~0'~313;~2 -1. Field of the Invention The present invention relates to ethylene-vinyl alcohol copolymer membranes which are of use as selective separation membranes and are particularly valuable as membranes for the dialysis of blood in artificial kidneys.
2. Description of the Prior Art Heretofore cuprammonium cellulose membranes have been widely employed in the dialysis of blood, but because of their inadequate permeability,there has been a need for new dialysis membranes. The desiderata of a membrane for artificial kidney applications include the following. Such a membrane should have a controlled permeability to water, a very high permeability to substances of the so-called intermediate molecular weight, i.e.
in the neighborhood o 300 to 6000, a comparatively low molecular welght dependence, and a high rejection for proteins and other biologically essential substances. To develop a membrane having such desired properties, the membrane character-istics of various high polymer materials have been investigated.
In this connection, we have found that ethylene-vinyl alcohol copolymers have excellent biological compatibilities, satisfactory antihaemolytic and antithrombogenic properties and such other properties as durability, chemical stability, heat-sealability, etc., thus making these materials suitable membrane materials for the dialysis of blood.
Hirofugi et al (see Japanese Patent Application Laid Open No. 113859/1974) have already succeeded in the production -~
of a permeable film from an ethylene-vinyl alcohol copolymer, but because they dissolved the ethylene-vinyl alcohol copolymer in a solvent mixture consisting of water and alcohol (e.g.
water-methanol, water-isopropyl alcohol, etc.) and wetcast the solution, the resultant film showed white turbidity and lacked - 1 - ~ :~
uniformity in structure, there being a thick skin or surface layer and a multiplicity of pores as large as 2 microns and upwards in the inner layers. This meant that, while the film had a high permeability to water, it was not sufficiently permeable to substances of intermediate molecular weight (e.g.
vitamin B12). Such characteristics, of course, do not make the film suitable for use as a membrane for blood dialysis applica-tions and even when it is to be employed for other separation functions, the range of usage is considerably limited. This is why, until now, ethylene-vinyl alcohol copolymer membranes have found only limited application.
There is therefore a need to overcome the afore-mentioned lack of homogeneity of prior art ethylene-vinyl alcohol copolymer membranes, and to eliminate the large pores therein, so that a membrane suitable for the dialysis of blood can be produced and the inherent biological compatibilities and structural and chemical characteristics of this category of :.
copolymer materials suitably employed. So long as one depends upon the production methods heretofore attempted, however, it is considered impossible to obtain a membrane having a properly controlled permeability to water, a high permeability to substances of intermediate molecular weight and a high rejection rate for proteins and other essential blood constituents.
Summary of the Invention It is therefore an object of the present invention to provide an ethylene-vinyl alcohol copolymer membrane which is of value as a selective separation membrane, and to provide a method for producing said membrane.
According to one aspect of the invention there is provided a separation membrane formed from an ethylenè.vinyl alcohol copolymer wherein said membrane-has a micropore structure which is substantially uniform throughout its longitudinal and : :
-transverse sectional areas and wherein its constituent particles have an average diameter in the range of 100 to 10,000 Angstrom units as electron-microscopically determined ~or a dry membrane and are bonded to each other to form a membrane that is substan-tially free from pores in excess of 2 microns in diameter.
According to another aspect of the invention there is provided a method for producing a separation membrane comprises dissolving an ethylene-vinyl alcohol copolymer with an ethylene content of 10 to 90 mole percent and a degree of saponification of not less than 80 mole percent in a solvent selected from the group consisting of dimethylacetamide, methylpyrrolidone and dimethylsulfoxide to a polymer concentration (C) in the range of 10 to 35 percent by weight and introducing the resulting solution into a coagulation bath to coagulate said solution into a shaped article wherein said coagulation bath contains water and from 0 to 50% of a solvent selected from the group consisting of dimethylacetamide, methylpyrrolidone and dimethylsulfoxide at a coagulation bath temperature (TC) within the range defined ;~
by the following relation: when 10 _ C < 25, 0 c T ~ C-10 (1) when 25 ~ C ~ 35, C-25 ~ T < C-8 (2).
