o biomass and 99 dec easing he numbe o eac o s. 100 101 1. Expe imen al 102 2.1 Ma e ials 103 Deionized wa e p oduced by Elix® Ad an age pu i ica ion sys em was used as eac ion 104 medium o un he expe imen s. The s anda ds used in a High Pe o mance Liquid 105 Ch oma og aphy (HPLC) analysis we e: cellobiose (≥98%), glucose (≥99%), xylose (≥99%), 106 galac ose (≥99%), mannose (≥99%), a abinose (≥99%), glyce aldehyde (≥95%), glycolaldehyde 107 dime (≥99%), lac ic acid (≥85%), o mic acid (≥98%), ace ic acid (≥99%), ac ylic acid (≥99%), 108 u u al (99%) and 5-hyd oxyme hyl u u al (≥99%) pu chased om Sigma. 0.01 N solu ion o 109 sul u ic acid (HPLC g ade) in Milli-Q® g ade wa e was used as he mobile phase in he HPLC 110 analysis. Sul u ic acid (≥96%) and calcium ca bona e (≥99%) supplied by Pan eac, Spain, we e 111 used as eagen s o he quan i ica ion p ocedu e o s uc u al ca bohyd a es and lignin [33]. 112 Also, Milli-Q® wa e was used in his de e mina ion. Holm oak wood employed as aw ma e ial 113 was collec ed in Spanish o es s. The wood was milled ob aining chips wi h a e age wid h o 2 114 mm and a e age leng h o 5 mm, as i is shown in Figu e S1 o Supplemen a y Ma e ial. 115 2.2 Analy ical me hods 116 The composi ion o he holm oak wood aw ma e ial, exhaus ed solid and ex ac ed liquo was 117 de e mined h ough wo Labo a o y Analy ical P ocedu es (LAP) om NREL [33, 34]. The 118 p ocedu e o solid samples consis s in quan i ying he s uc u al ca bohyd a es and lignin in he 119 biomass as ollows. A) The biomass was weigh ed be o e and a e being d ied in an ai d i en 120 o en a 105 °C o 24 hou s in o de o calcula e he mois u e con en . B) D ied biomass was 121 ea ed in a Soxhle equipmen wi h n-hexane, lea ing a solid ee o oils and o he ex ac i es. 122 C) 300 mg o d ied and ee-ex ac i es solid om s ep (b) we e hyd olyzed in 3 ml o 72% w 123 sul u ic acid solu ion a 30 °C o 30 min, in o de o b eak he bonds be ween biopolyme s and 124 he main solid s uc u e. D) The mix u e o oligome s ob ained in s ep (c) is dilu ed using 84 ml 125 o deionized wa e and hea ed a 120 °C o 60 min wi h he aim o hyd olyzing hemicellulose 126 and cellulose o ob ain hei co esponden monome s [35]. E) The solid is sepa a ed om he 127 solu ion by acuum il a ion. F) The o al mass o solubilized suga s was quan i ied as he 128 di e ence in weigh be ween he o iginal solid and he exhaus ed solid a e o en d ying a 105 129 °C in o en o 24 hou s. G) The exhaus ed solid is placed in a mu le a 550 °C o 24 h and he 130 emaining esidue was weigh ed be o e and a e his s ep o calcula e he insoluble lignin and 131 he ash con en o he sample. H) A liquid aliquo was analyzed wi h UV-Vis spec opho ome e 132 a 320 nm wi h ex inc ion coe icien o 34 Lg−1cm−1 [36] o calcula e he amoun o soluble 133 lignin. I) Ano he liquid aliquo was neu alized o pH ange 6 o 7, hen i was il e ed using a 134 0.2 µm memb ane and analyzed by HPLC de e mining he ca bohyd a es composi ion. This 135 p ocedu e is pe o med using a column SUGAR SH-1011 (Shodex) wi h a 0.01 N o sul u ic 136 acid solu ion as a mobile phase. To iden i y he soluble p oduc s, wo de ec o s we e used: 137 Wa e s IR de ec o 2414 (210 nm) and Wa e s dual λ abso bance de ec o 2487 (254 nm). In 138 o de o calcula e he amoun o ca bohyd a es, each ch oma og am was in eg a ed nume ically 139 by decomposing i in o a sum o 9 o 13 Gaussian peaks, minimizing chi squa ed unc ion o a 140 Le enbe g-Ma qua d -Fleche algo i hm [37]. Glycolaldehyde and Py u aldehyde esul ed o be 141 o e lapped, since he e en ion ime o hei s anda ds is ex emely close (11.99 s 12.24 142 minu es, espec i ely). So we e e o hem as glycolaldehyde-py u aldehyde. 143 The aw ma e ial con ained 1.6 % w . ex ac i es, 1.8% w . mois u e, 0.2% w . ashes, 24.2% w . 144 Klason lignin ( om which 4.0% co esponds o soluble lignin), 45.7% w . o hexoses, 23.9% 145 w . pen oses. The sum o all he componen s ep esen s he 97.4% o o al weigh , he 146 disc epancy is due o expe imen al e o s like he loss o solid ma e ial a e he eco e y a he 147 end o he expe imen s, o he inhomogenei y o he ma e ial which can ha e sligh ly di e en 148 composi ions depending on he analyzed aliquo ; in any case, i is inside he accep able 149 expe imen al e o . 