Note: Descriptions are shown in the official language in which they were submitted.
7 C
~ g~ ~ ~
_ckground of the Invention
Traditionally, used wood, paper and agricultural by-
products, such as sawdust, woodwaste, corncobs, straw, sugar
cane bagasse, newspaper and the like have been regarded essen-
tially as waste materials, and have been disposed of through
incineration or by other, similarly unproductive, means. It
is well known that the lignocellulosic constituents of such
materials can be hydrolyzed to produce more valuable products
which in turn can be converted into additional and different
valuable products; however, such operations are in limited use,
due largely to the relatively low returns on investment which
they have been capable of generating. The capital expenditures
required to design and construct the facilit,ies for carrying
out such recovery operations tend to be significant, thus
demanding that relatively high conversion rates be attainable
in order to justify the expense invo:Lved.
In U.S. Patent No. 4,201,596 issued May 6, 1980 to
John A. Church et al entitled "Continuous Process For Cellulose
Saccharification" and commonly assigned herewith, there is
described a method and apparatus for saccharification of cellu-
losic products in which the cellulosic constituents of typi-
cal waste products may be converted into glucose, furfural
and xylose. Such a process conveniently, rapidly and econom-
ically provides by acid cellulose hydrolysis, a hydrolyzate
which may be used as the raw material for the production of
more valuable products. For e~ample, as disclosed in U.S.
Patent No. 4,201,596, such hydrolyzates may be used as the
raw fermentable substance in a process for converting sugar
into ethyl alcohol.
~hile it has long been known that cellulosic hydro-
lyzate solutions ma~ be made fermentable for the production of
alcohols, prior procedures have been feasible only on a small
-- 1 -- ~ "~.,
~s~
laboratory scale and have not significantly developed beyond
this stage. Among the principal conditions contributing to
this state of the art has been the inordinately slow rates of
reaction with extremely low yield and the lack of predicta-
bility of conditions that would pernit fermen~ation with any
given hydrolyzate. Additionally, the economic considerations
inherent in the chemical conversion of sugar to alcohol have
been a limiting factor. For example, it is only theoretically
possible to obtain one unit of alcohol from every two units of
sugar present in the raw material. Losses in sugar content
of the raw material through process conditions, mechanical
processing, etc., serve to even further decrease the low
yields that it has been possible to obtain.
Cellulosic extractives and their decomposition pro-
~ ducts as fermentable raw materials have been particularly
; enigmatic to approach because of the wide variety of factors,
many of which are unknown, that adversely affect and, in many
cases, prevent the fermentation process. One factor that has
long been recognized in the art as significantly retarding the
development of a feasible fçrmentation process has been the
presence of materials in the hydrolyzate that act as toxins
or fermentation inhibitors. However, the toxins present in
any given hydrolyzate may vary considerably depending on its
processing history, its source, etc. Moreover, the problem
is further compounded by the fact that even after the particu-
lar toxins have been identified in a given hydrolyzate, their
action under any given set of conditions has been largely un-
predictable and fermentation has been difficult even under
special conditions. Various wor~ers in the art have suggested
that 1hese difficulties may be dependent on any number of
factors including processing temperatures, pH of the media,
the presence or absence of oxygen, the concentration and type
- 2 -
5~77
of toxin substance, the ratio of yeast cells to toxin sub-
stances, the physiological condition of the yeast cells, the
wide ~ariation in the toxicity of various substances on the
metabolism of the particular yeast, the oxidation-reduction
potential developed duriny reaction, and many other factors.
Discussions of the various difficulties of fermentation and
general factors influencing fermentation are found in many
sources in the literature.
Cumulative discussions are given by Harris et al in
"Fermentation of Douglas Fir Hydrolyzates by S. cerevisiae"
and Leonard et al, "Fermentation of Wood Sugars to Ethyl
Alcohol"; Industrial and Engineering Chemistry, Vol. 38, pp.
896 to 904, (1946) and Vol. 37, pp. 390 to 397, (1946), respec-
tively.
Other workers in the art include Eklund et at, "Acid
Hydrolysis of Sunflower Seed Husks for Production of Single
!~ Cell Protein", European ~ournal of Applied Microbiology, Vol. 2,
pp. 143-152 (1976) who disclose a method of hydrolyzing sun-
flower seed husks and degradation of the resulting hydrolyzates
to produce protein.
