Note: Descriptions are shown in the official language in which they were submitted.
1 325221
-- 1 --
This application is a division of co-pending
Canadian Patent Application Serial No. 535,279 entitled
VEGETABLE OIL EXTRACTION PROCESS, filed 22 April, 1987.
5This invention relates to the art of oil
extraction from a vegetable oil bearing material such as
soybean, corn and the like, and more particularly, to a
method for pretreating oil bearing vegetable material,
extracting the oil therefrom, and producing a superior
quality vegetable oil suitable for physical refining.
The invention is particularly applicable to the
processing of oil from soybeans and corn germ, but is
also applicable to many other vegetable oil bearing
materials such as cottonseed, peanuts, sunflower seed,
rape seed, fresh coconut meats or dried coconut meats,
palm fruits and palm kernels and the like. The process
of the present invention improves the extractability of
the vegetable oils from the oil bearing materials while
producing an oil product very low in phospholipids and in
- 20 mineral content such as, specifically, calcium, magnesium
and iron. The oil product is thus amenable to physical
refining. However, it will be appreciated by those
skilled in the art that the invention can be readily
adapted for use with other extraction processes as, for
~y 25 example, where similar methods are employed to obtain
:;~ other types of valuable constituent products.
'i Soybeans dominate the United States and world oil
and vegetable protein markets and, accordingly,
' conventional vegetable oil processing techniques are
predominantly directed to soy oil processing. Soy oil
and soy protein offer maximum benefit to the consumer at
a lower cost than can be obtained from any of the other
major oilseeds.
,...
.. " ~h
.,
:X
, .
''
:',
, .
~ ~ ,
.:: :: . ~ . ,
- - . , . -
.. : . ,, . ~ .
-2- 2 1
A wealth o~ information exi-sts describing the
conventional methods and equipment used in vegetable oil
processing. The commercially viable and successful
techniques for soy oil processing entail a number of
processing steps to extract the oil. Several techniques
exist for the extraction of oil including solvent
extraction, mechanical pressing, or a combination
thereof, although the dominant technique in commercial
use today is solvent extraction.
The crude oil extracted through these various known
techniques is a darl; colorcd, turhid liquid with an
unacceptable odor and flavor. The liquid needs
substantial further treatment to convert it to a bland,
stable and nutritious product that is useful in the
15 manufacture of shortening, margarine and salad and
cooking oils. (Crude oils from other oilseeds are
generally equally unacceptable as a food product and
equally need to be further treated.) This further
treatment consists of a number of steps tJhich
20 collectively may be called the refining process and
~hich typically include such steps as degumming,
neutralizing (alkali refining), bleaching and
deodorization. Refining is necessary to remove
phospholipids, free fatty acids, color bodies and other
25 constituents wllich either affect efficient e~ecution of
any subsequent processing steps and/or affect the
quality and the stability of the oil as a food product.
The crude oils produced by conventional solvent
; extraction and mechanical pressing methods from soybeans
30 typically contain high levels of phosphorus compounds
commonly called phospholipids, phosphatides,
phospilog]ycerides or gums in the range of 500 to 800 PPM
(parts per million measured as phosphorus) and small but
significant quantities of calcium, magnesium and iron.
.
. .
.-. . :
. ~ , . . ... . .. . .
, . .. ~.
. .. .
.
. . , ~: .
1 32522 1
As much as 30% of the above phospholipids may be
complexed ~ith calcium and magnesium. These are
commonly called non-hydratable phospholipids. In
addition, it is generally ~nown that prior methods of
5 pretreatment and oil extraction of soybeans are, in
fact, conducive to increasing the quantity of
non-hydratable phospholipids present in the crude oils
produced. The non-hydratable phospholipi~s generally
10 require a separate degumming step in the refining
process for their removal as will be discussed below.
; It is also well-known by those knowledgeable in the art
of refining crude oils that the varying quantities of
phospholipids in the crude oils may be attributed to
5 variations in the extraction processes themselves and to
the varying compositions of soybeans incurred during the
growing, the harvest and the storage of the heans.
, Since it is well-kno-~n that the presence of
phospholipids and certain trace metals are undesirable
~ 20 to the quality of the final food grade vegetable oil, it
1 is advantageous to reduce the level of these compounds
as ~uch as possible during the oil extraction processing.
The scope of processing steps referred to above,
i.e. degumming, neutralization (alkali refining),
~ 25 hleaching and deodorization are often collectively
i~ called "refining." In a narrower use of the word
"refining", it is often defined as the technique for
~ neutralizing the free fatty acids in the oil. As this
`~ is done l~ith alkali, the technique is also referred to
30 as alkali refining, or because of the use of chemicals,
as chemical refining. It should be kept in mind that
each processing step generally affects more than one
property of the crude oil. While neutralization
primarily reduces free fatty acid levels, gums are also
removed, the color may become lighter and some odor
:'',
::: : - : ....
-: , ~ .
' '
-:
1 325221
-4-
compounds may be removed. It is this propensity of a
particular processing step to affect A variety of oil
properties which ma~es it difficult to predict the
complete cause and ef~ect o~ the processing step and
thus is accountable Eor the inconsistent results
obtained from prior processing methods.
The typical ~nown vegetable oil refining process
involves several steps including a "degumming" step
- which essentially comprises addin~ water to the crude
oil and heating and agitating the mixture for a period
of time (approximately 10-30 minutes) and at
temperatures of typically S0-70 degrees Centigrade.
This mixture oÇ hot oil and water is subjected to
centrifugation wllerein the water and oil are separated.
In the process the hydrated phospholipids are separated
with the water. The resulting partially "degummed" oil
typically still contains a quantity of phospholipids,
including all the non-hydratable phospholipids. This
quantity may typically contain the equivalent to 10 to
120 PPM of phosphorus, however, this quantity varies
depending upon the precise degumming techniques and
conditions used.
The partially "degummed" oil produced in accordance
` with the above process may be further "degummed" to
25 remove the non-hydratable phospholipids by the addition
of certain chemicals (such as phosphoric acid) and water
and by again heating and agitating the mixture followed
by centrifuging. The "degummed" oil produced from this
step will typically contain a quantity of phospholipids
30 equivalent to 5-20 PPM of phosphorus.
The degummed crude oil from this second refining
step is further subjected to several additional refining
steps to remove other unwanted constituents such as the
~, free fatty acids, the color bodies and other materials
.,
, .,
. , . -
,. . . .
: , -
~ 325~2 1
-5--
that contribute unwanted flaYor, color and odor and
which cause flavor reversion. These steps are more
commonly identified ~s saponification of free fatty
acids, washing of the oil to remove the soaps,
5 neutralization and urther washing to remove excess
chemicals and soaps and further reduce the quantities of
phospholipids, bleaching to remove color bodies and some
additional quantities of phospholipids and, finally,
deodorization. Oil produced from all of these
10 extracting and refining steps is useful as a food
product but still contains phospholipids equivalent to
l-lO PPM of phosphorus.
It should be particularly emphasized and noted in
considering the subject invention that all of these
15 prior known processing steps, and in particular the
; degumming steps, are applied to a crude oil product
already extracted from the oil bearing vegetable
material. The steps are not applied to the material
itself but to the crude oil extracted from the material.
The capital cost associated with equipment to
practice these refining steps is very high. Chemical
refining involves many steps which are cuMbersome, is
captial intensive in that it requires substantial
equipment which is hard to maintain such as centrifuges
25 and filter presses, and is inherently characteri~ed by
oil losses as each of the refining steps produces a
residue which carries with it a certain quantity of
usable oil thus decreasing the yield of the salable food
~ product oil.