150 The amoun o C6 was calcula ed as he sum o glucose, cellobiose and uc ose concen a ions. 151 Xylose was he only C5 de ec ed. Ace ic acid was conside ed o come om he deace yla ion o 152 xylan du ing he ex ac ion p ocess o , as explained in he nex sec ions, om he hyd olysis o 153 py u aldehyde. The hyd olysis p oduc s om hexoses and pen oses we e mainly 154 glyce aldehyde, glycolaldehyde, py u aldehyde, lac ic acid, 5-hyd oxyme hyl u u al and in 155 some cases ac ylic acid we e de ec ed in e y low concen a ion. 156 The p ocedu e ollowed o analyze liquid samples consis s in he s eps (C), (D) and (I) 157 desc ibed abo e. In his case, he ca bon con en liquid solu ions was de e mined by o al 158 o ganic ca bon (TOC) analysis using a Shimadzu TOC-VCSH equipmen . E e y sample was 159 p e iously il e ed using a 0.2 µm sy inge il e and dilu ed 1:10 imes wi h Millipo e wa e . 160 The pH o he ou le s eam was measu ed online using an elec onic pH-me e (Nahi a model 161 903). 162 2.3 Expe imen al se up and ope a ion p ocedu e 163 The se up used in his wo k is shown in Figu e 1. The sys em consis ed in wo eac o s online 164 in eg a ed: 1) he ac iona ion eac o (R.1), whe e he C5 and C6 a e solubilized and pa ially 165 hyd olyzed; 2) he supe c i ical hyd olysis eac o (SHR), which con e s he soluble 166 compounds in o added alue p oduc s. The ac iona ion line is composed o a wa e deposi 167 (D.1), downs eam an Ame ican Lewa EK6 2KN high p essu e pump (P.1, maximum low a e 168 1.5 kg/h) p opels wa e h ough a p e-hea e (H.1, 200 cm o 1/8” SS 316 pipe, elec ically 169 hea ed by means o wo esis o s o 300 W) which ensu es an uni o m empe a u e a he eac o 170 inle . The eac o (R.1), a ube o SS 316, 40 cm leng h, 1.27 cm O.D., is hea ed by h ee la 171 esis o s o 300 W each, placed axially along a machined aluminum ba wi h 5.08 cm O.D. 172 Bo h, p ehea e and he eac o a e loca ed inside a o me ch oma og aphic o en HP5680. The 173 ou - low s eam om he ex ac ion line is mixed wi h he supe c i ical wa e s eam, en e ing in 174 a second eac o (SHR) (R.2). The supe c i ical wa e line is composed o a hea e (H.2), a ube 175 o 18 m, 1/8 in O.D. SS316 w apped a ound a b ass cylinde and hea ed by wo ca idges and 176 wo la esis o s, which p o ided adjus able powe o up o 10 kW, in o de o con ol he 177 empe a u e o his s eam. The wa e low was gene a ed by a Mil on Roy XT memb ane pump 178 (P.2, maximum low a e 6 kg/h). The SHR allows a as hea ing o he biomass s eam, which 179 is mixed almos ins an aneously wi h he supe c i ical wa e s eam, and a apid cooling o he 180 p oduc s, which akes place h ough a sudden expansion which e icien ly s ops he hyd olysis. 181 In his way, he eac ion ime could be p ecisely calcula ed, as he eac o wo ks iso he mally. 182 P essu e was con olled Mic o Me e ing al e 30VRMM4812 om Au ocla e Enginee ing 183 (V.4). The se ups o he wo eac o s we e p esen ed in de ail in p e ious wo ks [32, 38]. 184 An a e age amoun o 6.12±0.03 g o holm oak biomass was placed inside he eac o R.1 o 185 each expe imen . Two me allic il e s we e used (po e diame e ≈200 m), loca ed on he op 186 and bo om o he eac o , a oiding he elease o he solid du ing he expe imen s. A p essu e 187 es wi h cold p essu ized wa e was ca ied ou be o e e e y expe imen , wi h he aim o check 188 he p esence o leaks in he sys em. Then, he supe c i ical line was hea ed ensu ing he 189 unc ioning o he sys em a equi ed ope a ing condi ions. Once hese condi ions we e s able, 190 he pumps we e swi ched o and bo h, he p ehea e and he eac o R.1, we e hea ed up un il 191 he empe a u es eached he espec i e se alues. A e wa ds, bo h pumps we e swi ch on 192 again and he low and p essu e we e se o he desi ed condi ions, ze o ime is conside ed when 193 p essu e eached he desi ed alue. 194 A o al o 11 expe imen s we e pe o med (3 ac iona ions and 8 coupled eac ions), ob aining 195 a o al o 130 liquid and 11 solid samples, cha ac e ized wi h he me hods desc ibed abo e. Six 196 expe imen s we e pe o med a ying he empe a u e in he SHR om subc i ical (350ºC) up o 197 supe c i ical (400ºC) condi ions, main aining he p essu e a 250±10 ba . The eac ion ime in 198 his eac o was modi ied by a ying he wa e low- a e and changing he eac o olume (2.2 199 o 12.4 cm3); eac ion imes be ween 0.25 s and o 12 s we e es ed. Th ee di e en wa e lows 200 (11, 17, 26 cm3/min) we e es ed in he ac iona ion line, main aining cons an he a io wi h he 201 low o supe c i ical wa e s eam, o ge he desi ed condi ions du ing he u he hyd olysis. 