JPS5893798A - Purification of vegetable oil with membrane - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for refining vegetable oils by film. More specifically, the present invention relates to a method for removing shigami from crude vegetable oil using a porous membrane.

Gummies in vegetable oils produce undesirable odors and flavors, precipitate during storage, and inactivate white clay during the decolorization process, resulting in loss of white clay. , called degumming).

As a degumming method using a film of vegetable oil, it was already published in 1983.
The method of No.-14715 is known.

In this method, gum substances are separated and removed from vegetable oil dissolved in an organic solvent using an ultrapolar membrane, but since the pore size of the membrane is very small, the oil has a high resistance to permeate through the membrane. , the amount of permeated oil per unit membrane area and unit time is small. That is, there is a drawback that the degumming treatment capacity is low. Therefore, it is necessary to increase the membrane area, which increases the cost of degumming treatment, and there are also problems in that the device is not thick and requires a large installation space. Furthermore, as a means to recover from the decrease in the amount of permeated oil over time, fluid is allowed to flow through the membrane from the surface on the permeate side to the surface on the feed liquid side, that is, in the opposite direction to the normal liquid flow direction. Washing treatment is effective, but since the pore size of the ultrafiltration membrane is small, the amount of backwash fluid permeated per unit membrane area per unit time is small, and the peeling effect of the cake layer deposited on the M surface is weak. There is also the problem that a sufficient backwashing effect, and thus an effect of recovering the amount of permeated oil, cannot be obtained.

As mentioned above, the existing methods are extremely unsatisfactory.

As a result of intensive studies, the present invention has solved all of the above-mentioned drawbacks, and by using crude vegetable oil as a liquid to be treated and by selecting the membrane surface characteristics, an ultrafiltration membrane has been developed. Gums, which normally cannot be separated using membranes with small pores, can be separated using membranes with pores that are orders of magnitude larger than ultrafiltration membranes. The nine valves are designed to increase the amount of oil fed per unit, and the method and conditions for supplying the liquid to be treated to the membrane are devised to reduce the decrease in the amount of oil fed over time. That is, the present invention is an innovative degumming purification method using a vegetable oil membrane that has extremely high degumming efficiency.

That is, in the present invention, after ripening crude vegetable oil containing gum while stirring in the coexistence of 0.1 to 10% by weight of water, the average pore size is O, OSm3μ, and the critical surface tension of the membrane surface is 33dyn. Contact with the surface of a porous membrane of less than 7 cm, and apply a linear velocity of O, OSm/&cc-3m/S parallel to the membrane surface.
This is a method for refining vegetable oil using a membrane, which is characterized by flowing the plant oil at cc, concentrating and separating the gummy substance together with water, and obtaining a vegetable oil substantially free of gummy substance and water. Further, the present invention provides that, in the above method, by adding acid in an amount of 0.01 to 5.0 times the gummy content during aging of the crude vegetable oil, a more remarkable degumming effect is exhibited. A process for the membrane purification of vegetable oils which yields vegetable oils that are substantially free of gums, water and added acids.

41 targeted by the present invention [physical properties include rapeseed oil, soybean oil, linseed oil, safflower oil, sunflower oil, corn oil, cottonseed oil, sesame oil, bran oil, castor oil, olive oil,
These include camellia oil, coconut oil, palm oil, eno oil, hempseed oil, tung oil, kapok oil, tea oil, etc., but especially rapeseed oil, soybean oil, sunflower oil, safflower oil, cottonseed oil, and sesame oil. The gummy content of the crude oil used for lubrication is large;
This is suitable because the effect is remarkable. The vegetable oil may be a mixture with an organic solvent such as hexane or acetone, such as Mise 2, and it may also contain natural contaminants other than gummy substances contained in the raw materials for the vegetable oil, such as waxy substances and free substances. fatty acids, sulfur compounds, peptides, pigments, aldehydes,
It may contain ketones, etc., and may also contain trace amounts of contaminants that are intentionally or unintentionally mixed in during the oil extraction or refining process, such as alkalis, metal ions, inorganic and organic solid particles, etc. Good too.

The gum substance in the present invention refers to acetone-insoluble substances among natural contaminants contained in raw materials for vegetable oil, and
Its main component is IJ lipid.