Because of the high cost of equipment, the high
operating expense and the losses of valuable product
oil, there has been an emphasis and desire in recent
years to practice a technique commonly called physical
refining. In this technique a crude oil which has been
. . .
.
:`
., .
. : ~
'
1 325221
-6--
subjected to several pretreatment processing steps is
brought to an e]evated temperature (250 degrees
Centigrade or more) in a vessel or column operated under
vacuum. Steam is sparged into the oil during
treatment. Temperature and retention time conditions
are selected such that the free fatty acids and other
impurities and odiEerous compounds are volatilized and
distilled off. The treated oil is then typically cooled
and given a post bleach to further lighten the color of
the oil.
The capital cost and operating costs of a physical
refining step is for many crude oils considerably less
- than that of chemical refining. Oil losses are also
substantially ]ess because only unwanted impurities are
15 distilled off. Generally, very little post physical
refining treatment is necessary to produce the finished
- shelf product. Hence, physical refining is very
desirable to an oil processor.
However, a number of crude oils, includi~g crude
20 oils from soybean and corn germ extraction, require
substantial pretreatment steps before the physical
refining step can be applied. Most of these
pretreatment steps are associated with the removal of
hydratable and non-hydratable phospholipids from the
25 crude oil.
Physical refining does not remove significant
quantities of gums or phosphorus, nor does physical
refining remove the heavy metals (such as iron). The
presence of gums in excess of 6-20 PPM of phosphorus are
;~ 30 subject to breakdown during physical refining due to the
high temperatures employed and this causes unwanted
,~ flavor and color characteristics and causes acceleration
of flavor reversion or rancidity (in the case of soy
bean oil), as well as a reduction of oil stability (or
'' .
- - . .~: , - ,; . .: . ,
: ,
7 2 ~
shelf life) in other vegetable oils. The lower limits
of the presence of pllopholipids are not quite clear, but
it is well kno~n that there is a direct relationship of
flavor reversion and loss of shelf life due to the
5 presence of excessive quantities of phospholipids and of
heavy metals such as iron in all vegetable oils.
Therefore, the feed to a physical refining step should
not contain a quantity of phospholipids measured in
excess of 3-10 PPM measured as phosphorus. Thosè
10 knowledgeable in the art may agree that high levels of
phospholipids in the feed to the physical refining step
cause deep set color changes in the oil which are hard
to bleach out. The need for reduction of the
phospholipid level in corn and soybean crude oils
15 requires many o~ the prior art chemical refining steps
described earlier and thus much or all of the economic
incentive for physical refining is lost.
The application of physical refining is therefore
limited to those vegetable oils that are naturally of
20 such a quality as to have low limits of phosphorus
' (particularly the non-hydratable phospholipid form),
3 have a low iron content and, in addition, contain a
, level of free fatty acids dictated ~y economic
¦ justification to permit the full application of physical
25 refining or some modification thereof.
A major reason for not applying the physical
refining step to soybean and corn oil crudes is that
these crudes are high in phospholipids and in thè case
of corn oil contain much foreign solid matter such as
30 finely divided starch particles. High levels of
phospholipids in the crude affect the quality of the oil
and generally limits have been set on the maximum
phospholipid levels for physical refining of a crude
,''
.. . .
'~
~; -
1 325221
-- 8
oil. These requirements set by the refiners of crude oilrange from less than 5 PPM tmeasured as P) to less than
20 PPM.
As noted above, the reduction of the quantities
of phospholipids in soybean oil and corn oil crudes is
not an easy task because part of the phospholipids are in
a form generally referred to as non-hydratable
phospholipids or may be converted to this form under the
influence of certain constituents of the oilseeds or the
oil. The greater part of the phospholipids generally
referred to as hydratable phospholipids may be removed
readily by contacting the crude with water, salt
solutions, acidic or caustic solutions and the like and
then removing the agglomerations of hydratable
phospholipids by means of centrifuging. The removal of
the non-hydratable phospholipids is more difficult. The
`` non-hydratable phospholipids are complexes of calcium and
magnesium with phospholipids and the known removal
techniques depend upon chemical treatments to cleave the
bond between the calcium and magnesium groups and the
phospholipids, rendering the non-hydratable phospholipids
into hydratable phospholipids and preventing reattachment
i' of the calcium and magnesium group to the hydratable
phospholipids.
` 25 The present invention contemplates a new and
improved process which allows for the more efficient
processing of a better quality oil product and meal
product from an oil-bearing material.
As a result of the process according to the
present invention the phospholipids substantially remain
with the extracted solids. The extracted crude oil is
.; .
very low in phospholipids and may be physically refined
without any further pre-treatments.
In accordance with the present invention, there
is provided a process for extracting a low phospholipid
oil from an oil bearing material such as soybean, corn,
;i and the like to allow for the more efficient processing
i of a better quality oil and meal product. The method
comprises a series of steps. The first step can comprise
:
,~
,
, . - .
,
' ~: . . : ,
1 32522 1
g
pretreating oil bearing vegetable material to place it in
a condition for mixing and grinding. These pretreating
steps can comprise cleaning, drying, cracking or
dehulling. The material can then be ground and mixed
with a reagent for reducing the phospholipid content in
the oil extracted. The reagent comprises a reagent for
cleaving the chemical bond of the non-hydratable
phospholipids and thereby rendering the non-hydratable
phospholipids into hydratable phospholipids. A
, 10 complexing, precipitating or chelating agent can also be
added to the mixture to prevent reversion of the
hydratable phospholipids back to non-hydratable
phospholipids.
One benefit of the subject invention is an oil
processing method which produces a vegetable oil product
of substantially low and uniform levels in phospholipids
` regardless of the varying content of phospholipids in the
, vegetable oil material.
Yet another benefit of the subject invention is
that temperature and moisture conditions throughout the
system are such that the hydratable phospholipids stay
, with the cake or meal product of the process and thus the
, oil from the process is substantially free of hydratable
phospholipids. The temperature and moisture conditions
~ 25 prevent the conversion of hydratable phospholipids into
1 non-hydratable phospholipids. Suitable reagents may be
added in the system to convert non-hydratable
phospholipids into hydratable phospholipids and thereby
facilitate the phospholipid removal.
Other benefits and advantages for the subject new
vegetable oil extraction process will become apparent to
those skilled in the art upon a reading and understanding
of this specification.
The drawings and detailed description which
follow are directed not only to the process which is the
;3 subject of the present application, but also to methods
and apparatus which are the subject of co-pending
i Application Serial No. 535,279 and of two other co-
~ 32522 1
-- 10 --
pending applications which, like the present application,
are divisions thereof.
The invention may take physical form in certain
parts and arrangements of parts, and in certain steps and
arrangements of steps, the preferred and alternative
embodiments of which will be described in detail in this
specification and illustrated in the accompanying
drawings which form a part hereof and wherein:
FIGURE 1 is a schematic diagram of a plant formed
in accordance with the present invention for extracting
oil from an oil bearing material such as soybeans;
FIGURE 2 is a schematic diagram of a plant formed
in accordance with the present invention for extracting
oil from an oil bearing material such as corn germ;
FIGURES 3A and 3B comprise a block diagram
illustrating the process steps in the practice of the
present invention in extracting oil from a vegetable oil
material such as soybeans; and,
FIGURE 4 is a block diagram illustrating the
steps of a process in accordance with the present
i invention for extracting oil from an oil bearing material
such as wet corn germ, dry corn germ or wet/dry germ
mixtures.