Brief Description of the Drawin~ -Figs. 1 to 5 are electronmicrographs showing cross-sectional views of ethylene-vinyl alcohol copolymer membranes according to preferred embodiments of the present invention;
Figs. 6 to 7 are electronmicrographs showing the cross-sectional views of prior art ethylene-vinyl alcohol copolymer membranes;
Fig. 8 is a cross-sectional view showing the apparatus for measuring the permeability to water of membrane samples;
and Fig. 9 is a cross-sectional view showing the apparatus for measuring the permeability to vitamin B12 or uric acid of ~ .
~0'~38;~2 membrane samples.
Detailed Description of the Preferred Embodiments The ethylene-vinyl al.cohol copolymers employable according to the present invention may be, for example, random, block o,r graft copolymers. It should, however, be understood - 3a -~,..;
10';~38~2 that if the ethylene content of such a copolymer be less than lO mole percent, the resultant memhrane will have only reduced wet mechanical properties and may suffer a large dissolution loss. Should the ethylene content be over 90 mole percent, the membrane will have only reduced biological compatibilities and poor permeability characteristics. Thus, it is preferable that the copolymer have an ethylene content in the range of 10 to 90 mole percent and, for better results, in the range of 15 to 60 mole percent. Such an ethylene-vinyl alcohol copolymer is characterized in that, unlike polyvinyl alcohols, it hardly loses any amount of its constituents by dissolution, and is suitable for use as a membrane for the dialysis of blood. As regards the degree of saponification, to ensure adequate wet mechanical properties, the ethylene-vinyl alcohol copolymer should preferably have a degree of saponification of not less than 80 mole percent, and more preferably not less than 95 mole percent. Normally, a substantially completely saponified copolymer, i.e. a copolymer having a degree of saponification in excess of 99 mole percent, is employed. The ethylene-vinyl alcohol copolymer may contain copolymerizable comonomers such as methacrylic acid, vinyl chloride, methyl methacrylate, acrylonitrile, vinyl pyrrolidone, etc. in the range not exceeding 15 mole percent.
Also falling within the scope of the present invention are ethylene-vinyl alcohol copolymers obtainable by a cross-linking procedure either before or after casting, or otherwise forming into a shaped article. The ethylene-vinyl alcohol copolymer can be cross-linked by treatment with, for example, an inorganic cross-linking agent such as a boron compound or an organic cross-linking agent such as a diisocyanate or dialdehyde.
Alterna,tively, the functional hydroxyl groups of the vinyl ~38Z2 alcohol units may be acetalized by an aldehyde within limits not exceeding 30 mole percent, suitable aldehydes being formaldehyde, acetaldehyde, butyraldehyde, benzaldehyde or the like. The ethylene-vinyl alcohol copolymer employable in the present invention preferably has a viscosity of l.0 to 50 centipoises as determined by a B-type viscosimeter for a dimethylsulfoxide solution at a concentration of 3 weight percent at 30C. If the copolymer has a lower viscosity, that is to say a lower degree of polymerization, it does not provide a membrane possessing particularly good mechanical properties. Should the viscosity be higher than the above upper limit, the copolymer may not be easy to cast or to form by another method.
Suitable solvents for dissolving the ethylene-vinyl alcohol copolymer include monohydric alcohols such as methanol, ethanol, etc.; polyhydric alcohols such as ethylene glycol, propylene glycol, glycerin, etc.; phenol, m-cresol, methyl-pyrrolidone, formic acid, etc. and mixtures of the aforesaid solvents~with water. However, to obtain a blood-dialysis membrane having a good balance between water-permeability and solute-permeability characteristics, it is preferable to employ dimethylsulfoxide, dimethylacetamide or a mixture o~ these substances as the solvent. Dimethylsulfoxide, in particular, is desirable because ethylene-vinyl alcohol copolymers are highly soluble in this solvent. The solvent, particularly dimethyl-sulfoxide, may include other solvents such as water, methanol, isopropyl alcohol, dimethylformamide, etc.; other liquids which are highly miscible with the particular solvent; and/or inorganic salts, provided that they have precipitation points below 60C (the temperature at which the ethylene-vinyl alcohol copolymer completely diæsolved in the solvent starts being precip~tated as the solution is gradually cooled).