In the present invention, the gum quality is 0.1 based on the weight of vegetable oil.
When the crude vegetable oil is contained in an amount equal to or more than 10% by weight, the degumming effect is significant, which is preferable. If the amount is less than 0.1 times, there is no substantial degumming effect.

In addition, crude vegetable oil was mixed with water at a rate of 0.
It is necessary to ripen it in the presence of 1 to 10% by weight, preferably 0.5 to 5.0%, with stirring. If the water content is less than 0.1% by weight, there will be no substantial degumming effect, and if the water content exceeds 10% by weight, the oil delivery rate will decrease significantly, and depending on the situation, water and gum may permeate through the membrane. There is also something undesirable.

Furthermore, before stirring the crude vegetable oil in the coexistence of 0.1 to 1011 parts of water, add 0.01 to 3.0 times the acid concentration based on the gummy content. It has been found that a more significant degumming effect can be obtained by adding r:1k and aging the product while stirring. If the amount of acid added is less than 0.01 times, there is substantially no effect of adding acid, and 5. If it exceeds 0 times, there will be a lot of wasted acid, which is not desirable in terms of cost.

#It is desirable to add the acid under conditions where a large amount of water does not coexist in the crude vegetable oil in order to reduce the amount of acid used.
Therefore, when adding water to coexist with 0.1 to 10% of water by weight, it is preferable to add acid before adding water.

As the acid, it is preferable to use any one of phosphoric acid, sulfuric acid, nitric acid, boric acid, citric acid, oxalic acid, etc., or a mixture thereof.

It is estimated that the processing of crude vegetable oil described above coarsens the small gums that cannot normally be separated using a membrane with a small pore size such as an ultrafiltration membrane. We believe that the porous membrane shown below, which has an order of magnitude larger pore diameter than the previous one, made it possible to separate the cams.

Next, the characteristics of the porous membrane will be explained.

It has been found that in order to dehydrate the liquid to be treated using a porous membrane at the same time as degumming, it is necessary to use a porous membrane that is characterized by its pore size and surface characteristics.

The porous membrane used in the present invention has an average pore diameter of 0.05 to 3.
μ, preferably 0.1 to 1 μ. If the average pore diameter is less than 0.05μ, the amount of permeation per unit membrane area and unit time is small, so a large membrane area is required, which is undesirable.
If it exceeds 5μ, the gum quality and water permeation through the membrane will be significant, and the degumming and dehydration effects will be small, which is not preferable. Note that the average pore diameter shown in the present invention is measured by electron microscopic observation.

Furthermore, the critical surface tension (re) of the porous membrane is 33 dyn.
/crn. 52 dyn/
When it is more than m, the membrane permeation of gum and water is significant, and it is not preferable because the gum and water of the moxibustion cap are mixed into the purified vegetable oil. In other words, for solid particles, the pore size of the porous membrane generally determines the separation performance (tl), but for non-solid particles such as gum or water, the pore size alone does not determine the separation performance; It can be judged that the affinity of can also determine separability.

A porous membrane with a critical surface tension (rc) of less than 55 d-Yn/cm has a critical surface tension of 55 dyn/cI
11 or a combination of these resins, or a blend polymer or copolymer consisting of a combination of these resins and a resin that does not have the above critical surface tension, and the critical surface tension of the final resin is 35 dyn/ Formed from books less than c1n.

It is also possible to use a porous membrane in which the critical surface tension of the membrane surface is adjusted to have a value within the above range by means of coating or a chemical bond table on the porous membrane.

Specific examples of materials for forming the porous membrane include polyethylene, polypropylene, polybutene, polyisobutylene, polypentene, poly4-methylisopentene, and other borylfin and fluoride compounds containing one or more futune atoms; 4-fluoride; Ethylene-perfluoroalkyl vinyl ether copolymer; ethylene, propylene, butene,
Ethylene hydrocarbons or fluorine such as inbutylene, pentene, hexene, 1-fluoroethylene, vinylidene fluoride, 3-fluoroethylene, 4-fluoroethylene, 3-fluorochloroethylene, or Fi6-fluoropropylene Copolymers consisting of a combination of 10-substituted ethylene hydrocarbons containing one or more atoms, and combinations of polyethylene/and polypropylene, polyvinylidene fluoride, poly(4-fluoroethylene) or polystyrene, polypropylene and polyvinylidene fluoride, or polystyrene. Blend polymers such as a combination of 4-fluoroethylene, polyvinylidene fluoride and polysulfone, polyacrylonitrile, polyphenylene oxide, or poly4-fluoroethylene may be mentioned. Preferably polyethylene, polypropylene, ethylene-propylene copolymer, polyvinylidene fluoride, ethylene-tetrafluoroethylene copolymer, 4-7 fluorinated ethylene-6-fluorinated propylene copolymer, 4-fluorinated ethylene-perfluoroalkyl vinyl ether copolymer, and these. Blend polymers are used.