Referring now to the drawings wherein the
showings are for purposes of illustrating the preferred
and alternative embodiments of the invention only and not
for purposes of limiting same, the FIGURES show a
vegetable oil extraction process and assembly for the
production of a superior quality vegetable oil product
suitable for physical refining. Although, as noted
above, the invention is applicable to a wide variety of
vegetable oil bearing materials, its predominate
commercial concern is directed to soybean and corn and
will be particularly discussed with reference to these
oil bearing materials.
:
.
:: . . :.
1 32~22 1
11 --
~ ith reference to ~IGURES 1, 3A and 3B, the
invention~ as it is applicable to soybeans is
specifically illustrated. The first processing step for
a harvested soybean delivered to a processing plant
comprises receiving 10 and cleaning 12. Thereafter, the
beans may be stored as at 14 or transmitted for further
processing. Stcrage of the beans generally results in
higher phospholipid levels in the beans due to changes
in bean compositions during the storage period.
Beans in storage are dried to have a typical
1 moisture content of about 6% by weight. Further
`~ processing for oil extraction can entail drying or
tempering as at 16, cracking as at 18, and dehulling as
:~, at 20 using equipment commonly used for these purposes
and known to those ~nowledgeable in soybean processing.
The drying typically occurs at a temperature of 220-240
degrees F. (104-116 degrees C.) or lower to minimize
heat damage to the beans. Dehulling typically removes
approximately 3-4% of the weight of the conditioned
20 material. Dehulled beans are next comminuted or ground
' as at Z4 in a dry state to produce a ground bean
material suitable for mixing in a grinding mill and
, slurry preparation tank 26 (FIGURE 1).
A mixture is formed in the slurry preparation tank
25 26 comprised of several items. In the preferred form of
the invention, several chemical reagents are added as at
28. These reagents may be introduced dissolved in
water. Also, additional water may be introduced.
~i Typically, the total water added may comprise 15 weigllt
30 percent of the dry weight introduced in the case of
soybeans. In a preferred embodiment of the invention,
the slurry preparation tank 26 is also a cooking vessel
in which the soybean particles may be cooked or
1 32522 1
- 12 -
partially cooked, that is, heated for a period of time
to an elevated temperature in the presence of the water
introduced. In this cooking step enzymes, bacteria and
fungi are deactivated and the oil and the solids are
5 detoxified and pasteurized. Among the enzymes
deactivated are urease, various proteases and lipases
and the enzyme which promotes the conversion of
~ hydratable phospholipids into non-hydratable
- phospholipids. Preferably, a first reagent is added to
10 reduce the phospholipid content in the oil extracted
from the soybeans. The first reagent may be an acid.
The non-hydratable phospholipids in soybeans are
substantially magnesium and calcium phosphatidates which
upon treatment ~iith an acid are acidulated and converted
15 into a disassociated phosphatidic acid, which disappears
from the oil phase in the form of micelles in the water
~` phase or in hydrated form as liquid crystals. In other
words, the magnesium and calcium bonds linking the
non-hydratable phospholipids to the oil (lipids) are
20 ruptured by the acid and the non-hydratable
' phospholipids thus become hydratable. The resulting
magnesium and calcium complexes separate from the oil
phase and are now in the water phase. ~hen this
happens, certain salt complexes may be formed. The salt
25 complexes may be in several forms in the water phase,
that is, as a precipitate, in suspension, in solution or
in the form of a micelle.
However, this hydration or rupturing reaction is
reversible. As the water evaporates, the disassociated
30 non-hydratable phospholipids may return to the solution
- in the oil as non-hydratable phospholipids. Thus,
. precipating, chelating, blocking or binding agents can
be added, as explained more fully below, to prevent the
reverse reaction. Preferably, the reagent used for
:
'
: . . . ~ ., . ~ . . . . .
... ................................... . .
1 325221
- 13 -
rupturing of these chemical bonds comprises phosphoric
acid (H 3PO4 ) Alternatively, a reagent from the
following group could be employed: Citric acid
(HOOCCH 2C(OH)(COOH)CH 2C00H - H2 ), hydrochloric acid
5(HCl), potassium chloride (KCl), sodium chloride (NaCl),
sodium hydroxide (NaOH), disodium hydrogen phosphate
(Na 2HP04 ), potassium dihydrogen phosphate (KH 2 PO4 )~
acetic anhydride (CH 3CO)2O, sulfuric acid (H 2SO4 ),
sodium borate (Na2 B 4O7 ), and glycine (NH2 CH2 COOH).
The degree of rupturing of the appropriate chemical
bonds increases with increasing contact between the oil
and the water phase. Accordingly, in the grinding mill
and slurry preparation tank 26, the ground soybean are
comminuted and homogeneously mixed witll the slurry so
`15that the ground soybeans are ground to a preselected
particle size which facilitates the desired chemical
reaction in the heating step as will be in hereafter
more fully explained.
Since it is known that the above hydration reaction
20is reversible, it is necessary to lock-out the
reversible reaction. Accordingly, a precipitating,
binding, blocking or chelating reagent can also be mixed
into the grinding mill to ultimately sequester the trace
metals and/or bind the salt complexes to something
`25else. The binding reagent preferably comprises sodium
.~citrate (C 6Hs O 7Na3 -2H 2)~ but may also comprise
sodium chloride (NaCl), sodium acetate (NaC H 32 )~
sodium sulfate (Na 2 SO 4), sodium hydrogen sulfate
(NaHSO 4), trisodium phosphate (Na3 P04 ), EDTA
30((ethylenediaminetetraacetic acid, ((HOOCCH2 )2
NCH2 CH 2N (CH2 CO0H)2)), sodium floride (NaF),
sodium-oxalate (Na2 C 24 ), sodium-tartrate
(Na 2 C4 H4 6- 2H 2 ), sodium carbonate (Na2 CO 3 ) and
sodium pyrophosphate (Na4 P 27 )
;
-, -, ,
.,. , ~ . .. .
, ~ , ! ~ ',
.''"' ' ' ,''' " " ' ' ~ "" ' . ". . ", ' ''' '
.' '' : '
.'. ' '" " . ,
.,., ' .
1 32522 1
- 14 -
In addition, again as the water is evaporated, the
acidity of the remaining water and slurry changes and
thus certain water insoluble precipitates may return to
the solution. Certain reagents under certain conditions
comprising surfactants (anionic, cationic, nonionic) or
proteins are added to the mixture to control to some
extent the degree of acidulation. Preferably, the
surfactant or protein agent is selected from the group
consisting of ethoxylated fatty alcohol, oleylamine,
` 10 casein, pancreatin, soy yrotein and Na-soap.
In the preferred practice of the invention, all of
these reagents are added in the grinding mill. ~lowever,
it is within the scope of the invention to add only the
first and second reagents, the first reagent only, or no
15 reagents at all and still produce an improved product
oil.
Also added in the grinding mill 26 is a portion of
oil of preferably the same type as will be extracted
from the oil bearing material to form a pumpable
20 slurry. It is within the scope of the invention to
employ an oil other than that of the same type as will
be extracted from the oil bearing material. Oftentimes
an oil blend is desired in which case another type of
oil, either vegetable or animal, may be employed.
