Note: Descriptions are shown in the official language in which they were submitted.
- 2 -
Cellulose-con-taining ma-terials occur in nature
in large numbers and in great variety. Wood is an
example of such a known natural cellulose-containing
material. It consists essentially of cellulose
(a material mainly built up from glucose units), hemi-
cellulose (a substance mainly built up from pentoses and
hexoses) and lignin (a polymeric substance havingaromatic
ringswhich aresubstituted bymethoxy groups). ~ood is
utilized in many di~-ferent ways~ for example for the pro-
duction of heat (burning) and as a building material inthe furniture and building industry a~d the like; utiliz-
ation o~ wood by purely chemical means is also possible.
Chemical disaggregation processes~ which effect not
only the separation of wood into its constituents, hemi-
cellulose, cellulose and lignin, but also the degradationand modification ofthe latter 7 have alreadybeen known for a
longtime. As a rule, chemical processes yieldaqueous solutions
o~ monomeric, dimeric and oligomeric saccharides, which
may be subjected to subsequent hydrolysis to
glucose or can be subjected direct to fermentation to give
ethanol, or to concentrating or to evaporation to dryness.
E~amples of possible fields of use of the products thus
obtained are in the sphere of cattle feed additives or,
preferably1inthat o~ raw materials for fermentation.
II1 chemical processes for wood saccharification,
two principles have been used on a large industrial scale
in the past: the disaggregationof wood with concentrated
aqueous hydrochloric acid (Bergius-Rheinau-Udic) anddis-
a~Jsregation with dilute sul~uric acid (Scho~ler-Tornesch-
Madison); in this con-text see, ~or example, Ullmanns
Encyclop~die der ~echnischen Chemie ("Ullmann's
Encyclopedia of Industrial Chemistry"), 3rd Edition,
Volume 8 (1957) 9 pages 591 et seq.
The disag~regation ~cellulose-containing raw
materials with anhydrous hydrofluoric acid has also
already been investigated onseveral occasions. Here,
however, none of the processes hitherto disclosed has yet
led to a technically satisfactory solution o~ the problem.
German Patent Speci~ication 560,535 describes the disaggre-
~ationof wood w;th pure HF,in the form of liquid or vapor,
at low temperatures, the HF being recycled by being
evaporated or blown off and subsequently condensed.
As a continuation o~ this work, German Patent Speci~ication
585,318 describes a process Ior the disaggregation o~ ~ood Wi~
gaseous hydrogen fluoride, which operates in three stages
via absorption of HF on wood at 10-20C,~lisaggregation a.t
20 50C and desorption at 100-150C, it being possible
to dilute the ~ with a stream of inert gas. The out-
lay on cooling for condensing the HF has a disadvantageous
effect here, as has also the fact that in the condensa-
tion stage initially there is only a very non-uniform
distribution of the hydrogen fluoride on the reaction
material, a circumstance which can only be counteracted
by very long residence times or by greatly increasing the
amount o~ hydrogen fluoride used, otherwise the yields
are greatly impaired
German Patent Specification 606,009 describes an
extraction process using liquid E~which, however,
~ 3~7
requires large quan-tities of HF and has the disadvantage
that large quantities of heat have to be supplled in order
to evapora~e the hydrogen fluoride fro~l the extract and
extraction residue (lignin),and whichhavetobe removed
again in the subsequent condensation process. More
precise data concerning yields in processes of this ~ind
are to be found in Angew. Chem. 46 (1933) 113/7, where by
absorbing HF from the gas phase in an externally cooled
vessel at 0C 32% of sugar is achieved, relative to car-
bohydra-tes present, at a loading of 50% by weight of HF
relative to wood, and 86% of sugar, relative to ca~bo~
hydrates, is achieved at a loading of 100% by wei~h-t.
No further in~or~ation concerning the recycling of the
HF is given in this reference~
All these processes suffer from the disadvantage
that they consume large quantities of expensive hydro-
fluoric acid, that the recovery of HF from the reaction
products is very exponsive and that large losses of HF
take place in practice.