Critical surface tension is measured by the following method. In the case of a porous membrane formed from a uniform material, a uniform nonporous film is formed using the wood resin and used as a sample for measurement. If the porous membrane is not made of a uniform material, but the surface of the membrane is coated or chemically bonded with various chemical species, it may be made of a uniform material made of the same resin that forms the membrane. A porous film is formed, and the same porous film is coated on its surface.
The sample is coated or chemically bonded under suitable conditions and used as a measurement sample. The critical surface tension is the surface tension when the contact angle θ is 0°, so the contact angle of the above sample was measured with liquids with various surface tensions, and the separation method was used to determine θ = 0.
If the surface tension at the time of 6 is determined, the critical surface tension can be obtained.

The porous membrane of the present invention preferably has a porosity of 15 to 95%. If the porosity is less than 15, the amount of permeated oil will be small, and if it exceeds 95, the membrane will not be tightly packed, both of which are unfavorable.

Note that the porosity (P) is defined by the following formula.

ρb p = (1--) x 100 (ch) ρa・ρa: specific gravity of membrane without pores ρb = weight of porous membrane divided by its volume, thickness of porous membrane is 0.01 to 41111 preferable. In other words, the film thickness is 0
．． If the swelling is less than 01, the membrane strength will be low, and if it exceeds 4 IIm, the amount of oil fed will be small, both of which are unfavorable.

The module may be formed into a shape suitable for actual use, such as a hollow fiber shape or a hollow fiber shape, but a hollow fiber shape is preferable in terms of miniaturization of the module and simplification of its structure.

In the case of a hollow fiber membrane, the inner diameter thereof is preferably 0.1 to 10 m11. If the inner diameter is less than 0.1111I, it is undesirable because clogging with gum or the like occurs at the end face of the hollow fiber membrane.
If the thickness exceeds the threshold, the volume of the module using the membrane becomes large, which is not preferable.

As a method for forming a porous membrane that meets the above conditions, known methods can be applied, such as a melting method as disclosed in JP-A-52-70988.

Other methods that can be applied include the microphase separation method, the Masanobu method, and the radiation irradiation method.

Next, the conditions for supplying the liquid to be treated to the porous membrane will be explained.

The liquid to be treated is brought into contact with the porous membrane surface, and the edge speed is 0.05 m/sac % S m/s parallel to the surface.
ec, preferably 0.1 m/s~2 m/s
It was found that flowing with EC was necessary for efficient degassing treatment. If the edge speed is less than 0.05m/see,
This is not preferable because the amount of oil fed becomes small, and 5Tn/&
If ec'fc is exceeded, not only will energy consumption increase, but the liquid to be treated will need to be flowed under high pressure, which may cause membrane damage.On the other hand, in an attempt to avoid high pressure, the intermembrane l! Increasing i1 is not preferable because the module volume becomes extremely large.

The filtration pressure when the liquid to be treated comes into contact with the porous membrane is 0.01~
5 atmospheres is preferred. If the filtration pressure is less than 0.01 atm,
If the amount of oil fed becomes small and exceeds 5 atm, it is not preferable because IIIL is often damaged and 171c energy consumption increases.

The filtration temperature when the liquid to be treated comes into contact with the porous membrane is preferably 20 to 95°C if the vegetable oil is not a mixture with an organic solvent, and if the vegetable oil is a mixture with an organic solvent, the boiling point is 15°C. It is preferable to use an organic solvent having a temperature of 20°C or higher and 5°C lower than the boiling point of the organic solvent. If the filtration temperature is less than 20°C, the viscosity of the liquid to be treated will increase and the amount of oil fed will be small, so it is not preferable. , evaporation of water or organic solvent is significantly reduced, which is undesirable.