~0 ~38~2 In dissolving the ethylene-vinyl alcohol copolymer in the aforementioned solvent, the concentration of the copolymer should preferably be within the range of 5 to 50 weigh~ percent and more preferably within the range of 10 to 35 weight percent.
The temperature of the polymer solution should preferably be within the range of 0 to 120C, more preferably 5 to 60C. The polymer could be degraded at temperatures higher than 120C, while it wouLd be too viscous to be easily shaped into an article at temperatures below 0C.
The coagulating agent to be employed in the coagulation bath should preferably be an aqueous medium. The aqueous medium may be water alone or a mixture of water with not more than 50 weight percent of a water-miscible organic solvent, normally the same solvent as that used in the preparation of the copolymer solution or casting dope, or a system comprising such a medium plus an inorganic salt, such as sodium sulfate, dissolved therein.
To obtain a suitable membrane having good permeability characteristics according to the present invention, it is of vital importance to select suitable coagulation conditions. If the coagulation is accomplished under conditions as mild as practicable, the resultant membrane is such that it is trans-parent in the wet condition, being substantially free from large pores exceeding 2 microns in diameter and, instead, having micropores substantially uniformly distributed throughout its longitudinal and transverse sectional areas. The term 'mild conditions' as used herein means that the solution coagulates in a time of not less than 3 seconds, preferably of 5 seconds or more, as determined by the method hereinafter described. Thus, the coagulation time of the solution is determined by casting a solution of ethylene-vinyl alcohol copolymer onto a glass plate to obtain a film having a thlckness of 100 microns (as ::
measured in pure water) and measuring the time required for the solution to completely coagulate (i.e. the time at which the film can be stripped off the glass plate, without leaving a residue of uncoagulated polymer solution on the glass plate).
Therefore, in forming the copolymer solution into a shaped article, the solvent, the concentration and temperature of the polymer solution, and the composition and temperature of the coagulation bath should be selected according to the ethylene content and degree of saponification of the ethylene-vinyl alcohol copolymer employed so as to meet the above coagulation time requirement. Since the coagulation time varies with film thickness, trials can be performed under conditions leading to the attainment of a film having a thickness of 100 microns to find a coagulation time satisfying the above time requirement and these conditions are used for the actual production of separation membranes having various thicknesses. The actual - production of a membrane does not necessarily require a glass plate or any other support means but, even in such other cases, the coagulating conditions selected by the above procedure can be employed. The coagulation time thus selected is character-istically much longer than the time heretofore known in association with the prior art wet-casting process (in which the solvent is water-alcohol). With a water-alcohol system, the coagulation time as determined by the above procedure is some-where between 1 and 2 seconds and slower coagulation cannot be accomplished even by varying various other conditions. Of course, even when dimethylsulfoxide is employed as the solvent, unless the above conditions are satisfied, rapid coagulation takes place, failing to provide a practically useful membrane having a satisfactory balance between permeability to water and permeability to solutes. The factor to be particularly ,.
_ _ _ . _ _ _ _ _ _ ' ' , . -considered in achieving such mild coagulation is the coagulation temperature. When an ethylene-vinyl alcohol copolymer with an ethylene content of 15 to 60 mole percent and a degree of saponification of not less than 95 mole percent (preferably 99 mole percent or higher) is dissolved in a solvent based on dimethylsulfoxide to a polymer concentration of 10 to 35 weight percent and the resultant solution is extruded or otherwise contacted with a coagulation bath comprising water as a principal component, the preferred coagulation temperature is expressed by the following relations. Assuming that the con-centration of the copolymer is C and the coagulation temperature is TC, When 10 ~ C < 25, 0 - T ~ C - 10 --- (1) When 35 ~ C ~ 25, C-25 - T - C-8 --- (2) Thus, under conditions conductive to the above coagulation time, the coagulation temperature indicated is .. . .. .
According to the preferred contemplated mode of use, the ethylene-vinyl alcohol copolymer membrane is formed in the shape of flat sheet, tubing or hollow fiber, with or without the aid of a supporting device. The coagulation may be achieved by means of a plurality of baths, but in such a setup, at least the first of the coagulation series of baths must satisfy the aforementioned requirement.
