Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
This invention is in a new process for 'creating ligno-
cellulose with a solvent mixture comprised of water, methyl
alcohol, and a dissolved metal salt catalyst in a pressure vessel
at a temperature in the range 180C to 2~C to produce a chemical
pulp of fibrous material.
The process is effective to depolymerize and dissolve the
greater part of the lignin and at least part of the hemicellulose
materials of cell walls, and is particularly successful in producing
pulps having high fiber strength in which cellulose retains nearly
its natural, undegraded state. Such pulps are suitable for making
high quality paper, and for dissolving purposes.
Heretofore, in processes wherein wood is subjected to
hydrolysis by an acid catalyst in a solvent mixture made up of
water and an alcohol having two or more carbon atoms in a straight
chain at a temperature between 165C and 210C, the cellulose is
attacked rapidly, so that before the lignin and hemicellulose
constituents of cell walls are dissolved, extensive cellulose degr-
adation has occurred. Even when a buffered strong acid or relat-
ively weak dicarboxylic organic acid is used as catalyst to promote
hydrolytic solvolysis, the fibers when liberated have a degree of
polymerization reduced significantly below their natural condition,
so that paper sheet made from such pulps lacks the high tear
strength, burst strength, and breaking length properties desirable
for industrial uses.
While known processes have advantages of requiring very
short cooking time and yield a soluble lignin and dissolved sugars
having considerable value, it has remained a desirable objective
that liberation of cellulose fiber should preserve almost the
native strength of cellulose fibrils when the lignin content of
the pulp has attained acceptably low values. Another desirable
objective is that the process should recover virtually all of the
catalyst substance employed without losses. A further d~esirable
3~3~
objective is to avoid damage to polyglucan through attack by acids
and by water-miscible volatile organic solvents such as ethanol
and acetone, which have been discovered to rapidl~ degrade
crystalline cellulose at temperatures above 150C particularly
at the disordered regions.
The present invention overcomes the major deficiencies of
known chemical pulping processes employing acid-catalysed aqueous
solvent mixtures, and essentially consists in the method for
converting lignocellulose such as softwood, hardwood, or agricult-
ural residue to the form of separated fibers by cooking the ligno-
cellulose with a solvent mixture at least four times the weight of
material to be pulped containing water and methanol in the propor~
tion of 1 to 1 to 1 to 4, and containing from 0.005 to 1.0 moles
of a metal salt or a mixture of metal salts selected from the
chlorides and the nitrates of anv of the metals magnesium, calcium
and barium, at 180C to 210C for a time generally between 30
minutes and about 90 minutes.
The method of the invention is particularly effective when
the salt is magnesium or calcium chloride or nitrate at a molar
concentration between 0.02 molar to 0.05 molar and the ratio, by
weight, of water to methanol of the solvent mixture is about 3:7,
and the cooking is carried out at a temperature between 190 and 200C
When Spruce is cooked by the novel process a pulp is obtained which
retains appreciable amounts of hemicelluloses, yet has low residual
lignin content and a degree of polymerization (DP) comparable to
that of a pulp obtained bv the kraft process; the cooking time need
be only from 30 to 45 minutes to yield a pulp with Kappa number 63,
~3~5
TAPPL viscosity (0.5) of 28.2 centipoises, or a degree
of polymerization of 1550. Such pulp is far more easily
bleached from its as-cooked brightness of 45 to ~0~ to the
desired 80~ or higher brightness than is a kraft spruce
pulp of comparable residual lignin content with a much lower
starting brightness typically 35%.