Liquid containing gum and water concentrated by the membrane (concentrated fL
) The so-called partial circulation filtration method, in which the oil is brought into contact with membranes of the same degree of strength, is preferable because the amount of oil fed per membrane area can be reduced.

Furthermore, from the viewpoint of operability or effective utilization of one membrane, it is preferable that the concentration ratio (the ratio of the gum concentration in the concentrated liquid to the gum concentration in the liquid to be treated) is constant over time. For this reason, part of the concentrate pipe is discharged outside the system, and the remaining concentrate is mixed with an amount of unconcentrated liquid (liquid to be treated) that matches the amount of permeated liquid and the amount of discharged concentrated liquid, and is partially circulated and filtered. In this case, the concentration ratio is preferably 1.5 times or more. If it is less than 1.5 times, the concentration is low and the processing efficiency is poor, which is not preferable.

Figure 1 shows a conceptual diagram of an example of the partial circulation law. child\
So, Atl1 treated liquid, BFi concentrated liquid, cFi permeated liquid,
D is the discharge am fluid, Ed pump, F is the module, and the PII pump.

JII as a means to recover from the decrease in permeate volume over time.
Although backwashing treatment for K is known, the membrane of the present invention has a much larger pore size than the ultra-V membrane, so the backwashing effect is very significant. The backwash liquid at this time is I
The P liquid obtained by permeation through II is preferred.If the vegetable oil is a mixture with an organic solvent, it is also preferred to use the same organic solvent.

The present invention can also be applied to a multi-stage filtration method in which the concentrate is brought into contact with different membranes several times. Figure $2 shows a conceptual diagram of an example of the multi-stage F-proper method. Here, A is the liquid to be treated, B1 m
Bt is concentrated liquid, El #g is module, Pl
#p is a pump.

The present invention will be specifically explained below using examples.

Example 1 Polyglobylene 21.5 times! H, dibutyl phthalate (
55.5% by weight of DBP) and 23.0% by weight of finely divided silica were mixed in a kneader and then ground in a pulverizer.

This raw material was melt-extruded into a hollow fiber shape using a twin-screw extruder and a hollow nozzle, then cooled and taken off. 1.1.1-) The hollow fibers were immersed in dichloroethane to extract and remove DBP, dried, further immersed in a 40% by weight aqueous solution of caustic soda to dissolve the finely divided silica, and diluted with a dilute aqueous solution of caustic soda and water. It was washed and further dried to obtain a hollow fiber porous membrane.

This membrane has an inner diameter of 0.7m1%, a film thickness of 300μ, and an average pore size of 0.
．． 15μ, porosity 69%, r C = 29.0dyn/
It was a painting.

Using this hollow fiber membrane, a module with an effective membrane rki product of 0.1 was fabricated.

On the other hand, a mixed oil of 75 volumes of compressed rapeseed oil and 25 volumes of extracted oil was stirred and aged at 50°C while adding 1.7 weights of water to the oil little by little. liquid fjs
I did it. The gum content in this liquid to be treated was 1.1% by weight, and the water content was 1.8% by weight.

In addition, the gum substance was determined by the amount insoluble in acetone, and the water content was determined by the Karl Fischer method, and in the following examples (the same method was used).

The first Me module is brought into contact with the liquid to be treated, and the first speed Q
, 6 m/formula, average filtration pressure 1.0 atm, p filtration temperature 45C, and partial microvolume filtration at III reduction ratio S, O times, the gum quality in the obtained permeated oil was 0.66% by weight. , moisture Fia, TS heavy dryness, degumming effect, dehydration effect, and good stuttering effect. The amount of permeation at this time was 4j t/hr under steady filtration conditions.

Example 2 To a mixed oil of 75 volumes of pressed rapeseed oil and 25 volumes of extracted oil, a 50% heavy t-citric acid aqueous solution was added little by little at 40°C in an amount of 0.6% by weight based on the oil. The mixture was stirred, and then 1.5 volumes of water was added little by little to the oil, and the mixture was stirred and aged to prepare the treated gt. The liquid to be treated had a gummy substance #ilj by weight and a moisture content of '-IL8 by weight.

When this treated liquid was partially circulated* filtered using the same module and the same conditions as in Example 1, the obtained permeated oil contained gummy FiO of 18% by weight and water content of 0.15% by weight. A better degumming effect than in Example 1 was obtained by the acid addition treatment. At this time, the oil permeation amount in the steady filtration state is 4. I was an office lady/hr.