The structure of the membrane thus produced can be examined with a scanning electron microscope. In this method, ~; the dry membrane is frozen with liquid nitrogen and broken so that it exposes a fracture cross section. This sectional area ~ -is coated with gold to a thickness of 100 Angstrom units and examined under an electron microscope. Using an accelerating voltage~iof 20 KV, the secondary electron image can be observed - . ., , . ' : . . . ~. .: ' .
-, ,, . . ': ' . ' ' ' . :.
1(~738'Z2 and photographed.
A membrane subjected to the above electron-microscopic examination was prepared by dissolving a completely saponified ethylene-vinyl alcohol copolymer with an ethylene content of 33 mole percent in dimethylsulfoxide to a concentration of 20 percent and coagulating the solution in water at 5C to a thick-ness of 50 microns. The electronmicrographs of the resulting membrane are shown in Figs. 1 to 5. The electronmicrographs of Figs. 1 and 2 were taken at magnifications of 2400 times and 8000 times, respectively, using an electron-microscope (JSM-2;
manufactured by Nihon Denshi Kabushiki Kaisha). Figs. 3 to 5 are the electronmicrographs of the same membrane taken at magnifications of 12,000, 12,000 and 24,000 times, respectively, using an electron microscope (HFS-2; manufactured by Hitachi Seisakusho, K. K.). It will be seen from Fig. 1 that, at a magnification of 2,400 times, the membrane is substantially homogeneous throughout its cross-section, indicatlng that, at magnifications of this order, no porous structure is ascertainable.
When the membrane of the present invention is examined at higher magnifications, it is found, as from Fig. 2, that the membrane consists of small particles bonded together, suggesting that tiny gaps between the particles contribute to the excellent permeability of the membrane. This structure is more clearly apparent from Figs. 3 to 5. Fig. 3 shows the structure near the surface of the membrane; Fig. 4 shows the inner zone of the same membrane; and Fig. 5 is a electronmicrograph of the membrane taken at a greater magnification of the same inner zone.
These electronmicrographs show that the membrane according to this embodiment of the present invention has the following structure. Its constituent particles have an average 1~738'~Z
diameter substantially in the range of 100 to 10,000 Angstrom units, normally within the range of 500 to 5,000 Angstrom units, and are bonded to each other to form a self-supported membran-eous structure. The term 'average particle diameter' as used throughout this specification, including the claims, means the average of particle sizes throughout the membrane as found by electron micrographic observation.
As will be seen from these electronmicrographs, any two adjacent particles in many cases do not contact each other at a point but have a plane of contact in common, thus being bonded to each other to form a membrane whilst retaining their independent particulate identities. Where the shape Gf a particle has been distorted by its bonding to the ad~oining particle, the particle diameter is calculated assuming the intact shape of the particle that could be visualized if it were not bonded to the ad~acent particle but independently present.
The membrane shown in the electronmicrographs has an average particle diameter of about 2,000 Angstrom units.
Thus, while, as aforesaid, the constituent particles of the membrane have an average particle diameter in the range of 100 to 10,000 Angstrom units, the individual particles also respectively have a diameter substantially wihtin the range of 100 to 10,000 Angstrom units. It holds true, roughly speaking, that these particles are distributed substantially evenly in ; ~-the direction of the thickness of the membrane, although there is a tendency for the surface layers of the membrane to comprise comparatively large particles while the inner or core layers of the membrane comprise relatively small particles. Although some of the individual particles are too small to be discrete enough on electronmicrographs, such particles are not numerous and are,,disregarded in the computation of an average particle diameter.
As these particles are bonded together to form a membrane, a multiplicity of tiny gaps are created between the particles. The gaps vary in size and shape but the variations are by far smaller than those of the hitherto-known ethylene-vinyl alcohol copolymer membranes and it is apparently for this reason that the membrane according to the present invention displays permeability charac~eristics distinct from those of the prior art membranes.
