Effects of short-time vibratory ball milling on the shape of FT-IR spectra of wood and cellulose

Transcription

1 Virtionl Spetrosopy 36 (2004) Effets of short-time virtory ll milling on the shpe of FT-IR spetr of woo n ellulose M. Shwnninger,,*, J.C. Rorigues, H. Pereir, B. Hinterstoisser, Deprtment of Chemistry, BOKU University of Nturl Resoures n Applie Life Sienes, Muthgsse 18, Vienn A-1190, Austri Competene Centre for Woo Composites n Woo Chemistry (Woo K plus), Linz, Austri Centro e Estuos e Florestis, Deprtmento e Engenhri Florestl, Instituto Superior e Agronomi, Tp Aju, Liso, Portugl Deprtment of Mteril Sienes n Proess Engineering, BOKU University of Nturl Resoures n Applie Life Sienes, Vienn, Gregor Menel Strsse 33, A-1180 Vienn, Austri Reeive 3 My 2002; reeive in revise form 22 Jnury 2004; epte 18 Ferury 2004 Aville online 24 April 2004 Astrt Virtory ll milling is shown to hve strong influene on the shpe of FT-IR spetr of woo n ellulose, even if the smples re mille for only short time. The results show tht neither temperture nor prtile size or oxition proesses re the most importnt reson for the oserve hnges in the struture of the FT-IR spetr, ut tht the mehnil tretment itself is the min influening ftor. Stuies of the effets of temperture, prtile size, oxygen n mehnil tretment on hemil n struturl hnges re presente. The most onspiuous hnges in the spetr of ellulose n woo were oserve t wve numers 1034, 1059, 1110, 1162, 1318, 1335, 2902 m 1 n in the OH-strething virtion region from 3200 to 3500 m 1. It is suggeste tht these hnges re minly ssoite with erese in the egree of rystllinity n/or erese in the egree of polymeristion of the ellulose. # 2004 Elsevier B.V. All rights reserve. Keywors: Virtory ll milling; FT-IR; Woo; Pie ies; Cellulose; Mehnil tretment 1. Introution FT-IR spetrosopy using the tritionl trnsmission tehnique in KBr-pellets hs inresingly een use in woo hemistry to hrterise ellulose n lignin oth qulittively [1] n quntittively [2,3]. The preprtion of woo for nlysis requires milling followe y extrtion with orgni solvents n wter. Aoring to TAPPI T 257 om- 85, grining to prtile size to pss 40 mesh (0.4 mm) sreen is reommene. Heting of the mteril uring grining or regrining of the mteril shoul e voie. The rising prolem is, tht the resulting prtile size is too lrge for FT-IR spetrosopy n it is neessry to regrin the mteril [1 4] or to frtionte it [1,2,5]. On the other hn, it is known from TAPPI T 264 om-88 tht ifferent frtions my ontin vrying quntities of some woo onstituents, so tht their removl oul lter the hemil omposition. Regrining n lso signifintly lter the hemil omposition of the woo mteril. * Corresponing uthor. Tel.: þ ; fx: þ E-mil ress: mnfre.shwnninger@oku..t (M. Shwnninger). It is tken s ft tht the mehnil ehviour of woo is result of the struturl rrngement of the woo polymers [6], ut there is still lk of knowlege onerning the intertions n ehviour of ellulose, hemielluloses n lignin within the omposite woo uner the type of mehnil stress, ourring uring milling [7]. Alrey in the 1940s n 1950s, the influene of milling on the egree of polymeristion (DP) [8 10] n the rerystlliztion of groun ellulose [11] n on the roxyl ontent [12] ws investigte. Fix n Bötther [1] reporte on the influene of prtile size n onentrtion on FT-IR spetr of groun tropil hrwoo. Stuies of the effet of ll milling on the ultrstruture of woo ell wlls, y Murer n Fengel [13], showe tht the ompoun mile lmelle n the ell orners re most resistnt to milling. The seonry wll 1 (S1) is very sensitive to mehnil stress [14] s it is loosene t n erly stge of milling n is seprte from the ompoun mile lmell s well s from the S2 lyer. Fengel [14] oserve totl rekown of the ell-wll struture uring intensive ll milling. Fengel n Luwig [15] ompre vrious elluloses n foun tht the spetr eome similr in the fingerprint region fter intensive ll milling /$ see front mtter # 2004 Elsevier B.V. All rights reserve. oi: /j.vispe

2 24 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) This work fouses on the influene of short-time virtory ll milling on the shpe of the FT-IR spetr of woo n ellulose. The effet of virtory ll milling n of temperture, prtile size n oxygen re isusse in reltion to hemil n struturl hnges. The importne of stnristion of the methos use for smple preprtion prior to the olletion of spetr for spetrl tses n for the qulittive n, espeilly, quntittive nlysis of woo is isusse. A reommention for the preprtion of mille woo smples for spetrosopi nlysis is given. 2. Experimentl The smples, their preprtion n the investigtions performe re liste in Tle Preprtion of smples Smples A, B n C of ir-rie woo of Norwy sprue (Pie ies (L.) Krst) were ut with seteurs, mille with offee griner (to size of out 5 mm) n finlly with RETSCH Ultr Centrifugl Mill ZM 1000 with fixe ring sreen with 80 mm hole with. Speil re ws tken to voi heting the woo uring the milling proess. Smples of 3.5 g were Soxhlet-extrte with enzene-ethnol (2:1 (v/v); 4 h), wshe with ethnol (solute), n then extrte with ethnol (95% (v/v); 4 h) n finlly with hot wter oring to TAPPI T 264 om-88, exept tht the hot wter extrtion ws performe on mgneti stirrer with heting for 1 h using 750 ml istille wter. The extrte smples were ir-rie for 3 ys n susequently oven-rie to onstnt weight t 45 8C. Tle 1 A survey of the smples, their preprtion n the investigtions is given