German Patent 676,967 to Scholler (1939) describes
a method for clarifying xylose worts obtained by acid hydrolysis
of cellulose-containing substances for feed purposes or yeast
production by precipitating calcium phosphate and calcium
sulfate after heating to 65 to 100 together with centrifuging
and conducting the wort over oxidized metal fillings or large
surface area materials while the wort is at a pH of 4 to 7.5,
adding malt sprouts to the thus clari~ied wort and stirring
for several hours.
U.S. Ratent No. 2,203,360 dated June ~, 1940 to
Partansky discloses a method ~or improvin~ the fermentation
characteristics of acid wood hydrolyzates by treating the
-- 3
hydrolyzate with lime to adjust the pH to between 9 and 10,
aging for 1 to 2 days, reducing the pH with sulfuric acid to
pH 5, purifying the solution with activated charcoal, diluting
the solution to contain 40-70~ by volume of hydrolyzate, inocu-
lating the solution with yeast culture and fermenting for 2
days.
The prior art, as represented by the methods dis-
cussed above, is illustrative of the absence of a feasible com-
mercial process for fermentation of acid hydrolyzates to alco-
hol~ due to inordinately slow reaction times and low yields
and/or the lack of direction for obtaining the samPO A method
for readily and efficiently producing alcohol by fermentation
of sugaxs present in wood and wood-byproducts is a particularly
timely and significant development :in view of eurrent interest
in alcohol as a potential energy source available from renew-
able raw materials.
A primary object of this invention is to provide a
process for fermentation of sugars present in acid hydrolyzates
derived from lignocellulosic materlals.
Another object of the invention is to provide such
a process in which reaction times are relatively short, in
whieh fermentation may be effected at relatively high sugar
concentrations and in which control mechanisms are established
which permit predietability, reproduction of results with con-
sistency and production of end products of high value~
Another object of this invention is to provide a
process in which the acid hydrolyzate of cellulosic waste
materials may be converted into ethyl alcohol.
The ~ccomplishment of these and other objects will
be apparent from the descrip-tion of the invention which follows:
Summary_of the Inventlon
The foregoing and related objects of this invention
-- 4
7~7
are attained in a method for preconditioning acid hydrolyzates
derived from lignocellulosic materials, to negate the effect
of substances tending to inhibit the fermentation of such
hydrolyzates and to a process for the production of ethyl
al~ohol from glucose contained in such preconditioned acid
hydrolyzates. The hydrolyzate is preconditioned to remove
and/or reduce or otherwise negate the effect of inhibitory
substances to a level whereby the hydrolyzate may be readily
fermented to ethyl alcohol in substantially theoretical yield.
The preconditioning method broadly comprises the
steps of: (1) subjecting the hydrolyzate to steam to remove
furfural and other steam-volatile substances therefrom;
(2) adding sufficient calcium oxide to the steam-stripped
hydrolyzate, at room temperature, to adjust the pH to between
about 10 and 10.5, maintaining the resulting mixture at said
pH for about 1 to 3 hours and separating the hydrolyzate from
the resultant precipitate; (3) adding sufficient amounts of
a mineral acid to adjust the p~ of said hydrolyzate to about
5 to 7; and (4) adjusting the concentration of said hydroly-
zate to a glucose concentration of less than about 150 grams
pe~ liter to provide a solution fermentable to ethyl alcohol.
The fermentation process broadly comprises the
steps of: (1) preconditioning an acid hydrolyzate to negate
the effect of substances tending to inhibit the fermentation
thereof by subjecting the hydrolyzate to the preconditioning
method described hereinabove; (2) inoculating the precondi-
tioned hydrolyzate with yeast inoculum comprising from about
0.7 to about 7 dry weight percent of yeast cells per 100 grams
per liter of glucose in the hydrolyzate; (3) fermenting the inoculated
` 30 hydrolyzate at a pH of 5 to 7 for a period sufficient to con-
vert glucose to ethyl alcohol; and (4) recovering ethyl
alcohol from the fermentation mixture.
- 5 -
~! ~
~5~7~1~
In a preferred embodiment, yeast cells are recovered,
reconcentrated and recycled to a subsequent fermentation medium
comprising preconditioned, concentrated hydrolyzate.