According to the present invention, therefore,
there is provided a method for chemically converting
lignocellulose to the form of separated fibers which
comprises cooking lignocellulose with an aqueous solvent
mixture of at least four times the weight of the
lignocellulose r the mixture comprising from ]. to 4 volumes
of-methanol per volume of water and a metal salt catalyst
dissolved therein in a concentration between 0.005 molar to
1 molar, the salt being selected from the group consisting
of the chlorides and nitrates of magnesium, calcium and
barium, the cooking temperature bein~ between about 180C
and 210C and the cooking time being not longer than two
hours and sufficient to effect at least partial depolymeriza-
tion and dissolution of the lignin, the hemicellulose and the
other cell wall materials encasing the cellulose, and
recovering the separated fibers from the liquor residue.
- 2a -
csm/~
'~
~3~t~
The invention will be more particularly described
in and by the following stlt~men-ls of preferred modes of
carrying it into effect, together with tables and examples
of the method.
Wood-To-Liquor Ratio
The method preferably employs an amount of solvent
mixture which cannot be less than the void voiume of wood
fra~ments and must be sDfficient to provide solvent action for
li~nin and carbohydrates during hydrolysis, and hence should
be at least four times the wood weight or a greater weight
where effective convective or forced circulation of liquor
is essential. It has been found that the cooking action is
fully effective if the solvent mixture is in the range
between about 5 times the weight of wood and about 30 times
the weight of the wood, and that cooking with even greater
proportions, for example 50 to 100 times, delays the
separation of fibers to the extent that such proportions are
economically undesirable. Efficient cooking ;s realised with
wood: liquor ratios from about 1:5 to about 1:10 as may be
seen from Table I following.
T A B L E
PULP PROPERTIES, SPRUOE WOOD COOKED WITH AQUEOUS ~rH~NOL
__ AND CaC12 USING DIFFERENT WOOD/LIQUOl~ PROPORTIONS
CaC12 H20~ethanol Wood/ Temp. Time Yield Kappa TAPPI (0 5)
Molarity Liquor C min Wt. % No. Visc. cp
. __ . _ . _ _ ._ __ _._ - _ A . . . __
0.053 : 7 1 : 10 205 60 45.4 19 9
0.053 : 7 1 : 100 205 60 54.0 65 22
0.103 : 7 1 : 10 200 30 59.0 61 19
0.103 : 7 1 : 100 200 30 No fiber separation
.
~7 -- 3 -
csm/
s
Ratio of Water to Methanol
Even though methanol is known to be a poor solvent
for natural lignin or lignin polymer groups resulting from cleaving
of natural lignin by hydrolytic action, when a solvent mixture
is formulated with a water/methanol ratio between 1 : 1 and
about 1 : 4 and the wood/liquor ratio is about 1 : 10, at the
elevated temperatures of the process, near 200C, satisfactory
dissolution of depolymerized lignin as well as of sugars formed
by hydrolysis of hemicellulose is obtained.
The effectiveness of the process in providing high
yields of separated fiber from Spruce which is difficult to
delignify by aqueous organic solvent mixtures is enhanced by
employing the higher proportions of methanol, with correspondingly
higher TAPPI (0.5) viscosities but with correspondingly higher
retained lignin, as shown by Table II.
T A B L E I I
PULP PROPERTIES: SPRUCE WOOD COOKED 30 MIN. METHANOL/WATER
CONTAINING 0.16 Molar CONCENTRATION OF CaC12
COOKING LIQUOR COMPOSITION PULP YIELD KAPPA No. TAPPI (0.5)
Water vol. Methanol vol. Wt. Percent lignin Visc. cp
1 1.5 51 44 14.0
1 2.33 53 55 20.5
1 4.0 58 63 18.0
.
From the foregoing it will be seen that at the constant
salt concentration of 0.16 molar calcium chloride, a proportion
of 30 volumes water to 70 volumes methanol yields a pulp containing
the highest proportion of high polymer weight cellulose with an
excellent yield of separated fiber and at an acceptably low
lignin content.
The pulping effectiveness of the novel process and the
superiority of the pulp obtained may be contrasted with prior
observations on use of aqueo~s~methanol mixtures; see "The Cooking
Process", S. I. Aronovsky and R. A. Gortner, Industrial and
Engineering Chemistry 28,(1936) pp. 1270--76.