202 The eed composi ion o he SHR was analyzed by ca ying ou h ee ac iona ions wi hou he 203 second hyd olysis s age, a he same condi ions o empe a u es, low- a es and p essu e es ed 204 wi h he coupled eac ion. 205 The ac iona ion in he ixed bed eac o was pe o med in wo s ages ma ked by wo dis inc 206 empe a u es: 180ºC o ex ac he hemicellulose and 260ºC o emo e mos o he cellulose 207 ac ion om he biomass. The hea ing ime be ween bo h se poin s was in he ange o 5-10 208 min, while he low was empo a ily s opped o he expe imen unning a 26 cm3/min. In o de 209 o ollow he eac ion e olu ion, he pH o he ou le s eam was measu ed online sampling 210 e e y 1 minu e. Liquid samples (30-40 cm3) we e aken acco ding he pH a ia ions e e y 5 o 211 20 min o he expe imen a 11 cm3/min, and e e y 2 o 8 min o he o he expe imen s. The 212 o e all expe imen ime a ied om 110, 60 and 45 min o he uns a 11, 17, 26 cm3/min, 213 espec i ely (called he e as (1), (2) and (3)). A e he las sample was g abbed, he hea ing was 214 u ned o and he eac o R.1 was le o cool down o oom empe a u e wi h ai lux. Bo h 215 pumps we e se o ze o low and he sys em was dep essu ized. The solid was emo ed om he 216 eac o , il e ed and d ied 24 h a 105ºC o u he analysis. A e cleaning, he ixed bed 217 eac o was placed back, igh ened and he sys em was washed ou wi h deionized wa e . 218 219 3. Resul s and Discussion 220 3.1. Biomass ac iona ion 221 F om he analysis o he aw holm oak, he amoun o soluble ma e ial was 4.65 ± 0.03g, 222 co esponding o 72.1% o he biomass weigh . 3.02 ± 0.02g o his soluble mass we e 223 composed o hexoses (C6) and 1.58 ± 0.01g o pen oses (C5). The spa ial ime o he liquid (l), 224 is de e mined using he liquid low a e, he eac o olume and he a e age po osi y o he bed 225 (i0=0.457±0.01,  =0.948±0.019). The la e was calcula ed by means o Eq. (1), aking in o 226 accoun he ini ial and he inal ac ion o oid olume in he bed, due o he sh inking size o 227 he biomass pa icles, and also conside ing a cons an densi y o wa e [38] (since i s a ia ion 228 wi h empe a u e is less han 2%) and a cons an densi y o he holm oak wood (800 kg/m3, d y 229 co esponding o he i s s age o empe a u e in R.1. This ace ic acid exceeded he amoun 396 p oduced in he hemicellulose deace yla ion. The e o-aldol pa hway coming om xylose by 397 means o glyce aldehyde ou e, could explain he di e ence o ace ic acid ob ained di ec ly 398 om lac ic acid deca bonyla ion. This ex a amoun o ace ic acid could no be conside ed only 399 om he hemicellulose sou ce, since he e is also a la ge concen a ion o C6 in he i s ac ion 400 o he eed s eam (see Figu e 2 (a)). This C6 po ion could also con ibu e o he 401 glyce aldehyde ou e. Besides, ace ic acid could be ob ained di ec ly om lac ic acid 402 deca bonyla ion [15]. Bo h, glucose and xylose, a e able o p oduce lac ic acid by means o he 403 e o-aldol pa hway wi h glyce aldehyde and py u aldehyde as in e media ies (see Figu e 3). In 404 his sense, hese e o-aldol pa hways could explain he ex a amoun o ace ic acid ob ained a 405 longe esidence imes in he SHR. Figu e 4, displays he pH o he ou pu s eam a e he 406 ac iona ion s age (expe imen 1) and he coupled p ocess ac iona ion+hyd olysis 407 (expe imen s 4, 5, 6 and 8). The pH in he ou le s eam, a e SHR, was always lowe o he pH 408 o he ou pu s eam om he ac iona ion s ep i sel . compa ing he H+ concen a ion o he 409 expe imen s du ing he ime pe iod o he i s s age du ing he ex ac ion. This obse a ion 410 ag ees wi h he ac ha ex a amoun o ace ic acid was p oduced when a deepe hyd olysis was 411 pe o med (see expe imen s 5 and 8). A e his ime pe iod, no di e ence in he pH can be 412 de ec ed. Simila beha io was obse ed om expe imen s 6 and 8, howe e , in his case, la ge 413 amoun o o mic acid was obse ed compa ed o he expe imen s abo e men ioned in spi e o 414 he di e ence o o mic acid p oduced in bo h (see Figu e 2 (c)). 415 The p essu e change in he ange s udied, had no e ec on he chemicals dis ibu ion (see Figu e 416 2 (c), expe imen s 6 and 7). Unde he condi ions o expe imen 7, pKw is 11.9, calcula ed by 417 means o an empi ic equa ion [45]. This alue is qui e simila han pKw o he expe imen 6. In 418 his way, in spi e o densi y change, he same AVP dis ibu ion is obse ed. Longe esidence 419 ime o expe imen 7 explains he di e ence in oligome s wi h expe imen 6. The dis ibu ion o 420 he AVP o expe imen 6, is independen o he change in pKw, as was discussed abo e o 421 expe imen s 4 and 6. The highes yield o glycolaldehyde-py u aldehyde (calcula ed as mass o 422 p oduc /mass o soluble ma e ial in aw biomass) was ob ained o expe imen 7 (24.4%), 423 p obably due o he combina ion o highe H+ concen a ion and longe . 