Examples 3 to 7, Comparative Examples 2 to 5 Example 1 except that the amount of water used as the liquid to be treated was changed.
The seven types shown in Table 1 were made under the same conditions as above.

Partial circulation 1j was carried out using the same module and under the same conditions as in Example 1, except that the filtration temperature of these liquids to be treated was 50°C.
I passed the vi.

For each liquid to be treated, the gum quality in the permeated oil and the amount of oil permeated under steady offshore conditions are shown in Table 1.

From this result, the amount of coexisting water in the liquid to be treated is 0.1 to 1.
It can be seen that 0 weight is preferable in terms of degumming effect and oil permeation amount.

Example 8 High density polyethylene 22.9% by weight, dioctyl phthalate (DOP) 55.5wts, imitation powder silica 23
．． After mixing 5Nt in a mixer, it was pulverized in a pulverizer.

This raw material was melt-extruded into a flat film using a twin-screw extruder and a T-die, then cooled and taken off.

This membrane was immersed in 1.1.1-) dichloroethane and DO
After extracting and removing P, it was dried, further immersed in a 40 coul-1 caustic soda aqueous solution to dissolve fine powder silica, washed with a dilute caustic soda aqueous solution and water, and further dried to obtain a flat porous membrane.

This membrane had a thickness of 300μ, an average pore diameter of 0.12μ, a porosity of 70 Ls, and rC=31 dyn/cW1.

This flat membrane was attached to a Paio Engineering Co., Ltd. filter holder (UD-6, effective membrane area 0.08 m'), and using the same liquid to be treated as in Example 2, the rolling speed was 0.4 m/s.
Partial circulation filtration was carried out at EC, average filtration pressure of 1.0 atm, filtration temperature of 55° C., and concentration ratio of 5 times.

The resulting permeated oil had a gummy Fi of 0.12 heavy IIk-1 water #'i of 0.10% by weight, and had good degumming and dehydration effects. At this time, the oil supply amount Fi in the steady p excess state
Ivy at 2.2t/hr.

Examples 9 to 13, Comparative Examples 5 to 5 Degumming experiments for vegetable oil were conducted using the following membranes.

Membrane A: A mixture of 20 parts of polyvinylidene fluoride, dimethylformamide, and acetone ~ solvent (volume mixing ratio 20/80)
) Extrude the stock solution dissolved in 80 parts through a hollow nozzle,
A hollow fiber membrane was obtained by coagulating in water. This membrane has an inner diameter Q,
7111, film thickness 200μ, average pore diameter 0.02, a, r
c=25dYrl/1M1.

B film: 20% by weight of polyvinylidene dimethylacetamide fluoride and polyethylene glycol, 6
0 weight g'Is and 20 M amount'ls were mixed, extruded through a hollow nozzle, and coagulated in water to obtain a hollow fiber membrane.

This membrane has an inner diameter of 0.7+111°, a film thickness of 190μ, and an average pore diameter of 0. D 5 p, rc = 25 dyn 7 strokes.

C film: high density polyethylene 22.9% by weight, DOP53
．． After mixing and pulverizing 5% by weight and 25.5% by weight of fine powder silica, the mixture was melted and extruded using a twin-screw extractor and a hollow nozzle, and then DOP and silica were removed to obtain a hollow fiber membrane. This membrane has an inner diameter of 0.7 mm, a thickness of 300 μm, and an average pore diameter of 0.7 mm.
12p, rc=31 dyn/z.

D membrane: The same hollow fiber membrane as in Example 1, with an inner diameter of 0.71
111, film thickness 300μ, average pore diameter 0.15μ, rc=2
It was 9 dyn/-.

E membrane: ethylene-tetrafluoroethylene copolymer 26
．． 7l each! After kneading and pulverizing DOP 60.0 volume t % and fine powder silica 13.3 volume, the mixture was melted and extruded using a knee extruder and a hollow nozzle, and then DOP and silica were removed to obtain a hollow fiber membrane. Ta. This film is within @0.7<em>m
a, film thickness 500μ, average pore diameter 0.5μ, rc = 26
, 5 dyp, I cm.