General Disclosure
The process of this invention utilizes a combination
of steps and conditions which are interrelated and interdepen-
dent for the successful achievement of the objectives of the
invention. This interrelationship will best be seen from the
following description of the effect or function of each parti-
0 cular sequence within the context of the total process.Hydrolyzate Raw Material
The invention may be successfully realized with any
hydrolyzate derived from the acid hydrolysis of lignocellu-
losic material. Such lignocellulosic material may be selected
from a wide variety of materials including wood and paper and
particularly used paper and wood by-products such as sawdust,
wood waste, straw, sugar cane bagasse, rice hulls, newspaper
and the like. Such materials may be hydrolyzed in the presence
of an acid catalyst by methods well known in the art to pro-
vide a suitable hydrolyzate raw material for use in the pro-
cess.
The hydrolyzate raw material provided will vary in
sugar content and other components depending on the conditions
under which it has been produced. This can be best understood
by a consideration of the chemistry involved in acid hydroly-
sis stated for the sake of illustration in simplified terms.
When cellulosic material is heated with dilute aqueous acid,
glycosidic bonds which connect individual anhydroglucose units
to one another in the cellulose molecule are cleaved by acid
catalysis and one molecule of water adds to each anhydroglucose
unit to fornl one molecule of glucose as illustrated by the
idealized equation:
-- 6
7~ .
H-~
(C6 ~llo s)n ~ n 2 ~ ~ 6 12 6
Glucose is inherently unstable in hot acid solutions and can
lose three molecules of water to yield 5-hydroxymethylfurfural
(HMF) ac-oxding to the equation:
H+
6 12 6 ~ ~ C6 H6 3 + 3H2O
HMF in turn is unstable and can add two molecules of water to
yield levulinic acid and formic acid:
H~
C6 H6 3 + 2H2 ~ ~ C5 8 3
Other very complex reactions also occur in which it is believed
HMF condenses into dark insoluble residues known as humins.
Lignin breakdown products such as vanillin or other aromatic
compounds may also be present. Additionally, the bonds of the
hemicellulose molecule are cleaved to produce free molecules
of xylose from xylan. Cex~ain reaction conditions will favor
formation of glucose or xylose and accompanying decomposition
; products.
An especially preferred method and apparatus for
producing suitable acid hydrolyzate raw materials for use in
this invention is that disclosed and claimed in U.S. Patent
No. 4,201,596 entitled "Continuous Process for Cellulose
Saccharification" referred to hereinabove. ~s disclosed
therein, whereas the xylan conversion to xylose occurs at
relatively low temperatures, the cellulose conversion to glu-
cose best occurs under more severe conditions. It has not
been possible to produce maximum amounts of both xylose and
glucose in a one phase method due to the fact that the
5~77
xylose dehydrates to furfural under the conditions which
most efficiently effect the conversion of cellulose to glucose.
The method of U.S. Patent No. 4,201,596 defines the conditions
which favor production of glucose/furfural and which minimize
degradation of glucose to HMF and levulinic acid. Since, as
discussed and illustrated further hereinbelow, furfural, HMF
and levulinic acid are each toxins to the fermentation organ~
ism, the present invention most preferably utilizes a hydro-
lyzate raw material obtained under such conditions that mini-
mize, to the extent possible, the presence of such toxin sub-
stances.
Such conditions in general provide for acid hydro-
lysis, in the presence of steam, of cellulose feedstock having
a solids c~ntent of from about 20 to 45 weight percent at
temperatures within the range of about 190 to 225 C and pres-
sures of about 200 to 400 psi with residence times in a reac-
tion zone of about 1 to 10 minutes. In the reaction mass,
the optimum amount of water after steam injection is about
75 to 80 weight percent and virtually any strong mineral acid
can be employed to catalyze the hydrolysis reactant, sulfuric
acid being normally the acid employed in amounts of about 1
to 3 percent based on the total weight of the reaction mass.
In this preferred process for production of hydrolyzate raw
material, the reaction mass will be subjected to an abrupt
pressure reduction whereby a fraction of the hydrolyzate
vaporizes and may be recovered. This fraction will normally
comprise furfural and acetic acid.
It will be understood that the above description is
for purposes of illustration of the preferred mode of obtain-
ing an acid hydrolyzate that is especially suitable for usein the present invention. The method of the invention may
utilize acid hydrolyzates from any source since it is a
-- 8 --
feature o~ the invention that the preconditioning method will
serve to reduce or remove certain of the toxin materials to
- tolerable levels and/or to otherwise negate the effect of
such materials without substantially adversely affecting the
glucose present in the hydrolyzate.