3 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) Woo smple B n smple of ellulose power from sprue (Fluk 22181, fire length mm) were seprte into severl grnulometri frtions using Retsh Anlytil Sieve Shker AS 200. The woo frtion etween 45 n 63 mm, n the ellulose frtion whih psse the 45 mm sreen were retine for susequent investigtions. Holoellulose ws prepre from smple C oring to Gierer [16,17] n Hs et l. [18] using pereti i. The ontent of resiul lignin ws out 2%. Fumri i (MERCK 269) n mlei i (SIGMA M-0375) were use s otine. Mille woo lignin (MWL) ws provie y Prof. O. Fix n ws use without milling Bll milling of smple A n prtile size effet Aliquots of woo smple A of out 200 mg were mille for (0 60) minutes in RETSCH mixer mill MM2 using 5 ml steel milling up with two spheres, 7 mm in imeter n weighing 1390 mg, with mss rtio (woo/sphere) of Eh milling ws repete on seon liquot. The liquot mille for 30 min ws seprte in RETSCH Anlytil Sieve Shker AS 200 n further y hn into severl grnulometri frtions Bll milling of smple B, ellulose n temperture effet Smples B, the ellulose n smple of mlei i were mille with virtory ll mill (Perkin-Elmer), using 7.5 mg of eh smple in the 1.2 ml steel milling-up with stinless steel-ll hving imeter of 5 mm n mss of 500 mg, with mss rtio (woo/sphere) of Eh milling ws repete on seon liquot. The melting points were etermine with thermomirosope (LEICA VM TG). Smple B ws hete in n oven t tempertures of 45, 75, 105 n 135 8C, for 10 min to enle possile hnges ue to therml tretment to e ssesse Influene of oxygen To voi the influene of oxygen, milling were rrie out in whih the virtory ll mill (Perkin-Elmer) n the milling-up with the smple (smple B n ellulose) were put into ox whih ws thoroughly purge with nitrogen for out 3 h. The ox ws riefly opene (while still eing purge) to enle the milling-up to e lose, n the up to e mounte on the virting mill. The lose ox ws then purge for n itionl 30 min n the smples were mille up to 5 min FT-IR n t proessing KBr pellets of smples A were prepre y mixing ( ) mg of woo power with 200 mg KBr (spetrosopi gre) in virtory ll mixer for 20 s. The 13 mm imeter pellets were prepre in stnr evie uner pressure of 75 kn m 2. IR-spetr were reore using Bio-R FTS 165 spetrometer with DTGS etetor t 4 m 1 resolution. The spetr were rtioe ginst ir n 64 sns were tken per smple. The verge spetr for eh ll milling step ws use for the evlution. FT-IR spetr of woo smple B, ellulose, mlei i n fumri i were reore on Bruker FT-IR spetrometer (EQUINOX 55) equippe with DLATGS etetor. The 1.5 mg smples were ilute with 200 mg KBr (Alrih; 22,186-4; FT-IR gre) n mixe in the milling-up of the virtory ll mill (Perkin-Elmer) without ny ll for 10 s. Inserting the steel ll n milling for further 15 s followe this. The 13 mm KBr pellets were prepre uner vuum in stnr evie uner pressure of 75 kn m 2 for 3 min. Thirty-two sns were ollete per smple t spetrl resolution of 4 m 1, n the ollete spetr were rtioe ginst ir. The verge spetr of four mesurements (repete ll milling n two spetr for eh ll milling step) were use for the evlution. No sorne sle is given in the figures, euse they re shifte prllel to the wvenumer xis n/or normlise. The smple mss ws hosen to otin trnsmission spetr with minimum of out 15% trnsmission with respet to ir, whih were trnsforme to the sorption spetr shown in the figures presente. Post-spetrosopi mnipultion ws kept to minimum. The woo spetr were only shifte prllel to the wvenumer-xis so tht the minimum etween 2000 n 1 m 1 ws equl to zero, s suggeste y Zvrin et l. [19], n were normlize to the mximum t 1510 m 1, s suggeste y Fix n Bötther [1]. Seon erivtive spetr were otine y pplying the Svitzky Goly [20] lgorithm. The ellulose spetr were selineorrete so tht the minimum etween 2000 n 1 m 1 ws set equl to zero n they were normlize to the highest n, s suggeste y Fengel n Luwig [15]. Spetr use for the investigtion of rystllinity were seline-orrete using the ruern-metho. 3. Results n isussion 3.1. Spetrl hnges of woo ue to ll milling Fig. 1 shows n intensities in the infrre spetr of sprue woo (smple A) tken fter 0 60 min milling uner mient onitions. The most onspiuous hnges re t wvenumers 1034, 1059, 1110, 1162, 1318, 1335, 2902 m 1 n in the rnge from 3200 to 3500 m 1. Bn ssignments oring to the literture n n shifts re liste in Tle 2 ([12,15,21 39]). During milling, the reltive n-heights (height of n mesure

4 26 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) (A) h (B) h Fig. 1. Infrre spetr of sprue woo smple A. (A) Full rnge spetr of smples A; (B) spetr of woo in the rnge from 3700 to 2700 m 1 n (C) from 1600 to m 1 (C). Different times of ll milling: () without ll milling, () 1 min, () 6 min, () 8 min, (e) 10 min, (f) 14 min, (g) 30 min, n (h) 60 min. from lol seline using n jent vlley) erese n the OH-strething virtion region eomes roer. Bns t 1110, 1318, 1335 m 1 n the OH/CH 2 vleny virtion rnge ( m 1 )seemtoeverysensitive to mehnil stress. After 60 min, the minimum etween 1034 n 1059 m 1 lmost isppers. This orrespons to the n shifts for the two mxim (Fig. 1C spetrum h). The ns t 1110, 1318, 1335 n 2902 m 1 lso ispper. In the CH 2 vlene virtion rnge, two new ns emerge t 2892 n 2927 m 1 (Fig. 1B spetrum h) Prtile size effet The more intensive the ll milling, the higher is the perentge of fine mteril in the smples. This my explin the inrese in sorne vlues in the spetr. Fix n Bötther [1] otine omprle results. A 30 min ll-mille liquot of woo smple A h prtile istriution s follows: 32% prtiles >80 mm, 10% etween 63 n 80 mm, 45% etween 40 n 63 mm, tres etween 20 to 40 mm n 13% <20 mm. In greement with Murer n Fengel [13], who foun mostly prtiles with