A more advanced process is described i~ Austrian
Patent Specifioation 147,494, where the existing state of
the art is described as follows: 'IIf the reaction is
carried out using highly concentrated or anhydrous
hydrofluoric acid in a liquid or gaseous state at low
terQperatures, the ~egradation of the wood only proceeds
in a very non-unifo~n manner and there~ore incompletely.
Fir~tly, at suoh low t~mperatures the distribu -
tion of the hydrogen fluoride, wll:ich is pre~ent ~s a
fine r~ist in the air , is ~rexy non-uniform ~
t3'J
. 5 ~
the more so as the air ~Jhich is presen-t
makes i-t difficult -to achieve a unlforrn reaction. On
the other ha~d,it is known that,in thesaccharjlication of
~oodwith conce~trated hydrogen fluoride,not onlyin aliquid
but also ina gaseousstate, the particles of wood react
rapidly at their surface with the concentrated hydrogen
~fluoride, form a hard, rather impenetrable skin and shrirLk
whereby the further penetration of the gas into the interior
is impeded. In addition, the penetration of the wood
particles is already rendered difficu3t by the air present
in the cells~ Thus an outer crust which encloses
unsaccharified material and prevents further saccharifi-
cation is formed very rapid]y. In order to remove
these drawbacks, i-t has already also been proposed to
carry out the disaggregation with conce~trated li~[u~d hydro- -
fluGric acid using an extraction process or to pre~ent the
formation of crusts by mixing ine~t gases into the hydro-
fluoric acid, in order thereby to achieve a more uniform
andmore complel;e disa~gregation. Ho~ver, the extraction pro-
cess operates with a disproportionately large excess ofhydrofluoric acid and the reaction material retains large
quantities of hydrofluoric acid without preventing the
~orrnation of cr~sts, with ali its disadvantages.
Although dilution with inert gases can somewhat reduce
the formation of crusts, it can never elimina-te thls and lt
also cannot result in the gas penetra-ting uniformly into
the interior of the wood, since the wood is indeed filled
with air. This is because, as is known, wood is only
composed to a very small ex-ten-t o~ ligneous matter and its
:L I ~ ;3~'ît
-- 6 --
largest component by far is air, which is presen-t between
and within -the wood cells. A virtually anhydrous wood
is composed, for example, OL approx. 15% of ligneous
matter and approx. 85% of air. Since the wood cells
are extremely small in relatlon to -the size of a piece of
wood, however thoroughly it may have been comminuted, the
airconten-t plays a predominant part, even in ~he case of
sawdust."
Hardening of the surface of wood particles also
appears to have been observed in saccharifying wood with
aqueous mineral acids, such as aqueous hydrochloric or
sulfuric acid, since9 for example, in Z. Angew. Chem. ~7
(1924) 221 the substances present in wood, such as lignin~
mannan, galactan and the like, are described as "incrus-
tants" ~hich should be removed,if possible, prior tothe actual wood saccharification process, additionally
because of interfering degradation products (furfural,
acetic acid, formic acid and -the like) For -their
remo~al itwouldhave beenpossible -since thesel'incrus~ants"
wereknown tobe hydrolyzable -toconsidertheuse ofa kind of
~'pre hydrolysis" with dilu-te mineral acid at eleva-ted
temperature and, if appropriate, eleva-ted pressure, also
in the case of wood saccharification by means of hydrogen
fluoride. However, considera-tion had not been given
to a pre-hydrolysis of this type; instead, i-t was
suggested by Hoch and Bohune~c that in order to avoid the
disadvantages described above a vacuum of approx. 30mm Hg
be employed for ~rood saccharification with hydrogen
~luoride [Austrian Patent Specification 147,494 and
J~ 3i7
Patent or ~ddition 151,2~ the Hoch and Bohunek wood
saccharification process using hydrogen fluoride is also
described in the journal "Holz Roh~ und ~Jerkstoff" 1,
pages 342~344 (19~8)].