Metal Salt Selection and Concentration
The process exhibits a high tolerance to large
variation in the molar concentration of the metal salt used, at
constant cooking conditions. A hardwood such as Aspen may be cooked
to a fully separated fiber state using a concentration of calcium
chloride as low as 0.005 molar, in aqueous methanol at 1:10 wood/
liquor ratio, in a time well below one hour, while at a concentr-
ation as high as 1. molar a fully separated fiber state is
reached in about 15 minutes. Hardwoods generally may be cooked
with lower salt concentrations of any of the preferred catalysts
than softwoods require; for example, Aspen cooks with calcium or
magnesium chloride or nitrate less than 30 minutes yield fully
cooked fragments at salt concentrations 0.05 molar to 0.10 molar.
Softwoods such as Spruce will generally require a higher salt
concentration between 0.05 molar to 0.20 molar, and in certain
instances improved fiber separa~ion may be gained by a concentration
approaching 0.5 molar or higher~ and by addition of HCl.
The chlorides and nitrates of calcium, magnesium and
barium have been found to be effective when used at the concentra-
tions mentioned, for a wide range of woody plant materials. The
preferred salts are the chlorides of magnesium and calcium, whichhave the advantages of lowest cost, and being tolerated well in
fermentation processes to which the liquor residue may be fed to
convert oligosaccharides dissolved in the liquor following recovery
of the pulp and removal of lignin and methanol. Nevertheless
the nitrates are effective catalysts, although their salts
represent a higher cost. Table III shows representative data
on pulp properties for one hardwood, Aspen, and one softwood,
Spruce, when cooked with aqueous methanol at 200C for various
metal salts, with a constant wood/liquor ratio of 1 : 10.
1~3L41 5
TABLE III
PULP PROPERTIES OF WOOD SPECIES COOKED WITH METHANO~-H2O AT 200C
USING SELECTED METAL SALTS AND CONCENTR~TIONS.
CATALYST COOKING PULP LIGNIN TAPPI
WOOD SALT MOLFS TIME, YIELD,KAPPA (0.5) VISC. D P
min % NUMBER cp
~lgC12 0.01 30 62 27 20 1320
Q CaC12 0.01 30 63 30 21 1360
o
CaC12 0.10 25 61 16 18.7 1280
z;
~ MgC12 0.10 25 59 15 19 1300
BaC12 0.01 30 69 46 Poor fibre separation
MgC12 0.05 30 59 51 17 1200
MgC12 0.10 30 54 29 18 1270
a
g CaC12 0.05 30 66 60 20 1320
-
CaCl 0.10 30 54 35 19 1300
2 ~
Mg(N~3)~0.10 45 57 55 23 1410
P~ -
~ Ca(NO3~0.10 45 58 62 29 1570
All wood:liquor ratios 1:10.
In the foregoing the degree of polymerization (DP) values are
derived from TAPPI Standard viscosity measurements and use of
nomogram published by Rydholm, p. 1120. Excellent fibre separations
were obtained by use of both chlorides and nitc~t~s of magnesium
and calcium with low residual lignin and reten~io~ o~ significant
amounts of hemicelluloses.
EFFECTS OF DURATION OF COOKING
.
Da~/as ~re presented for a set of pulps -cooked t~ith liquor
~of water:me~an-olxatlo~ 3:7 and wood liquor ratio of 1:10
using 0.01 molar CaC12 as catalyst. Both spruce and aspen woods
were cooked for extended durations. The residual DP o(f the pulps
is observed to decrease with lengthened digestion times, however
excellent fibre separations were observed and relatively good
~3~
TAPPI (o.5) viscosi-ty values and DP values are retained at low
residual lignin Kappa number values, Eollowing 35 minutes digestion
of the softwood and after about 20 minutes digestion of the
hardwood, as shown in Table lV.