Such hydrolyzate raw materials will in general, how-
ever, comprise glucose, furfural, 5-hydroxymethylfurfural,
acetic acid, formic acid, and levulinic acid and will have a
p~ of less than about 1.5 and preferably of about 0.5.
Preconditioning of the Hydrolyzate
The hydrolyzate raw material as received is a con-
glomerate of chemical substances. Many of such substances
act as inhibitory agents or toxins to the yeast while many of
such substances are unknown in identity and effect. It is
possible that the presence of such substances, known or un-
known, may exert a cumulative affect on the fermentation mech-
ism or yeast culture. It is also possible that some of the
substances may be combining synergistically to inhibit either
the particular yeast organism or other mechanism involved in
the fermentation. Thus, while there is necessarily a degree
of uncertainty as to exactly how the objectives of this in-
vention are realized, it is believed that the preconditioning
method renders the hydrolyzate fermentable either through
removal of toxin substances or through conversion of at least
a portion of such substances to non-toxin forms.
Several materials that are known toxins have been
found to be present in the acid hydrolyzates derived from
the lignocellulosic materials utilized herein. Their effect
has been quantified to enable elimination or at least mini-
mization of the same. The effect of such substances may beseen from the results of the following experiments in which
an anaerobic culture of S. uvarum was employed at about 0.7
- g _
~` .
7~
dry weight percent cell concentration with acid hydrolyzates
under the conditions indicated and employing identical inocu-
lum and fermentation media. Glucose sugar determination was
made using a Beckman Glucose Analyzer.
Where fermentation was achieved or attempted, the
steam-stripped, CaO pre-treated hydrolyzate was neutralized
with HCl as the neutralizing agent.
Cell concentration as referred to herein is deter-
mined by optical densit~ or dry weight measurements.
1. Acetic and Formic Acids
Both acids are toxins to the alcohol-producing
yeast. Acetic acid is present in the hydrolyzate in concentra-
tions of about 3 to 4 g/l while formic acid is present in
amounts of about 8 to 9 g/l. The toxic effect of these acids
can be negated by conducting the fermentation at a pH of abcut
5 to 7, preferably about 5.5 to 6.5. Fermenting below about
pH 5 to 6 does not negate the effect of the toxins while
fermenting above pH 7 is unfavorable for ethanol production.
The parameters may best be illustrated by the results from the
Eollowing experiments in which a newspaper hydrolyzate was
preconditioned in accordance with the invention and analyzed
! for Eormic and acetic acids. A control solution of pure glu-
cose was also provided. Identical nutrients in identical
amounts were added to each of the solutions. Each of the
solutions was inoculated with 0.7~ yeast cells and allowed to
ferment for 18 hours. The results were as indicated in Table
1.
-- 10 --
T~LE I
Fermentation of ~lucose in Presence of Formic and
_
Acetic Acids at pH 4.0 and pH 7~0
g/l
pH 4.0 pH 7.0
-
A. Control Experiment
Initial Glucose 40.0 40.0
Added Formic Acid 9.0 9,0
Added Acetic Acid 4.0 4.0
Final Glucose 41.0 0
Final Ethanol 0 18.5
% ~ield (based on Glucose) 0 46.3
B. Newspaper Hydrolyzate
Initial Glucose 36.4 39.8
Contains Formic Acid3.0 3.0
Contains ~cetic Acid3.0 3.0
Final Glucose 36.4 0
Final Ethanol 0 18.9
Yield o 47.5%
2. 5-Hydroxymethylfurfural_(HMF)
HMF is a strong inhibitor of yeast growth. However,
this material can be destroyed or degraded by CaO treatment
at room temperature at a pH o~ about 10 to about 10.5 without
adversely affecting the glucose. The pH range is believed to
be critical herein since at a pH below about 10, the effect
of HMF is not negated while at a pH above about 10.5, the
su~ar product is unstable. It has been discovered that CaO
treatment of the hydrolyzates at pH 10 to 10.5 results in
rapid depletion of HMF during the first two hours and levels
off after that period. For example, it was observed that
approximately 63~ of the HMF is removed in 1 hour at pH 10.5.