Disadvantages of these process~ are the difficulties
of industrial implementation which necessarily occur when
operating in vacuGI and also the dif~iculties caused
by the relatively complicated manner in which the reactions
are carried out. A drawback from which all the processes
suffer is the formation of mixtures of pentoses and hexoses
as a result of simultaneous hydrolysis of the hemicelluloses
and o~ the cellu]ose of the wood.
A further problem is the removal of the acetic acid
formed in the hydrolysis of hemicellulose, which renders
difficult the circulation o~ the ~F as loss-fr~e as
possible, and also the fact tha~ the pentoses are easily
decomposed to give ~urfural.
It has now been found, surprisingly, that the
above-described disadvantages of the stat~ of the art
Ean be avoided and that it is possible to ~accharify
cellulose readily , if the v~getable materials
are disaygregated with anhydrous, gaseous ~F, not in their
natural form, but in the form of 'tcelloli~nin" which is
obtained a~ter a pre-treatment.
"Cello'ignln" is to be lmders-tood here as meaning
vegetable materials, such as wood, straw, bagasse and
similar raw materials having been subjected to a pre~
hydrolysis -~hich is-in itsel~ known.
This pre-hydrolysis, which is in itsel~ know~, of
L~ 3'~
3 --
wood comprises a relatively short treatment, with very
- dilute mineral acid,at elevated temperatures and
pressures, in which essentially the pentosans and hexosans
present in the hemicelluloses are split as far as the
monomer units, such as, for example, xylo~e or mannose
Depending on the reaction condi-tions, the latter can
subsequently be isolated as such or undergo further
changes, for example dehydration to give furfural or
hydroxymethylfurfural (compare IJllmann, loc. cit.,
Volume 7 (1957), page 711)~ In addition to fermentation,
the reduction of xylose to xylltol may be mentioned as a
further example of the industrial utilization of hemi-
cellulose degradation products It is thus possible
to obtain valuable products ~rom wood by pre-hydrolysis
~rior to using the disaggregation process according to
the invention.
Cellolignin is also to be understood ~ere as
meaning paper material (for example was-te paper)
which has a low content of hemicelluloses. In the pre-
hydrolysis of wood,its structure is la~gely retainedbut the cellolignin which can thus be obtained has a much
more crumbly and porous nature compared wi-th the natural
state~ so -that HF, including mixtures with air or ano-ther
inert carrier gas, can penetrate readily without encrus-
tation o~ the sur~ace taking place, It is not necessaryto operate in vacuo.
A further important advantage of the use cf cello-
lignin instead of nati~e wood is that the reaction
material is then appreciably simpler to handle from the
'7
_ 9
point of view of process technology This is due, on
the one hand, to the fact that, compared with wood o~
the same particle size, cellolignin has an apparent
volume which is only approx. half as large and i~ thus
exhibits an appreciably smaller degree of shrinking when
disaggregated with hydrogen fluoride gas, which makes matters
considerably easier, for example in dimensioning reactors.
Secondly, reaction material composed of cellolignin
remains pourable and free-~lowing, even when charged
with hydrogen fluoride, whilst re~ction
material composed of native wood has a strong
tend~ncy to gum up due to r~si~ous constituent~,
and al~o because of cleavage products of the hemicellu-
loses, a~d is difficult to con~ey.
Naturall~, such a tendency to gum up also impedes
the desorption of hydrogen fluoride, particularly if the
latter process step is intendedto takeplace rapidlyand as
quantitativelyas possible. When celloli~nin is used as the
substrate, however, this is possible withou-t difficulties.
Furthermore, in this process it is no longer
nec~ssary to separa-te the mixtures of sugars ~ormed in
the hydrolysis of hemicellulose from -the oligomeric glucose
structural units ~ormed in -the hydrolysis of celluloseg
or from glucose, which makes it possible to utilize these
various sugars more easily in ~ermentation~
It is also an advantage that no acetic
ac,d and furfural are fornrd Ln the disaggregation of cello-
lignin, so that the ~i can be circulated without
having to condense these components. Difficulty in
.~.. .q ~ 4
- 10
separa-tion and losses of ~ are thus avoided.
