CA1096374A - Method of rendering lignin separable from cellulose and hemicellulose in lignocellulosic material and the product so produced - Google Patents

Abstract

TITLE

A METHOD OF RENDERING LIGNIN SEPARABLE FROM CELLULOSE
AND HEMICELLULOSE IN LIGNOCELLULOSIC MATERIAL AND
THE PRODUCT SO PRODUCED

INVENTOR
Edward A. DeLong ABSTRACT OF THE DISCLOSURE
Lignocellulosic material is packed in particulate form in a pressure vessel, steam is passed into the pressure vessel at a pressure of at least 500 psi to bring the ligno-cellulosic material to a temperature in the range 185 to 240°C
(preferably in the range 200 to 238°C, more particularly 234°C) in less than 60 seconds (preferably less than 45 seconds) to plasticize the lignocellulosic material. As soon as the ma-terial is plasticized it is extruded through a die to atmo-sphere where abrasion in the die and explosive decompression forms a particulate material like potting soil that rapidly sinks in water and from which the cellulose content is large-ly utilizeable, e.g. as a feed for ruminants. During extru-sion the softened material is subjected to severe deformation and then severe mechanical shock thus fracturing the linkages between lignin, the hemicellulose and the cellulose. The temperature range and rapid rise thereto by the material substantially minimizes degradation of the fragile hemicellu-lose. The process according to the present invention con-verts lignocellulosic material into a substrate which is extremely easy to treat by enzymes, either as a feed for ruminant animals or for conversion to sugars and alcohols and for the solvent extraction of lignin.

Description

~t91~37'~
This inven-tion relates to a method of renderiny lignin separable from cellulose and hemicellulose in ligno-cellulosic material and the product so produced.
In this specification, rendering lignin separable from hemicellulose and cellulose in lignocellulosic material means -that the lignin is capable of being separated from the hemicellulose and cellulose by, e.g., organic solvent ex-craction.
~ n United States Patent Number 3,212,932, dated October 19, 1965, ~'Selective Hydrolysis of Li~nocellulose Materials'~, R. W. Hess r A . M. Thomsen, F. Porter and J. W.
Anderson~ there is described a two stage hydrolysis process for producing pentose sugars and hexose sugars in separate fractions from lignocellulosic materials so that each may be applied as desired in the formulation of stock feed or each may be applied separately to the various other commer~
cial uses to which pentose and hexose sugars are applicable.
The lignoceliulosic materials which may be employed in the selective h~drolysis process of Hess et al comprise those 2n classes of lignocellulosic materials which stem from plant growth processes and are readil~ available as waste by-products ~f various industries. Thus they may comprise oat hulls, corn stalks and bagasse. In particular, however, they comprise the woods of various species of trees. The lignocellulosic material recluires no special treatment pre ~iminary to its use in the Hess et al process, althoucJh it should be reduced to a finely divided state if it already ~s nct present in this condition. ~'hus wood may be employed to advantage in the form of sawdust, wood shavings, thin hip flakes, and the like.
In the first stage ~drolysis of the Hess et al pi-ocess the lignocellulosic material is heated with a first , ~ .

i3 ~'4 aqueous liquor of sucn a nature, and under such conditions, as to hydrolyze selec-tively the hemicellulose content of the li~nGce~lulose, forming pentose sugars as the desired produc.. The premixed lignocellulose and aqueous liquor are intLoduced into a suitable pressure vessel. This may be eithf~r a continuous or batch pressure reactor provided with means for heating the charge to the predetermined tem-perature at the predetermined pressure. ~s stated before, this may be accomplished by direct steam injection.
'0 Within the reactor the pressure upon the change is incre~sed as rapidly as possible to a value from 100-700 psi preferably from 250-600 psir the temperature being increased contemporaneously to the corresponding levels for saturated steam. These conditions are maintained for a relatively brief period of time, sufficient only to convert substantially and selectively the hemicellulose content of the liynocelluiose to pentose su~ars. In the average case this requi~es but from 0.3~10 minutes, the time beiny in substantially inverse re~ation to the temperature applied.
That is, the higher the temperature, the shorter the time, and vice versa.
As a result, there is formed a first liquor pro-duct containin~ a substantlal proportion of pentose sugars, 'ogether with a small amount of volatile oryanic acids, such as acetic acid, as well as t~le residual mineral acid if a mineral acid is included in the first instance. There also is formed a first solid residue which is substantially free frorn pentosans and contains predominantly undegraded lignin and undegraded cellulose.
The pressure next is reduced preliminar~ to sepa-ration of the liquor and solid residue products. Whereas the time required for pressure reduction by prior art 1~9637~

procedures has been very long, i.e. of the order of several hours, it is important for the success of the presently described process that it be kept at a very low value.
Thus t ere is a substan-tiall~ instantaneous reduction of pressur_ resulting in what is termed herein a "flash blow-down". Where a continuous reactor is employed, the blowdown time is bu-t a few secondsO Where a large bath reaction is used, the blowdown time is but a few minutes.
~uch a rapid reduction in pressure has several significant effects.
First, it rapidly stops the hydrolytic reaction.
This in turn minimizes the production of hexose sugars from degradation of the cellulose. It also minimizes pro-duction of lignin degradation products and prevents the decomposition of the desired sugar products.
Secondly, the flash blowdown evaporates some of the water which is present. The resultant steam then may be employed to advantage in a heat exchange with the mate-rial charg d in the reactor.

2~ Third, the flash blowdown flashes o~f acetic acid, formis acid, or other organic volatiles which may have been ~ormed as by-products of the reaction. There thus is pro-vided a built-in operation for separation and removing im-purities from the reaction products.
Fourth, the flash blowdown explodes the particles of the solid residue. This ma~es them porous opening them up for more efficient treatment in the second hydrolytic stage r In ~he second stage hydrolysis of the Hess et al process the second stage reactor conditions are moreistre-nuous than those prevailing in the first stage reactor.

They have as their object the conversion of the cellulose 7~

t:o hexose sugars without inducing undue degradation of the lignin.
Accordingly the charging liquor to solids ratio is main~ained within the broad range of from 1:1-5:1, pre-ferably ~rom 1~1-3:1. The mineral acid concentration of - the liquor treating agent is maintained at a level of from 0O3-3~0~ by weight.
The reactor is heated indirec-tly or preferably by the direct injection of steam, until a pressure of 150 to lC 900 psi preferably from 400~800 psi and corresponding tem-peratures for saturated steam, are reached.
The reactor is maintained under the foregoing con-ditions for a time which is substantially inverse relation to ihe temperature, i.e. the higher the temperature the shorter the time and vice versa. During this time, which ls within the range of from 0 3 to lQ minutes, the cellulose content of the charge is converted substantially selectively to hexose sugars, leaving a solid residue containing pre-dominantly unhydrolyzed ligrlin.
~0 As in the first stage, it is highly desirable to terminate the reaction abruptly in order to evaporate ex-cess water, in order to flash off any organic volatiles which may be present, and in order physically to modify the lignin residue so that it may be filtered and handled more easily.
For these reasons the cha-ge of the reactor is sub~ec~ed to a flash blowdown as by passing it continuously to a blowdown cyclone apparatus. This reduces the pressure to atmospheric pressure in a matter of but a few seconds.
~0 The steam from the blowdown apparatus is vented while the solid product is washed with water in a second stage extractor. The operation of this extractor results