T A B L E I V
PULP PROPERTIES FOR WOOD COOKS IN H20-MET~ANOL 3:7, 200 C
WOOD/LIQUOR RATIO 1:10, SALT 0.10 Molar CaC12
Duration of Pulp yield, KAPPA TAPPI (0.5) p
SpeCles cook, Min. Weight % Number visc., cp D
72 103 (no fiber separation)
62 63 28.2 1550
SPRUCE 35 56 46 18.2 1275
52 42 15.3 1160
99 (no fiber separation)
73 61 25.0 1450
ASPEN 20 63 22 20.5 1340
61 16 18.7 1280
Effect on Fiber Separation of Temperature, and Cooking Range
Softwoods when cooked by the process of the invention
exhibit a doùbling of the reaction rate, as expressed by the
duration of cooking required to attain a specified residual
lignin content, generally below KAPPA number 45, for each 10 C
rise in temperature within the effective range of temperature
bounded by an upper scorching temperature limit of about 210C
and a lower delignification threshold at about 180C, at which
the digestion time interval does not exceed about 90 minutes.
As may be understood by reference to Table I and the
following Table V, cellulose damage increases for a hardwood
such as Aspen as temperature increases above 18G, which is a
lower temperature limit at which delignification and fiber separ-
ation is assured within 40 minutes. An upper digestion tempera-
ture of 210set by onset of scorching need never be reached, as
the degradation of cellulose for practical cooking durations
-- 7 --
~IL3~
of about 25 minutes above 190C will dictate usual].y not exceeding
the latter temperature.
Table V below reports analyses of Spruce and Aspen
pulps cooked with 0.1 molar CaC12 at different temperatures and
for varying durations, in aqueous methanol of 3:7 volume ratio
with wood/liquor ratio 1 : 10.
T A B L E V
PULP PROPERTIES FOR WOOD COOKS IN H20-METHANOL 3:7
WOOD/LIQUOR RATIO 1 : 10, SALT 0.10 Molar CaCl
Species Duration of OTemp. Pulp Yield, KAPPA TAPPI (0,5)
Cook, Min. C Weight % Number Visc. cp DP
180 63 48 41 1750
ASPEN 40 190 57 9.4 21 1360
200 53 6.0 12.5 1050
190 63 61 28 1500
SPRUCE 60 190 59 53 26.4 1470
190 56 40 23.2 1410
200 59 6~ 19.0 1300
Because each lignocellulosic material represents a different
composition and character of its lignin, hemicellulose, and other
cell wall constituents, the practice of the invention will
necessarily require some experimentation with a given material
to obtain the optimum pu.lp properties. Some guida~ce may be
obtained by reference to Table VI which records pulp analyses
of cooking conditions for seven different species which yielded
high quality pulps typical of good results, althou~h it is to
be understood that these conditions are illustrative of good
practice rather than optimal conditions. The Table includes
handsheet strength properties of interest after the pulps were
beated to 300 Csf, sheet testing being carried out according to
prescribed TAPPI standards. The pulps were treated only by
a hot aqueous methanol wash to remove precipitated lignin, and
further washed and suspended .i.n water hefore beating and handsheet
formation.
8 --
~.~3~
o o o o o
, ~ o o ". o ",
o P~ o ~ o ~ ~
~1 U~'~D Ll ) O CO a~
1~1 ~ ~1 ,_1t`l ~1 ~1
U~
h L~ o ~`I`1
~ ~9L~) Lr)
am
ms~
U~
u~ a
C) E~ r~
P; O
o O ~
-rl ~ O O O O O
U~ Z ~ ~ O O O o O
~ ~ ~ ~ CO ~ ~ r_
H E~ a) ~
E~ ~ S~ a)oO oO ~ o o
~; ~q ~ ~1
~ m
o ~ ~ O O O O O
a) ~ o o1-- o o
C~CO O
$ ~ ~
E~ O ~ ~; ~ ~ ~ ~r
~ ~ m
p: ~ U~
m
u~ m
Z E-~ ~~)Lt') 00~ 0 0
~ O ~ ~ ~ O t-- CO L~)
m z P~ ~ ~
h
~1 OOLl')O O O O
~D~n ~1 ~ d
~)
~J~1 ~1~1 ~1~1 ~1
t` I Lr)
H ~ ~ .