424 A di e en AVP dis ibu ion is obse ed in he expe imen 9, whe e lac ic acid is he mos 425 abundan p oduc and ace ic acid is deple ed compa ed o he expe imen 4 and 6 (see Figu e 2 426 (d)). This inding could be explained by he sho o he mix u e a high empe a u e, 427 condi ions in which he eac ions a e s opped be o e a e lac ic acid p oduc ion in he e o- 428 aldol ou e, inhibi ing he ace ic acid o ma ion. This selec i i y seems o ake place in he SHR 429 mainly du ing he i s s age o empe a u e in he eac o R.1. A e ha , he o ma ion o lac ic 430 acid in he SHR is educed. This selec i i y seems o ake place mainly du ing he ime pe iod o 431 he i s s age o he ac iona ion, because a e ha , he o ma ion o lac ic acid as well as o 432 glycolaldehyde-py u aldehyde is lowe . The highes yield o lac ic acid was ound a 433 Expe imen 9 (25.5%). The wa e low inc ease in he i s eac o has no clea e ec on he 434 p oduc ion o e o-aldol compounds (see eac ions 10 and 11 in Figu e 2 (d)). Unde hese 435 condi ions, he oligome s b eakup seems o become slowe , since hei amoun is enla ged 436 ela ed o he monome ic suga s. In bo h cases, he e o aldol pa hways a e ollowed p oducing 437 glycolaldehyde-py u aldehyde and lac ic acid wi h simila yields. 438 The combina ion o many a iables in luencing he dis ibu ion o a la ge numbe o 439 p oduc s, in ol ed in a complex eac ion pa h as he desc ibed in Fig. 3, is ha d o be easily 440 explained. Fu he mo e, as was men ioned abo e, we a e dealing wi h he hyd olysis o a eal 441 biomass, in which o he componen s could be in luencing he obse ed beha io . 442 3.3 Hyd olysis kine ic model 443 Aiming o analyze u he he esul s ob ained by he coupled sys em, a kine ic model o he 444 second eac o is p oposed in his sec ion. This model akes in o accoun he solubilized biomass 445 composi ion ed o he second eac o . I was specially ocused on he ime pe iod co esponding 446 o he i s s age o he solubiliza ion (a he condi ions o expe imen 1) since his s ep p oduces 447 an ou le s eam wi h highe amoun o he chemicals o in e es (see las wo columns o Table 448 1). The bigge added alued compounds p oduc ion om s age 1 is due o he ac ha he 449 ope a ional empe a u e was a ound 180 ºC, which means a lowe deg ada ion. The eac ion 450 pa hway p oposed in his case, showed in Figu e 45. I is a simpli ied e sion o he eal 451 hyd olysis desc ibed in Figu e 3. The modelling was done by he ansien egime mass balances 452 o each compound in he luid: oligome s, suga s and p oduc s (Eq. (3)). Mo eo e , he 453 ollowing assump ions ha e been conside ed: (1) he eac ion o de o all he kine ics is 1 o 454 he biomass compound and p o on concen a ion in wa e , (2) he e a e no di usional e ec s in 455 luid phase, (3) kine ic cons an s ollows A henius’ law and (4) he eac o wo ks a he same 456 empe a u e a any poin . Rega ding kine ics, a con en ional exp ession was used including he 457 e ec o he concen a ion o wa e p o on since i is a hyd olysis p ocess (Eq. (4)). 458 𝛿𝐶𝐿𝑗 𝛿𝑡 =𝑟𝑗−𝑢 𝐿·𝛿𝐶𝐿𝑗 𝛿𝑧 (3) 𝑟𝑗=C𝐻+ ·∑∝𝑖,𝑗·𝐾𝐿𝑖·𝐶𝐿𝑖 𝑖=𝑁 𝑖=1 (4) (4) 459 3.3.1 Nume ical esolu ion 460 Eq. (3) is a se o 6 pa ial di e en ial equa ions (PDE) which has o be disc e ized o ob ain a 461 se o o dina y di e en ial equa ion (ODE). The esolu ion o his se o ODEs was pe o med 462 by he Runge-Ku a’s me hod wi h a 8 h con e gence o de and he disc e iza ion by coupling 463 o hogonal coloca ion me hod on ini e elemen s [46]. The i ing o he expe imen al da a 464 cons i u es an op imiza ion p oblem. Due o i s complexi y, i was p e iously seeded by manual 465 i e a ion, and hen, op imized by he Nelde -Mead-Simplex me hod. Mo eo e , as he inle 466 concen a ion o he hyd olysis eac o was a iable and he oligome p ope ies changed wi h 467 ex ac ion ime [47], he p oblem was op imized a e e y expe imen al poin . Finally, he 468 solu ion was e iewed in o de o ensu e he physical meaning o he pa ame e s. The objec i e 469 unc ion was he minimiza ion o he Absolu e A e age De ia ion (A.A.D., Eq. (5)) o 470 oligome , suga and p oduc s concen a ion a he SHR ou pu . 471 472 473 474 𝐴.𝐴.𝐷.