3~74 in separating the hexose su~ar liquor from the cellulose containing lignin residue, which is recycled or passed to waste.
Thereafter the liquor is -~reated with a neutrali-zing aqent such as ca~.cium carbonate and any appropriate filter aid, more water is added if necessary and the mix-ture filtered.
While the ~less et al process is a useful contri-bution to the art the applicant has found that the hydro-lysis, which renders the pen-tose su~ars and hexose sugars separable from the lignin in lignocellulosic materials, does so at: the expense of loss of useful hemicellulose sub-stances present in lignocellulosic materials from plant growth products. ~he acid action of the hydrolysis that is necessary to break down the lignocellulose, progressively des~roys the fermentation value of the hemicellulose, so that an undesirably high proportion of t-he hemicellulose is rendered indigestible by ruminants in the period neces-sary for hydrolysis of the lignocèllulose to have occurred.
There is a need to provide a method of increasing the accessi~ility of cellulose in lignocellulosic materials wherein the lermentation value of the hemicellulose is largely preserved~
In United States Patents Numbers 3,667,961, dated June 6, 1972, "Process for Improving Digestibility of Feed-stuffs for Ruminant Animals"/ J. W. Al~eo and 3,817,786, dated June 18, 1974, "Equipment for Converting the Physical and Complex Molecular Bond Structures of Natural Feedstuffs for Ruminant Animals to Different and Less Complex Molecu]ar 3~ Bond Structures Thereof", J. WO Algeo, equipment and a pro-cess are described for subjecting ruminant animal feedstuffs to high pressure steam in a pressure vessel and effecting ~A

9637'~

- explosive release thereof to the atmosphere through a cen-tral valve for collection and storage of sponge-like end products which are characterized by their capacity to take up watQr and digestive juices readily when in~ested by a ruminart ar.imal. For cellulosic feedstuffs such as all common roushages including ha~s, straws and the like, as well as for lignocellulosic feedstuffs such as cottonseed hulls, rice hulls, nut shells or coffee grounds the steam pressure in the pressure vessel is rapidly raised to 5Q0 to l,000 psi and the temperatures achie~ed therein approxi-mate 470 E (243C) to 546F (285C). This process may be described as a thermo-mechanical process for treating lig-nocellulosic materials. It is also applicable to other substances such as grain feedstuffs which are not lignocellu-losic and for these substances a temperature range for the steam of between 350F ~177C) and 400F (204C) is used.
While the Algeo process is undoubtedly a useful contribution to the art it has been found by the applicant thal raising the material to a temperature range of 470F
(243C) to 5~6F ~285C) tau~ht by Algeo drastically reduces the fermentation value of the hemicellulose. As previously stated, a need exists for a method of xendering lignin sepa-rable from the cellulose and the hemicellulose contained in li~nocellulosic materials of plant growth process, which will, for some uses of t11e product so produced, largely preserve the fermentatio1l ~a~ue of t:he hemicellulose therein and at the same time will increase the accessibility of the alphacellulose therein ko enz~matic digestion.
In this specification plant growth species means ~Q a soft stemmed or hard stemmed organism wherein organic substances have been produced by photosynthesis. This ma-terial is commonly xeferred to as biomass.

~6~

3~74 Lignocellulosic materia],s of plant grow-th pro-cesses are created in nature as a three-dimensional com-posite in which the cellulose, hemicellulose and the lignin polyr,lers are intimately mixed in a complex structure which varies rom one plant growth species to another and with age for a p~rticular specles~ Lignin is an amorphous three-climensicnal polymer, each macromolecule of which is built up from several hundred phenyl propane units linked in various ways. The lignin, probably in compound with the iO hemicellulose, permeates the matrix of cellulose microfibrils in the cell walls and largely fills the spaces between the cells. X~ray crys-tallographic evidence suggests that be-cause the cellulose microfibrils constitute a separately crystalline phase, the lignin is more intimately cross~
linked with the hemicellulose than with the alphacellulose.
Dry cellulose of plant growth has a softening or glassification temperature of 230~ and because of its cry-stalline nat~re, this temperature is not much reduced by the ?resence o~ moisture. In contrast, the hemicellulose 2~ and lignin of plant growth soften at 120-180C and because these materials are non-crystalline, their softening tempe-ratures ~ary with moisture and pressure. The pol~mers o~
plant growth are formed in definite morphological patterns and are interconnected by covalent bonds which can be li-kened to "spot welds" and it is these `'spot welds" which produce the well-known and persistent association between lignin and the hemicelluloses of plant growth. In a spot ; welded composite of macromolecules of plant growth, the thermal softening is clominated by the behaviour of the 3C cellulose component. ~ven when sorbed moisture lowers the transition tem~erature of the amorphous constituen-ts, the material retains its rigidity up to temperatures near 3'7~

230 C (4~.6 F) because of the over-riding effect of the cellu-lose frame-wor~ which is no-t plasticized by water molecules.
The lignocellulosic comF)osite of plant growth re-presents a hindrance to an enzyme or rumen microflora trying to break down cellulose and hemicellulose to simple sugars.
Li~nocellulosic materials ~ary in their degree of cross-linkage. In order -to improve digestibility these lignin-carbohydrate bonds, chemical and/or physical need to be broken.
~t is another object of some embodiments of the present invention to provide a thermo-mechanical process for rendering :Lignin separable from cellulose and hemicellu-lose in lignocellulosic material with negligible loss of the fermentation value of hemicellulose substances.
~ccording to one aspect of the present invention there is provided a method of rendering lignin separable from cellulose and hemicellulose in lignocellulosic material, comprising:
a) packing the lignocellulosic material in a divided, exposed,moist form in a pressure vessel having a closed extrusion die outlet, b) xapidly :~i.lling the pressure vessel ~tith stecm at a pressure of at least 500 psi to bring the lignocellulosic material to a temperature in the range 185 to 2~0& in less thall 6~ seconds to the.rrnailv soften the lignocellulosic material into an extrudable condition, and c~ as soon as the said extrudable condition has been attained,opening the extrusion die outlet and instantly extruding the lignocellulosic rnaterial in the said extrudable c~nditionfrom the pressure vessel through the extrusion die outlet to atmosphere so tha-t the said material issues from the ext.rusion die in particulate form with lignin therein 3~4 xendered into particles substantially in the range 1 to 10 microns and separable from the cellulose and hemicellulose the particulate lignin and cellulose beiny together in disasso iated form having the appearance of potting soil, a maior portion of -~he lignin bein~ soluble in methanol or et'lanol and being thermoplastic, the cellulose being in the form of crystalline alpha cellulose microfibrils and suitable for dic3estion by microorganisms and ei-~ymes.

Preferably the lignocellulosic material is ex-:lO truded on to an impinging surface.
In preferred embodiments of the present invention the lignocellulosic material is brought to a temperature in the range 200 to 238C in less than 45 seconds.
In some embodiments of the present invention the process is repeated at least once more with the lignocellulo-sic material at a temperature no greater than that at which the lignocellulosic material was treated for the first tIme in the s-teps b) and c)O
In some embodiments of the present invention the ~0 lignin is sclvent extracted from the particulate material that has issued from the extrusion die, preferably by a solvent selected from the ~xoup consisting of ethanol, methanol and acetone.
Wet alcohol insolubles may be separated from the particulate material b~ alcohol prior to lignin ex-traction therefrom by acetone.
Water solubles may be extracted from the particu-late material b~ water before or after lignin has been so~vent extracced therefrom.