Furthermore, the membrane according to the present invention is substantially skinless; that is to say, it has no dense and thick surface layer. While, as will be seen from Fig. 3, there is occasionally a very thin skin (on one side only, having a thickness of about 1 percent based on the overall thickness of the membrane), the membrane may be said to be substantially skinless because the presence of a skin layer of this order does not interfere with the permeability of the membrane to any significant extent.
Structural views of the inner layers of prior art ethylene-vinyl alcohol copolymers are æhown in Figs. 6 and 7.
Figs. 6 and 7 are electronmicrographs taken at the magnification of 2,400 times and 8,000 times, respectively, using an electron microscope, JSM-2 of Nihon Denshi Kabushiki Kaisha. This prior-art membrane was prepared by dissolving an ethylene-vinyl alcohol copolymer similar to the above in a solvent mixture (7:3) of methanol and water and forming the resultant solution into a membraneous article. The concentration of the solution, and the composition and temperature of the coagulation bath employed were the same as those employed for the production of the membrane shown in Figs. 1 to 5 involving the use of dimethylsulfoxide as the solvent. As will be seen from Fig. 6, the pri4r-art membrane reveals a porous structure even at a 1~738Z2 magnification of 2,400 times, having many large pores (diameter more than 2 microns). This structure is more apparent at a magnification of 8,000 times. Thus, comparison of the membrane according to the present invention with the prior art membrane at once shows a marked difference in micro-structure. Whereas the prior art membrane contains a large number of pores larger than 2 microns in diameter, the membrane according to the present invention is substantially devoid of pores exceeding the above pore size limit.
The prior art ethylene-vinyl alcohol copolymerm~mbrane is produced by a forming process involving the employment of a solvent mixture of water and an organic solvent such as methanol, isopropyl alcohol or the like, but because of the inadequate solubility of the copolymer in such a solvent mixture, the coagulation time is of necessity short irrespective of what conditions are employed. Thus, it is impossible to effect a slow coagulation such as that feasible in accordance with the present invention. It follows, then, that the prior art .: .
membrane contains many pores larger than 2 microns in diameter and, although it is not evident in the electronmicrographs, there is a thick skin on the surface (a dense skin layer as thick as about 3 percent or more based on the overalll thickness of the membrane). Therefore, the prior art membrane is inhomo-geneous, shows a white turbidity and fails to exhibit the ~ -desired permeability characteristics.
Because it is formed by using the aforementioned particular solvent under the conditions defined hereinbefore, the membrane according to the present invention is substantially free from pores larger than 2 microns in diameter, It is a membrane having a substantially homogeneous micro-structure which is usually transparent in the wet condition and displays .
the characteristics desired in a separation membrane and, particularly, in a membrane for the dialysis of blood. Thus, the membrane according to the present invention usually has a permeability to water of 10 to 200 x 10 16 cm2, a permeability to vitamin B~2 of not less than 0.8 x 10 7 cm2 per second and, in addition, has the mechanical strength required of a membrane for the dialysis of blood. Usually, there is a good balance between permeability to water and permeability to vitamin B12 The membrane formed as above can be rinsed with water at a low temperature not exceeding 50C, if required. The membrane may be maintained in wet condition without being dried and, if required, may be sterilized before use. However, the storing of the membrane in water before and after use is a dis-advantageous factor in transportation and in assembling membranes into a module. It is therefore desirable to produce a dry film retaining the advantageous properties mentioned above. For this purpose, the wet membrane just after formation can be dipped into a water-miscible organic non-solvent in order to replace the aqueous solvent present on the surface aDd/or inside the membrane with the non-solvent and, thereafter, the membrane can be dried at atmospheric or reduced pressure and at a temperature below the glass transition point of the ethylene-vinyl alcohol copolymer, preferably in the neighborhood of room temperature. By the above procedure may be obtained a dry permeable membrane retaining the desired permeability characteristics. Suitable organic solvents for this purpose, include lower aliphatic alcohols or ketones of 1 to 5 carbon atoms, such as methanol, ethanol, amyl alcohol, acetone, methyl ethyl ketone, diethyl ketone and so forth. Acetone is par-ticularly desirable.