=∑1 𝑛·|𝑋𝑒𝑥𝑝−𝑋𝑠𝑖𝑚 𝑋𝑒𝑥𝑝 |·100 𝑛 𝑖=1 (5) 3.3.2 Expe imen al da a i ings 475 In o de o alida e he model, only expe imen s 4, 6 and 9 we e used because hey we e 476 ca ied ou a simila esidence imes and h ee di e en empe a u es (see Table 1). Fo 477 expe imen 4, he da a a ex ac ion ime o 9 and 14 min we e no conside ed because hey do 478 no ollow he endency ixed by he se o he h ee expe imen s used (4, 6 and 9). Mo eo e , as 479 each expe imen was ca ied ou independen ly, he inle o he eac o was assumed o ha e he 480 same composi ion ha expe imen 3 1 bu wi h TOC p o ile o he i ed expe imen (4, 6 and 481 9). I is also ema kable ha he olume ic low was he addi ion o he p o ided low by he 482 wo pumps o all he expe imen s (see Figu e 1). The de ia ion be ween he model and he 483 expe imen al da a is a ayed in Table 2 and o expe imen 6 i also can be seen in Figu e 6. The 484 model was able o ep oduce success ully he hyd olysis o solubilized biomass, being he 485 a e age A.A.D. 21.14 %, 37.37 %, 18.41 % and 7.24 % o ins an hemicellulose and cellulose 486 oligome s, suga s C6, suga s C5 and hei deg ada ion p oduc s ( he added alue p oduc s 487 espec i ely o AVP). These disc epancies changed o 27.46 %, 7.61 %, 9.31 % and 3.99% 488 espec i ely when cumula ed alues we e used. Taking in o accoun hese las alues, i can be 489 checked ha he highes e o is in he es ima ion o he oligome s mass, which can be caused 490 by he ac ha he expe imen al da a we e ob ained by he di e ence be ween he TOC and he 491 sum o he o he compounds (suga s and AVP). Mo eo e , he de ia ion be ween he 492 expe imen al and simula ed TOC was also calcula ed in o de o check ha he mass 493 conse a ion law is ollowed. Fo all he cases, his mass balance de ia ion esul in ze o pe cen 494 wi h h ee signi ican igu es (0.00%). The kine ic cons an s and he s oichiome ic coe icien s 495 (𝐾𝐿𝑖 and ∝𝑖,𝑗 in Eq. (4), espec i ely) had o be ob ained om i ing. Rega ding o ∝𝑖,𝑗, i was 496 always 1 less o he inal p oduc s coming om hemicellulose oligome s since hey a e 497 composed by pen oses and hexoses [48, 49]. These i ed pa ame e s, which a e shown in Table 498 4, equi e a deepe analysis and hey a e discussed in he nex sec ion. 499 3.3.3 Analysis o kine ic pa ame e s 500 The dependence o he kine ics pa ame e s wi h empe a u e was p o ed. The eg ession 501 coe icien (R2) acco ding he A henius’ heo y was highe han 0.84 o all he cases (see 502 Table 3). No change in he kine ic beha io was obse ed h ough he c i ical poin (see Table 503 3) like does in he hyd olysis o mic oc ys alline cellulose [44, 50] acco ding o he commen ed 504 in sec ion 3.2.2, since he e is no simul aneous solubiliza ion in he SHR. Howe e , a 505 dependence o he kine ic beha io was obse ed wi h he ex ac ion ime, since he a es o 506 p oduc ion o aluable chemicals is dec eased a e he maximum o solubilized mass is eached 507 (see Figu e 7 (a) and (b)). This obse a ion could be ela ed wi h he in luence o some o he 508 chemicals p oduced by he u he hyd olysis o suga s on he hyd o he mal hyd olysis. In 509 addi ion, i is also in e es ing ha a e his change, he kine ics o he suga ans o ma ion end 510 o hei ini ial alue while he kine ic o he oligome b eakdown g ows exponen ially. This 511 di e ence would be o igina ed by he changes in he molecula weigh o he ex ac ed 512 oligome s and he ac ha hey would be ans o med mo e quickly i he molecula weigh is 513 lowe . Mo eo e , i can be seen ha empe a u e can compensa e his nega i e e ec , being 514 negligible o oligome s a 400ºC (Figu e 7 (c)). 515 O he in e es ing esul is he e olu ion o he a io be ween he ou kine ic cons an s. In 516 sec ion 3.2.1 i was indica ed ha suga ans o ma ion is as e han oligome clea age in 517 subc i ical condi ions and lowe in supe c i ical wa e . This beha io ag ees wi h he ob ained 518 om he i ings, bu only be o e he ime o maximum o ex ac ion (Figu e 7 (a) and (c)). So, 519 om his poin , he changes in molecula weigh and he aw ma e ial ans o ma ion makes he 520 oligome clea age always g ea e . As he oligome composi ion changes wi h ex ac ion ime, 521 he kine ic canno be ep oduced by a ypical A henius’ kine ic. So, wo equa ions unc ion o 522 his ime ( e) a e p oposed, one o he p e-exponen ial ac o oligome clea ing (Eq. 46) and 523 o he o he ac i a ion ene gy suga u he eac ions (Eq. 57). 