The solvent extracted lignin may be in particulate form wi.th particles substantially in the range 1 to 10 microns, have a puritv of the order of 93 weight ~, be readily reac-3 ~ ~

tive chemically, be thermoplastic, and have an infrared spectrum approachlng that of so-called native lignin.
Hemlcellulose may ~e extracted from the lignocellu-- losic material hy hydrolysis with hot water prior to packing the ligno_ellulosic material in the pressure vessel.
~-eferably when the lignocellulosic material is wood, Ihe pressure vessel is rapidly filled with steam at a pressure in -the range 600 to 700 psi to minimize hydrolysis.
More par-t~.cularl~ when the lignocellulosic mate-rial is wood, the pressure vessel is preferably rapidly filled with steam at a pressure in the region 650 psi.
Preferably when the lignocellulosic material is annual plant material r the pressure vessel is rapidly filled with a steam at a pressure in the range 50Q to 600 psi, more particularly in the region 5S0 psi.
; In some embodiments of the present invention, the lignocel:Lulosic material :in .finely divided form ;.s moist wit~ water, in the region of fibre sa-turation, when packed in the pressure vessel~
In some embodiments of the present invention, when the li~nocellulosic material is annual plant substance, the pressure vessel is rapidly fi~led with steam at a pressure in the region of 5~0 psi, after approximately ~0 seconds the steam pressure is rapidly increased to a higher pressure and the material is released for extrus.ion immediately upon achieving said higher pressures.
According to a second aspect of the present invention, there is provided the product in particulate form produced by the method according to the present invention.
The product according to the present invention has the desirable characteristics that a significant portion of the alphacellulose that was present in the lignocellul.osic material is now accessible in a digestible for~ for - :L O

1~J~ ~3 7L~

microorganism diges-tion while minimal decrease in the fermentation value or content of hemicellulose that was present in the lignocellulosic material is found to have occur-ed. Tnus the product is use:Eul:
1) as a feed for ruminants, ii) for the enzymatic production of sugars, iii) ~or the conversion to alcohol using microbio-logical or chem~cal or both means.
The product of the present invention also has the desirable characteristics that:
a) a major portion of the sugars of the hemicellu-lose are removable therefrom by, for example, hot water, and b) a major portion of the lignin is removable therefrom by solvent extraction, for example, using ethanol, methanol or acetone leaving a cellulose rich material for use, for example, as pulp in a cellulose derivative process or to be directly solubilized using an appropriate cellulose solvent.
In this specification "lignoceliulosic material"
includes sucn plant growth materials as oat hulls, corn stalks, bagasse, wheat straw, rice straw, oat straw, barley straw, and woods of various species, particularly hardwoods.
In the accompanying drawings, Figure 1 is a sectional side view of a pressure vessel having an extrusion dle outlet and which has been used ~o verify the present invention, Figure 2 is a graph showing the rate at which re-ducing sugar is released by hydrolysis from products produced by the extrusion process, 3~ Figure 3 is a graph of enzyme digestibility of re-ducing substances such as glucose produced by different tem-peratures at which the lignocellulosic material is extruded, oll--~'.

37~

Figure 4 is a graph of -the enzyme digestibility of the sugar content produced by different temperatures at which the li(3nocellulosic material is extruded, Figures 5 to 8 are scann.ing electron micrographs of products according to the present invention produced using the apparatus shown in ~igure 1, Figure 9 is a flow diagram for solvent extraction of lisnin from treated aspen wood, I~'igure lC is a flow diagram for solvent extraction of lignin from treated barley straw, Figure 11 is a flow diagram for solvent extraction of li~nin from treated aspen wood or barley straw using ace-tone as the sol~nt, and Figure 12 is a flow diagram of the acetone ~xtrac-tion of lignin from, for example, treated wood or straw.
In Figure 1 there is shown a pressure vessel '; 2, a star-shaped extrusion die outlet 4 r an ext:rusion die closure plug 6, a loading end closure flap 8 and steam inlet ori.ices 10 ,o 12~
The pressure vessel 2 has a bottle neck portion 14 leading to the die 4 and entry ports 16 and 18 fnr tempera-tu.re probes (not shown)~
The front end of the pressure vessel 2 containing the die outlet 4 has a flange 20 to whlch is sealed a curved impinqing tube 22 which graduall~y reduces .in cross-section in a downstream directionO The curved lmpinging tube 22 has a spindle inlet sleeve 24 provided with a flange 26. A
pneumatic ram 28 is attached to the flan~e 26 and has a die closure plug 30 mounted on t'he spindle 32 of the ram 26.
The rear end of the pressure vessel 2 is sealed to the remainder by flanges 36 and 38 and has the loading end closure flap 8 hinged thereto by a hinge 40 and sealable ~2-'~
~ .