At pH 10.25, approximately 2 hours is required to remove the
same amount of HMF while at pH 10.75, some glucose is concomi-
tantly destroyed. There~ore, in the preferred embodiment of
the invention, hydrolyzate is treated with sufficient CaO,
at ro~m temperatu~e and with stirring, to maintain the pH at
about 10.5 for about 1 to l.S hours.
-- 11 --
,~ .
27~
The effect of the HMF on the yeast may be seen from
the following experiments in which yeast growth as a function
of HMF concentration in the hydrolyzate was determined at a
pH within the fermentation range of 6 to 7. The doubling time
is that time it takes for a cell to reproduce itself and was
determined by optical density measurements.
H~F, g/l Doubling Time, hours Growth Rate, h-
0 4.6 0.15
3 9.7 0.071
11.7 0.059
7 13.6 0.051
Growth Rate = ln 2
Doubling Time
3. Levulinic Acid
Levulinic acid inhibits yeast growth at concentra-
tions of 10 g/l or greater at pH 6-7.
Levulinic Doubling Growth
Acid, g/l Time, h. Rate, h-~
0 4.6 0.150
7.0 0.099
20.4 0.034
18.7 0.37
22 0.32
No sequence in the preconditioning step is believed
to negate the effect of this toxin material. Fermentations
are obtained, it is believed, because of the removal or nega-
tion of the effect of other toxins present which may have a
commulative or synergistic effect with the levulinic acid.
- Additionally, as discussed further hereinbelow, levulinic-
insensitive yeast also provide an alternative means for further
negating the effect of this toxin.
4. Furfural
Furfural is toxic at concentrations in excess of
5.0 g/l. At concentrations between about 3 to 5 g/l, it has
been found to markedly inhibit yeast growth.
- 12 -
~4~7~
FurfuralDoubling Growth Rate,
g~l _Time, hours (h) h-l
0 4.6 0.150
2.0 4.2 0.165
4.0 7.4 0.094
5.0 - O
Furfural is readil~ eliminated from the hydrolyzate
either b~ steam-stripping or calcium oxide treatment or both.
The effect of steam-stripping and CaO pretr~atment
may be illustrated by the following experiments in which hydro-
lyzate at pH of about 6.8 was admixed with an anaerobic culture
of Candida utilis yeast at about 0.7% dry weight cell concen-
tration in a shake flask and observed for fermentation of
glucose after 16 hours. The results were as tabulated below
in Table II.
TABLE II
A. Effect of Steam-Stripping CaO
Treated Hydrolyzate
Glucose after
Glucose Furfural HMF 16 h growth
Treatment g~l g/l g/l g/l
A. None 44 10.0 4~6 44
`~ B. CaO 44 5.0 1.6 44
C. Steam-stripping of B 44 1.2 2.2 0
D. Adding furfural to C 44 4.2 2.2 ~ 0
B. Effect of CaO Treatment on Steam-
Stripped Hydrolyzate
Glucose after
Glucose Furfural HMF 16 h growth
Treatment _ g/l g/l g/l g/l
A. None 44 10 4.6 44
B. Steam-stripping 42 1.2 4.4 42
C. CaO treatment of B 41 0.1 0.1 0
r
It will be seen from the above, that neither CaO
30 treatment alone nor stripping with steam alone is effective to
render the h~drol~zate fermentable although both treatments
reduce the furfural content. It was found that addition of
- 13 -
furfural to CaO treated and steam-stripped hydrolyzate to a
concentration nearly equal to that after CaO treatment but
before steam-stripping did not inhibit the growth. This indi-
cated that the steam-stripping may be removing some additional
unknown inhibiting material.
It will be seen from the above that the steps of (1)
steam-stripping; (2) treating with CaO at a pH of 10 to 10.5;
and (3) fermentation at pH 5 to 7 are critical to the success-
ful operation of the preconditioning stage of the process.
Various additional steps may be interjected between
the essential steps of the preconditioning method, if desired.
Thus, in the preferred mode of the invention, the hydrolyzate
having a pH of about 0.5 is partially neutralized with lime-
stone or ammonium hydroxide to a pH of about 4 prior to steam-
stripping. While the hydrolyzate may be steam-stripped either
at low pH as received or at pH 4 after partial neutralization,
partial neutralization prior to steam-stripping is desirable
to reduce corrosion of equipment. This step necessitates an
additional filtration step to remove precipitated material
and may be omitted in the event equipment is employed that is
not readily corroded or whenever corrosion is not a signifi-
cant concern. Where the hydrolyzate is neutralized to pH 4
employing calcium carbonate, etc., the resultant precipitate
may be incinerated after recovery from the hydrolyzate to
provide the fuel for generation of the steam for the process.