O V
_ Lr)Lt~ L~
~1 H ~ ~ ~ ~ ~ ~ O
o ~ P~ c~
Z ~ P~ u~
.
E~ ~
a) ~r o
h ~R. .
m o E-~ ~, ~cn (~l~r Oli~ O Ll')
~ ~ ~ ~ ~ ~ ~ ~7
m ~ ~ Z
ZH ~H ~
E~ ~ ~ o\o
o ~ .~
O ~ ~ ~ 1-- co,~ ~ ~ ~ ~
c~ ~ ~' Lr~ LnLr,,,~ ~ ," ",
P~ ~
oo ~
~: o o ,~ o o o o
P; O ~~ ~ ~ ~r
0 H E~ ~
~ .
u~ ~
H P~ 0 0 O O O O O O O
~ ~ E~~
01 ~.
~ H ~ u~ o Lf~ OC:~Lt~ O
O i l ~) ~~1 ~-~ 'P O ~1
P~; (~J O . . .~ . . .
t~O O o O O O O
~ 0 ~ .'
(D ~
Pulping With ~queous Methanol and ~etal Salt, ~cidified b~ HC1.
It is known that for rapid bulk delignification in acid-
catalysed aqueous organic solvent mixtures large initial proton
concentrations are required. ~ further improvement of the present
invention for rapid wood delignification and pulping is obtained
by mild acidification of the solvent mixture containing the metal
salt catalyst earlier described. The addition of hydrochloric acid
in concentrations of 0.0005 to 0.006 Normal to methanol-wa-ter
solutions containing 0.05 M CaC12 was found to improve the rate of
delignification significantly without significantly affecting the
cellulose, by the time that fiber-separation of the wood was evident.
The acid enhances the initial bulk delignification and
hydrolysis of hemicelluloses by providing high initial proton
concentrations at a time when the cellulose is still well protected
from the hydrogen ions by the encrusted lignin since in its native
structure cellulose is less aceessible to protons when swelling
by solvents is restricted by the lignin-hemicellulose matrix.
The functions and effects of the metal salt catalysts are
essentially unchanged from their roles as described hereinbefore,
the metal salt catalyst serving both as a proton-generating agent
as well as providing protection to the cellulose especially at the
later stages of the cooking against degradation by hydrolytic
solvolysis.
A study of the effects of metal salt catalyst alone, and
when combined with very minor concentrations of hydrochloric acid,
was made by a series of six cooking operations using a laboratory
scale pressure vessel measuring about 20 cm by 8 cm diameter, of
stainless steel. Air-dry wood chips containing about 8% moisture
were made up into 10.0 gram batches, the wood being Spruce, not
30 previously subjected to any treatment nor elevated temperatures~
The cooking liquor was prepared using 3 volumes of distilled water
and 7 volumes of methanol, 100 grams of the solution being loaded
~- - L 0
into the vessel at room temperature, the initial batch ~nd all
subsequent batches having added thereto an amount of calcium
chloride (CaC12) as a solution in the water component used as
cooking liquor, to provide a concentration in the li~uor 0.05 Molar.
The first batch was cooked by heating the sealed vessel
in a pre-heated oil bath held at 200C. It was noted that at the
end of 7 minutes, the vessel temperature had reached 200C. The
cooking time, counted from immersion time in the bath, was extended
to 40 minutes. The vessel was then chilled by removing from the
bath and running cold water over it, for one minute, when the vessel
was opened. The liquor was decanted, and the pulp removed. The
removed pulp was washed with a hot aqueous methanol containing no
additive, and drained. The washed pulp was mechanically disintegra-
ted to make a dispersion in water about 4% by weight.