~ F;37~

therewith by a clamp 42.
In operation the loading end closure flap 8 is opened and the pressure vessel 2 is loaded wi,th lignocellu-losic m._terial in a divided form with the die closure plug 30 closiny the die outlet 4O ~ rod (not shown) is used to pack the lignocellulosic material in the pressure vessel 2.
With the pressure vessel 2 completely filled with lignocellulosic material the die closure plug 30 is sealed by the pneumatic ram 28 and the closure plug 8 is sealed to the rear end 34 by the clamp 42 and then the pressure vessel is filled with steam at a pressure of at least 500 psi, preferabl~ in the ran~e 500 to '700 psi, and at a suf-ficient temperature to raise the temperature of the ligno-ce~lulosic material to a temperature in the range 185 to 240C, preferably 200 to 238C and in particular 234C, in less than 60 seconds (preferably less than 45 seconds), to i p~.as.,icize the hemicellulose and the lignin in the lignocellu-losic ~aterial s~ that the li~nocellulosic material is thermallv so'tened into an extrudable condition, by injecting steam into the steam inlet orifices 10 to 12 from a source (not shown). The temperature po~ess (not shown) in the ports 16 and 18 are used to monitor the temperature of the lignocellulosic material in the pressure vessel 2 to determine when the lignocellulosic material has reached the chosen t.empera'ure.
As soon as the lignocellulosic material in the pressure vessel 2 reaches the desired temperature the pneu-matic ram 28 is actuated to with~raw the closure plug 30 and more or less instantaneously open the die outlet 4 to atmosphere so that the lignocellulosic material is extruded through the die outlet 4 in the plasticized condition~and at the extrusion pressure and is flashed to atmosphere pre-ferably in milli-seconds along the curved impinging tube 22.
This sudden release to atmosphere extrudes the lignocellulosic material in the plasticized condition at -the pressure pre-ferably in the region of 500 to 700 lbs per square inch and produces a particulate material having the appearance of pot~_iny soil which stains the fingers brown and has a high e~ough specific gravîty to sink like a stone in water.
~ hile the curved impinging tube is no-t essential it has the advantage of utilizing some of the extrusion force to further comminute the lignocellulosic material in addition to the comminution obtained by extrusion.
~ possible explanation of the present invention is that with the iignocellulosic material in the plasticized ~ondition ~he lignin, hemicellulose and cellulose are softened and thereby structurally weakened so that durin~ extrusion the softened material is subjected to severe mechanical shock which has the effect of fracturing the cross links between the lignin, the hèmicellulose and the cellulose. As the wea~ened and fractured lignocellulosic materi~l emerges from the die outlet 4 the instantaneous drop in pressure causes the wea~ened and fractured lignocellulosic material 0 to e~plode, thereby degradillg the hemicellulose t-o a low mo~ecular weight and at the same time causing substantial -de~radation of the main fibre structure, which in turn ex-poses substantial areas of the alphacellulose micro~ibrils7 for, in some uses of the present invention, attack by the fenrlentation a~ent.
The basic crysta:iline structure of the cellulose microfibrils is retained. A very high degree of disassoci-ation of the lignin is acllieved. When tlle process is car~
ried out at a temperature above the upper temperature of ~43C the xylose component decomposes, probably to furfural, and valuable digestible matter is lost from the substrate.
A1SOF decompos tion causes toxicity to microorganisms. One ~4-~G3 ~ 4 objective of this process is -to make the alphace~lulose highly accessible to the microorganism minimizing the ef-fect of Aydrol~sis on the hemicellulose, thus it is impor-tant that the time to achieve temperature should not ex-; ceea 1 mlnute and better s-till, should be in the ranse of 30 to 50 se~onds. Pressure is not critical except as it affects th, time to achieve the optimum temperature of 220-230C. he process can be repeated at least once more at the same or lcwer temperatures and will produce a further increment or increments in improvement in the digestibility.
Tests to verif~ the present invention were carried out using apparatus shown in Figure 1 having the following dimensions:
Inter,nal diameter of major portion of the pressure vessel 2 = 7~ .inches (190.5 mm~, Smallest internal diameter of bottle-necked por-tion of the pressure vessel 2 = 3~ inches t82.5 mm), Length of the interior of the pressure vessel 2 -45 lnches (1,1~ m), Bore of each steam inlet 10 and 12 = ~ inch (19 mm), Bore of steam inlet 11 = 2 inches (50,8 mm), The orifice of the die outlet was star-shaped with five equall~ spaced points giving a cross-sectional area of 0.8 square inches (11.6 square mm).
In the following Tahles I to V the temperature is the temperature of the material in the pressure vessel 2 when the closure plug 30 is removed, and the steam pressure is the steam pressure in the ~ressure vessel 2 at this mo-ment, The time is the time-in which temperature of the lig-nocellulosic material was raised to the temperature indicated and then extruded from the pressure vessel 2. The tables show the results of tests of-products of the present ~15-invention compared with tests carried out at temperatures that are greater than those according to the present inven-tion~ In the tests digestibility was evaluated in two ways, in vitro and ln vivo testiny with ruminant animal microflora, and secondly by treatment with a cellulase enzyme and the results are given in Tables I to V respectively.
In all of the tables the same lignocellulosic pro-duct testecl is desîgnated in the same manner, for example, in each test A designates the same lignocellulosic material and so on.

--:l6 a) a~
o o ~ o C~
a) ~q ~ cq .~ I U~ O U~ O Lt~

a) H ri ~ te 1~
~ .~
O ~ ~ CO ~ ~ O ~ CoU~ O
E~ ~ ~) 3 ~ ~D r- t` o ~ I` ~o ,, Z; ~ ~ .~
H E~ I¢ a 0\O
~; ~
~ Z ~ .~
H ~ a) ~1 ~1 O
E~~ c~ ~ O ~ co ~
~ ~ O ~ ~o ~ o .~J
E~ ~1 :~ ~ ~ Q rl GJ
o ~ E~ o ~L~
H W ~C . ~1 :qO E~ ~q ~q ~ ~ . ,~
E-l ~ ¢ ~ ~ ~) ~ N 1` r~
~!q 0 3 O I N N ~1 ~ `1 t'`JN ~1 1~1 ~ o o ô ô u) o o ô I
tn O e ~
H H E~ O O
~ .~
2 ~ r-l cq ~ ~ m c~ a w E~
Cw, ~q o ~ - ~
~ o Qq ~Lq o ~ 3 ~ ~ E~
'~ ~o ~n ~c ~63~
It will be seen from Table I that with material temperaLures above about 240C there is a drastic drop in the alp:~acellulose digestibility and as this represents approximately 65~ by weight of the total digestible carbo-hydrate present in the starting material it is necessary for this to be high.
Ana_ysis shows (see Figure 4 where sugar content, arbitrary units, is plotted against temperature in C) that the alphacellulose has not been destroyed by this treatment, therefore the loss in digestibility in vitro by animal rumen is caused by some extraneous factor.

~8~

. . , G:37~

CHEMICAL CHARACTERIZATION OF TREATE:D WOOD C~IIPS
The ~ain objective of the following study was to compare the effects of the process under different reaction condit ons. T~e experiments conducted were of an exploratory nature and the samples used were those obtained in tests A
to D, and a]so processed aspen and processed barley straw which were fed to sheep in a digestibility trial. These tests covered a broad range of reaction conditions. The results of the analysis are displayed in Tables II to IV.
In general, the results of the study will be seen to indicate that these samples can be classified into two categories:
lo Wood that has been treated according to the present invention at temperatures below 240C
(A, B and C).
2. Wood that has been treated at temperatures above 240 C (D and Wood as Fed).
The samples in the first category appear to be a better source for carbohydrate material than the second ca-tegorv because the high temperatures required to produce the second group have resulted in the extensive decomposi-tion of pentose sugars present in the wood.

~.i,3637i~

METI~ODS
-Sample ~reparation The samples (A, B, C, D and Wood as Fed) were dried al 10~ overnight to a cons-tant weight. Samples (10 a ) ~ere extracted with water using Soxhlet extractors for 18 hrs. The wood fibre was dried and the weight of extra^ted substance was determined. The water extract was diluted to 250 ml and analyses were performed on aliquots of this extract.
Analytical Methods Used on Samples Total carbohydrate was determined by the phenol-sulfuric acid colorimetric method (M. Dubois et al, Anal.
Chem. 28 (3) 350, (1956)); reducing sugar was determined by the Somogyi micro copper titrimetric method (Methods in Carbohydrate Chemistry, Vol. I, page 380, 1962, Acade-mic Press Inc.); and total sugars by gas chromatography were determined after hydrolysis using Tappi method T249 pm-75.