5 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) (C) h Fig. 1. (Continue ). istint imeters of 8 n 50 mm fter ll milling, the lrgest mounts of prtiles were foun in the mm frtion n in the frtion <20 mm. The spetr of these frtions re shown in Fig. 2. Chnges in the shpes of the ns n e seen lerly, the most prominent eing the ispperne of the pek t 1110 m 1 (Fig. 2B). The intensities of the polyshrie peks in the fingerprint-region (see Tle 2 holoellulose n ellulose), reltive to the hrteristi lignin pek (1510 m 1 ) use for normliztion (Fig. 2A n B), re higher in the <20 mm frtion n lower in the n mm frtions thn >80 mm frtion. This is proly ue to reltive enrihment of the orser frtion with ompoun mile lmelle n ell orners [13], whih re the prts of the ell most resistnt to ll milling [13]. The ifferenes in the prtile size o not suffiiently explin the oserve hnges in the spetr in Figs. 1 n 3. A omprison of the spetrum of the ll-mille frtion with prtiles lrger thn 80 mm with tht of non-ll-mille smple of the sme prtile size showe tht the spetrl hnges our even without size reution (Fig. 2C). Moreover, omprison of non-mille mm frtion with the mm frtion of the 30 min ll-mille smple gve the sme result, s shown in Fig. 2D (p. lso 3.1 Fig. 1) Temperture effet After 60 min of milling (smple A) the temperture of the milling up inrese notiele. An inrese in temperture of the woo uring milling is expete [10]. This les to the question of whether n inrese in temperture n use the ifferenes in the spetr oserve in, e.g. Fig. 1. In orer to voi the temperture effet, new ll milling set-up with shorter milling time ws hosen. Using smller milling up n smller mss rtio (woo/sphere) of 0.015, it ws possile to otin spetrum (Fig. 3A spetrum g) fter only 2 min milling whih ws omprle to the spetrum (Fig. 1A spetrum h) otine fter 60 min milling. The hnges in n intensities in the spetr of smple B tken fter 5 s to 2 min ll milling (Fig. 3A C) re lmost ientil to those oserve for smple A. To hek the temperture inrese tht ours uring the milling proess, hemil onversion retion ws hosen. Pelouze [40] reporte tht mlei i is onverte into fumri i if the temperture rises few egrees ove 130 8C. Therefore, 7.5 mg mlei i were mille in the milling up for 6 min. No hnges were oserve in the FT-IR spetrum n no hnges in melting point, srutinise with thermomirosope, were oserve, n it ws onlue tht the temperture remins elow 130 8C uring ll milling. Fengel [33] oserve no signifint hnges in the profile of the FT-IR spetr uring n 8-h therml tretment of ellulose, even t tempertures up to 150 8C. No signifint therml effet is therefore expete uring the ll milling of ellulose uner the onitions use here. However, n influene on lignin n hemielluloses nnot e exlue. Pein [4] oserve hnge in the shpe of the spetr from sprue woo uring short-time therml tretment. Therefore, liquots of smple B were thermlly trete in n oven t 45, 75, 105, n 135 8C, respetively, for 10 min. An inspetion of the spetr (Fig. 3D) showe no ifferenes. Therefore, temperture effet ws exlue.

6 28 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) Tle 2 Pek wve numers of ns efore n fter ll milling n their ssignments oring to the literture Wvenumer (m 1 ) rnge of mxim Bn origin (ssignment) with omments MWL referene [72] ellulose woo holoellulose ellulose fter 5 min milling woo fter 10 min milling woo fter 30 min milling woo fter 60 min milling Free OH [73] [64] 3560 Asore wter wekly oun [66] [64] O H streth [72] Vlene virtion of H one [63] OH-groups (intrmoleulr) O(2)HO(6) intrmoleulr in ellulose [67] [64] O(3)HO(5) intrmoleulr [64] in ellulose [67,68] O(6)HO(3) intermoleulr [64] in ellulose [66 68] Vlene virtion of hyrogen one OH-groups [63] C H streth in methyl n methylene groups [72] methoxyl C H strething [74] Asymmetri CH 2 vlene [63] virtion CH 2,CH 2 OH [66 70,73], [64] in ellulose from C6 [75] Symmetri CH 2 vlene [63] virtion C¼O vlene virtion of etyl- or COOH-groups [63] C¼O streth in unonjugte ketones, ronyls n in ester groups (frequently of rohyrte origin); onjugte lehyes n roxyli is sor roun n elow 1700 m 1 [72] C¼O streth; in onjugte p-sustitute ryl ketones; strong eletronegtive sustituents lower the wvenumer [72] Asore wter [63] Aromti skeletl virtions plus C¼O streth; S > G; G onense > G etherifie [72] Aromti skeletl virtions; G>S[72] CH 2 of pyrn ring symmetri sissoring; OH plne eformtion virtion C H eformtions; symmetri in CH 3 n CH 2 [72] 1460 Asymmetri C H ening from methoxyl group [72] 1435 HOC ening [76] 1430 HOC in plne ening of lohol groups [77] Aromti skeletl virtions omine with C H in plne eformtion [72] [63] CH 2 sissoring [63]