The cooled residue with precipitated lignin was dumped into
an equal volume of cold water, whereupon about 70% of the dissolved
lignin removed fro~ the wood was available as a precipitate. The
liquor was filtered and evaporated to further precipitate lignin and
yield a syrup containing the calcium chloride salt. The washed pulp
was analyzed by atomic absorption techniques, and residual cation
was observed to be insignificant as compared to the cation originally
present in the woodO
Five further batches were cooked each including concentra-
tions of HCl in addition to CaC12, for varying cooking times. Data
for the solvent mixtures and pulp analysis are shown in Table VII,
showing high DP values and TAPPI viscosities at low Kappa numbers.
T A B L E V I I
Conc. Conc. Solution Cooking Pulp Kappa TAPPI ( ' 5)DP
HClCaCl2 pH Time Yield Number Visc. cp
NM before after mln Wt. % _ _
-- 0.05 5.8 3.6 40 54 44 20 1320
0.001 0.05 2.8 4.1 40 56 58 30 1600
0.002 0.05 2.6 4.0 40 54 48 21 1360
0.002 0.05 2.6 3.9 35 58 57 24 1440
0.004 0.05 2.5 3.7 40 53 37 23 1420
0.004 0.05 2.5 3.9 ~5 52 30 19 1300
Material: SPRUCE WOOD chips 11 -
THE COOKING PROCES_
The process must be carried out in a pressure vessel
capable of withstanding internal pressure of the order of 30
atmospheres corresponding to the vapor pressure of the lowest-
boiling component, methanol, at a maximum tempera-ture at least
205C, whether stationary batch type of cooking or continuous
cooking is practiced.
The material of which the vessel is made requires to
be non-corrodible under process condi-tions, in which the liquor
pH is observed to range from an initial value of about pH 4.25
after the solvent mixture has impregnated the charge of wood chips,
to a slightly smaller pH at the end of the cooking.interval, which
increase in acidity is ascribable to the generation of small
amounts of organic acids by conversion of lignocellulosic
components at the elevated temperature conditions, and by hydrol-
ytic action.
The weakly acidic character of the liquor may be
understood to arise when wood is impregnated by the solvent mixture
by cation exchange phenomena, which, per se, are well kno~n:~rom
the literature; see for example, H. Masters, "Reactions of
Cellulose with Sodium Chloride and other Neutral Salt Solutions"
p. 2032, J. Am. Soc. 121, 1922 wherein reference is made -to observed
acidic effects.with solutions of barium chloride and calcium
chloride, ordinarily neutral to litmus paper in aqueous solution.
The formation of a metal complex by the cation formed upon dissoc-
ation of the dissolved metal salt and subsequent exchange with
carboxylic groups present, as depicted by the following relation,
releases two protons for each bivalent metal ion which participates
in the ion exchange:
LIGNOCELLULOSE _ COOH + M =======LIGNOcELLuLOsEcooH-M
-~ 2(H3O)
The hemicellulose and lignin constituents which carry a multiplicity
-.12 -
of funetional groups may be presumed to be the chief sources
of proton formation by such complexing. Table VIII presents
data recording pH measurements made with several effective
eatalyst salts according to the invention at several wood/liquor
proportions, illustrating aeidifieation of the solvent mixture
upon steeping several speeies with water, with aqueous salt
solutions, and with cooking liquor.
The eooking proeess, reeovery of the pulp fibers and
of dissolved products, and recovery of methanol and of metal salt
are equally successful and feasible when a minor amount of the
strong acid catalyst is ineluded in the cooking liquor. The use
of the auxiliary eatalyst, sueh as HCl, is particularly advanta-
geous when cooking certain Gymnosperms ~ notably Spruce wood.