20~

~ ~J~63 ~7~

.~, .,~
-1 V .,, .
o~ o~ di o~ o\o u~ ~ o ~ I
~ ~, ~ . . C
0 ~ ~o O ~ O 00 ct~
N ~ ~ ~ ~ O
,' + + + ~ ~ ~
O
O ~-~1 ~
~1 3 ~) ,C

tn ~ o~o oWd~ o`o\ ~:
E~ ::~ >1 0 ~ 1`Cl~ O
~3 U~
'CS~ ~ ~ O
a o Z ~0 ~ ~ O R
C
~n ~Q
~ ~' ~ .~
O R :~
,_ O ~ O di~ d. 'CS ~
1:1 C.) o o ~ r o O ~) aJ h r~
H 3 ~1 .
I~,7 H _ ~: ~
H u~ ~ ¦ o o oo ~ _I O h W E:l X ~ r o ~ ~ 5 ~ ~ ~ ~ O ~.) i~ 1~ 0 ~ rC 0 X ~ ~) tn 0 ~\ o\ A~ d. A --I 3 h 1~ 5 ~ ~ ~O (~ ~r ~ 11~ ~J
a~ , ~ u~ ~ o t~ a)o ~ 1 0 tD
L ,. ~: O ~ r~ n ~ ~ ~5 ~ O o S~ ~ .~

~ ~ , F~ ~a Q
5;
~ I . .-, o o~ 7 ~,q O ~ C~ ~ 7 ~ L-~ Q
u~ E~ ~ ~ 0 -H
U~ ~- C) ~ ~1 0 ~7 ,0, S~ ~7 ~
~ S~ ~ ~ ~ ._ ~r7 X ~
z; a) ~ O O O L~ a) L' ~; ~ S~ . ~
a) ~ ~ n ~0 ~
rA ~ a) ' ~ a~ ~ o F:~~hr5 ~C ~ r .~ ~n ~ 0 o v Q ~ rc~ C., r~
~ ~ m u a U 8r~
~n p 1 ~1 1¢ * +

~--2].--~i~

~$~63~

~ABLE III
.
ANALYSIS OF HOT ~ATER EXTRACTAE~LES

SAMPLE TOTAL ~ BY WEIGHT ~ E~Y WEIGHT OF q'OTAL % BY ~IGHT
0~ CARBOHYDRA~2 1 REDUCING SUGAR 2 OF SUGARS BY GAS
_ _ _ _ CHROMATOGR~PHY 3 A 56.7% 27.7% 36.6~
B 47.5% 25.8% 40.4%
C 44.5% 27.4% 36.1%
D 24.2% 23.6~ 15.8%

Wood as 9.1% ~.o% 12.4 fed 1 Total carbohydrate determined by Phenol-Sulfuric acid colo-rimetric method using glucose as a reference sugar.
2 Reducing sugar determined by Somogyi Micro Copper Method.
3 Total sugar determined after hydrolysis by gas chromatogra-iry ~ y i ~-ih~d T2~9 e~--5 ~22 3~7~

o'~ d~ o`
O ~D O ~ U~

E~
o~ ~ d~ ~
o u~ u~ O ,~ ~ a~
~ ~ ~ ~ O 1~ D X

X U~
rL~

o o`P o~.o o~o o~
Z; E~ V Ln t~
~ H X r-l ~ l r~l O O
1_~ IY Z t~
a) ,¢ 1-1 ~ C~
Q ~, o ~ a) O~o ,~0 d~ d d~
. I 1:~ ~` ~D O O
O ~:: ~
E~ ~
1:1:l . r~ ~ 0,O ~,0 G~O 0,O 0~O 1 a) H ~ O a~ W ¦ o X ~ ,~ ,1 ~r ~
o ~ a) r~ o ~,~ 0,O 0,O ~,,0 0,O
:~ .~ o ~
Q ~ o I
~c m ~ ~ ~o ~n ~o a~

~3--~?9 6~7 DISCUSSION
The pH values of the samples ranged from 3.6 to .1. This range is considered normal for samples from wood substar.ce treated with high pressure steam (600 psi). Under the reaction condit:ions listed in the introduction the acetyl groups present in the wood would be expected to undergo hy~
drolysis releasing acetic acid which would decrease the pH
(i.e. increase the acidity). This acidity would assist in hydrolyzing the lignocellulosic material and increasing the water extractable material in the wood.
The processes are shown to have solubilized 18 to 34% of the wood substance (Table II). ~ood is norma~ly com-posed of 2 to 3% water solubles. The samples in categor~ 1 (A, B and C) have water extractables ransing from 29.0 to 34.7~ whereas samples in category 2 (D and Wood as Fed) have only 18.1 to 21~6%. The lower percentages of extrac-tables in category 2 have resulted from the decomposition of carbohydrate material (pentose sugars) because of the high reaction temperature.
For samples in category 1, the reducing swgar content of the extract is approximately half o~ the total Garbohydrate content (Table III). This indicates that the carbohydrate material present has an averag~ degree of polymerization of 2. For samples in category 2, the redu-cing sugar content of the extract is nearly equal to the total carbohydrate contentn This indicates that all the solubilized carbohydrate material has been converted to monomers.
The gas chromatographic analysis of sugars pre~
sen-t in the water extractables is tabulated in Table IV.
All samples tested have approximately the same quan~ity of hexose sugars ~mannose~ galac-tose, glucose) but samples D

-2~-lr~ 7~

and Wood as Fed have much lower pentose sugar contents (ara-binose and xylose) than the other samples which were treated at lower temperatures. These results indicate that a-t a tempera-ture between 238C and 246C there exists a transi-tion point (ranye) for the destruction of pentose sugars.
Above this ~emperature pen-tose sugars are destroyed by de-composition to furfural. Thus, wood fibre treated above this -transition temperature has a lower fraction of water soluble extractables than samples treated below this tran-sition point because the furfural produced by the decompo-sition of pentose suyar is ;'flashed off" during the explosion process. The decomposition of hexose sugars does not ap-pear to be a problem at the temperatures tested mainly because of their higher stability to thermal decomposition.
~estruction of pentose sugars above the transition temperature reduces the quantity of sugars available for di-gestion by approximately 10% based on treated wood weight.
Analysis of Lhe water soluble extractives shows that above the transition temperature only 2 to 3~ of fibre weight is composed of solubilized carbohydrate (e.g. D and Wood as Fed) bllt below the transition temperature 10 to 14% of fibre weight is composed of solubilized carbohydrate (e.y. ~' B
and C). This weight (sugar) loss was caused by the destruc-tion of xylose and arabinose. This loss of 10~ of wood weight carbohydrate material is supported by further stu-dies, ~ood as Fed (treated at 246C) was found to have a final reducing sugar value of 33.8% after diyestion for 24 hrs with Maxazyme (a commercial Trichoderma viride cellu-lase) whereas all other wood samples (treated below 238C) had reducing values between 42.8% and 47.4%. Sample D was not tested in this mannerO
Treated wood samples were hydrolyzed with 2N H2SO4 ~Z5-37~