7 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) Tle 2 (Continue ) Wvenumer (m 1 ) rnge of mxim Bn origin (ssignment) with omments MWL referene [72] ellulose woo holoellulose ellulose fter 5 min milling woo fter 10 min milling woo fter 30 min milling woo fter 60 min milling CH eformtion virtion [63] Aliphti C H streth in CH 3 ; not in O Me [72] 1368 symmetri C H ening from methoxyl group [74] OH plne eformtion virtion [63] (1322) Phenoli OH; S ring plus G ring onense; (i.e. G ring sustitute in pos. 5) [72] CH 2 roking virtion [63] CH - eformtion [63] G ring plus C¼O streth [72] OH plne eformtion, lso COOH [63] C C plus C O plus C¼O streth; G onense > G etherifie [72] OH plne eformtion [63] C O C symmetri vlene virtion [63] Aromti C H in plne eformtion; typil for G units; wherey G onense > G etherifie [72] Asymmetri in-phse ring strething, C C n C O strething oring to [65,71,77,78] Ring symmetri vlene virtion [63] C O eformtion in seonry lohols n liphti ethers [72] C O vlene virtion [63] [63] minly from C3 O3H [77] C lky O ether virtions, methoxyl n -O-4 [79] C lky O ether virtions in guiol [80] Aromti C H in plne eformtion, G > S; plus C O eformtion in primry lohols; plus C¼O streth (unonj.) [63,72,77] C O vlene virtion [63] Pyrn ring virtion [63] C H out-of-plne; romti [72] Anomere C-groups, C 1 -H eformtion, ring vlene virtion [63] C H out-of-plne in position 2, 5 n 6 of G units Due to gluomnnn [81] Pyrn virtion [63] 715 Roking virtion CH 2 [82]; in ellulose I [83] 670 C OH out-of-plne ening moe [71] Wvenumers (m 1 ) of ns tht re visile only s shouler re written in itlis. The wvenumer rnges for ns given in the first olumn were ssigne to ellulose (fourth olumn) y Fengel [63] n Fengel n Luwig [64] referring to [65 71] KBr pellet preprtion Sine 5 s of ll milling of woo mel (smple B) hs n influene on the spetr (p. Fig. 3A C), possile influene of the KBr pellet preprtion use for woo smples B ws heke refully. Spetr of pellets prepre s esrie in Setion 2.5 were ompre with the spetr of pellets where KBr ws mille for 15 s efore the smple ws e. These ltter smples were mixe with the KBr in the milling up without the steel ll for 15 s. No spetrl ifferenes were foun. This inites tht this short-time ll milling of the ilute smple (1.5 mg smple: 200 mg KBr) i not

8 30 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) (A) (B) (D) (C) Fig. 2. Infrre spetr of 30 min mille sprue woo smple A fter seprtion into severl frtions of prtile sizes. (A) Full rnge spetr n (B) spetr of woo mel in the rnge from 1600 to m 1 for ifferent frtions. () >80 mm, () (63 80) mm, () (40 63) mm, () <20 mm; (C) thik otte line >80 mm frtion of the ll-mille smple, thin soli line >80 mm frtion of the non-ll-mille smple A; (D) spetr of the (40 63) mm frtion of () non-ll-mille n () 30 min ll-mille woo smple A. influene the smple spetrum n tht KBr sors the energy impt uring milling Influene of oxygen A further milling uner nitrogen tmosphere ws performe to stuy the influene of oxygen on the spetr, ering in min tht possile oxition proess oul our uring the smple tretment. No ifferenes etween the spetr of the smples (woo smple B n ellulose) ll-mille uner nitrogen n those mille uner mient onitions were oserve (t not shown). In ft, no extensive oxition ws expete, sine Forziti et l. [41] showe tht even fter 8 h ll milling ellulose h een oxiise to only negligile extent Cellulose ll milling The most strongly ffete regions in the spetr of woo inlue those ns tht re minly ssigne to

9 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) polyshries espeilly ellulose (p. Tle 2). These ns re in the virtion rnge of intr- n intermoleulr hyrogen one OH-groups (etween 3500 n 3200 m 1 ), in the fingerprint region (p. Tle 2) suh s those t 1336, 1318, 1162 n 1110 m 1,ninthe rnge of C O vlene virtions from 1060 to 1015 m 1. Cellulose power tht h psse 45 mm sreen ws therefore ll mille for 0 5 min using the sme onitions s for the woo mel to tre the spetrl hnges. Two effets were oserve (p. Fig. 4 n Tle 2): hnges in the shpe of the spetr over pproximtely the whole rnge from 3700 to 2 m 1 n from 1500 to 600 m 1, n shift in position of the mxim of some ns. Fig. 5 shows the hnges in the FT-IR spetr of ellulose n woo (smple B) ue to ll milling. All the hnges in the ellulose spetr s result of the milling re in the hrteristi ellulose ns of woo. No oservle hnges in the hrteristi lignin peks were reognise, suggesting tht the hnges in woo spetr s result of the ll milling re minly ue to hnges in the ellulose. (A) g (B) g Fig. 3. Infrre spetr of ll-mille sprue woo smples B. (A) Full rnge spetr of woo; (B) spetr of woo in the rnge from 3700 to 2700 m 1 n (C) from 1600 to 600 m 1. Different times of ll milling: () without ll-milling, () 5 s, () 10 s, () 15 s, (e) 30 s, (f) 1 min, (g) 2 min; (D) smple B thermlly trete for 10 min in n oven t 45 8C (), 75 8C (), 105 8C () n 135 8C ().

10 32 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) (C) g (D) Fig. 3. (Continue ). Avery wek ro n rises t 1098 m 1 in the spetr of mille ellulose (Fig. 4C n Fig. 5B). This n is lso visile s wek shouler in the woo spetr. From norml oorinte nlysis of ellulose I [42] it is expete tht n shoul pper roun 1097 m 1. Wiley n Atll [43] stte tht this n (1095 m 1 ) erives from C C n C O strething motions. Chnges in ellulose ue to ll milling hve een ssoite with erese in the egree of rystllinity [9,44] n in the egree of polymeristion, epening on the milling time [8,41,45 47]. An inresing numer of lehye groups might e oserve ue to erese in the egree of polymeristion (DP), ut they oul not e seen in the ellulose spetr. Either the numer of lehye groups ws too smll to tre hnges in the spetr or the ellulose power (fire length mm) use for this investigtion ws lrey epolymerise so tht ll milling i not le to ny further epolymeristion. This ws lso foun y Stuinger n Dreher [8], who mille ellulose from sprue with n initil DP of 430. Even fter 7.5 h milling, no erese in DP ws oserve.