F membrane: Vinylidene polyfluoride 26.7 capacity - 1DOP6
Knead 0.0 volume tht, 13.3 weight of fine powder silica,
After pulverization, the mixture was melt-extruded using a twin-screw extruder and a hollow nozzle, and then DOP and silica were removed to obtain a hollow fiber membrane.

This membrane has an inner diameter of 0.7cm, a thickness of 500μ, and an average pore diameter of 3.
μ, rc = 25 dyn/fM.

G film: The above F film was stretched 5 times in the longitudinal direction. This membrane is
Inner diameter 0.5mm, film thickness 155μ, average pore diameter 5μ, r C
” It was y1 at 25 dyn/.

H membrane: Ethylene-vinyl alcohol copolymer (ethylene content: 33 mol, degree of saponification 99% or more) 25.5 vol. mol. After melt extrusion fc using a shaft extruder and hollow nozzle, deglycerate 5 phosphorus with warm water, and further 4
A hollow fiber membrane was obtained by removing the silica using a 0 weight - caustic soda aqueous solution. This membrane has an inner diameter of 0.7 mm, a film thickness of 500 μ, an average pore diameter of 0.15 μ, and rC==33 dyn/cr.
It was n.

A module with an effective membrane area of 0.1 was made using the various hollow fiber membranes mentioned above, and the same liquid to be treated as in Example 2 was used, except that the amount of water added was 4.0 wt. s
ec, average p overpressure 1.0 atm 1. Filtration temperature 40℃,
Partial circulation 11 filtration was carried out at a concentration factor of 4 times.

Table 2 shows the results of the gum quality and water content in the obtained permeated oil and the amount of oil fed in a steady state of excess p.

The gum content in the liquid to be treated is 1.1% by weight and the water content is 4.1% by weight.
5 weight.

Comparative Examples 6-7 Degumming experiments on vegetable oil were conducted using the following membranes.

IM: Nylon 6,10 to 15? After dipping into nitric acid at a concentration of /10Dad, it was poured, immersed in a coagulation bath to which oil sulfate had been added several times, and washed with water to obtain a flat membrane. This membrane has a thickness of 200μ, an average pore diameter of o, 17μ, and r
c=57, Odyn/cm.

J membrane: 15% by weight of vinyl chloride, 10% by weight of polyethylene glycol, and 85% by weight of dimethylformamide were uniformly mixed and cast, dried, and immersed in water to obtain a flat membrane.

This membrane has a thickness of 200μ, an average pore diameter of 0.10μ, and rc=
39, Odyn/cm.

The above 2ai flat membrane was made by Bio Engineering Co., Ltd. 7 (
kfi-hokf-(UD-6, effective membrane area 0.08m
') and using the same liquid to be treated as in Examples 9 to 13.
Linear speed: 0.4 m/sec, average filtration pressure: 1. ot/c
11. Partial circulation JJ at a filtration temperature of 40°C and a concentration ratio of 4”
I did IFm.

Table 3 shows the results of the gum quality and water content in the obtained permeated oil.

Table 3 Average pore diameter is 0.05 to 3μ, critical surface tension rc
is less than 35 dyn/6H, the amount of oil escaping, the degumming effect,
It can be seen that this is preferable in terms of dehydration effect.

Example 14 Using the same membrane as in Example 11 except that the inner diameter was 1.5ill, a module with an effective membrane area of 0.1mm and an effective length of 25α was produced, and the same liquid to be treated as in Example 2 was used. , the average p
Partial circulation filtration was performed at an overpressure of 4.0 atm, a filtration temperature of 45° C., and a concentration ratio of S and O times, while varying the linear velocity as shown in Table 4.

At any line speed, the gum quality in the permeated oil was 0.18 times the filter and the moisture was 1jO015 times -, which did not change, but the amount of oil escaping under steady filtration conditions and the durability time until pinholes appeared in the membrane was influenced by the linear speed, and the results shown in Table 4 were obtained. That is, if the speed is less than 0.05 rn/sec, the amount of oil escaping will be small and it is not preferable.
If the linear velocity exceeds 0.05 tn, the durability of the film will be poor and the energy consumption will also increase.
It can be seen that more than /sec and less than jam/sec is preferable.

Table 4 Example 15 Vegetable oil was purified under the same conditions as in Example 2, except that backwashing with F liquid was carried out for 10 seconds at a differential pressure of 2.5 atm, and the process was repeated for 715 minutes.