at 100C to determine if a chemical method could be used to correlate digestibility with the rate of carbohydrate re-lease on hydrolysis. The results of this study are shown in Fig~-e 2 where the ~ by weight reducing sugar by weight of cry ~roduc. (RS) is plotted against the digestibility time in minu.es (TM).
In Figure 2 the graphs designate:
x - ~ is sample A in Table IV
~is sample B in Table IV
O ~O is sample C in Table IV
+ is sample D in Table IV, and o ~ is wood was fed (Aspen) in Table IV.
Samples, D and Wood as Fed, have reducing values slightly lower than the other three samples probably because the pentose sugars in these two samples have been decomposed and are not availab1e for ~drolysis. The difference in fi-nal reducing value (2 hrs~ and the rate of sugar release in-dicate that approximately the same amount of cellulose is exposed in the wood fibre of each sample (i.e. the accessi-bility of each fibre to acid is the same for all samples).
Since the ac-essibility of the cellulose is the same, it may be expected that the digestibility of the fibre will be the same but this ma,y not necessaril~r be true since the ap-proach of a cellulose enzyme to a cellulose fibril may be-much more restricted than an acid.
CONCLUSIONS
1. The treatment of wood chips with high pressure steam re-sults in hydrolysis of acetyl groups and degradation of hemicellulose present in wood, releasing acetic and uronic acid.
2. Treatment of hardwood chips by the extrusion process renders 18 to 34% by weight of the wood substance hot water extrac-table.
3. Solubilized carbohydrate material appears to exist in the form of monomers and dimers.
4I The Applicant's tempera-ture of approximately 240 C ap-pears tc, be the transition point, above which substantial decompo,ition of the pentose sugars takes place.
5. De:,truction of pentose sugars above this transition point ~educes the quantity of sugars available for di-gestion by approximately 10% of the treated wood weight.
6. The portion of "exposed" cellulose on the surface of the exploded fibre appears to be roughly the same in all samples tested.
An enzyme diqestibility characterization of treated wood chip~
The following Table V gives the results of measure-ments of enzymatic digestibility by Trichoderma viride cellu-lase Gf an aqueous suspension of the product.
!`

~7--~ri~6374 s ~ orl ~ .Dt~ ~ 0~ 0 r~ I t o ~ <
1:$ r~ rC~1~_tNCO~r) ~r ~ C~
c3 ~) Q)~ ) t~l ~) r~
~-.~ a .rl a~ C) rc 5 ~ r-l ~~' ~ l l H Q rl ~ U~
~ 5~
F~ ~ r~
a u~ ~,~o ~ ~ t~ t~ t~ ~ ~ o ~~ o o t ~ o ~ -U~ ti\ NO ~r~ D U-) U~ S~ r~ lr~
r aJ O
O
~; rC~ 51 U~
1~ O o Q) U~
~1 ~
o E~
H

t m~:
~ o U~ rl _ ~:1 t~ U7 ~ S-l ~ ~r 4J' ~ d' ~r ~r a a ~ ?~
E-~ r Q Q;
U~ al ~ O ~--I tlO r-¦ ~I r-¦ L~

1~ h ~) t'`l rl 1` ~ ~r t~l t` ~ ~ J
U~ h N t~ r-l tN t~l t~l r~l (~

O t~ o o u~ o o Ln ~ ~ ~ D rr~ D 5 h O ~ ~
O

E~ m ~ ~ #
u u~
~:: ~ ~ O 3 3 ~
R- 0 41 5-1 h h h ~q C~ 1~ ~ J) U~

A

37~

Enzymic digestion conditions: 1~ by volume of aqueous suspension Pulp consistency: 0.5% by volume of aqueous suspension pH of aqueous suspension: 5.0 Temperature of aqueous suspension: 30C
Incubation: continuous mixing Time: 24 hrs, (1 hr) Figure 3 is based on the results shown in Table V
and shows the $ by weight of dry product of reducing sub-stances (RS~) derived such as, for example, glucose, plotted against the temperature ~TEr~C) at which the lignocellulosic material was extruded.
In Figure 3 the ~raphs designate:
after digestion by digestion only x - before digestion Figure 4 is based on the results shown in Table V
and shows the sugar contents in arbitrary units (SC) after ~0 enz~ic digestion plotted against the temperature (TEMP C) at wnlch the li~nocellulosic material was extruded.
In Figure 4 the graphs designate:
- - hexoses - pentoses ~ ... .
; The Figures 3 and 4 show that the ~ield of hexoses remains unaffected by the temperature at which the ligno-cellulosic material is extruded ~hile the yield of pentoses starts dropping rapidly towards zero at temperatures yreater than 240C. These results compare very well with the results 3~ given in the previous tests.

In Table V, Test F was double processed at both temperatures shown and this further improved the digestibility.

~ ~J~)~i374 Tests have shown that a die having an orifice cross-sectional area of 0.8 square inches works well. The shape of the die orifice is not impor-tant but a star-shaped ori-fice ;~as been found to give a large boundary to cross-sectional area so that a large amount of mechanical deformation occurs with good fracturing of the cross-links between the lignin, the hemicellulose and the cellulose.
~ preferred tempera-ture range is 200 to 234C, particularly 234C.
The following morphological analyses were conduc-ted on treated aspen wood and barley straw:
~ Freeze-drying, to determine the moisture con tent of the samples while maintaining their structure.
- Optic microscopic observation of the freeze-dried products.
- Boiling water extraction for 2~ hours.
- Stereotypic diffraction by Y~rays of the origi-nal product, of the freeze-dried product, and of the extrac-tion ~esidue.
- Observation of the freeze-dried result by a scanning electrorl microscope.
lo Freeze-Drying SAMPLE MOISTURE CONTENT
Aspen Wood 48O8% by weight ~arley Straw 45~3% by wei~ht After freeze-drying, the dark brown colour of each of the two samples~brightens and reveals grayish reflections.
The strong and acidic odor is characteristic of carbonized sugars and carbohydrates.
2. Optic Microsco~ic Observations of Freeze-Dried_Samples The fibrillar structure of cellulose is recog-nizable, but the ibres are distinctly slashed. It appears i374 - that the fibres have been striated or split during treatment;
their characteristic fea-tures (lumen, punctuations) are still visible. To recognize the leafy nature of wood used, by the presense of wood vessels, is possible.
Extraction with H2O, 100 C, 24 hrs -Tllis extraction is done with the help of a soxhlet tube, under water reflux a-t 100C for 24 hours. In fact after 24 hours, the extraction appears total since the ex-tracted solution in the last cycles is limpid.
~spen Wood 30.0% extracted ~arley Straw 26.8~ extracted Note: The samplels characteris-tic odor has lessenèd in.the extracted residues, but the colour seems identical. The carbohydrates and the weak chains of lignin have passed into solution, but the residue contains much more lignin, insolu-ble in water at 100C. In brief:
Residue -- cellulose + iong chains of hemicellulose + lignin Extract = carbohydrates from hemicellulose hydrolysis + al-read~ carbonized sugars + a few traces of lignin The cold extracted solution is coloured, but also contains a large percentage of non-dissolved matter.

4~ X~Ray ~nalysis Cellulose is found in cr~stalline form in the crude products, the freeze-dried products and the extracted resi-dues. This confirms the presence of long cellulose chains (in the form of more or less slashed fibers after sustained treatments).
There was no significant difference between samples of processed ~spen Wood and Barley Straw.