11 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) Crystllinity It is expete tht the milling proeure will ffet the rystllinity of the ellulose. Crystlline ntive ellulose I is ompose of two llomorphs, ellulose I n ellulose I [48]. ellulose minly ontins ellulose I. A hrteristi n in the OH region is reporte to e ple t 3270 m 1 [39], ut it is hien in the ro OH strething virtion n. The region etween 700 n m 1 is lso known to ontin hrteristi ns of the I s well s of the I form [39]. The seon erivtives of the spetr of sprue woo smples provie n t 715 m 1 (Fig. 6), whih lies within this region. This n eomes smller with inresing milling n it isppers fter 120 s of ll milling. This is true for other ns s well. The smll shouler t 750 m 1 ssigne to ellulose I lso isppers. Severl ttempts hve een me to hrterize n quntify the egree of rystllinity of pure elluloses using the intensities of ertin ns in the infrre spetr [49 53]. (A) f e (B) f e Fig. 4. Infrre spetr of ll-mille sprue ellulose smples. (A) Full spetr of ellulose; (B) spetr of ellulose in the rnge from 3700 to 2700m 1 n (C) from 1500 to 600 m 1. Different times of ll milling: () 0 min, () 0.25 min, () 0.5 min, () 1 min, (e) 2.5 min, n (f) 5 min.

12 34 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) (C) f 665 e Fig. 4. (Continue ). The rtios of n-heights t 1429 n 894 m 1 (A 1429 / A 894 ) [50,51] n t 1372 n 2 m 1 (A 1372 /A 2 ) [45,51] hve een use s reltive mesures of ellulose rystllinities. Evns et l. [52] foun tht the re rtio A 1372 /A 670, ompring the omine res of three peks t 1370, 1335 n 1315 m 1, further lle the 1370 m 1 pek, to tht of the pek t 670 m 1 my e use to etermine the reltive ellulose rystllinity in woo n krft pulps. As the lignin-erive pek t 1270 m 1 ffete the setting of the seline for the jent 1370 m 1 pek, the lignin-free spetr were use [52]. The sutrtion of spetrum of MWL (non-ll-mille prior to spetr olletion) from the sprue spetrum, until the re of the lignin-erive pek t 1510 m 1 eme zero [52], le to spetrum tht ws lmost ientil to the spetrum of holoellulose (Fig. 7A). Fig. 7B shows the regions use to tre hnges in the reltive egree of rystllinity uring ll milling. The pek res elow the spetrum within the limits were etermine s shown in Fig. 7C. The n t 1110 m 1 ereses in re uring ll milling reltive to tht of the rnge from 1430 to 1270 m 1 n it ws therefore inlue in these investigtions. Using the re of the n t 670 m 1 s referene [52] i not represent the hnges visile in the spetr (Fig. 8A B). A etter representtion ws otine for oth woo (sprue minus MWL) n ellulose using the rtios of the res from to 1289 m 1 (inluing oth jent vlleys) n from 1143 to 1089 m 1 to the re elow the n t 898 m 1 s referene (Fig. 8C D). Fig. 8D shows the rpi erese in rystllinity uring the initil stge of ll milling, whih is in greement with the literture [45,48 52,54]. The n t 898 m 1 ws the most suitle referene sine this n shows no positionl shift ue to ll milling. The vrition in n height n re ws less thn 10% (the slight hnge oul e prtly expline y erese in prtile size [1]), wheres the n t 667 m 1 vrie in oth height n position (Fig. 6). Shultz et l. [54] hve shown from Diffuse Refletne Infrre Fourier Trnsform (DRIFT) spetr tht the rystllinity of ellulose n est e estimte using the n height rtio 1430 m 1 /1403 m 1, whih gives liner reltion to rystllinity inies etermine y X-ry iffrtion. Liner reltionships with high orreltions (r > 0:97) were foun etween the n height rtio 1430 m 1 / 1403 m 1 in ellulose n the n re rtios (1110 m 1 /898 m 1 n 1289 m 1 /898 m 1 )in ifferene spetr (sprue woo minus MWL) n in sprue woo spetr. This will e pulishe in etils elsewhere. It n e ssume tht the n-re rtios in sprue woo IR spetr n e use to tre the egree of rystllinity. The results foun for ellulose n woo uring ll milling inite more rpi erese in the egree of rystllinity in woo Lignin Woo ontins smll mount of reltively stle free rils [55]. A known effet of ll milling is n inrese in the ourrene of these reltively stle free rils in lignin n the pperne of lile free rils in ellulose, lthough to less extent (out 1 2% ompre to woo mel or MWL) [55,56]. Hon [55] onlue tht the mehnil egrtion of ellulose is muh less severe thn tht of lignin.

13 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) (A) (B) Fig. 5. Infrre spetr of ll-mille sprue woo smples B n ellulose from sprue. (A) Spetr of woo n ellulose in the rnge from 3700 to 2700 m 1 n (B) from 1600 to 600 m 1. Different times of ll milling: () ellulose 0 min, () ellulose 5 min, () woo 0 min, () woo 2 min ll milling. In the seon erivtive spetr (spetr not shown), ifferenes were foun t 1662, 1269, 1220 n 1030 m 1. These ns re influene y the numer of C¼O groups ppering (p. Tle 2). The inrese in nheight in the seon erivtive spetr uring ll milling up to 2 min for ns t 1722 m 1 (out 25%), 1662 m 1 (out 15%) n 1220 m 1 (out 20%) my therefore e use y oxition retions s onsequene of the formtion of swith mehnorils (phenoxy rils n ron rils). The hnge in the n t 1722 m 1 (unonjugte ronyls [57,58]) is ue to hnges in the hemielluloses. The ns t 1269 n 1030 m 1 re ffete y hnges in the ellulose use y the ll milling. The hnges re of similr orer in the first se ut muh greter in the ltter. Free rils in lignin re forme minly y the levge of ether ons leing to phenoxy rils, ut lkoxy rils n lso e proue when hemil ons in the lignin-rohyrte omplex re roken [55]. The linkges inlue oth C ryl O n C lkyl O ons, whih re