Normally, prior to incineration, the filter cake will be wash-
ed to remove and recover sugars, with the washings being added
to the hydrolyzate filtrate.
Neutralizing the hydrolyzate with ammonium hydroxide
has the added advantage of supplying a nutrient which the
fermentation microorganism can use for its growth while at the
same time raising the pH of the hydrolyzate to preventcorrosion.
- 14 -
.,~,
Steam-stripping is accomplished preferably by in-
jecting steam into the hydrolyzate in an amount sufficient
to maintain the hydrolyzate at a temperature of about 95 to
105C. Conveniently, the hydrolyzate may be passed through a
countercurrent extractor to remove steam-volatiles. In this
technique, steam is introduced at the bottom of the column
and the hydrolyzate is introduced at the top and collected in
a vessel at the bottom of the column. Steam-volatile toxins
are removed in the steam which is condensed and collected in
a separate vessel.
After steam-stripping, the hydrolyzate is treated
with sufficient CaO to maintain the pH between about 10 and
10.5 for a period of about 1 to 3 hours at room temperature
after which the precipitate is removed by any convenient means
including filtration, centrifugation, etc.
The hydrolyzate, after neutralization with a mineral
acid and removal of the resultant precipitate, is fermentable
at this stage, the rapidity of the reaction having been found
to be dependent on the concentration of the sugar solution~
the particular yeast strain employea, the yeast cell concen-
tration and the mineral acid used to neutralize the condition-
ed hydrolyzate.
Effect of Mineral Acid
The hydrolyzate is neutralized to a pH of about 5
to 7, and preferably 5.5 to 6.5 after CaO treatment employing
a mineral acid, e.g. hydrochloric acid, sulfuric acid, phos-
phoric acid, etc. Phosphoric acid is especially preferred for
several reasons. Neutralizing with hydrochloric or sulfuric
acid results in a turbid solution which is of no consequence
in batch fermentation. However, in a continuous culture fer-
mentation, yeast cells must be recycled from the effluent
stream back to the fermentor and must be reconcentrated prior
- 15 -
;`'
ii2~
to such recycling. The use of phosphoric acid as the neutral-
izing acid results in clarification of the hydrolyzate and
ther~by enhances reconcentration and recycling of the yeast
cells. Even more significantly, as discussed further herein-
below, it has been discovered that the use of phosphoric acid
results in more rapid fermentation and when employed in combin-
ation with a height concentration of yeast cells, results in
extremely rapid fermentation rates with concentrated hydroly-
zates making it possible to realize theoretical yields after
fermentation for as short a period as 1 to 3 hours.
Effect of Concentration of Hydrolyzate
It was discovered that preconditioning of the hydro-
lyzate by CaO pretreatment, steam-stripping and addition of
mineral acid, e.g~ HCl neutralization to pH 6 to 7, lead to
fermentable solutions that became progressively more difficult
to ferment as the glucose concentration was increased over
50 g/l. It appears that concentration of the solution also
increases the level of other inhibitory substances in the
hydrolyzate beyond the tolerable level. To counteract this
inhibition of concentrated solutions, different yeast and
yeast concentrations were evaluated as discussed further here-
inbelow. However, to illustrate the effect of the concentra-
tion step, typical results obtained with anaerobically propa-
gated Candida utilis at a cell concentration of 0.7~ with
sawdust hydrolyzate neutralized with HCl and fermented for
about 16 hours may be seen from results of the following
experiments.
- 16 -
27~
TreatmentGlucoseGlucose after Growth
g~
. None 44.0 44.0
B. CaO and stripping 44.0
C. Concentrating B 100.0 100.0
D. Diluting C to70.0 49.5
E. Diluting C to60.0 0
F. Diluting C to50.0 0
Attempts to ferment hydrolyzate solutions concentrated
to 150 g/l or grea-ter have not been successful even when em-
ploying the means discussed below.
Effect of Yeast Strain, Cell Concentration and
Neutralizing Acid
' '
The above-results were obtained in shake flasks using
Candida utilis as the yeast culture and HCl as the neutraliz-
ing acid.
Other yeast strains were tested and evaluated for
effect on the fermentablity of the preconditioned hydrolyzate
when neutralized with HCl.