5. Scannin~_Electron Microscope In Figures 5 to 8 there are shown electron micro-graphs of the free~e~dried products of Table V and they 37~1L

represent: Wood as Fed ~x 1,350), Wood as Fed (x 2,640), Straw as Fed (x 6~380) and Straw as Fed (x 2,760), respec-tively. These micrographs demons-~rate the high accessibi-lity ol the alphacellulose microfibrils, as well as the degree of lignin separation, demonstra-ted by the spheres of lignin sI~read throughout the material.
From the Figures 5 to 8 of each of the freeze-dried samp]es, it will be observed that lengthwise torn fiber fragments are visible. The spherical particles, probably lignin, have been expelled from the fibers and have been e~ected as small droplets, during processing.
Conclusions from the Morphological Analyses The greater accessibility of fibers due to "tears"
and cleavages of the secondar~ wall explains the increased enzymatic accessibilit~ which occurs in the processed mate-rial. In fact, the specific surface area of fibers distinct-~ ly increases (wa~l cleavage~
- This cleavage seems to be due to the bursting of the r.iddle lamella; the lignin is then expelled, and thereby ~0 disassociated from the walls.
On the other hand, at the mlcrofibrillar level, the cellulose crystallinity is preserved as deduced from X-ray diffraction anal~sis.

A further test was conducted to investigate the feasibility of leaching of lignin from treated aspen wood and barley straw using ethyl alcohol, and to measure the lignin y~elds of such extractions.
As shown in Figure 9, treated wet aspen wood (55%
by weight moisture), designated WAP, was equally divided into two portions~ Each portion was extracted three times with 95% ethanol at room temperature (methanol may also be ~2-P~'' ~ !3~3~7L~

used as well as o-ther related sol~ents). The mix-ture was stirred continuously in a plas-tic container with an air operated impeller. The extraction duration and volume of ethanol used were as follows:
1st Extraction: 12 liters, 0.5 hr 2nd ~xtraction: 12 liters, 2 hrs 3rd Extraction: 7 liters, 2 hrs The dark alcohol extract (fraction No. 1~, desig-nated Fl was in the form of wet alcohol solubles and was ~0 filtered, centrifuged in a centrifuge designated C, and then concentrated down to a thick black solution. Sedi-ments from the centrifuge bottles were washed with fresh ethanol and extracted with about 500 ml of benzene in a separatory funnel. A green fat (fraction No. 6), desig-nated F6, was obtained from the benzene layer and a gra~
pcwder (fraction No. 7), designated F7, was obtained from the water layer upon concentration.
About 3 liters of acetone was slowly added to the concentrated alcohol extract swirled in a flask. The mix-~0 ture was let stand overni~ht and then filtered to obtain a damp acetone insoluble precipitate (fraction No. 2), desig-natea F2. The acetone filtrate was concentrated down to about 1 liter and slowly added to 14 liters of water and stirred in a plastic container. The water insoluble pre-cipitate (fraction NoO 3~ r desi~nated F3, was filtered off - and extracted twice with a total of 5 liters of petroleum ~ther to yield a small wax (fraction No. 8), designated F8.
The water solution was evapora-ted to about 1.5 liters and extracted twice with 2 liters of ethyl acecate to obtain 3~ the ethyl acetate soluble fraction ~fraction No. 5).desig-nated F5,~ and the fraction remaining in water (fraction No.
4), designated F4.

h ~9~ 7~

~ Initially, the wet weig~lt of the wet aspen wood product (~AP) was 3,252 g and the oven dry weight was 1,463 ~.
The following Table VI gives comments on each frac.ic~ obtained from the process described with reference to Fig~re 9.
TABLE VI
_ . .,_ total Oven dry oven dry weight exploded Fraction _ wood Comments Pl 884.0 60.4 carbohydrate rich F2 94~5 6c5 carbohydrate rich 3 272.5 18O6 lignin rich F4 115.3 7.9 carbohydrate rich F5 81.3 5O6 lignin rich F6 14O~ 1.0 F7 0.6 0O04 , F8 0 9 0.06 _ _ ~ similar leaching procedure was Garried OUL for treated wet barley straw, designated WBSP in Figure 10.
In Figure 10, the same fractions are designated ~y the same references as those iIl Figure 9 and the previous description is generally relied upon to describe them and how they were obtained.
The initial wet weight of the wet barley product - was 173.4 g and the oven dry weight was 88.97 g.

The initial extraction was in two stages using 1000 ml of 95% ethanol in each stage.
The concen-trated alcohol extract of fraction No.
1 (Fl) was concentrated to almos-t dryness and acetone was added for fraction No~ 2 (F21~
~34-7~

The acetone filtrate of fraction No. 2 (F2) was concen.rated to about 120 ml precipita-ted and was added to 1.8 l~t~es of water.
The ethyl acetate soluble fraction No. 5 (F5) was extracted twice with ethyl acetate.
The following Table VII gives comments on each fraction obtained with the process described with reference to Figure ]0.
TABLE VII

,_ Weight ~
Oven dryof total we~ghtoven dry Fraction __ Comments F 55.76 65.9% carbohydrate rich F2 7 D 67 9.1~ carbohydrate rich F3 11.53 13~6% lignin rich (93%

~' 5~32 6O3% carbohydrate rich ! 4.32 5.1%. lignin rich hignin was probably present in very small amounts ~0 in fractions 1, 2 and 4. Infrared spectra of these frac-tions had absorption characteristics of carbohydrate rich materials. Fractions 3 and 4 were rich in li~nin. The Klason lignin contents of fraction 3 for barley straw was 92~9~ by weight, and for WAP was 93.1~. These fractions had the infrared absorption characteristics of native lig-nin. It may be concluded that a very large frac~ion of the lignin content of the exploded materials can be easily extracted with ethyl alcoholO Disassociation of the lignin from the cellulose and hemicellulose was probably caused by the method of depolymerizing lignin according to the present invention.
Since ethanol and acetone can be readily recovered ~35-~`i , 37 in this simple extraction, it provides a very efficient method to obtain lignin.
Lignin i5 disassociated from the lignin carbo-hydra-~e complex (LCC~ after the treatment of, for example, wood and s~r~w by the process. A further test was made to show tha-t most of this native li~nin can be readily extrac-ted with acetone.
In this tes-t, 70.5 g of treated, wet aspen wood (55. 4?6 moisture, oven dry weight 31.44 g) designated AW
was extracted three times with acetone in a 500 ml beaker.
About 250 ml of acetone was used in each extraction. The mixture was stirred with a glass rod for abou-t 5 min and then filtered. The third extract was clear and light in colour. The extracts were combined and concentrated to about 150 ml in a rotary evaporator. Thus acetone inso~u-bles (AI) were obtained and acetone solubles (AS~ in frac-tion l tFl)- Near the end of the evaporation care was ta~en to avoid overh~ating the solution which would cause the lig-nin suspension to coagulate and form a sticky mass. The con-centra-'ed extract was slowly poured into 1 litre of cold water (10C) and stirred in a container. The lignin water mixture was then filtered through a slow flow filter paper and dried at room temperature and 6.55 g of dry lignin rich powder of water insoluble precipitate (WIP) was obtained in fraction 2 (F2). This represented 20.8% of the total dry material leaving a fraction 3 (F3) remaining in water as water soluble (WS)O
Similar extraction was carried out with the treated barley straw designated BS. 97.0 g of wet exploded BS
(50.0~ moisture, 48.5 g oven dry weight) was extrac-ted three times with a total of 1050 ml. The extract was con-centrated down to about 200 ml and slowly poured into ~36-,~`

~7~ ~ 37 ~

lltre of cold water. 7.2 g of dry lignin rich powder was obtained, which was about 14.8~ of the total dry material.
The ratio of solvent volume to material weight was arbit~arily chosen so that the mixture was not too thick .o stir. A concentr~tion of about 12 g DM/100 ml liquid mixtllre was probably thin enough for stirring with a g'-ass rod, Since the extracted lignin was quite soluble in acetone and the leaching was fast, one would expect a very large stage efficiency. However~ the equilibrium data for the acetone extraction is needed to determine the requirement of solvent quantity and the number of stages. Because of the presence of water in the exploded mate~Ial, some low molecular weight hemicellulose was also leached out. Other much smaller amounts of wax, phenolics and extractives were also extracted. The wax;(in AW only) sta~ed with the lignin rich fraction and represented less than 1~ of the total material. Figure 11 shows the leaching sequences of AW and BS.
The following Table VIII gives the amounts of each fraction and comments upon them.

