Porous separation membrane

Abstract

The present invention addresses the problem of providing a porous separation membrane that, even after a long-term use, shows little decrease in protein permeability.　The porous separation membrane is obtained: by having an asymmetrical structure in which one surface of the membrane is a dense layer and the other surface is a coarse layer; and by having a biocompatible polymer supported in the membrane. In a TOF-SIMS surface analysis on a cross-section that includes the dense layer and the coarse layer, (1) and (2) are satisfied. (1) The smallest value of the standardized intensity of ion signals derived from the biocompatible polymer in the coarse layer is at least 0.15 times the greatest value. (2) The standardized average intensity of ion signals derived from the biocompatible polymer in the dense layer is at least 2.0 times the standardized average intensity of carboxylate ion signals of the coarse layer.

Description

[Technical field]

[0001] The present invention relates to a porous separation membrane. [Background technology]

[0002] Porous separation membranes are suitable for membrane separation, which screens out substances in liquids based on the size of the pores, and are used in a wide range of applications, including medical applications such as hemodialysis, blood filtration, and plasma separation, and water treatment applications such as home water purifiers and water purification. Furthermore, in recent years, biopharmaceuticals, especially antibodies such as immunoglobulins, have become widely used due to their high therapeutic effect and few side effects. Since antibodies are produced by living organisms such as animal cells, in order to use them as pharmaceuticals, it is necessary to separate and purify only the antibodies from the many impurities, and porous separation membranes are used for separation and purification.

[0003] Among these, as a method for removing viruses, separation by the sieving effect of membrane filtration using a separation membrane is effective, since it has little effect on active ingredients such as antibodies and can remove even chemically resistant viruses. This separation membrane for removing viruses is required to have high separation performance and to prevent leakage of viruses, and is also required to efficiently permeate and recover proteins such as antibodies, which are active ingredients, at a high recovery rate.

[0004] As such a virus removal membrane, for example, a porous hollow fiber membrane for virus removal (see Patent Document 1, for example) has been proposed, which is made of two components, a polysulfone polymer and a copolymer of vinylpyrrolidone and vinyl acetate, has an inner diameter of 150 μm to 300 μm, a membrane thickness in the range of more than 50 μm to 80 μm, and has an asymmetric structure with a dense layer on the outer layer. However, in order to prevent clogging due to protein adsorption to the porous hollow fiber membrane, it is necessary to prevent protein adsorption in the dense layer, but the technology of Patent Document 1 does not solve this problem. Therefore, there have been proposed a porous hollow fiber filtration membrane (see, for example, Patent Document 2) which contains a polysulfone-based polymer and a hydrophilic polymer, has a gradient asymmetric structure in which the average pore size of the pores increases from the outer surface to the inner surface, has a hydrophilic polymer content of 6.0 to 12.0 mass%, and has a ratio of the hydrophilic polymer content of the outer surface to the hydrophilic polymer content of the membrane of 1.20 to 1.60; and a separation membrane made of a polymer, which has a functional layer on one surface of the membrane, and has a peak area percentage of carbon derived from ester groups measured by X-ray photoelectron spectroscopy (XPS) of 0.1 (atomic number %) or more and 10 (atomic number %) or less, and has a peak area percentage of carbon derived from ester groups measured by X-ray photoelectron spectroscopy (XPS) of 10 (atomic number %) or less on the opposite surface of the functional layer (see, for example, Patent Document 3). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] International Publication No. 2013 / 012024 [Patent Document 2] International Publication No. 2016 / 117565 [Patent Document 3] International Publication No. 2009 / 123088 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the porous hollow fiber filtration membrane disclosed in Patent Document 2 is formed from a membrane forming solution containing a polysulfone-based polymer, a hydrophilic polymer, and a solvent, so the hydrophilic polymer on the dense layer surface is still insufficient, and there is a need to further suppress protein adsorption in the dense layer. In addition, the separation membrane disclosed in Patent Document 3 has ester groups localized on the functional layer surface for the purpose of suppressing protein adhesion, but the effect of suppressing protein adsorption in areas other than the functional layer surface, such as the support layer, is insufficient. In view of the problems of the conventional technology, the present invention aims to provide a porous separation membrane that suppresses protein adsorption inside the separation membrane and has little decrease in protein permeability even after long-term use. [Means for solving the problem]

[0007] In order to solve the above problems, the present invention provides a porous separation membrane having an asymmetric structure in which one surface is a dense layer and the other surface is a coarse layer, and which supports a biocompatible polymer, and in which a cross-section including the dense layer and the coarse layer is subjected to surface analysis by TOF-SIMS, the porous separation membrane satisfies the following (1) and (2): (1) The minimum normalized intensity of the ion signal derived from the biocompatible polymer in the coarse layer is 0.15 times or more the maximum. (2) The normalized average intensity of the ion signal derived from the biocompatible polymer in the dense layer is more than 2.1 times that in the coarse layer. Effect of the Invention

[0008] The porous separation membrane of the present invention is capable of maintaining protein permeability even after long-term use by inhibiting adhesion of proteins to the inside of the separation membrane. [Brief description of the drawings]

[0009] [Figure 1] 1 is an example of a total secondary ion image taken by TOF-SIMS of a cross section of a porous separation membrane of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] The present invention will be described in detail below. In this specification, the range "to" indicates a range including the lower and upper limits.