No modification of the cooking parameters or of the liquor need
be made when using the very low proportions represented by 0.0005 N
to 0.008 N HCl, i.e. not over 0,30 gm HC1 per dm of solvent
mixture.
- 13 -
~:q ~ â
o O--. l l l . l l l l .
5 o m u. ~r ~
:q~Z ~r co ~D
o O ~ . l l l . l l l l .
~50~ ~ ~ ~
m u, z ~ ~ er
o o ~ . l l l . l l l l .
~OU O ~ ~
o o u . l l l . l l l l .
Z 5~o~
H O O O U 10 l l I ~D I I l I N I
H H ~ O :~ In ~r ~
D O ~ ~1 ~ ~ r` co ~ ~7 o ~1 ~ ~ In
~Uu u~ ~r ~r ~r ~ ~r ~:r In ~t' ~r ~r ~r ~r
U~ ~
::C
OO U~ CO ~ o~ o~
O ~ . . . l . l l l l .
~:X ~ U~ Lr~ Lr) U'l
L~ ~ ~D ~ r~
O O U . . . l . l l l l .
~oc,) u~ ~r ~r ~r ~r
O O I~ ~D
~ . . . l . l l l l . l l l
..1 )o$ ~ ~1 O O O O Oo O O O O
~ ~ 1 l .. -1 Ll') ~1 ~I 10 ~1 11~ ~1 ~ Il') ~1
0~ l 1~ r-i r-i ~1 ~i ~ ~i ,_i ~1 ~1 ~1 ~i
~o ~ a h
o 8 o ~
z u~ l ~ o~ ~ ~ o
t~ r¢ X ~ C) (~
1~ K ~ H ~4 F~
-- 14 --
~3~
The insensitivity of any of the pre~erred metal salts
herein disclosed to temperatures of 200C or hiyher allows the
process to be carried out with heat exchange apparatus arranyed
to heat the vessel or the solvent mixture which is circulated,
throughout the duration of the cooking interval. The cooking
vessel may be of any form suitable for single stage or dual or
multi-stage processes whether by batching or in a continuous
manner, as desired for the recovery of specific byproducts.
For recovery of oligosaccharides formed by hydrolysis
of hemicelluloses with least heat losses by degradation, means
must be provided to drain and exchange part of the cooking liquor
throughout the cooking interval. Lignin of low molecular weight
state is not significantly affected by the process conditions
since the cook durations are short and acid concentration is too
low to effect recondensation. Due to the successive hydrolyses
and a degree of alcoholysis which may temporarily alter the
solubility of lignin macromolecules, suitably depolymerized
molecular aggregates as recovered from residual liquors after
30 minute cooks have measured molecular weights in the range
from about 320 to 12,000 with an average weight about 3600.
Delignification may be arranged to take place in
digesters, vessels, or extraction towers commonly used in
pulping of wood. In any case it will be desirable to remove
from the digester such amounts of liquor and at predetermined
rates as will allow removal of methanol, lignin and oligosacch-
arides by steam stripping, flash evaporation, filtration/
centrifuging, thickening, crystallization and replenishment of
water, methanol, and metal salt.
The removal of pulp from the pressure vessel in a
batch operation may be arranged to initially drain the hot
liquor from the pulp and thereby lowering the vessel pressure,
15 -
~3~
and washing the pulp with a hot wash liquid which is ac~ueous
methanol, followed by rinsing with cold water. The hot liquor
and the wash liquid may be steam-stripped to recover methanol,
and to precipitate lignin.