The gum quality in the obtained permeated oil is 0.18 heavy lids - 1 moisture Fi
O,151:! Difference between t% and 'Q example 1t n Nakatsu*ka,
Feet'PP: Average oil permeability in filtered state is 7.2 l/h
r was improved to 1.8 times the result of Example 1, and a remarkable backwashing effect was observed.

Example 16 Vegetable oil was purified under the same conditions as in Example 12, except that ν liquid backwashing was carried out for 30 seconds at a differential pressure of 2.0 atm at a bottle rate of once/30 minutes.

The gummy Fi 0.15 weight 1-1 water content in the obtained permeated oil was 0.12 weight -, which was the same as in Example 12, but the steady p
Average oil permeation amount Fi8. Ol/hr and Example 12
The result was improved by 2.0 times, and a remarkable backwashing effect was observed.

Example 17 - To a mixture of 40 volumes of pressed oil of rapeseed oil, 4 volumes of extracted oil and 20 volumes of n-hexane, at 1.40°C, water of 1.25 parts by weight relative to the oil was added little by little. While stirring and aging, a liquid to be treated was prepared.

The gum quality and water content in this liquid to be treated are each 1 to 1 oil.
．． They were 53% by weight and 1.30% by weight.

This liquid to be treated was subjected to partial circulation filtration using the same module and under the same conditions as in Example 1.

The gum quality and water content of the obtained permeated oil are respectively 0.0% relative to the oil.
The degumming effect and the dehydration effect were good. The amount of transparent sleeves (mixed oil with hexane) at this time was 5.5 t/hr in a steady p state.
Met.

Example 1B A mixture of 40 volumes of pressed rapeseed oil, 40 volumes of extracted oil and 20 volumes of n-hexane was heated at 40°C for 50 minutes.
Add 1.0!1% by weight of oxalic acid aqueous solution to oil little by little while stirring, then stir while adding 3.1% by weight of water to oil little by little, age and process. I made a liquid pipe. The cam quality and water content in this treated layer were 1.32% by weight and 3.6% by weight, respectively, relative to the oil content.

This liquid to be treated was subjected to partial circulation 11 filtration using the same module and the same conditions as in Example 12.

The gum quality and water content of the obtained permeated oil are 0 relative to the oil.
．． 10 fold tS, 0. o7ju Jlt1? ``The reeds, de-camming effect, and dehydration effect were good.The amount of oil passed at this time (
Mixed oil with hexane) is 5.8t/h in steady filtration state.
It was r.

In the same case as above, a further 10 seconds of n at a differential pressure of 2.5 atm.
- When backwashing treatment with hexane was carried out at a frequency of once/20 minutes, the average oil permeation amount (mixed oil with hexane) during steady filtration was as large as 1lt7.5l/hr.
A remarkable backwashing effect due to n-hexane was observed.

Example 19 Water of 0.5 weight tS was added little by little to sunflower oil at 40° C. while stirring, and then rapidly cooled to 5° C. to prepare a liquid to be treated. The gum content in this liquid to be treated was 0.31% by weight, and the water content was 0.5% by weight.

This treated liquid was subjected to a degumming and dehydration test using the same module as in Example 1.

P filtration conditions are linear velocity 0.5 m/scc, average filtration pressure 1.0
Partial circulation at atmospheric pressure, filtration temperature of 25°C, and concentration ratio of 8.0 times.
It's filtration.

The gummy content of the obtained permeated oil was less than 1% by weight, and the water content was 1% by weight, indicating a good dehydration and degumming effect. The amount of oil flowing at this time is 1 in the steady filtration state.
．． It was 7t/hr.

Example 20 50 weight t* of an oxalic acid aqueous solution was added to soybean oil at 40°C in small amounts of 0.5% by weight based on the oil while stirring, and then 3.5% by weight water based on the oil was added little by little. While stirring and aging, a liquid to be treated was prepared. The gum content in this treated liquid was 1.42% by weight, and the water content Fi was 3.8% by weight. A degumming and dehydration test was conducted on this liquid to be treated using the same module and the same p-filtration conditions as in Example 1.