The treatment of the oil bearing material by
grinding in the presence of hot oil conditions the
oilseed to release oil with a much lower energy
requirement in the later oil extracting steps such as
centrifuging and screw pressing. This lowered energy
30 requirement in the extracting steps means that less
horsepower is required per ton of seed being processed
and thus less heat damage is done to the product oil.
:.,
.
~.
. :.,; . . . . .... . . .. . . .
:- . ; ~: .: . :
~ 32522 1
- 15 -
Also added to the mixture in the grinding mill is
water to achieve a moisture level of at least 15% of the
dry weight of the slurry in the slurry preparation
tank. The addition o~ water is desirable so that the
5 slurry can be properly cooked at a later processing
step.
With continued reference to FIGURE 1, it is noted
that the oil and water added to the slurry preparation
tank is obtained through system recycle operations.
In one commercial embodiment of the invention, the
slurry preparation tank level is controlled such that
under steady state conditions a ratio of preferably 2.5
weight parts of recyled oil to 1 weight part of soybean
solids (bone dry basis) is maintained. To the slurry is
15 added 10-30 weight percent of water based upon the
weight of dry soybean solids introduced. As noted
before, this water may be condensate from other plant
` processing steps, with or without demineralized or
distilled water makeup. The various liquid or solid
20 reagents comprising acids, bases, salts and others which
are added to the slurry preparation tank 26 are added to
.j enhance the quality of the end product oil with an
emphasis on phospholipid and trace metal removal. One
desirable system of additives is one pound per hour
25 t0.454 kg/hr) of concentrated (85%) phosphoric acid and
one pound per hour ~0.454 kg/hr) of sodium citrate per
1,700 pounds per hour (772 kg/hr) of soybean solid
feed. The slurry preparation tank is continuously
agitated to promote wetting of the soybean solids by
30 water and to promote proper dispersion of the reagents.
The temperature of the slurry in the tank, without any
external application of heat, is approximately 150
degrees F. (66 degrees C.) and follows from the mass
flow rate, specific heat and temperature of the soybean
.,
i~
.
. .,
.
~ ~ . - . . . - ~ ,
:.
. :, - - . , ~
.:. ~ - -
. , ~ - .
~- 1325221
- 16 -
solids food, the recycle oil stream and recycle water
(condensate) and make-up water streams. A function of
the slurry preparation tank is to provide a degree of
cooking to the soybean solids and to partially or wholly
5 deactivate all enzymes affecting such properties as the
stability of the oil and meal and the enzyme or enzymes
which control the conversion of hydratable phospholipids
" to non-hydratable phospholipids and to detoxify and
sterilize said oil and meal. The degree of cooking
10 depends upon both temperature and retention time of the
solids in the tank. The temperature can be increased by
applying external heat to the tank. ~laximum temperature
of the slurry in the slurry preparation tank could match
the maximum temperature in the evaporator pump described
15 below (typically 185-210 degrees ~., 85-99 degrees C.).
The maximum retention time in the slurry preparation
tank 26 may be established by the design of the tank.
Another function of the slurry preparation tank 26
is to condition the soybean solids for the release of
20 oil in the subsequent processing steps and the
extraction steps in the centrifuge and the screw press.
` Yet another function of the tank 26 is to provide an
enclosure for operation in a reduced oxygen atmosphere
~, by introduction of a nitrogen or other inert gas
25 atmosphere.
~` The slurry is pumped througl- a sizing mill (not
shown) which preferably should be a Reitz disintegrator
or equivalent. The mill should include a screen such
that a desirable particle size distribution is achieved,
30 typically 10 weight per cent plus 20 mesh; 82 weight
percent plus 40 mesh. A feed pump 30 feeds the sized
slurry to a falling film evaporator 32.
!i,
,,'
'
. ' '
'"',
'' ''
'",;,
. ~ I
.' ' : . ' ' ' ' :
'`' ' .' ' ' `' "'
':',. ~ ' ' ' ~' ' ' '
1 325221
The evaporator 32 is operated under vacuum
conditions (for example, 25 inches of mercury or 635
millimeters of mercury) to limit temperature exposure of
the oil in the system. The evaporator is preferably
operated in combination with the recycle pump 34 to
promote proper film formation in the tubes of the
evaporator heat exchanger and to insure optimum heat
transfer conditions. The evaporator sump temperature is
typically 185-210 degrees F.(85-99 degrees C.); the
10 vapor temperature is approximately 150 degrees F. (66
degrees C.). The evaporator sump is sized such that
retention time in the evaporator 32 may be controlled to
between 20 and 40 minutes.
Onc function of the evaporator is to remove
15 substantially all of the water introduced in the slurry
preparation step witll the soybean feed and the recycled
condensate and make-up water streams. A small amount of
water in the soybean solids must remain for effective
separation of oil from the solids in the centrifuge and
,r, 20 the screw press.
Another function of the evaporator 32 is to complete
if necessary, the cooking initiated or partially
completed in the slurry preparation tank 26, i.e.
complete the deactivation of enzymes and the
25 detoxification and sterilization of the oil and meal.
Still another function oE the evaporator 3Z is to
complete the beneficial reactions between the reagents
and the phospholipids initiated in the slurry
preparation tank Z6.
Yet another function of the evaporator 32 is to
. complete the conditioning of the soybean solids to
improve the release of oil from the solids in the
~', extraction step in the centrifuge and the screw press.
. .,
:'
., .
. . .
,::
, ~.
.,-:
: -~
.... .
s~
.,
,.:: - . , , ., : .
, ,; . . , -
: s
,................................. , : .~ ,
:,. . . . ~; .: ,
, .
1 325221
- 18 -
The dried slurry is discharged to a high gravity
decanter centrifuge 3G such as may be commerically
obtained from the Sharpless Division of Pennwalt
Corporation, Plliladepilia, Pa. and from other centrifuge
manufacturers. The solids in the feed to the centrifuge
should contain 3-4 weight percent of moisture on an
oil-free solids basis. Water condensed from the
evaporator 32 in condenser 38 is employed in later
processing steps.
The solids obtained from centrifuge 36 are commonly
referred to as centrifuge cake and will contain 25-35%
oil by weight. The cake is conveyed to a screw press 40
wherein the solids are pressed to a 3-4 weight
percentage of residual oil in the press cake.
Because of the pretreatment of the soybean solids in
the slurry preparation tank and grinding mill 26 and in
the evaporator 32, the screw press 40 requires
substantially less power to press out the oil (typically
1.5 hp metric ton per day instead of the 4 hp metric ton
per day required for pressing soybeans in the
conventional solvent extraction pretreatment
i technique). The cake from the process is a very light
tan colored product. The oil from the screw press 40 is
conducted to the feed stream to the centrifùge 36 in
' 25 order to remove press fines from the press oil stream.
The oil from the discharge of the centrifuge is
conducted to a recyle tank 42 for either return to the
. slurry preparation tank 26 or as product oil which is
communicated to a wash tank 44. The temperature of oil
:, 30 in the recyle tank is approximately 185-200 degrees
, F.(85-93 degrees C.). The product oil from the recycle
~ tank 42 typically contains 1-2 ppm (parts per million)
f of phospholipids (measured ~s elemental phosphorus) and
:) trace metals such as calcium, magnesium and iron. The
. :,
:.. ~,. . . .