524 P=C·| emax −A· e|B (46) 525 526 P=D+ E 1+e(F·( e−G)) (57) 527 528 Whe e P e e s o bo h, he ac i a ion ene gy (Ea/R) and he na u al loga i hm o he p e- 529 exponen ial ac o (ln(k)). In Eq. 46, he pa ame e C is he na u al loga i hm o he p e 530 exponen ial ac o o ac i a ion ene gy a he maximum ex ac ion ime (𝑡𝑒𝑚𝑎𝑥) and pa ame e s 531 A and B in oduce he e ec o he changes in he s uc u e and eac ion medium. A would be 532 ela ed wi h he s ong o he compound agains i s deg ada ion by hyd olysis. B would be a 533 measu e o how s uc u e o eac ion medium can accele a e o es ain he deg ada ion. In Eq. 534 57, D is he p e exponen ial ac o o he ac i a ion ene gy a he ime whe e he bigges 535 solubiliza ion akes place, E and F, a e he pa ame e s ha conside he ole o he s uc u e and 536 eac ion medium and G is he ime when he maximum ex ac ion is eached. In his case, E 537 would ep esen how he medium o he s uc u e can enhance he hyd olysis o hinde i . F 538 would be he compound esis ance agains deg ada ion. 539 Finally, he e olu ion o he hexoses con en in hemicellulose oligome s is ep esen ed in 540 Figu e 7 (b). I can be obse ed ha he a io be ween hese alues g ows wi h ime. This esul 541 was expec ed because hexoses would make he dissolu ion mo e di icul and would explain he 542 ac ha in expe imen 5 he ex ac ion was as e han 4 and 9 expe imen s (see Figu e 7 (c)). 543 Mo eo e his esul ag ee wi h he da a epo ed by o he au ho s [51]. 544 3.3.4 Simula ed expe imen s 545 As i was men ioned in sec ion 3.3.2, only expe imen s 4, 6 and 9 we e used o alida e he 546 model. Expe imen s 5 and 8 we e no conside ed because hei eac ion ime we e much highe , 547 which implies almos a o al con e sion a he eac o ou le . Howe e , i checked i he model 548 was able o ep oduce hei beha io . The esul o he simula ions a e p esen ed in Figu e 8, 549 being he absolu e de ia ion a ound 4% o bo h expe imen s. The e o e, he model can p edic 550 success ully he hyd olysis a bo h low (0.2 – 1.0 s) and high (11.1-12.5 s) esidence imes. 551 3.3.5 Model limi a ions 552 F om he esul s showed in he h ee p e ious sec ions, he model was able o success ully 553 ep oduce he expe imen al beha io o he se -up. In ac , his model can be used o any o he 554 lignocellulosic biomass because o he ac ha i has been de eloped o a gene al biomass 555 hyd olysis pa hway. Howe e , i is limi ed o p ocesses whe e soluble lignin is low and when 556 he aim is o ep oduce he o e all beha io o a solubilized biomass s eam hyd olysis ins ead 557 o an analysis o each indi idual compound. Fu he mo e, his model can be also adap ed o 558 p ocesses whe e he inle s eam is a iable in ime. 559 Conclusions 560 A new p ocess coupling ac iona ion and hyd olysis s eps was de eloped. By means o his 561 p ocess, up o 64.2% o eed Holm oak wood was solubilized mainly as oligome s o hexoses 562 and pen oses and suga s wi h a small ac ion o e o-aldol compounds. The low a io o he 563 amoun o oligome s o monome ic suga s in he ou le s eam could be explained by a simila 564 beha io han in he case o pu e cellulose hyd olysis: he a e o monome s hyd olysis is highe 565 o he oligome s b eak up in subc i ical condi ions, bu his endency is e e ed a supe c i ical 566 empe a u es. 567 The main p oduc s o he u he hyd olysis in he second eac o we e glycolaldehyde, 568 py u aldehyde and lac ic acid. Yield ( ela ed o he amoun o soluble suga s in he aw 569 biomass) o 24 w % o Glycolaldehyde-Py u aldehyde was ound a long eac ion imes (350ºC, 570 160ba and 8,6 s) and 25 w % o lac ic acid was ound a sho eac ion ime bu high 571 empe a u e (400ºC, 250 ba and 0.23s). An inc easing amoun o ace ic acid was obse ed a 572 he highes esidence imes (e.g. 12 s). 573 The dis ibu ion o p oduc s is ela ed wi h a combined eac ion hyd olysis pa hway o 574 cellulose and hemicellulose in ol ing oligome clea age o monome s, isome iza ion s eps and 575 wo compe ing pa hs: Re o-aldol condensa ion and dehyd a ion. The in luence o he wa e 576 densi y and he amoun o ions H+ coming om he dissocia ion p ocess is no clea as i is in 577 he case o he hyd olysis o pu e cellulose, in which he glucose dehyd a ion is highly inhibi ed 578 and e o aldol pa hways clea ly a o ed a empe a u es and p essu es abo e he wa e c i ical 579 poin . In he p esen wo k, p oduc s coming om e o-aldol pa hs as well as p oduc s o 580 dehyd a ion a e obse ed in bo h condi ions: sub and supe c i ical. Finally, a gene al kine ic 581 modelling o he hyd olysis eac o was p oposed. This model could ep oduce he expe imen al 582 da a o suga and added alue p oduc s wi h de ia ions lowe han 10%. Besides, he calcula ed 583 kine ic pa ame e s ep oduced he changes in oligome and suga con e sion when he 584 hyd olysis is pe o med in supe c i ical condi ions ins ead o in subc i ical wa e . This model 585 can be applied o any o he lignocellulosic biomass wi h a low con en o soluble lignin. 586 The main ad an age o his combined p ocess consis in p o iding a lique ied biomass 587 s eam o a selec i e hyd olysis eac o gi ing he alo iza ion o he aw ma e ial a oiding he 588 cos ly g inding o pa icles om se e al millime e s o less han wo hund ed mic ons needed o 589 pump i in a wa e s eam belonging o a high p essu e p ocess. 