14 36 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) g f e 2n erivtive Fig. 6. Chnges in intensity n positionl shifts in the seon erivtive spetr of sprue woo smples B uring ll milling. The spetr re shifte prllel to the wvenumer xis. () 0 s, () 5 s, () 10 s, () 15 s, (e) 30 s, (f) 60 s, n (g) 120 s. expete to give IR strething ns t n m 1, respetively [59]. In lignin, the most importnt C lkyl O ether virtions (methoxyl n -O-4) were ssigne to the m 1 region n the C ryl O virtions to the m 1 region [35,59]. Sine the spetrl regions in whih ns ttriutle to ether ons n e seen re ominte y ellulose (polyshries), it ws not possile to ssign hnges to etheron levge in lignin. Stuinger n Dreher [12] foun erese in DP from 1000 to uring intensive ll milling of Aies l Mill. n Ling et l. [47] reporte erese in DP from 1172 to 253 within 2 h milling for Pinus mssonin. The presene of oxygen uring ll milling elertes seonry retions. The trnsient mehnorils ret reily with oxygen to give peroxy rils, n these re ple of inititing oxitive hin retions [55,60]. The formtion of peroxie rils ws reognise [55], ut they ey rpily t mient temperture. No strething virtion n (O O) of peroxies, ppering in the rnge etween 1000 n m 1 [59], oul e seen in the spetr s it is very wek n is therefore esily hien y the others. Furthermore, the expete onentrtion of peroxies is extremely low. Termintion n eomposition retions of the trnsient mehnorils ppere to le ultimtely to the formtion of hromophori groups suh s quinonoi strutures (ompnie y emethyltion of the methoxy groups), ronyl groups, n oule ons in the lignin mromoleule [55]. A slight inrese in ronyl ontent [61,62], n inrese in phenoli hyroxyl ontent n erese in moleulr weight hve een reporte s result of virtory ll milling [61]. The smll inrese in the numer of ronyl groups foun uring this short-time virtory ll milling is in greement with the literture [55,60 62], ut the influene on the shpe of the FT-IR spetr ws negligile n none of the possile hnges in lignin struture reporte y the other uthors oul e tre Comprison etween woo n ellulose The hnges visile in the infrre spetr ue to ll milling of woo n ellulose were very similr, ut the muh fster erese in n intensity uring the milling of woo requires some explntion. Compre to ellulose, lignoellulose seems to e more esily onverte into n morphous stte. Using seon erivtive spetr, the resolution of the ns is enhne. The n t 3335 m 1, whih is ssigne to O3-H...O5 0 intrmoleulr hyrogen ons (Tle 2), ws foun in woo, holoellulose n ellulose, n the n ptterns etween 3480 n 3220 m 1 were lmost ientil, exept for remrkle ifferene in the se of ellulose in the n t 3268 m 1 (Fig. 9) tht is ssigne to O6-H...O3 intermoleulr hyrogen ons in ellulose I [39]. In the erivtive spetr of woo, n ws oserve t 3280 m 1, ut no ns were foun etween 3360 n 3220 m 1 in either MWL or Klson lignin. A ro n ws however foun t out 3275 m 1 in the seon erivtive spetrum of holoellulose. It is plusile to suggest tht the shift from 3268 to 3275 m 1 is use y hemielluloses, sine holoellulose onsists of

15 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) A B C e f g h Fig. 7. The infrre sorne spetr, the spetr proessing n the selete wvenumer regions for tring the reltive hnge in the egree of rystllinity uring the ll milling proess. (A) () sprue woo (smple B), () mille woo lignin from sprue, () ifferene spetrum (sprue woo mille woo lignin), () spetr of holoellulose from sprue with out 2% resiul lignin; (B) (e) ifferene spetrum (unmille sprue woo minus MWL), (f) ifferene spetrum (sprue woo fter 2 min ll milling minus MWL), (g) ellulose without ll milling, (h) ellulose fter 5 min ll milling. oth ellulose n hemielluloses, ut it is not ler whether the shift to higher wvenumer (from 3275 to 3280 m 1 etween holoellulose n woo), n therefore higher energy, is result of tensions within the polymer woo. It is remrkle tht the ll milling les to erese in the n t 3280 m 1 in the erivtive spetr of woo, wheres the n t 3268 m 1 in the ellulose spetr is itionlly shifte to 3260 m 1 (Fig. 9). After longer milling time the spetr of woo n ellulose eome lmost ientil. Fig. 10 shows for omprison the seon erivtive spetr of unmille n mille sprue woo n ellulose. The n t out 2968 m 1, ssigne to n symmetri streth virtion of CH 2 from the CH 2 OH group in ellulose [22,26,30], ppers s shouler in ll the spetr n eomes smller in the se of the mille smples. On the other hn, the n t 2878 m 1, whih erives from symmetri streth virtion of the CH 2 group [22,26,30], inreses. A similrity in ehviour to tht of the ns t 3335 n 3280 m 1 (3268 m 1 for ellulose) is ovious.

16 38 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) (A) Cellulose re-rtio -1300/670 Cellulose re-rtio 1110/ (B) 8 re-rtio -1300/670 8 Are-rtio re-rtio 1110/670 Are-rtio Cellulose re-rtio -1289/670 Cellulose re-rtio 1110/670 re-rtio -1289/ re-rtio 1110/ Time (s) Time (s) (C) Cellulose re-rtio -1300/898 Cellulose re-rtio 1110/898 4 (D) Cellulose re-rtio -1289/898 Cellulose re-rtio 1110/898 re-rtio -1300/898 re-rtio -1289/898 Are-rtio 3 2 re-rtio 1110/898 Are-rtio 3 2 re-rtio 1110/ Time (s) Time (s) Fig. 8. Rtios of res elow the selete ns in the infrre sorne spetr within the hosen rnges (A D) showing reltive hnges in the egree of rystllinity uring ll milling of sprue woo smple B (sprue woo minus MWL) n ellulose. In ition, the inresing n t 2940 m 1 (n(ch) [22]) shows shift to lower wvenumers. The n t 2898 m 1, whih ws use for the etermintion of the egree of rystllinity y Nelson n O Connor [45], isppers. In summry, the results show tht FT-IR spetrosopy n e use to ssess the struturl hnges in woo ourring uring to short-time virtory ll milling n mehnil tretment in generl. When FT-IR spetr of woo n ellulose re presente, it is therefore neessry to report f e n erivtive Fig. 9. Chnges in intensity n positionl shifts in the seon erivtive spetr of sprue woo smples B n ellulose uring ll milling. () 0 s (woo), () 15 s (woo), () 120 s (woo), () 0 s (ellulose), (e) 30 s (ellulose), (f) 300 s (ellulose).