.1~

t-- ~
~ ~ .c ~ ~ s o ~
O h ~J h ,1 ~ ~: h ~ S~ .C h ~;1 S
G ~) ~ t) ~ ~ U
O ~ .5~ S~ I G ~C ~I
~ 5`1 Q G Q h ~ 5-1 G ~ h 0 ~1 S-l~ ~ S-l t~) S-l t ) 0~1 C) C~ ~I t) ~J
O
a d,d, o~ d~ di ~P
O ~ rl ~CO O ~
d~ G O o o a) ~ID
S ~0 .~ .
~ 3 H S
H
~ 'a) t ~ 3 Ou~ o ~_1 ~ ~ o ~I o Q ~ ~ ~ ~ ~ o ~: ~ o ~ ~r o E~ ~ ~ ~ ~

.
S:~

0 h h 1~ 4 h h r ~
~ ~ _ U~
m h E~ . . . _ _ S37 L~

Klason lignin determinations of the lignin rich fractions indicated that the lignin rich fraction contained about 93~ nin. Therefore a minimum of 90% lignin yield can be ~xpected from the leaching of wood and straw, pro-cess~d ~lccording to the present invention, with acetone.
The solvent can be readily recovered with little heat re-quiredr which makes the process even more attractive. Ano-the~ advant:age is tha-t the material which has been extracted with acetone is almost free of lignin, therefore its value as a fermentation substrate or feed stock is increased.
Figure 12 shows a simplified schematic diagram of the ace-tone extraction of lignin from, for example, treated wood and straw.
In Flgure 12, treated material M is fed to ex-tractors ES and fresh acetone (ethanol or methanol~ FA is fed to a tank T. Acetone from the tank T is continuously fed to the extractors ES and overflows therefrom to a first evaporator EV1 while an underflow from the extractors ES flows to a first filter F1. The filtrate from the first filter F1 is passed to the first evaporator EV1 while the residue fror. the first filter F1 is dried in a drier D to yield a carbohydrate rich solid Sl. ~ecovered acetone from the first evaporator ~Vl and the drier D is returned to the tank T. The residue from the first evaporator EVl is mixed with water W in a precipitator P. The mixture from the precipitator P is passed to a second filter F3 from which filtered li~nin FL of >90~ purity is obtained.
Residue from the second filter E'2 is passed to a second evaporator EV2 where a carbohydrate rich solution S2 is ob-tained while the evaporated products therefrom are passed to a distillation unit DU from which water ~2 is separated from acetone and the ace-tone returned to the tank T.

-39~
.~

7~

The lignin produced in accordance with the present invention by solvent extraction from the treated material can be ~hermall~ plasticized and extruded as lignin fila-ments, and the lignin filamen~s can be carbonized in -the preser,c~ of heat to form carbon fibers. Carbon filaments produced in this manner may be used in, for example, air or water filtration.

~ 0~

Claims (15)

1. A method of rendering lignin separable from cellulose and hemicellulose in lignocellulosic material, comprising:
a) packing the lignocellulosic material in a divided, exposed, moist form in a pressure vessel having a closed extrusion die outlet, b) rapildy filling the pressure vessel with steam at a pressure of at least 500 psi to, by means of the pressurized steam, substantially, all of the ligno-cellulosic material to a temperature in the range 185 to 240°C in less than 60 seconds to thermally soften the lignocellulosic material into an extrudable condition, and c) as soon as the said extrudable condition has been attained, opening the extrusion die outlet and instantly extruding the lignocellulosic material, in the said extrudable condition, from the pressure vessel through the extrusion die outlet to atmosphere so that the said material issues from the extrusion die in particulate form with lignin therein rendered into particles substantially in the range 1 to 10 microns and separable from the cellulose and hemicellulose, the particulate lignin and cellulose being together in disassociated form having the appearance of potting soil, a major portion of the lignin being soluble in methanol or ethanol and being thermoplastic, the cellulose being in the form of crystalline alpha cellulose microfibrils and suitable for digestion by microorganisms and enzymes.

2. A method according to claim 1, wherein the ligno-cellulosic material is extruded on to an impinging surface.

3. A method according to claim 1, wherein the pres-sure vessel is rapidly filled with steam to bring the ligno-CLAIMS (Cont.) cellulosic material to a temperature in the range 200 to 238°C in less than 45 seconds.

4. A method according to claim 1, wherein the pres-sure vessel is rapidly filled with steam to bring the ligno-cellulosic material to a temperature of 234°C in less than 45 seconds.

5. A method according to claim 1, wherein the steps a) to c) are repeated at least once more with the ligno-cellulosic material at a temperature no greater than that at which the lignocellulosic was treated for the first time in steps b) and c).

6. A method according to claim 1, wherein the extruding of the lignocellulosic material to atmosphere is accomplished in milli-seconds.

7. A method according to claim 1, wherein lignin in native form, having a purity of the order of 93 weight %, is solvent extracted from the particulate material by a solvent selected from the group consisting of ethanol and methanol and the lignin so extracted is readily reactive chemically, is thermoplastic, and has an infrared spectrum approaching that of so called native lignin.

8. A method according to claim 1, wherein the ligno-cellulosic material is wood and the pressure vessel is rap-idly filled with steam at a pressure in the range 600 to 700 psi.

9. A method according to claim 1, wherein the ligno-cellulosic material is hardwood.

10. A method according to claim 1, wherein the ligno-cellulosic material is annual plant material and the pres-sure vessel is rapidly filled with steam at a pressure in the range 500 to 600 psi.

CLAIMS (Cont.)

11. A method according to claim 1, wherein the ligno-cellulosic material is an annual biomass and the pressure vessel rapidly filled with steam at a pressure in the region 550 psi.

12. A method according to claim 18, wherein the ligno-cellulosic material in finely divided form is moist with water in the region of fibre saturation when packed in the pressure vessel.

13. A method according to claim 1, wherein the ligno-cellulosic material is an annual plant substance, the pres-sure vessel is rapidly filled with steam at a pressure in the region of 550 psi, after approximately 40 seconds the steam pressure is rapidly increased to a higher pressure and the material is released for extrusion immediately upon achieving said higher pressures.

14. A method according to claim 1, wherein the ligno-cellulosic material is wood, the pressure vessel is rapidly filled with steam at a pressure in the region of 650 psi, after approximately 40 seconds the steam pressure is rapidly increased to a higher pressure and the material is released for extrusion immediatetly upon achieving said higher pressures.

15. The product in particulate form when produced by the method claimed in claim 1.