S. uvarum at 0.7~ cells at dry weight was observed
to ferment 100 g/l hydrolyzate in about 98 hours and 50 g/l
hydrolyzate solutions in about 19 hours while S. cerevisiae
(Baker's yeast3 fermented 100 g/l hydrolyzate in about 42
hours to a 50.8 g/l ethanol concentration. A levulinic acid-
insensitive strain of S. -~varum was produced and isolated by
exposure of the cells to levulinic acid in chemostat culture.
This strain at 0.7% concentration was found to ferment 100 g/l
hydrolyzate in about 50 hours resulting in 49.3 g/l ethanol.
It was then discovered that use of phosphoric acid
as the neutralizing acid had a definite positive effect on
the rate of fermentation of the concentrated hydrolyzates.
Thus, phosphoric acid treated hydrolyzates at 100 g/1, when
fermented with either the parent strain S. uvarum or with the
levulinic acid-insensitive strain of S. uvarum, both at 0.7%
cell concentration, resulted in yields o~ 47.7 g/l ethanol in
- 17 -
~-~
.
~52~
15.5 hours while Baker's yeast at the same concentration re-
sulted in 46.5 g/l ethanol in 11.5 hours.
Even more rapid fermentation to theoretical yield
is possible when using higher yeast cell concentrations as
will be illustrated by the results obtained and listed below
in Table III.
TABLE III
Effect of Cell Concentration on 100 g/l Phosphoric
Acid-Neutralized Sawdust Hydrolyzate-Baker's Yeast
Cell Dry Weight Time Required Ethànol
% Hours Yield, %
0.7 20 50.0
1.5 12 48.1
2.0 8 50.6
: 2.5 5.6 50.2
3.0 3.0 46.8
5.0 1.5 47.8
7.0 1.25 48.4
The interrelationship and interdependence of the various steps
of the preconditioning stage of the process as well as the
effect o~ the particular neutralizing acid and yeast cell con-
centration may be readily appreciated from a consideration of
the above experiments.
he function of the CaO treatment to effectively
remove or degrade HMF in periods as short as 1 to 3 hours, the
effect of neutralizing to pH 5 to 7, the effect of the con-
centration of the hydrolyzate and yeast cells and the achieve-
men~ of extremely rapid reaction rates of concentrated hydro-
lyzates when neutralized with phosphoric acid are each signifi-
cant factors that are unexpected and appear to serve a vital
function in the context of the overall process.
Provision_of Fermentation Medium
Following the preconditioning method, the hydroly-
zate is ready for fermentation by either a batch or continuous
culture fermentation process under aerobic or anaerobic cell
- 18 -
~527~
propagation conditions.
Inoculum o~ the various yeast strains may be develop-
ed by any method well known in the art. Any yeast may be em-
ployed that is capable of growth in the fermentation medium.
As discussed above, satisfactory results have been obtained
with members of the genus Saccharomyces, such as S. uvanl ,
S. uvarum modified to be levulinic-insensitive, S. cerevisiae
(Baker's yeast), etc. with Baker's yeast being especially pre-
ferred. Satisfactory results have also been obtained with
C utilis at glucose concentrations up'to about 60 g/l. This
particular yeast has not been found to be effective at higher
glucose concentrations.
Suitable microbial growth nutrients may be added to
the hydrolyzate and to the inoculum development medium as
desired including phosphorous and nitrogen in ihe form of
phosphate, ammonium, urea, etc. When phosphoric acid is the
neutralizing acid, phosphorous nutrient is added during the
neutralization step. Additionally, when phosphoric acid is
employed as the neutralizing acid, fermentation may be realiz-
ed by addition of urea as the sole additive nutrient source.Partial neutralization with ammonium hydroxide prior to steam-
stripping also adds nitrogen as a nutrient source. Other
mineral salts, trace elements, vitamins, etc. including ammon-
ium sulfate, magnesium sulfate, sodium chloride, calcium
chloride, potassium phosphate, biotin, folic ac'id, inositol,
niacin, p-aminobenzoic acid, riboflavin, thiamine, urea, etc.
may be added to the hydrolyzate as growth nutrients, as desir-
ed.