17 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) n erivtive Fig. 10. Chnges in intensity n positionl shifts in the seon erivtive spetr of sprue woo smples B n ellulose uring ll milling. () 0 s (woo), () 120 s (woo), () 0 s (ellulose), () 300 s (ellulose). the proeure in etil to voi ny misinterprettion ue to hnges rising uring smple preprtion, n to report n re rtio or n-height rtio representing the egree of rystllinity. The stnristion of the methos use for qulittive n espeilly for quntittive nlysis is essentil. the methos use for qulittive n espeilly quntittive nlysis n for the preprtion of spetr to e use in spetrl tses. Therefore, we wish to invite lortories intereste in prtiipting in the neessry investigtions leing to stnr proeure to ontt us. 4. Conlusions Virtory ll milling of woo h mjor influene on the shpe of FT-IR spetr, even though the smples were mille for only short time. Bns erese or ispper uring ll milling, resulting in loss of informtion. The most ovious hnges re ttriutle to erese in the egree of rystllinity n/or egree of polymeristion of the ellulose. This is ue minly to the primry mehnil tretment n in lesser extent to the finer mteril proue. Seonry effets like n inrese in temperture, oxition or ril retions were foun to e of minor importne. For the preprtion of FT-MIR spetr of woo mel, we reommen the following proeure. Mill the woo s fine s possile, tking speil re to voi heting the mteril. Dilute out 1.5 mg smple with out 200 mg KBr, mix it in the milling up of virtory ll mill without the ll for 10 s. Then the ll n mill this mixture for 15 s. Prepre KBr-pellets uner vuum in stnr evie uner pressure of 75 kn m 2 for t lest 3 min. Collet t lest 32 sns per smple t spetrl resolution of 4 m 1 n tke the verge of t lest two ientil spetr for further evlution. We elieve tht the prolems resse re very ommon ones. This emphsises the importne of stnristion of Aknowlegements This work ws rrie out with finnil support from the Austrin Siene Fun (FWF), Reserh projet P BIO, the EU (EU-projet Gemini-QLK ) n the Austrin Competene Centre for Woo Composites n Woo Chemistry (WOOD K plus). The uthors thnk the prtners of the EU-projet GENIALITY (Contrt FAIR CT ) for proviing mteril n Professor Oskr Fix from the Institute for Woo Chemistry n Chemil Tehnology of Woo of the Feerl Reserh Centre for Forestry n Forest Prouts in Hmurg for kinly proviing the MWL. The uthors thnk Dr. Eliseth Wineisen from the Holzforshung Münhen of the Tehnishe Universität Münhen for fruitful isussion. Referenes [1] O. Fix, J.H. Bötther, Holz ls Roh- un Werkstoff 50 (1992) [2] R. Meer, S. Gllgher, K.L. Mki, H. Böhler, R.R. Meglen, Holzforshung 53 (3) (1999) [3] J. Cost e Silv, B.H. Nielsen, J. Rorigues, H. Pereir, H. Wellenorf, Holzforshung 53 (6) (1999) [4] H. Pein, Holztehnologie 23 (2) (1982) [5] J. Rorigues, O. Fix, H. Pereir, Holzforshung 52 (1998)