A further advantage is -tha-t it ls possible to
absorb ~ onto cellolignin above the boiling point of HF,
so that external cooling is no longer necessary. It
5 was also surprising~ tha~ yields of ,gO% o-f glucose or
oligomeric glucoseg relative to the cellulose employed in
the cellolignin, are achieved in a simple mar~er in the
process according to the invention, -the sugars produced
being high-grade in quality, that is to say almost
colorless.
The invention therefore relates to a
process for obtaining water-soluble saccharides (glucose
or oligomeric glucose) from cellulose-containing material
by treating the latter with gaseous hydrogen fluoride -
Optiona11y - diluted with an inert gas - at temperatures
between about 20 and 120C 7 preferably between about 40
and 80C; the process comprises subjecting cellolignin
- to a treatment with hydrogen fluoride.
Cellolignin is to be understood here, as defined
above, as a ma-terial composed largely of cellulose and
lignin.
In view of the state of the art, in which the
most recent process to be developed toa majorextent
~Hoch and Bohunek, loc. cit.) attempts -to remove -the dis-
advantages associated with the non-uniformdisaggregation and
with the formation ofencrustations,by usingthe expensive
vacu~nmethod- although thefact that hemicelluloses canbe
hydrolyze~ readilywas knoYm (Oesterr. Chem.-Zeitg 40,
5 et seq. ~1937)) the use of pre hydrolyzed material was
in no way obvious. .It was, therefore, rath~r
surprising that this measure, which lies inthe opposite
directiontothatsuggested bythe state of the art,
permits uniform and problem-free saccharification of
wood and materials similar -to wood.
The cellolignin which, in accordance with the
in~ention, is particularly suitable for degradation to
give water-soluble sugars is ob-tained by pre-hydrolysis
of natur~ cellulose-containing material (wood, straw,
bagasse and the like) with dilute a~ueous mineral acid,
preferably dilute hydrochloric or sulfuric acid. As
already pointed out in describing the state of the art,
the process of pre-hydrolysis is known in wood saccharifi-
cation and is also to be found in fairly recent literature,
such as Ullmanns Encyclop~die der technischen Chemie
("Ullmann's Encyclopedia of Industrial Chemistry")~ 3rd
. Edition, Volume 8 (1957), pages 591-595,and also in the
book by W. Sandermann, t'Chemische Holzverwertung"
e Chemical Utilization of Wood"), Bayrischer
Landwirtschaftverlag, Munich 1963, page 253.
It comprises a relatively brief treatment of -the
natural starting material with a ve~ dilutemineral acid
at an elevated temperature (preferably bet~een about
100 and 160C)andunder aneleva-ted pressure (prefe~ably up
to about 1~ atmospheres) 3 in the course of which essen
tially the pentosans and hexosans present in the hemi-
celluloses are split as far as the monomeric units
(x~ylose, arabinose, mannose and the like~. Depending
on the reaction condi-tions, the latter can then be isolated.
- 12 _
as such or undergo further changec, for example to give
furfural and the like by dehydration.
They are preferably ernployed as raw materials for
fermentation or for the production o~ xylitol
waste pa~er of lcw hemicellulose content is also very
suitable for use as the s~ting material in the process of the invention.
The disa~greaation process accordmg to the inve~tion
can, for example, be effected either by bringing -the pre-
treated material (cellolignin or, for example, paper
1~ shredder material) which has been dried to a moisture
content of 0 to about 20%, preferably about 2~5%, and has
been comminuted if required, into contact with HF gas,
optionally mixed with air or another inert carrier gas
discontinuously in a suitable stirred vessel made of a
material resistant to hydrogen fluoride, or by passing
a gas mixture con-taining HF, advantageously in a conveying
equipment, in countercurrent toa con-tinuous stream of the
substrate to be disagqreyated.