, A dry membrane retaining the above-mentioned 10'73~;3Z'~
permeability characteristics may also be obtained in such a manner that, instead of replacing the water with an organic solvent, the membrane may be freshly formed and treated in the wet condition with a polyhydric aliphatic alcohol having 2 to 4 carbon atoms or an adduct of 1 to 20 moles of ethylene oxide to such a polyhydric alcohol in an aqueous, alcoholic or other solution and at a temperature of not more than 50C and, thereafter, the membrane can be dried at a temperature not in excess of 50C. In such cases, the resultant membrane contains about 20 to 120 percent of the polyhydric alcohol or polyhydric alcohol-ethylene oxide reactant based on the ethylene-vinyl alcohol copolymer which, however, may subsequently be easily -removed by rinsing prior to dialysis after being built into a modular unit. Suitable polyhydric alcohols having 2 to 4 carbon atoms include ethylene glycol, diethylene glycol, propylene glycol, 1,3-butanediol, 1,4-butanediol, glycerin and so forth, glycerin being particularly preferred. It is also -possible to incorporate such a polyhydric aliphatic alcohol into the wet-coagulation bath so that the membrane will contain said alcohol as the membrane is formed.
The separation membrane according to the present invention is normally put to use as a flat sheet or tube with a thickness in the range of 10 to 100 microns. It may also be formed and used in the shape of a hollow fiber, which may measure about 50 to 1,500 microns in outer diameter and about 10 to 300 microns in wall thickness.
While the ethylene-vinyl alcohol copolymer membrane according to the present invention has properties which, as mentioned hereinbefore, are particularly beneficial for use as an artificial kidney membrane for the dialysis of blood, it is also us~ful as a filtration and separation medium for bacteria, 16)'~38ZZ
proteins, viruses and colloidal substances and may also be used for other dialytic or ultrafiltration purposes.
The following Examples are intended to further illustrate the present invention without limiting its scope described hereinbefore and set out in the appended claims.
Example 1 An eth~lene-vinyl alcohol copolymer having an ethylene content of 33 mole ~ and a degree of saponification of not less than 99 mole % was dissolved in dimethylsulfoxide to prepare a solution of 24~ concentration at a temperature of 40C.
This solution was formed into a membrane in a coagulation bath comprising water, the membrane being 50 microns thick.
The membrane thus freshly formed and in wet condition was tested for permeability to uric acid, vitamin sl2 and water.
The results are set forth in Table 1. The properties of the cuprophane membrane currently available on the market for artificial kindney use are also shown in Table 1. It will be obvious from these data that the membrane according to the present invention is considerably superior to the conventional membrane for the purposes of blood dialysis.
The permeability behaviors of these membranes against water, uric acid and vitamin B12 were determined by the follow-ing procedures.
(i) The permeability to water of each membrane was determined by the apparatus illustrated in Fig. 8 at 37C and 100 - 300 mm Hg and the permeability coefficient k was calculated by means of equation (3).
k = VL~/tA ~P(cm ) --- (3) where V: volume of permeated water (cm3) L: thickness of the membrane (cm) ~: viscosity of water (~/cm.sec) ' 1~)73~ Z
~ t: permeation time (sec.) A: area of the membrane (cm2) ~ P: measuring pressure (g/cm .sec2) (ii) The permeabilities to solutes such as vitamin Bl2 and uric acid were determined by means of the apparatus illustrated in Fig. 9 at 37C and the permeability constants P were calculated -by means of equation (4). The concentrations were measured by ultraviolet spectrometry. -- L S l _ C2/Cl 2 P (l ~ ln~ V C (cm /sec.) --- (4) Vl V2)At ~l ~ 2/Vl . 2/C
Where L: thickness of the membrane (cm) ;
A: area of the membrane (cm2) Cl: the concentration of the solute in the chamber l aftcr t seconds C2: the concentration of the solute in the chamber 2 after t seconds Vl: volume of chamber l V : volume o chamber 2 (At t = 0, 1: the solute side, 2: the pure water side) .
Table l . _ . .