In a preferred embodiment, inoculum for batch fer-
mentation is deyeloped by inoculating a loopful of cells froma slant on ~ medium containin~ about 2.0% glucose, 1.0%
peptone, and 0.3~ yeast extract (hereafter rcferred to as
-- 19 --
t~
YPG mediu~). Medium thus inoculated is incubated with sha~ing
for 24 hours at 32C after which it is transferred into 900
milliliters or additional YPG medium, incubated with shaking
for 6 to ~ hours and transferred to a fermentor containing 9
liters of ID medium comprising 90 g/l glucose, 7.65 g/l yeast
extract, 1.19 g/l ammonium chloride, 0.01 g/l magnesium
; sulfate, 0.05 g/l calcium chloride, and 0 2 mls./l of GE 60 AF,
an antifoam agent (available from General Electric Co~). Cells
are aerobically propagated at p~ 6-7 under 1,000 rpm agitation
and 1 vvm air flow for 16 to 20 hours after which the cells
may be recovered by centrifugation or equivalent means and
employed in the desired concentration to inoculate the hydro-
lyzate.
Cells may also be developed from a continuous culture
whereby cells are separated from ethanol product removed from
the fermentor. In this step, recovered cells are reconcen-
trated and recycled under conditions that preserve the metabo-
lic state of the cell, the volume in the fermentor and the
constant value of the cell concentration in the fermentor.
Cell separation from the ethanol product may be accomplished
by various means including gravity settling, centrifugation,
ultrafiltration, etc. Preferably, the cells are recovered and
recycled through the use of dual output streams emanating from
the fermentor. For example, ethanol and yeast cells are
metered out in a first output stream at a rate determined by
a sensor in the fermentor which determines when the cell con-
centration has exceeded a desired upper limit. A second out-
put stream removes ethanol and yeast cells to a cell recycler
comprising a suitable membrane r for example a microporous
30 filter as employed in ultrafiltration, which retains the cells
but allows the ethanol and unmetabolized medium to penetrate
permitting recovery of substantially cell-free ethanol. The
- 20 -
2~7~7
membrane permits continuous cell concentration from which cellsmay be recycled to the fermentor as needed as fresh precondi-
tioned hydrolyzate streams are fed for fermentation.
The following illustration will serve to illustrate
a batch fermentation in accordance with the invention.
Illustration of a Preferred Embodiment
Acid hydrolyzate having a pH of about 0.5 was pro-
duced from sawdust by acid hydrolysis in the presence of steam
and sulfuric acid in a reaction zone maintained at a tempera-
ture of about 190 to 225C under pressure of about 200 to
- 400 psi. The hydrolyzate contained about 50.2 g/l glucose,
8.9 g/l furfural, 3.6 g/l hydroxymethylfurural, 6.5 g/l
levulinic acid, 9.7 g/l acetic acid and 4.8 g/l formic acid.
The hydrolyzate was partially neutralized with suffi-
cient ammonium hydroxide to a pH of about 4 after which the
resultant precipitate was removed and the partially neutra-
lized stream fed to a countercurrent extractor where it was
subjected to steam at a rate of 4 l/h during which steam
volatile materials including furfural were removed and collect-
ed. 14 g/1 of CaO was added to the steam-stripped hydrolyzate
material to adjust the pH to about 10.5, the mixture was
stirred at room temperature and maintained at pH 10.5 for
about 1 hour after which the resultant precipitate was remov-
ed. The hydrolyzate was neutralized to pH 5.5 to 6.5 with
1.5 to 3.0 ml/l of phosphoric acid and the resultant precipi-
tate was removed. The neutralized hydrolyzate was then con-
centrated to a glucose concentration of about 100 g/l by
heating at 35C under vacuum of 28 inches Hg using a continu-
ous evaporator.
After cooling to room temperature, 0.1% urea was
added to the concentrated hydrolyzate which was next fed to
the fermentor together with a sufEicient amount of preformed
- 21 -
inoculum comprising Baker's yeast aerobically propagated on an
agar slant in ~PG medium to give a dry cell concentration of
about 3 to 3.5 weight percent.
The mixture was anaerobically fermented for about
1.5 to 2 hours after which about 50 g/l of ethanol was recover-
ed.
Yeast cells were recovered by centrifugation and
transferred to a subsequent fermentation batch. Satisfactory
results were obtained for several transfers.
It will be apparent to those skilled in the art that
various changes may be made without departing from the spirit
and scope of the invention and that the invention is not
limited to the preferred embodiments that have been described
and illustrated hereinabove.
`:
- 22 -