As a result Q~ the heat of reaction, whichisliber-
Z0 ated spontaneously, the temperature rises and can be keptwithin the desired rangebetween about 20 and 120C, pre-
ferably be-tween 40 and 80C, by carrying out the reaction
in a suitable manner, such as, for example,bydilutingwi-th
inert gases
~5 The contac-t of the substrate wi-th hydrogen fluoride
gas is maintained until one part by weight o~ the material
has taken up about 0.2 to 3.0, preferably about 0.4 to
0.8, parts by ~eight of hydrogen fluoride.
It is then advantageous to continue the reaction
.~ '7
by choosing, depending on the nature of the substrate and
on the condi-tions of HF absorption, a residence time which
is adequate to achieve a high yield. Longer resi-
dence times are not disadvantageous, bu-t have no advan-
tage either. The reaction times can be between 15minutes and several hours. Preferred reaction con-
ditions are those in which a residence time of about 1
hour is not e~ceeded.
The subsequent HF desorption can be carried out in
accordance with the state of the art by warming the re-
action material andlor by evacuation or by treatment with
a stream ofan inertgas ~for example nitrogen, air9 C02 or
a rare gas) of suitable strengthg again with or without
simultaneous warming ând/or evacuatingO ~he hydro-
gen fluoride ~hus recovered can be isolated by condensa-
tion or can be reacted directly with fresh substrate so
as to produce a circulation of gaseous hydrogen ~luoride.
The further processing of the material which has now been
disaggregated ("saccharified~) can also be carried out in a
manner which is in itsel~ known, as described, for example,
by K Fredel~agen and G. Cadenbach, Angewandte Chemie 46
(1933), pages 113 to 117. That is to say, for example,
the material is extracted with hot water, insoluble lignin
is fil-tered off, the small quanti-ty of hydrogen fluoride
~dhering is neutralized in the filtrate wi-th oalcium
carbonate or calcium hydroxide and the mixture is concen-
tra~ed.
In the procedureaccor~ing tothe invention, the
quantity o~ "~,rood sugar" (or"straw sugar"and thelike)
obtainedafterdryingthe residuefrom evaporation is inall
casesmore tha~about 90~Oof the cellulos~present in thesub-
strate (calculated on dry substance).
Because OI the high ~'sugar" yield,the exceptivnally
simple and smooth performance of the process ~increased
porosity of the substrate and thus easier penetration of
HF) and also the energy-saving absorption of hydrogen
fluoride (no cooling or vacuum required), the invention
constitu-tes a not inconsiderable advance in this field.
The oligomeric glucose structural units can be
employed for further utilization (fermentation to give
ethanol, concentrating or evaporation and use as cattle
feed additives or as raw materials ~or ~ermentation and
the like) in the fcrm in which they are pro~uced or they
can also be subjected in a manner which is in itself kno1~n
to further hydrolysis to give monomeric glucose.
The invention will now be illustrated in greater
detail by means of the following examples:
500 g of spruce cellolignin (59% of cellulose +
~l~ of lignin) of particle size approx. 2 mm were
placed ina cylindrical 2 l vessel made o~ transparent
polyethylene, fitted with a stirrer, a thermometer and a
gas inlet, andweretreated with a mixture of air and hydrogen
~luoride gas,which is prepared by passingair overliquid hydro~
gen fluoride at 20C (waterbath)~ In the course o~
this treatment, the material ~Jas s-tirred slowlyand it turned
dark brown. The air stream and the vapori~ation o~ ~
were regulated in such a way that the internal temperature
did not exceed 70C.
- 15 -
After 300 g of hydrogen fluoride had been
absorbed, care was taken to maintain an internal tempera-
ture of 50°C for 30 minutes. The hydrogen fluoride
was then expelled by passing in warm air, with continued
stirring. In so doing, part of the heat of desorption
required was also supplied by external heating. The
desorption was continued with a continually increasing
temperature until a hydrogen fluoride content of about 5%
in the substrate had been reached. The material was
then transferred to a fluid bed dryer and hydrogen fluoride
was blown off until a residual quantity of approx. 0.5%
had been reached. The HF/air mixture thus produced
could be used directly for further batches.