Permeabilities Sample Uric acid ¦ Vitamin Bl2 ~ater (cm2/sec x 107) ¦ (cm2/sec x 108) (cm2 x 10l6) ~ B
Ethylene-vinyl alcohol copoly-11.6 35.1 llO
mer membFane _ _ Cuprophane 5.9 8.4 7.5 3Q membrane r~
Then, the ethylene-vinyl alcohol copolymer membrane was subjected to an elution test. The results are shown in Table 2, together with the corresponding data on the C ~
~u~rop~al~ membrane. The above elution test was performed in the following manner. The sample membrane was cut to 1.5 cm square, and 2 grams of the specimens were heated together with 100 ml of distilled water at 70C for predetermined time periods. Ten ml. of the extract was taken, and following the addition of 20 ml of a 0.01 N-aqueous solution of potassium permanganate and 1 ml~of a 3N-aqueous solution of sulfuric acid, the extract was boiled for 3 minutes and, then, allowed to cool. Then, 1 ml of an aqueous solution of potassium iodide (10 wt. %) was added, whereupon iodine was liberated to turn the solution from violet to reddish yellow. This liberated iodine was titrated with sodium thiosulfate and the difference from the blank was taken as the amount of potassium permanganate consumed. One ml of an aqueous solution of starch (1~) was added as an indicator. Of course, a fresh extract was used for each of the elution runs.
Table 2 Consumption of KMnO4, ml Sample 1st elution 2nd elution 3rd elution __ (1 hr.) (2 hrs.) (2 hrs.) _ thylene-vinyl alcohol 1.12 0.32 0.11 copolymer membrane ~ ........................ . .
membrane 15.80 0.87 0.31 It will be seen from Table 2 that, compared with the cuprophane widely employed nowadays as a dialysis membrane for artificial kidney use, the ethylene~vinyl.alcohol copolymer , 10738'ZZ
membrane according to the present invention yields only reduced amounts of extracted substances.
The in vitro blood compatibility of the ethylene-vinyl alcohol copolymer membrane was evaluated in the following manner. In the first place, an antihaemolysis test was performed as follows. The sample membrane was cut into a square 2 cm x 2 cm, which was then laid on the bottom of a glass test tube f 18 mm diameter. In the test tube was put 3 ml of a 10% suspension of red blood cells and the tube was left standing at 37C
for 49 hours. Thereafter, the suspension was centrifuged and 0.2 ml of the supernatant was taken and diluted to 10 ml.
Then, the absorbance at 413 nm was measured to determine the amount of haemoglobin (relative amount) produced by haemolysis.
The result was 0.66 for the present membrane, in constrast to 0.76 for the control cuprophane.
The anti-coagulation test was performed by a procedure similar to the :kinetic method of Imai-Nosé. Thus, the ACD blood ; of a dog was put on a wet specimen in a dish and an aqueous solution of CaC12 was added to the suspension to initiate the coagulation reaction at 37C. After 5 minutes and 30 seconds, ~;~ the clot of blood accounted for 40 weight % (with the weight of the glass as 100), in contrast t~o the corresponding value of ~ro ~ ~
42% for the control ~oprGp~hm~ It is thus clear that the blood compatibility of the ethylene-vinyl alcohol copolymer membrane according to the present invention relatively superior to that of the conventional membrane.
ExamPle 2 Ethylene-vinyl alcohol copolymers with an ethylene content of 33 mole % and that of 45 mole %, respectively, both having a 30l degree of saponification of not less than 99~, were each dissolved .
in methanol-water, n-propanol-water or dimethylsulfoxi.de, and each solution was formed into membranes under various conditions. These membranes were tested for permeability characteristics. The results are set forth in Table 3, ~ ' ~ ~
a) a) ':
3 .~ $ Q ~ :~
o ~ ~ o ,,_ ~ __ ~ .~ ~
~1 ~ ~r ~D CO 1` ~ Ln O O
o o o o o o ,, ~U
m~ x ~ X .
~ C~N ~ ~ ~1 ~`1 ~ ~ O O ~0 O O
a) ~ E~-l ~ LO o 1` ~ r u ~11 11~ 0 O
E~ ~ .__ .__ )r~; .,~ o . U) 11') LO
~: ~ aJ ta o ~rl 1~ a) ~ N t~
-~ Q~ 0 u~
, a) _ - _ ~ 1~ ~
E-~ ~ _ ~ o u~ u~
~ ~ .
a)-rl ~ o o Lr Ln ~ o ~ o ~ ~-- N N N N N : ~ N N N
O )-I ~1 0--C~ ~ O ~.q l ~O O O ':
. ~ ~ ~ D O 1` ~-rl .