: ,~ , -, : . ,., -
1 325221
-- 19 --
product oil is communicated from the recycle tank q2 to
the wash tank 44 in which distilled water or condensate
is introduced. The wash tank is agitated with an
agitator 46. The wash tank 44 is sized for a retention
time of oil and water of at least five minutes. The oil
and water mixture from the wash tank is pumped to a
centrifuge 48 by a pump 50. The centrifuge 48 may
comprise a three phase high gravity horizontal decanter
centrifuge as is commercially available from the
10 Sharpless Division of Pennwalt Corporation,
Philadelphia, Pa. or a high gravity di~c centrifuge such
as is available from Alfa Laval Gorporation of Tumba,
Sweden or Westphalia Corporation, West Germany. The
water phase separated in the centrifuge 48 is returned,
; 15 iE necessary with makeup distilled water, to the slurry
preparation tank 26. The wet solids discharge from the
centrifuge 48 which contains the phospholipids, various
soaps, starches and solids of unknown composition may be
disposed or added to the feed of the screw press 40.
20 The product oil is typically filtered from the
centrifuge and yumped by pump 52 to a physical refining
assembly (not shown). The product oil from the
centrifuge 48 shows no turbidity and there is no
detectable content of phosphorus, calcium, magnesium and
25 iron. The oil is ready for physical refining.
With particular reference to FIGURES 3A and 3B, the
method employed in the assembly of FIGURE 1 is
illustrated. After the soybeans have been ground or
comminuted as indicated by the grinding step at block
30 24, the mixture is fed to the grinding mill and slurry
, preparation tank. The reagents are added as indicated
by block 28 and mixed with water at block 60. The water
may comprise condensate from the evaporator at block 66,
or from the wash water at block 72 with demineralized or
.:;.. ~ .
... .
~ ~.
~:.
:',
, :: . . . - . . :
.. -: . . . . .. . .
.. ... . .
1 325221
- 20 -
distilled water makeup or, alternatively, an independent
demineralized or distilled water source may be
employed. Before grinding step 62, an oil o preferably
the same type as the oil to be extracted from the oil
bearing material is added to the slurry to facilitate
the grinding step and to produce a pumpable slurry. The
oil is preferably a recycled oil obtained from the
centrifuge 36. The mixture preferably has a moisture
level of at least 15% of the dry weight of the slurry.
Providing a suitable moisture level in the slurry is
important to enable the necessary cooking of the slurry
in a low pressure, low temperature environment. The
slurry is next cooked or heated as at 64 in the slurry
preparation tank 26 and in the evaporator 32 (FIGURE l).
` 15 The cooking step accomplishes several results.
First, it allows for the first reagent to rupture the
magnesium and calcium bonds linking the non-hydratable
phospholipids and thus rendering them hydratable;
second, it allows the second reagent ~the precipitating,
20 binding or chelating reagent) to bind the resulting
magnesium and calcium complexes to lock-out the
reversion of the phospholipids into a non-hydratable
form and consequently also reduce the trace metal
content in the oil ultimately extracted from the slurry;
25 third, it allows for the deactivation of the naturally
occuring enzymes such as lipase and urease or other
enzymes which may be toxic or cause toxic products to
; form in the slurry or which may control the conversion
of hydratable phospholipids into non-hydratable
;~ 30 phospholipids; fourth, it provides sterilization or
:i~ pasteurization of the slurry to deactivate certain
bacteria and fungi; and, fifth, it substantially
i insolubilizes proteins in the slurry. It should be
noted that in order to accomplish enzyme deactivation,
, :j
. .
, ;
- .
.~
,,. ,~. . . .
.. ~... . , ~- - .,
.,
1 32522 1
- 21 -
pasteurization, rendering selective ~ungi and
microrganisms llarmlcss and toxic destruction, the slurry
must be cooke~ Eor a preseiected time at a preselectcd
temperature witll a preselected moisture content. It has
5 been experimentally Eound that the invention may be
;successfully l~racticed by cooking under a partial vacuum
and in a reduced oxygen atmosphere at temperatures no
higher than approximately 99 degrees C. (210 degrees F.)
for a time preferably within a period of 20-40 minutes.
l0,NOt only does such a low temperature/low pressure
cooking operation accomplish the ~esired results, but it
also avoids damage caused by the conventionally used
higher temperatures to the ultimately resulting oil and
meal products.
After the slurry has been properly cooked, it is
subject to evaporation 66 where water condensate is
removed from the slurry Eor communication ~ack to the
slurry preparation tank 26 or the oil discharge wash
tank 44. The partially evaporated slurry is next
20 communicated to a centriEuge 36 (FIGURE 1) for oil and
solid phase separation by the step of centrifuging 68.
The solids or centriEuge cake generally contains 25-35
oil by weight which is extracted as at 70 typically by
pressing. The oil discharge Erom the extracting step is
25 conducted to the Eeed stream oE the centrifuging step 68
in order to rcmove press fines Erom the press oil
stream. The product oil Erom the cent r ifuge is a
isubstantially improved crude oil product which typically
contains only 1-2 ppm oE phospholipids (measured as
;,30 phosphorus) and a minimal amount of trace metals such as
calcium, magnesium and iron. The product oil is washed
~as at 72 with distilled water or condensate, filtered as
;at 74 or centriEuged Eor separation oE the water and
,
, .
"
.,.
:.
',
.
.::
~, , - :: .:
., : . -
:,.- , - .:
. . : . -
~: .
.
.
,..... ~ , . . . -
1 32522 1
- 22 -
solids residue ~rom the oil, physically refined as at
76, cooled as at 78, and ultimately stored as at 80.
The wash water and solids residue separated from the oil
in the washing and fil~ering steps may be sewered or
disposed; alterna~ively, the wash water may be fed back
to the mixing stcp 60 WlliCIl can take place in the slurry
preparation tank 26 (FIGURE 1). The wash water can be
filtered and thus demineralized water could be fed bac~
to the slurry ?reparation tank. The wash water is also
10 advantageous since it retains heat and thus reduces the
energy requircments Eor the treatment process. As noted
above, the finislled product oil shows no turbidity and
there is a very low content (usuaily less than l PPM) of
phosphorus and generally no detectable content of
` 15 calcium, ma~nesium and iron.
The meal cake obtained Erom the extracting step 70
is a high protein useful meal product which can be
ground as at 82 to a commercially salable product, dried
and stored as at 84.
; 20 Examples for Soybeans
The following bench scale examples were performed to
. prove the subject invention.
Four hundrcd grams ~dry basis) oE soybeans were
~' obtained Erom a conventional soybean processor company
25 and cleaned, cracked and partially dehulled. The
soybeans are commercially available Erom Cargill Inc. of
Decatur, Illinois or other suppliers. The approximate
proper~ies of the soybeans were as follows:
.... . .
19.8% oil on a 10% moisture basis
4730 PPM phospholipids as elemental phosphorus (P)
11.92% moisture
. .
..
"~,:
.
.~,
.
., .
: .
~ -
.
~.
1 325221
- 23 -
The oil in these soybeans when extracted with hexane
(a commercial solvent) typically contained:
l.Z~ free fatty acid (FFA)
. 5% moisture
597 PPM of P
The above soybeans were mixed into 1200 ml of
semi-refined soybean oil commercially available from the
Procter and Gamble Company, Inc. Cincinnati, Ohio and
other suppliers. The approximate properties of this oil
10 l~ere as follows:
0.3% FFA
.7 Red (Lovibond scale)
3.0 PPM of P
The mixture (or slurry) was introduced into a heavy
15 duty "Waring" type blender of the type commerciallY
available from Vitamix Corporation of Olmsted Falls,
Ohio under the trade name or VITAMIX*3600. The slurry
was mixed at the lowest speed setting for five minutes.