590 591 Acknowledgemen s 592 The au ho s acknowledge he Spanish “Minis e io de Economía y Compe i i idad 593 (MINECO)” and FEDER unds, P ojec BioF aHyne y CTQ2015-64892-R and he egional 594 go e nmen (Jun a de Cas illa y León), P ojec Re e ence: VA330U13 o unding. MEng. 595 Gianluca Gallina wishes o acknowledge he Spanish “Minis e io de Economía y 596 Compe i i idad (MINECO)” o he schola ship/p edoc o al con ac BES-2013-063556. MEng. 597 Al a o Cabeza would like o hank o he Spanish “Minis e io de Educación, Cul u a y 598 Depo e”, aining p og am o uni e si y p o esso s ( e e ence FPU2013/01516) o he esea ch 599 aining con ac . 600 601 Abb e ia ions and symbols 602 Ac onyms 603 604 A.A.D.: A e age absolu e De ia ion. 605 Olig: Hemicellulose and cellulose oligome s. 606 C6/OligC5: a io hexoses o hemicellulose oligome s. 607 G eek le e s and symbols 608 609 A-G: Pa ame e s o kine ics cons an es ima ion. 610 ∝𝑖,𝑗 : S oichiome ic coe icien o he compound “j” o he eac ion “i”, dimensionless. 611 𝐶𝐻+ : Concen a ion o he p o ons, mg/L. 612 𝐶𝐿𝑗 : Concen a ion o he compound “j”, mg/L.Ea/R: Ac i a ion ene gy, K. 613 ε: Po osi y o he bed, dimensionless. 614 ε : Po osi y o he bed, calcula ed a he end o he expe imen , dimensionless. 615 εa : A e age po osi y o he bed, be ween he beginning and he end o he expe imen , 616 dimensionless. 617 εo: Po osi y o he bed, calcula ed a he end o he expe imen , dimensionless. 618 𝐾𝐿𝑖 : Kine ic cons an , min-1. 619 k: P e-exponen ial ac o o he kine ic cons an , mg-1·min-1. 620 L: Leng h o he eac o , m.m0: ini ial mass o he solid in he eac o , g. 621 m : inal mass o he solid in he eac o , g. 622 m(i) (RM): o al amoun o componen (i) in he aw ma e ial, ex ac ed by acid hyd olysis and 623 de ec ed by HPLC analysis, g. 624 Mw(i): molecula weigh o componen i, g/mol. 625 𝑀𝑤𝐶 : molecula weigh o he a ca bon a om, g/mol. 626 msol o (RM): o al amoun o soluble compounds in he aw ma e ial, ex ac ed by acid 627 hyd olysis and de ec ed by HPLC analysis, g. 628 N: Numbe o compounds, dimensionless. 629 n: To al numbe o expe imen s, dimensionless. 630 n(i): Numbe o ca bon a oms in he soluble componen i, dimensionless. 631 P: Calcula ed kine ic pa ame e , ac i a ion ene gy o he na u al loga i hm o he p e- 632 exponen ial ac o 633 R2: Coe icien R2, dimensionless. 634 : a io be ween he molecula weigh o he soluble compounds ex ac ed and he molecula 635 weigh o he a oms o ca bon, dimensionless. 636 (i): a io be ween he molecula weigh o he soluble compounds ex ac ed and he molecula 637 weigh o he a oms o ca bon o compound i, dimensionless. 638 𝑟𝑗 : Reac ion a e o he compound “j”, mg/min·L. 639 u: Liquid eloci y in he eac o , m/min. 640 : Residence ime in he SHR, s. 641 𝑡𝑒: Ex ac ion ime, min. 642 𝑡𝑒𝑚𝑎𝑥 : Maximum ex ac ion ime, min. 643 𝑥𝑖𝐸𝑋𝑃 : Expe imen al alue o he i ed a iable. 644 𝑥𝑖𝑆𝐼𝑀 : Simula ed alue o he i ed a iable. 645 z: Coo dina e along he leng h o he eac o , dimensionless. 646 647 648 649 650 651 652 D.1 D.2 P.2 P.1 H.1 H.2 R.1 R.2 V.4 V.1 V.2 H.3 D.4 D.3 V.3 V.5 Wa e Sample Deposi P Wa e P T T T P T pH k1: Cellulose oligome b eakup kine ic cons an , k2: Hemicellulose oligome b eakup kine ic 842 cons an , k3: Suga s C6 hyd olysis kine ic cons an , k4: Suga s C5 hyd olysis kine ic cons an . 843 Figu e 8. Compa ison be ween he expe imen al and simula ed da a o AVP in expe imen 5 844 and 8. 845 Symbols a e he expe imen al da a and ull lines shows he p edic ion o he model wi h 846 op imized kine ic pa ame e s o each da a. 847 848 Figu e 1. 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 Figu e 2. 872 873 874 875 876 877 878 879 0 1 2 3 4 5 0.0 0.5 1.0 1.5 2.0 020 40 60 80 100 0.00 0.05 0.10 0.15 0.20 020 40 60 020 40 99.8% TOC HPLC Soluble mass [g] 81.2% (C5+C6) o al=88.3 % w 80.2% 81.5% 73.1% 80.3% C6 C5 Oligo C6 Oligo C5 Suga s [g] (C5+C6) o al=88.4 % w (C5+C6) o al=84.0 % w (3)(2) (1) Glice aldehyde Pi u .+Glycolaldehyde Fo mic acid Lac ic acid Ace ic acid 5HMF AVP od [g] (a) ime [min] 880 881 882 883 884 885 886 887 888 889 0 1 2 3 4 5 0.0 0.2 0.4 0.6 0.8 1.0 020 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 020 40 60 80 100 TOC HPLC 70.1% 88.3% 81.2% 108.6% C6 C5 Oligo C5+C6 Soluble mass [g] Suga s [g] AVP od [g] xOligo=86.5% xOligo=99.9% Glyce aldehyde Fo mic acid Py u .