The contents of the reactor were then digested for
15 minutes with approx. 2 1 of hot water, were suction-
strained and rinsed with a little water. The dark brown
filter residue weighed about 250 g after drying and was
thus composed of 82% of lignin and 18% of still not disaggregated
cellulose. The filtrate was rendered alkaline, while still
hot, with technical calcium hydroxide, the excess of
hydroxyl ions was neutralized with carbon dioxide, and
calcium fluoride and calcium carbonate were filtered off,
if necessary by means of a filtration aid. The clear,
slightly yellow, neutral solution was evaporated to dry-
ness in vacuo. This gave approx. 250 g of slightly
yellowish wood sugar, corresponding to 85% of the theore-
ical yield. The product gave a clear solution in water
and contained between 2 and 10% of monomeric glucose, the
remainder being composed of oligomeric glucose.
- 16 -
Examples 2-13
A jacketed -tube, resistant to hydrogen fluoride,
of length 30 cm and internal width 4 cm in a horizontal
position was filled about half full with 30 g of cello-
lignin of particle size 1-2 mm, and was closed at both
ends with bored rubber stoppersO Cne of tw~-~n steel
tube, perforated overthe~ entire length, was placed in
the layer of cellolignin and the other in the free ~ce above
the latter These tubes led to the exterior on both
sides through holes in the sealing stoppers and
were used for the introduction or r~al of~/air mixture, respectively.
This made it possible to treat the cellolignin wi-th gas
at right angles to the surface of the bed. The material
was allowed to absorb hydrogen fluoride and appropriate
heating was used to ensure an internal temperature of 50C
during the subse~uent residence time. Hot air, instead
of the HF/air mix-ture, was then blown through the bed for
15 minutes~and the reaction material thus obtained9 which
had been freed from the bulk of the hydrogen fluoride, was
worked up as described in Example 1.
The yields corresponding to differen-t quantities
of HF absorbed and residence time are sho~rn in -the
following table.
- 17 -
EYampleH~ absorbed Residence time Yield of wood
No. ~g] [minutes] ~ugar
~f theor. Yalue
2 9 120 4 24
3 11 120 6 35
~t 15 120 ~9'5 56
16 120 11 65
6 17 120 12 71
7 19 120 15 ~8
10 8 20 120 ]6 94
9 19 10 11 65
19 20 13 77
11 19 25 14 82
12 19 30 15 ~8
1513 19 60 15 88
Example 14
In a long, horizontal tube made of material resis-
tant to hydrogen fluoride in which a free-flowing soli.d
can be propelled continuously by means of a screw conveyor,
20 a hydrogen fluoridelcarrier gas mixture was passed in counter-
~ent to a contm~ous charge of.celloli~nLn at such a rate that the
material at the E~ inlet end of the -tube had a content of
approx. 600/o o~ HF, relative to cellolignin,whilst only
pure carrier gas flowed out at the cellolignin inlet end.
25 The material was continuously discharged at the HF inlet
end, while fresh cellolignin was recharged at the opposi-te
end. After having pas~ed a re~idence tisne section
for half an hour , the ma$erial discharged ~a
freed from hydrogen fluoride by blowing the latter off~
30 and the IIF rich gas mixture thus obtained was fed bac~
!
3t7
-- 18 --
into the reaction tube. The disaggregated cellolignin
was worked up in the manner described in Example 1. The
yield of wood sugar amounted to approx.85% of the
theoretical value.
Example 15
150 g of shredded newspaper material were treated
with a hydrogen fluoride/air mixture in the manner des-
cribed in greater detail in Example 1. After a resi-
dence time of one hour at 50C the reaction mixture
was freed from the hydragen fluoride, down to a residual
content of 2%, by passing in a stream of warm air, and
~e dark-colored residue was digested with hot water.
After filtering and drying, this gave 50 g of insoluble
material, mainly composed of lignin. The filtrate
was neutralized with calcium hydroxide and the calcium
fluori~ was filtered off with suction. The residue from
the evaporation of the filtrate weighed 80 g and con-
tained approx. 10% of monomeric glucose (remainder:
oligomeric glucose).