~i rl O S~ O
o ~ ~ a) ~
o ~ ~ ~ : : : ~ ~ k ~ ~ : ' - ~ ~ ~ ~ 3 ~ ~
~o l . .
r~ In ~ In ~
O ~ ~ ~ ~2 rl : ~ ~ ~
o ~3 o~
3 0 ~I N ~ ~ 1~ D 1`
O l l l l l l l l l l Z
(Note) 3-1 to 3-8: Temperature of polymer.solution 60C, 3-9 to .
3-10:40C coagulation ~ath: water; the thickness of membranes:
It will be apparent from the above results that when membranes are formed under conditions such that the coagulation is completed in times less than the limit of 3 seconds as defined in this specification, the resultant membranes do not satisfy the permeability requirements for membranes for the dialysis of blood. In contrast, when formed under conditions corresponding to a time of not less than 3 seconds, the membranes have high permeability to water and to vitamin B12.
It is also obvious from Table 3 that the use of a solvent mixture of alcohol and water does.not provide a membrane with a good balance between permeability to water and permeability to vitamin B12.
Example 3 Ethylene-vinyl alcohol copolymers (degree of saponification:
99 mole % or more) with an ethylene content of 33 mole ~ and that of 45 mole %, respectively, were each dissolved in dimethyl-sulfoxide and the resultant solutions were respectively extrudedthrough dies into a coagulation bath (water) to prepare membranes.
The relation of coagulating conditions to permeability ~.
characteristics is shown in Table 4.
~o ,, ~X ~ ~D O O 1~ 0 a~
_ o~ o er~ o ~ ~
~` , ,~ . -~o _Ix ~q ~ ~ u~ In .~ ~ ~ r~
~ ~o ~ X
~ ,~ c, 11~ a) Lt~ ul O O N ~ U)~
~ ~ ~i _~ N ~ I
E~ I ., ~:: h .
~ ~ ~0 _ Ir) u) O ~ 3 ,4 0 ~ W 0 ~ ~DO ~ N
.~1 O ~ ~ 1 ~ U -I ~J P. . ' ~ :~ 1 ~':', ' I ~ ~- O O O o O O O O .
~0 ~ ~
~1 I ~ O_ .
O .~ o ~ a~ ~ O O O O ~ O
o o ~ ~ o N ~ N
~ ~ O ~n-- .
_ _ . ~ ~
. ~ ~
Z' r~ ~.
' ' ' .
, * "33" and "45" stand ~or the ethylene contents of the ethylene-vinyl alcohol copolymers employed.
It will be seen from Table 4 that where the coagulation temperatures are high as in 33-3, 45-2 and 45-4, the permeabi~ities to water of the membranes tend to be too high.
.. . . . , . . . , . , . ~ .
Example 4 The membrane of No. 33-2 of Example 3 was further after-treated to evaluate the possible degradation of the membrane characteristics. The results are shown in Table 5.
Table 5 Sample ~ondition Permeabilities No. treatment Uric acid Vitamin B12 Water ~cm /sec x 10 ) (cm /sec x 108) (cm2 x 1016) ~ ~ __ 33-2 Membrane 10.5 35 146 just form ed and we -33-2-a Acetone 9.5 35 146 replace-ment, fol lowed by drying at room temp.
33-2-b Dipped in 10.0 35 144 20% a~ue-ous gly-cerin, followed by drying tetmpm ' It will be seen from Table 5 that neither acetone replacement followed by drying at room temperature nor glycerin treatment followed by drying at room temperature results in no degradation of permeability performance, thus invariably giving rise to dry membranes retaining the excellent permeabi-lity characteristics. This result is additional evidence of -the usefulness of the ethylene-vinyl alcohol copolymer membrane according to the present invention for blood dialysis purposes.
-., - 2~ -.
~ ,:. - , :