The nature of the hydrolysis reactions during the
cooking operation may be deduced from an analysis of the spectrum
of oligomeric and monomeric sugars present in the residual
liquor. Such analysis reveals that the liquor is directly
suitable, or after a mild secondary hydrolysis, for fermentation
and recovery of alcohol. The origin oE the major part of any
glucose in solution can be traced to hemicellulose, indicating
that degradation of cellulose structure is minor, as shown in
Table ~X. The data presented shows, as a progression, the
gross analysis of Aspen and Spruce wood constituents; the
distribution of constituents folowing the cooking; and a
comparison of carbohydrate analyses Eor the whole wood and the
pulp and the cooking liquor.
T A B L E IX
_
C 0 ~5 P O S I T I O N A N A L Y S I S
Species Holocellulose lignin extract-
Weight % Wt % ables
ASPEN WOOD 77.4 19.7 2.9
SPRUCE WOOD 72.3 _26.5 1.2
P O S T - C O O K I N G C O M P O S I T I O N
CookingPulp Yield ~PPI Dissolved Sugars
AspEN PULP T2i5me, Min 6~ 2.1 V sc;
LIQUOR 16.3 9.1
SPRUCE PULP 45 52 2.9 19
LIQUOR 23.0 8.9
C A R B O H Y D R A T E A N A L Y S I S
Species Glucan Xylan Galactan Arabinan Mannan Uronic Unaccounted
Anhydr.
ASPEN WOOD 57.9 13.2 0.6 0.2 3.4 1.0
ASPEN PULP 53.1 2.8 0.1 trace 2.2 0.06) 9 0
LIQUOR 0.4 7.2 0.5 trace 0.8 0.2 )
SPRUCE WOOD 49.9 5.7 1.8 1.1 11.9 0.8
SPRUCE PULP 43.1 2.0 - ~ 2.5 trac~ 14.7
LIQUOR 1.7 1.4 1.0 0.6 4.7 0.1
_ __. ___
- 16 --
-
~3~5
CONSIDERATIONS OF PROTON SOUl~.CES
Delignification of lignocellulose b~ the aqueous
methanol-metal salt system of this invention is believed to
proceed by combined alcoholysis and proton catalysed
hydrolysis that reduce the size of native lignin molecules
by bond breaking within the macro-molecules and between
the lignin-hemicellulose complexes. A limited reaction
of methanol in association with water of the solvent mixture
may play a role in temporary and permanent modification of
lignin functional groups at high temperature to increase
the rate of lignin dissolution. The simultaneous removal
of carbohydrates and cleavage of the lignin-carbohydrate bond
facilitate a widened pathway for passage of fragmented lignin
from the cell wall. As both lignin and t~e less stable
carbohydrates are removed by a similar mechanism and require
substantially lower activation energy in hydrolysis than
the cellulose, this limited proton source and the possible
protective effect of the exchanged and adsorbed cations is
believed to be responsible for minimization of the degradation
of cellulose as observed in Table IX.
Cations of magnesium, calcium and barium appear to
have the ability of restricted catalysis of delignification
while restraining cellulose degradation. The sources of
protons produced in the system comprised of impregnated
lignocellulosics and the catalysed aqueous solvent mixture
are considered to result from phenomena of hydrolysis and
alcoholysis of the solvent-cation complexes, cation exchange
on uronic acids, carbonyl, ester and ether functions on
both carbohydrates and lignin and finally through increased
dissociation and complexing of weak organic acids (acetic,
formic etc),formed during the course of cooking, due to the
- 17 -
x
alteration of their activity coefficients by adding the alakli
earth metal salts to the solution as predicted by the Debye-
Hùckel theory. A combination of these factors is held
responsible for the pH drop when the catalyst-containing
aqueous methanol solution is added to lignocellulosics as
demonst~ated in Table VIII. The proton activity is
further enhanced by the relatively high temperature and
pressure. It was found most advantageous~to work
at the higher temperatures (180 to 200C) as in this case the
reaction times are sufficiently short to favourably influence
the kinetics of delignification while cellulose degradation
remains suppressed. There were no reprecipitation or
secondary condensation tendencies of either lignin or the
hemicelluloses observed under these conditions.
- 18 -