The gum content in the obtained permeated oil was 0.04% by weight, and the water content was 0.
．． At 07 weight, a good degumming and dehydration effect was obtained, and the amount of oil passed at this time was 3.8 in the steady filtration state.
It was l/hr.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram showing an example of the partial circulation p-appropriate method according to the present invention, and FIG. 2 is an explanatory diagram showing an example of the multistage two-filtration method according to the present invention. 1°゛～Do Noisy Hayabusa Bow P April 20, 1980 Commissioner of the Patent Office Mr. Tsumada Shimada 1 Incident Q Indication Patent Application No. 1988 567190640 2 Title of Invention Method for Refining Vegetable Oil by Membrane 5 Person Making Amendment Relationship to the case / Patent applicant (003) Asahi Kasei Kogyo Co., Ltd. Postal code 105 5th floor, Toranomon Industrial Building, 2-29 Toranomon-chome, Minato-ku, Tokyo Detailed description of the invention in the specification 6 Description of amendments amend the description as follows. 11), page 6, lines 2 to 5 r In the present invention...
・Easy degumming effect. ” to be deleted. (21% "Also" on page 6, line 6 is corrected to "in the present invention". (311, page 7, line 7, "phosphoric acid, oxalic acid, etc." (41, "Permeation amount" on page 8, line 2 is corrected as "oil penetration amount". (5), JH page 5, line 2. "5 xi/5lcj is changed to "3m/5acJ". Correct. (61, Correct "20" on page 15, line 14 to "10". (7), Correct r20cJ on page 13, line 17 to [10CJ. (8), Page 13, line 18. Correct ``20CJ'' to ``1OCJ.'' (9) Correct ``small'' in line 6 of page 14 to ``large''. Change ``permeation amount'' in line 5 of page 17 to ``oil permeation amount.'' Correct it. 0% by weight.''" is corrected to "polyvinylidene fluoride, dimethylacetamide." Sen, page 21, line 20, "22.9% by weight -" is corrected to "25.0% by weight" (US, No. 2
On page 2, line 2, "twin-screw extractor" has been corrected to "twin-screw extruder." a? ), "13.3% by weight" on page 22, line 19 is corrected to "15.S capacity". (IL page 24, line 4, “Table 2” is corrected as “Table 2”. In field a, page 26, line 12, “855% by weight” is corrected to “75 weight rice cake.”) GL M page 26, line 19 Row or r 1.Of/a+IJtr
1. Correct it as ofi pressure J. (21), page 34, insert the following statement between lines 1 and 2. Example 21 After adding 2.0 weight of activated clay to the permeated oil obtained in Example 16 and decolorizing it at 120°C for 20 minutes. Sample P was obtained.For comparison, using the same crude oil before membrane treatment, it was subjected to eight degumming steps and a deoxidation step using the conventional centrifugation method under the same conditions as above. Color was evaluated using a FLg1 Vibond colorimeter using a 5.4 Tsuru cell. Chlorophyll was evaluated using CX-4704. Light exposure characteristics are values after 6 hours of irradiation with a fluorescent lamp.S Compared to the conventional centrifugation method, the present invention can omit the deoxidizing step that consumes chemicals. Furthermore, it is an excellent purification method with simplified steps and low energy consumption). The liquid quality of the refined oil obtained is comparable to that produced by the conventional centrifugation method.

Claims (1)

[Claims] tl+ Crude vegetable oil containing gummy matter, water content is 0.1~
After aging in the coexistence of 10% by weight with stirring for 9 hours, the average pore size is 0.05-5μ and the critical surface tension of the m surface is S3.
It is brought into contact with the surface of a porous membrane of less than dyn/z, and is flowed parallel to the membrane surface at a linear velocity of 0.05 m/sec to 3 m/sac to concentrate and separate the gums together with water. A method for refining vegetable oil using a membrane, which is characterized by obtaining a vegetable oil that does not contain water. (2) Crude vegetable oil containing gummy substances has a gummy content of 0.
．． Add 0.1 to 3.0 times the acid and add 0.1 to 10 times the amount of water.
After aging with stirring in the coexistence of % by weight, the average pore size is 0.
．． 05~5μ, the critical surface tension of the membrane surface is S 5 dyn
/ cm droplets are brought into contact with the surface of the porous membrane, and flowed parallel to the membrane surface at a linear speed of 0.05 m/see to 3 m/sa...gummy matter is concentrated and separated with water, and the gum substance is substantially A method for refining vegetable oil using a membrane, which is characterized by obtaining a vegetable oil that is free of high quality and water.