The mixed and ground slurry was then introduced into
; 20 a flask. After the flask was sealed the mixture was
agitated with a laboratory agitator at 180 RPM. .
,vacuum of approximately Z9 inches of mercury (737
;, millimeters of mercury) was pulled on the flask. The
flask and the contents were immersed in a constant
;~j 25 temperature water bath maintained at approximately 99.5
degrees C. The flask was connected to a laboratorY
glassware condenser and water was evaporated from the
slurry.
~, The batch of ground soybeans and soybean oil would
boil at approximately 68-75 degrees C. and the
temperature would remain level at approximatelY 68-75
degrees C. until a substantial portion of the water had
` * trade-mark
~,,
.,
:'' ' ' ~' , ' . , -
-
.
~........... : ' ~ ': ' ' '
1 325221
- 24 -
been evaporated. At that point, the temperature would
start to rise sharply and asymptotically approach the
temperature of the water ~ath. I~hen the temperature
reached approximately 85-90 degrees C., the vacuum was
5 broken and the slurry sample was poured into a Buechner
funnel lined with filter paper (Watman No. 5).
The filtered oil was collected in a flask. The oily
filter cake, containing approximately 45-50 weight
percent of oil was put in a press cage in which the ram
10 of a Carver hydraulic laboratory press moved to compress
the oily cake to a degree wherein the remaining press
cake would only contain approximately 10 weight percent
of residual oil. The oil separated from the cake was
mixed with the ~iltered oil. A typical sample of the
15 oil showed:
- 2.5 % FFA
3.0 to 3.5 Red (Lovibond scale)
Z30 PPM of P
The test was repeated several times using the oil
20 from each preceeding test, but new samples of 400 grams
of soybeans were introduced each time. Since in each
test the 400 grams of soybeans contained approximately
80 grams of oil and as the press cake still contained 27
grams of oil at 9.15 weight percent, the original soy
25 oil was diluted with 53 grams of new oil originating
from the soybeans. Thus, to replace the original oil
sample of 1200 ml (10~0 grams) multiple tests similar to
the above are required until the phosphorus content
asymptotically approached that of the oil in the
30 soybeans used. After seven cycles as described above
the P in PPM was 627.
.~
:
. . - .
- :: . . - - . : . . .
.. - . . ..
. . .
.. : . . .: : .
:,;
1 32522 1
- 25 -
The above oil sample was used as a "bench mark" to
determine the number of cycles required in the prac~ice
of the invention hereinafter described. Four hundred
grams of the above soybeans were used, to which 1200 ml
5Of the above commercially available semi-refined oil was
mixed in the VITAMIX blender previously described and
the mixture was subjected to mixing and grinding at the
high speed setting for 10 minutes. Three grams of
concentrated (85%) phosphoric acid was added at the
lOonset of the mixing. Subsequently, (after 2 minutes)
three grams of laboratory grade sodium citrate was added
to 40 ml of distilled water. The solution was added to
the mixing slurry and the mixing was continued for an
additional 8 minutes. The mixed and ground slurry was
15introduced to an agitated flask. The flask was sealed
and approximately 5 inches of mercury (127 millimeters
of mercury) was pulled on the flask. The mixture was
heated to 90 degrees C. and maintained at that
temperature for 20 minutes. Then the vacuum was
20increased to approximately 29 inches of mercury (737)
millimeters of mercury) vacuum. The water bath
temperature was maintained at approximately 99 degrees C.
~ The batch of soybean and soybean oil would boil at
'.J'' 68-75 degrees centigrade. The temperacure would remain
:~ 2slevel at approximately these temperatures until a
substantial portion of the water was evaporated. At
~ that point the temperature would rise sharply and
:~ asymptotically approach the temperature of the water
'1 bath.
When the temperature reached approximately 85 to 90
degrees centigrade, the vacuum was broken and the slurry
:~,
sample was poured into a Buechner funnel lined with
filter paper (Watman No. 5).
''
-
, -
. , . ~ ,
.
1 325221
- 26 -
The filtered oil was collected in a flask. The oily
filter cake, containing approximately 45 to S0 weight
percent of oil, was put in a press cage in which the ram
of a Carver hydraulic laboratory press moved to compress
5 the oily cake to a degree wherein the remaining press
cake would only contain 10 weight percent of residual
oil. The oil separated from the cake was mixed with the
filtered oil. A typical sample of the oil showed:
0.8 % FFA
3 Red (Lovibond scale)
1-2 PP~I o~ P (phospholipids)
The test was repeated until the original sample had
substantially disappeared and had been replaced by oil
from the subsequent quantities of soybeans introduced.
15 It was found in the series of tests that the free fatty
acid content and the red color (Lovibond scale) would
asymptotically approach the free fatty acid content and
the red color of the oil in the soybeans, i.e.
approximately 1.2% FFA and 3 Red. However, the
: 20 phospholipid content of each subsequent oil sample would
:' stay constant within a range of approximately 1 to 2 PP~I
of P.
A 1000 ml sample of the soybean oil from the above
tests was washed with 50 ml of distilled water and the
25 mixture was intensively mixed and heated ~o 70 degrees
~ C. for 10 minutes under 5 inches (127 mm) of vacuum.
i~ After mixing the oil it was centrifuged in a laboratory
`1 centrifuge for 10 minutes at 6000 times gravity. This
centrifuging substantially removed all of the water. A
30 whitish-brown solid precipitate was formed in the water
. and the precipitate was judged to be iron, calcium and
magnesium complex salts. The oil had some turbidity
which disappeared at the 60-70 degrees centigrade range.
'.'. .
.,
,,.' : , , , .:
~.; .' :' . : . "- .
.,',''.' ' . .
1 32522 1
- 27 -
The oil sample was cooled to 50 degrees centigrade
and filtered in a Beuchner funnel with Watman No. 5
ilter paper.
The washed, centrifuged and filtered oil sample was
- 5 no longer cloudy and the content of P was judged to be
in the 0 to 0.5 PPM range (AOCS Official Method Ca
12-55)
The wash water contained 35 PPM of phosphorus.
Also, the oil sample did not show any detectable
content of calcium, magnesium, iron or other trace
metals.
Washing samples from other tests in the test series
showed no statistically significant departure of the
15 test results from the earlier wash test, i.e.
phospholipids measured as phosphorus were barely
detectable ~0 to 0.5 PPM range) and trace metals such as
calcium, magnesium and iron could not be detected.
A 1000 ml sample of the above washed oil was
; 20 introduced into a laboratory bench scale physical
refining assembly which was operated under 29.5 inches
of mercury vacuum ~750 mm). The sample was heated to
91 252 degrees C. (485 degrees F) and high temperature
steam was sparged into the oil sample through a special
25 steam dispenser. The sample was subjected to this
physical refining treatment Eor 6.5 hours. The sample
was then cooled and judged to contain the following:
0.01-.02~ FFA
:~ 0.00 PPM phosphorus
300.5 Red ~electronic color meter)
~. 0.0 Peroxide value
`: 0.002% Moisture
In another series of tests the sodium citrate
reagent was replaced with 3 grams of sodium sulphate and
35 processed as before. Oil samples from this test showed
an average of 1.0-2.0 PPM of phosphorus.