+Glycolaldehyde Lac ic acid Ace ic acid 5HMF ime [min] 377º, 251ba , 11.1s 383º, 245ba , 1.1s (b) (5) (4) 890 891 892 893 894 895 0 1 2 3 4 5 0.0 0.2 0.4 0.6 0.8 1.0 020 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 020 40 60 80 100 020 40 60 80 100 TOC HPLC 70.7% 85.9% (7) 79.8% 103.0% xOligo=85.7% C6 C5 Oligo C5+C6 Soluble mass [g] Suga s [g] AVP od [g] xOligo=98.7% xOligo=99.7% 350ºC, 240ba , 12.5s 356ºC, 162ba , 8.3s352ºC, 241ba , 2.1s (c) Glyce aldehyde Fo mic acid Py u .+Glycolaldehyde Lac ic acid Ace ic acid 5HMF ime [min] (8) (6) 93.9% 74.6% 896 897 898 899 900 901 902 903 904 905 0 1 2 3 4 5 0.0 0.2 0.4 0.6 0.8 1.0 020 40 60 020 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 020 40 TOC HPLC 77.3% 102.7% (10) 68.5% 83.3% xOligo=99.3% C6 C5 Oligo C5+C6 Soluble mass [g] Suga s [g] AVP od [g] xOligo=87.2% ime [min] 398ºC, 260ba , 0.24s 401ºC, 252ba , 0.24s 396ºC, 249ba , 0.23s (d) Glyce aldehyde Fo mic acid Py u .+Glycolaldehyde Lac ic acid Ace ic acid 5HMF (11) (9) 82.7% 71.1% xOligo=74.6% Figu e 3. 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 Figu e 4. 922 923 924 925 926 927 020 40 60 80 100 120 2 3 4 5 6 (1) (4) (5) (6) (8) pH ime [min] O O H Fo mic Acid O H O H O O H O H O H O O O H O O H O O H O H O O C H 3 O H O H O O H O H O H O H O H O O H O O H O H O H O H O O H O H O H O H O H O O H O D - Glucose  - Gluco py ano se 5 - HMF Gl yce aldehyde Dihyd oxy ace one Py u aldehyde Glycolaldehyde O C H 3 O H O H Lac ic Acid C5 Oligome s O H O C H 3 O O O H Fo mic Acid Le ulinic Acid C6 Oligome s O O Fu u al + Xylose Ace ic Acid O HO F uc ose Figu e 5. 928 929 930 931 932 Figu e 6. 933 934 935 936 937 938 939 940 Figu e 7. 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 C6 Oligome s C5 Oligome s Suga s C6 Suga s C5 Added Value P oduc s H+ H+ H+ H+ H+ k1 k2 k3 k4 10 20 30 40 50 0.0 0.4 0.8 1.2 1.6 2.0 2.4 TOC AVP TOC, AVP [g] 0.0 0.2 0.4 0.6 Oligome s C5 C6 Oligome s, C5,C6 [g] ime [min] 30 31 32 33 33 34 35 10 20 30 40 50 34 36 38 a) c) b) ln ki k1 k2 k3 k4 ime [min] 0.0 0.1 0.2 0.3 0.4 10 20 30 40 50 0.0 0.2 0.4 0.6 0.8 1.0 C6/Oligo C5 [g/g] 4 6 9 d) e) mass solubilized [g] ime [min] Figu e 8. 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 10 20 30 40 50 0.0 0.2 0.4 0.6 0.8 5 8 AVP [g] ime [min] Table 1. 978 979 Exp T [ºC] P [ba ] 1 [s] QSHR 2 [cm3/min] MBTOC3 [%] xOligome s4 [%] Y1AVP5 - Y2AVP6 - 4 383.7 ± 5.1 245.7 ± 4.6 1.06 36.0 92.2 86.5 0.008 0.079 5 377.2 ± 3.5 251.9 ± 5.9 11.15 38.5 105.9 99.9 0.247 0.281 6 352.5 ± 4.4 241.3 ± 3.7 2.10 35.2 89.3 85.7 0.004 0.109 8 349.9 ± 2.4 239.6 ± 4.2 12.50 35.8 103.1 98.7 0.233 0.132 9 396.1 ± 3.6 249.1 ± 5.1 0.23 36.8 103.6 99.3 0.440 0.254 10 401.2 ± 2.8 252.2 ± 3.9 0.24 90.1 93.0 87.2 0.481 0.278 11 398.3 ± 3.0 259.9 ± 3.4 0.24 106.2 91.2 74.6 0.530 0.228 1 : eac ion ime in hyd olysis eac o , 2 Flow a e in he SHR, 3 Global mass balance o he coupled p ocess, 980 4Con e sion o oligome s om hemicellulose and cellulose, 5,6 Yields o added alue p oduc s in he ime pe iod o 981 he i s and second s age o empe a u e du ing ac iona ion Yi=massi / mass soluble ma e ial in aw biomass 982 983 984 Table 2. 985 ADD % Ins an aneous Cumula ed Expe imen Oligome s1 C62 C53 AVP4 Olig1 C62 C53 AVP4 4 21.70 21.88 21.92 7.38 26.85 10.78 4.98 7.52 6 20.58 29.11 22.16 6.27 6.99 2.15 14.82 1.78 9 * 61.04 11.15 8.07 48.53 9.90 8.15 2.66 A e age 21.14 37.34 18.41 7.24 27.46 7.61 9.31 3.99 5 * * * 5.77 * * * 4.94 8 * * * 0.93 * * * 1.02 A e age * * * 4.65 * * * 3.32 1 Oligome s om hemicellulose and cellulose, 2 Suga s C6, 3 Suga s C5,4 deg ada ion p oduc s. * Compound no 986 de ec ed. ADD% o o al o ganic con en was 0.0. 987 988 989 990 991 992 Table 3. 993 994 e1[min] k1a k2b k3c k4d R2 19 0.88 0.87 0.99 0.999 26 0.88 0.86 0.99 0.98 35 0.87 0.85 0.99 0.97 44 0.93 0.89 0.9999 0.96 A e age 0.89 0.87 0.99 0.97 a Cellulose oligome clea age cons an , b Hemicellulose oligome clea age cons an , c Suga s C6 hyd olysis kine ic 995 cons an , d Suga s C5 hyd olysis kine ic cons an . 1 F ac iona ion ime. 996 997 998 Table 4. 999 k1a ln(k)e Ea/R k2b ln(k)e Ea/R - [K] - [K] A 1.04 1.03 A 1.08 1.08 B 0.04 0.06 B 0.03 0.05 C 106 46,543 C 109 48,553 k3c ln(k)e Ea/R k4d ln(k)e Ea/R - [K] - [K] A 3.26 2069 A 2.85 1834 B 22.05 24.22 B 21.25 20.15 C 1.35 4.01 C 1.70 3.58 D 87.32 35,238 D 89.63 36,081 a Cellulose oligome b eakup cons an , b Hemicellulose oligome b eakup cons an , c Suga s C6 hyd olysis cons an , 1000 d Suga s C5 hyd olysis cons an , e Na u al loga i hm o he A henius’ p e-exponen ial ac o , Ac i a ion ene gy. 1001 1002