18 40 M. Shwnninger et l. / Virtionl Spetrosopy 36 (2004) [6] A.-M. Olsson, L. Slmén, Nori Pulp Pp. Res. J. 3 (12) (1997) [7] M. Åkerholm, L. Slmén, Polymer 42 (2001) [8] H. Stuinger, E. Dreher, Berihte 69 (1936) [9] E. Steurer, K. Heß, Zeitshrift für physiklishe Chemie 193 (1944) [10] K. Heß, E. Steurer, Zeitshrift für physiklishe Chemie 193 (1944) [11] P.H. Hermns, A. Weiinger, J. Am. Chem. So. 68 (1946) [12] H. Stuinger, E. Dreher, A.A. Ekenstm, Berihte er eutshen hemishen Gesellshft 69 (1936) [13] A. Murer, D. Fengel, Holzforshung 46 (5) (1992) [14] D. Fengel, Holz ls Roh- un Werkstoff 24 (1966) [15] D. Fengel, M. Luwig, Ds Ppier 45 (2) (1991) [16] J. Gierer, Holzforshung 36 (1) (1982) [17] J. Gierer, Holzforshung 36 (1) (1982) [18] H. Hs, W. Shoh, U. Störle, Ds Ppier 9 (19 20) (1955) [19] E. Zvrin, S.J. Jones, L.G. Cool, J. Woo. Chem. Tehnol. 10 (1990) [20] A. Svitzky, M.J.E. Goly, Anl. Chem. 36 (1964) [21] D. Fengel, Holzforshung 47 (1993) [22] K. Tshiro, M. Koyshi, Polymer 32 (8) (1991) [23] E.P. Klutsky, S.S. Gusev, Polym. Si. U.S.S.R. 22 (1981) [24] N.V. Ivnov, E.A. Korolenko, E.V. Korolik, R.G. Znkov, Zurnl Priklnoj Spektroskopii 51 (1989) [25] C.Y. Ling, R.H. Mrhessult, J. Polym. Si. 37 (1959) [26] H.J. Heiger, Infrrotspektroskopie, Grunlgen, Anwenungen, Interprettion, Akem. Verlgsges., Frnkfurt/Min, [27] L.J. Bellmy, The Infr-re Spetr of Complex Moleules, Chpmn & Hll, Lonon, [28] C.Y. Ling, R.H. Mrhessult, J. Polym. Si. 39 (1959) [29] O. Fix, Holzforshung 45 (Suppl.) (1991) [30] V.P. Pnov, R.G. Znkov, from D. Fengel, Proeeings of Celluon Conferene, 1993, Lun Sween, 1988, pp [31] W.E. Collier, V.F. Klsinsky, T.P. Shultz, Holzforshung 51 (1997) [32] B. Hinterstoisser, M. Åkerholm, L. Slmén, Crohyrte Res. 334 (2001) [33] D. Fengel, in: Proeeings of the Celluon Conferene, 1993, pp [34] Y. Mréhl, H. Chnzy, J. Mol. Strut. 523 (2000) [35] W.E. Collier, T.P. Shultz, V.F. Klsinsky, Holzforshung 46 (6) (1992) [36] L.P. Gol mn, V.M. Reznikow, Khim. Ispol z. Lignin, 1974, pp [37] R.H. Mrhessult, Pure Appl. Chem. 5 (1962) [38] J. Dehnt, R. Dnz, W. Kimmer, R. Shmolke, in: J. Dehnt (E.), Ultrrotspektroskopishe Untersuhungen n Polymeren, Akemie- Verlg, [39] J. Sugiym, J. Persson, H. Chnzy, Mromoleules 24 (1991) [40] J. Pelouze, Lieigs Ann. Chem. 11 (1834) [41] F.H. Forziti, W.K. Stone, J.W. Rowen, W.D. Appel, J. Res. Ntl. Bureu Stn. 45 (2) (1950) [42] J.J. Cel, K.H. Grner, J.L. Koenig, J. Blkwell, J. Chem. Phys. 62 (3) (1975) [43] J.H. Wiley, R.H. Atll, Crohyrte Res. 160 (1987) [44] N. Stuir, I. Smit, M. Stuir, A. Tonej, A. Jnosi, J. Shurz, P. Zipper, Holzforshung 52 (5) (1998) [45] M.L. Nelson, R.T. O Connor, J. Appl. Polym. Si. 8 (1964) [46] M.L. Nelson, R.T. O Connor, J. Appl. Polym. Si. 8 (1964) [47] X.H. Ling, L.Z. Gu, E.Y. Ding, Woo Si. Tehnol. 27 (6) (1993) [48] R.H. Atll, Siene 223 (1984) [49] D. Fengel, Holzforshung 46 (4) (1992) [50] R.T. O Connor, E.F. DuPré, D. Mithm, Textile Res. J. 28 (1958) [51] L. Ferrús, P. Pgés, Cell. Chem. Tehnol. 11 (1977) [52] R. Evns, R.H. Newmn, U.C. Roik, I.D. Sukling, A.F.A. Wllis, Holzforshung 49 (1995) [53] U. Rihter, T. Kruse, W. Shempp, Angew. Mkromol. Chem (1991) [54] T.P. Shultz, G.D. MGinnis, M.S. Bertrn, J. Woo Chem. Tehnol. 5 (4) (1985) [55] D.N.-S. Hon, Appl. Polym. Symp. 37 (1983) [56] T.N. Kleinert, Ds Ppier 24 (10) (1970) [57] H.L. Hergert, in: K.V. Srknen, C.H. Luwig (Es.), Lignins. Ourrene, Formtion, Struture n Retions, Wiley, New York, Chihester, Brisne, Toronto, Singpore, 1971, pp [58] B.S. Furniss, A.J. Hnnfor, V. Rogers, P.W.G. Smith, A.R. Tthell, Vogel s Textook of Prtil Orgni Chemistry, Longmn Sientifi n Tehnil, NY, [59] W. Gottwl, G. Whter, IR-Spektroskopie für Anwener, WILEY- VCH, Weinheim, [60] B. Råny, K.P. Krinst, E.B. Cowling, S.Y. Lin, At Chem. Sn. 23 (1969) [61] H.-M. Chng, E.B. Cowling, W. Brown, Holzforshung 29 (5) (1975) [62] A. Björkmn, B. Person, Svensk Ppperstining 60 (5) (1957) [63] D. Fengel, M. Luwig, Ds Ppier 45 (2) (1991) [64] D. Fengel, Holzforshung 47 (1993) [65] K. Tshiro, M. Koyshi, Polymer 32 (8) (1991) [66] E.P. Klutsky, S.S. Gusev, Polym. Si. U.S.S.R. 22 (1981) [67] N.V. Ivnov, E.A. Korolenko, E.V. Korolik, R.G. Znkov, Zurnl Priklnoj Spektroskopii 51 (1989) [68] C.Y. Ling, R.H. Mrhessult, J. Polym. Si. 37 (1959) [69] H.J. Heiger, Infrrotspektroskopie. Grunlgen, Anwenungen, Interprettion, Akem. Verlgsges., Frnkfurt/Min, [70] L.J. Bellmy, The Infr-re Spetr of Complex Moleules, Chpmn & Hll, Lonon, [71] C.Y. Ling, R.H. Mrhessult, J. Polym. Si. 39 (1959) [72] O. Fix, Holzforshung 45 (Suppl) (1991) [73] V.P. Pnov, R.G. Znkov, from D. Fengel, Proeeings of Celluon Conferene, 1993, Lun Sween, 1988, pp [74] W.E. Collier, V.F. Klsinsky, T.P. Shultz, Holzforshung 51 (1997) [75] D. Fengel, in Proeeings of Celluon Conferene, 1993, pp [76] B. Hinterstoisser, M. Åkerholm, L. Slmén, Crohyrte Res. 334 (2001) [77] Y. Mréhl, H. Chnzy, J. Mol Strut. 523 (2000) [78] H. Stuinger, E. Dreher, A.A. Ekenstm, Berihte er eutshen hemishen Gesellshft 69 (1936) [79] W.E. Collier, T.P. Shultz, V.F. Klsinsky, Holzforshung 46 (6) (1992) [80] L.P. Gol mn, V.M. Reznikow, Khim. Ispol z. Lignin, 1974, pp [81] R.H. Mrhessult, Pure Appl. Chem. 5 (1962) [82] J. Dehnt, R. Dnz, W. Kimmer, R. Shmolke, in: J. Dehnt (E.), Ultrrotspektroskopishe Untersuhungen n Polymeren, Akemie- Verlg, [83] J. Sugiym, J. Persson, H. Chnzy, Mromoleules 24 (1991)