.. - .
,
:,
, . ; ' ~ "'~' ''
- '
1 32522 1
- 2~ -
In another series of tests the phosphoric acid
~reagent was replaced with 3 grams of acetic anhydride
and the sodium citrate reagent was replaced with sodium
acetate.
In still another series of tests ~he phosphoric acid
reagent was replaced with mono-hydrated citric acid
crystals.
`. In all these series of tests with the various
reagents, the oil samples contained an average of
10 1.5-3.0 PPM of phosphorus. However, when washed witll 50
` cc of distilled water the phosphorus content was judged
to be less than l PPM.
With reference to FIGURES 2 and 4, the subject
`, invention as it is applicable to corn germ will be
15 specifically discussed. l~ith particular reference to
j FIGURE 4, it may be seen that the invention is
applicable to either a dry/wet corn germ mix, a wet corn
germ, or a dry germ. The water level of the germ is
adjusted as at 100, 102 to obtain a germ slurry which is
20 properly cookable. Generally, the water level in the
mixer should be at least 15~ by weight. The oil and
reagent chemicals are mixed and ground as at 104 to form
a comminuted and homogenous slurry. The reagents
comprise the same reagents which are used in the soybean
25 processing illustrations. After the slurry has been
agitated to achieve a thorough mixing, the slurry is
` heated or cooked as at 106 for a preselec~ed period of
- time under a partial vacuum at a temperature no higher
than approximately 99 degrees C. The cooking step
30 accomplishes essentially the same results as the cooking
step for the soybean processing, that is, rupturing of
: the calcium and magnesium bonds to render the
- non-hydratable phospholipids hydratable, lock-out of a
'
".'
. .
.,:
:,
;................................ ,
.
-
1 32522 1
- 29 -
reversion reaction as the water is evaporated, enzyme
deactivation, pasteurization and protein
insolubilization. After cooking, the slurry is subject
to partial evaporation as at 108 and oil extraction as
5 at 110 by centrifuge or a filter. The condensate from
the evaporator can be either directed to a storage
mixing tank or an oil product washing tank. The product
oil from the centrifuge step 110 can be segregated into
first and second portions. The first portion can be
lO directed back to the slurry mixing tank. The second
portion can be washed as at 112, filtered or centrifuged
as at 114 to remove solids residue, physically refined
as at 116, cooled as at 118 and stored as at 120 as a
food quality oil. The cake from the centrifuging step
15 110 can be further processed as by extraction step 122
to remove a substantial portion of the residual oil.
, The press oil from the extraction step 122 can be
~, conducted to the feed stream to the centrifuging step
110 to remove fines from the oil stream. The cake is
20 subsequently comminuted and ground as at 124 and dried
, and stored as a meal product.
Examples for Corn Oil
Example No. 1
Four hundred grams (dry basis) of corn germ were
25 taken from a wet milling process. Typically, the sample
would contain 50% water, i.e. 800 grams of wet germ
; would contain 400 grams of dry germ and 400 grams of
water. A dry sample of the germ would typically contain
45-50 weight percent of corn oil. Corn oil removed from
30~the sample by means of extraction with a solvent such as
hexane, typically had the following properties:
.
:
,~ ' ' '''~ ~ '
1 325221
- 30 -
2.7% FFA
7.6 Red (Lovibond scale)
700 PPM of P
The four hundred Fram sample of corn germ was mixed
with l200 ml of MAZOLA*brand corn oil, a refined,
bleached and deodorized corn oil, commerciallY available
~from CPC International Inc. of Englewood Cliffs, New
Jersey. The approximate proper~ies of the MAZOLA corn
' oil were as follows:
:
0.02~ FFA
0.3 Red (Lovibond scale)
3 PPM of P
The slurry was introduced in the mixing and grinding
container of a VIT~IIX Super 3600 mixer (Note: The
laboratory equipment used in this example is the same as
described in the preYious example for soybean). Three
grams of concentrated (85%) phosphoric acid was added to
the batch and the slurry was first mixed for two minutes
at the lowest speed setting of the Vitamix.
20 Subsequently, the slurry was ground for a period of ten
minutes at the highest speed setting.
; The mixed and ground slurry was then introduced in a
flask. After the flask was sealed, a Yacuum of
approximately 29 inches of mercury (737 mm) was pulle~
25 on the flask. The water bath temperature was maintained
at 95-lOO degrees C.
The batch of ground corn germ and corn oil would
boil at approximately 68-75 degrees C. ant the
temperature would remain leYel at approximately 68-75
30 degrees C. until substantially all of the water had been
evaporated. At that point, the temperature would start
to rise sharply and asymptotically approach the
temperature of the water bath. When the slurry
, * trade-mark
. , .
~'`
. :, . . .
:. ~
., ~ . . .
1 325221
- 3~ -
temperature reached approximately 85-90 degrees C., the
vacuum was broken and the slurry sample was poured into
a Buechner funnel lined with filter paper (Watman No. 5),
The filtered oil was collected in a flask. The oily
5 filter cake, containing approximately 45-50 weight
percent of oil was put in a press cage in which the ram
of a Carver hydraulic laboratory press moved to compress
the oily cake to a degree wherein the remaining press
cake would only contain approximately lO weight percent
,10 of residual oil. The oil separated from the cake was
;~mixed with the filtered oil. A typical sample of the
oil showed:
2.5% FFA
~3.0 to 3.5 Red ~Lovibond scale)
:` 15 1 to 2 PPM of P
;!
~,~ In each test 400 grams of corn germ (dry basis) was
used containing approximately 200 grams of oil. As the
:~ press cake still contained 22 grams of oil at 10 weight
percent residual oil content, the original MAZOLA corn
20oil sample was diluted with 178 grams of new oil
:~ originating from the corn germ. Thus, to replace the
original MAZOLA corn oil sample of 1200 ml (or
approximately 1,080 grams) required multiple tests
similar to the one described above. Each test used 1200
2sml of oil from the previous test. The test was repeated
until the original MAZOLA sample had substantially
disappeared and had been replaced by oil from the
subsequent quantities of corn germ introduced. It was
found in the series of tests that the free fatty acid
30content and the red color (Lovibond scale) would
asymptotically approach the free fatty acid content and
the red color of the oil in the germ, i.e. approximately
2.7 weight percent FFA and 3 to 3.5 Red. However, the
i
':
~,,,
~'
.. , , ' : -
~ ~ .
. .
," ' ' . ' ~ .
1 325221
- 32 -
phospholipid content of each subsequent oil sample would
stay constant within a range of approximately 1 to 2 PPM
measured as P. After approximately fifteen subsequent
tests the FFA and Red color no longer varied. I~owever,
5 there was still no change of the phospholipid contents
of the samples, which remained within the 1 to 2 PPM of
- P range.
In another series of runs, three grams of
concentrated phosphoric acid reagent (85%) and 3 grams
10 of sodium citrate reagent were added. Samples from
these runs also showed 1 to 2 PPM of P.
The oil samples prepared were slightly cloudy. This
cloudiness is generally attributed by those skilled in
the art of producing corn oil to the presence of finely
15 divided, dehydrated starch particles, which are carried
with the corn germ, because existing processes for
separating the corn germ from the corn starch cannot
PL event typically 2 to 12 weight percent of starch from
. remaining with the germ.
; 20 A 1000 ml corn oil sample from a test was washed
with 50 grams of distilled water. The wash solution was
intensively mixed with the oil sample using the low
speed setting of the VITAMIX mixer. After the mixing,
~ the oil was centrifuged. A precipitate formed and the
- 25 oil was no longer cloudy. The washed and decanted oil
sample showed that substantially all phospholipids had
been removed as the phosphorous content was judged to be
in the O to 0. 5 PPM range (AOCS Official Method Ca
12-55). Also, the sample did not show any detectable
30 content of calcium, magnesium or iron and other trace
metals.
v Washing samples from other tests in the test series
showed no statistically significant departure of the
test results from the earlier wash test, i.e.
.
. .
.
'' . - ~ ' '
.,.......... . ~ ~ .
~' ''
, . , . . , _
1 325221
- 33 -
` phospholipids measured as phosphorus were barely
detectable (0 to 0.5 PPM range) and trace metals such as
calcium, magnesium and iron could not be detected either.
Example No. 2
With reference to FIGURE 2, a pilot plant 128 with a
. capacity of 2000 lbs./hr ( 908 kg/hr) of wet corn germ
containing 50% of water by weight on the average was
operated to produce high quality corn oil. The initial
charge to the system was a semi-refined corn oil
i lO produced by CPC International Inc. with the following
approximate properties: 0.25% FFA, FAC red 2 to 3, 20
PPM of P. The properties of the corn oil intrinsic to
the corn germ used was as described in the previous
example.
FIGURE 2 shows the equipment of the pilot plant. A
metering, variable speed screw conveyor 130 was
~; calibrated to feed approximately 2000 lbs./hr (908
kg/hr) of wet germ to the system. The wet germ was
introduced into a slurry preparation tank 132. This
20 preparation tank 132 was partially filled with corn
oil. At the initial start-up, the tank was filled to
the required level from a tank containing the
; semi-refined oil described above. Once the plant was in
operation, part of the oil separated in a centrifuge 134
. 25 and a screw press 136 was recycled to the slurry
preparation tank 132 as discussed below. The quantity
of oil in the slurry preparation tank 13Z was maintained
such that under steady state conditions a slurry
composition of approximately 3.5 parts of oil by weight
30 to 1 part of dry corn germ solids by weight was
maintained. The slurry from the slurry preparation tank
was pumped through a fixed hammer mill (not shown) of
the type known as a Rietz disintegrator, which may be
:.
-
~
:: .
,: . :
: , - .
1 325221
- 34 -
commercially obtained from Bepex Corporation of
Minneapolis, Minnesota. This hammer mill typically
sized the particles to a distribution of 10 weight
percent + 20 mesh, 82 weight percent 1 40 mesh.
J 5 The slurry with the sized particles was pumped to an
evaporator 138. The temperature of the slurry feed was
approximately 150 degrees F. (66 degrees C.); the feed
~, rate was 2000 lbs./hr (908 kg/hr) of sized wet corn germ
~` suspended in 3500 lbs./hr (1589 kg/hr) of oil
10 (approximately 12 GPM or S4.6 li~ers per minute). The
evaporator 138 was a single effect falling film
evaporator. To maintain proper film formation in the
tubes and good heat transfer conditions, the slurry in
~, the evaporator 138 was recycled to the tube nest of the
'''J' 15 evaporator at a high rate of flow. The evaporator was
operated with a vacuum of approximately 25 inches of
mercury ( 635 mm) in the vapor space. Vapor
temperature was approximately 150 degrees F. (66 degrees
C.); the slurry temperature in the sump was maintained
20'at 190-210 degrees F. (88-99 degrees C.) range. Slurry
levels in the sump were maintained such that retention
times of the slurry in the evaporator ranged from 10-30
minutes. The dried slurry from the evaporator sump was
pumped to a horizontal decanter type centrifuge 134.
25 For optimum separation of the solids in the centrifuge,
3-4 weight percent of moisture was maintained in the
solids as measured on the basis of oil free solids. The
, oil content of the centrifuge cake was in the range of
35-55 weight percent. This centrifuge cake was
; 30 subsequently pressed in the screw press 136 to separate
substantially the balance of the oil. The oil remaining
~ in the press cake was typically in the range of 4-6
- weight percent. The oil from the press was added to the
feed stream to the centrifuge 136 to separate solid
:.
. ~ .
:.~
~?~.
,: . .
,:. -: ~
."..
1 32522~
- 35 -
fines which escaped with the oil through the oil
discharge openings in the barrel of the press. The oil
from the centrifuge flowed to a recycle tank 140 from
where 3500 lbs./hr (1589 kg/hr) of oil was pumped back
to the slurry preparation tank 132 to continue the
process and 474 lbs./hr (215 kg/hr) of oil was pumped
out as product oil.
As the test runs proceeded, the initial charge of
oil was replaced by an oil originating from the corn
10 germ and the free fatty acid content would level out at
approximately 2.7 to 3.0 percent by weight and the red
color would approach 3 on the Lovibond scale.
In one series of test runs dilute sulfurous acid
(approximately O.lN) was added to the feed stream of the
15 centrifuge at a rate of approximately 0.5 GPM (2.27
liters per minute). Oil samples were analyzed and
showed approximately 6 PPM of P (phospholipids
determined as P).
Approximately 5 lbs./hr (2.268 kg/hr) of 85%
20 concentrated phosphoric acid was metered into the slurry
preparation tank in another series of test runs. Oil
samples from the centrifuge typically averaged 1 to 2
PPM of P.
In still another series of test runs 0.5 GPM (2.27
25 liters per minute) of a dilute phosphoric acid solution
was injected in the centrifuge feed. This dilute
solution was prepared by mixing 5 lbs (2.268 kg) of
concentrated (85%) phosphoric acid into 250 lbs. (113.25
kg) of water. Oil samples taken from the centri~uge
30 discharge showed an average of 4-8 PPM of P.
In another series of test runs 5 lbs./hr (2.268
kg/hr) of 85% concentrated phosphoric acid and 5 lbs./hr
(2.268 kg/hr) of sodium citrate were added in the slurry
~ preparation tank. Samples showed 1-2 PPM of P.
':
; . . . . ` . .
: - . -: . : . . :
.- .
- .. ~
-: - ~ ,;-
: - . .
.. . ~ . .
^ . :: - ,. : -
- .:, :: - . .
1 325~2 1
; - 36 -
.
Oil samples from the three distinct series of runs
were washed with ~istilled water. The oil samples
measured 1000 ml. The wash water and the oil sample
were intensely mixed in a Waring blender for a period of
5 five minutes and then centrifuged. A sample taken from
~, the centrifuged oil was filtered. The filtered sample
~ showed no turbidity and the phosphorus content was
i judged to be less than 1 PPM of phosphorus. As before,
~ the P content of the oils was measured according to AOCS
'~ lO Official ~ethod Ca 12-55. This sample was also judged
.~ to be free of calcium, magnesium and iron.
The invention has been described with reference to
the preferred embodiments. Obviously, modifications and
alterations will occur to others upon the reading and
15 understanding of the specification. It is our intention
to include all such modifications and alterations
~ insofar as they come within the scope of the appended
j claims or the equivalents thereof.
:,~
. .
.~
. ~
,'''
.`i '
, ~
: ;.
A,~
,~
'~
' ". ;~" ' . ' ` ' ~ ' ' "' ' . ' ~ '
'~":~. , " , . '
.' ~ ~, ' '
`'~ ' ' ~ , ' "
,.~ , .
".'.,' ' ~ ~" '