[0011] The porous separation membrane of the present invention (hereinafter, sometimes simply referred to as "separation membrane". For convenience of explanation, the membrane before carrying the biocompatible polymer described later may also be simply referred to as "separation membrane".) has an asymmetric structure in which one surface is a dense layer and the other surface is a coarse layer. In other words, it has a structure in which the pore size on one surface side is smaller and the pore size on the other surface side is larger. Here, in the present invention, the dense layer refers to a layer with a small pore size present on the surface side with a smaller pore size, and the coarse layer refers to a layer other than the dense layer. By having an asymmetric structure, there are a region with a small pore size that contributes to the separation of the substance to be removed and a region with a large pore size that has a low water permeation resistance, so that it is easy to achieve both separation performance and water permeability. For example, in the case of a hollow fiber membrane, the inner surface may be a dense layer or the outer surface may be a dense layer, but it is preferable that the inner surface is a dense layer from the viewpoint of easily adjusting the pore size of the dense layer.

[0012] In the separation membrane of the present invention, separation of substances is mainly performed in the dense layer, which tends to have a large resistance to the permeation of a treatment liquid such as water. In the present invention, particularly in a separation membrane used for separating proteins and viruses, the dense layer is preferably a layer that does not have pores with a pore size of 130 nm or more, and when the dense layer satisfies such conditions, the thickness of the dense layer is preferably 10 μm or less, more preferably 5.0 μm or less, from the viewpoint of improving permeability. On the other hand, the thickness of the dense layer is preferably 0.05 μm or more, more preferably 0.1 μm or more, from the viewpoint of improving separation performance.

[0013] Here, in the present invention, the thickness of the dense layer in an embodiment that satisfies the above conditions can be determined by observing the cross section of the separation membrane, for example, the cross section perpendicular to the axial direction in the case of a hollow fiber membrane, at 10,000 times magnification using a scanning electron microscope (SEM) and analyzing the captured image using image processing software. Specifically, the captured image is first binarized by determining a threshold value so that the structural parts are bright and the other parts are dark. Then, in the separation membrane, the area of ​​1.3×10 where the diameter is 130 nm if the shape is assumed to be a perfect circle is determined. 4 (nm 2 ) or more is identified as the dense layer, and the average thickness of the dense layer in the cross section is calculated. More specifically, the measurement is performed by the method (1) described in the Examples below. However, if the structural part cannot be distinguished from the other parts due to the difference in contrast in the image, the parts other than the structural part may be painted black before image analysis. In addition, as a method of eliminating noise, the noisy parts may be painted white.

[0014] From the viewpoint of reducing the permeation resistance and improving the permeability, the separation membrane of the present invention preferably has a structure in which the pore size gradually increases from the dense layer side toward the surface side of the coarse layer. Also, from the viewpoint of the strength of the separation membrane, it is preferable that the coarse layer does not have macrovoids, which are oval or teardrop-shaped void regions where solid parts of the membrane are missing.

[0015] The separation membrane of the present invention is supported with a biocompatible polymer (hereinafter, may be referred to as a "coating polymer"). By supporting the coating polymer, it is possible to suppress protein adsorption and improve protein permeability.

[0016] The biocompatible polymer refers to a polymer that has the effect of suppressing adhesion of biological components, especially proteins, contained in blood, plasma, urine, etc. Although not particularly limited, examples include hydrophilic polymers such as polyethylene glycol, polyethyleneimine, polyvinyl alcohol, polyvinylpyrrolidone, and their derivatives, 2-methacryloyloxyethyl phosphorylcholine (MPC) and its derivatives, 2-methoxyethyl acrylate (PMEA) and its derivatives, and polymers containing a monocarboxylic acid vinyl ester unit, which will be described later.

[0017] A monocarboxylic acid means a compound consisting of one carboxy group and a hydrocarbon group bonded to the carbon atom of the carboxy group, that is, a compound represented by "R-COOH" (R is a hydrocarbon group). The hydrocarbon group R may be either an aliphatic hydrocarbon group or an aromatic hydrocarbon group, but from the viewpoint of ease of synthesis, an aliphatic hydrocarbon group is preferable, and a saturated aliphatic hydrocarbon group is more preferable. Examples of the saturated aliphatic hydrocarbon group include those having a straight chain structure such as an ethyl group, an n-propyl group, an n-butyl group, an n-pentyl group, and an n-hexyl group, those having a branched structure such as an isopropyl group and a tertiary butyl group, and those having a cyclic structure such as a cyclopropyl group and a cyclobutyl group. In addition, the aliphatic chain may contain an ether bond or an ester bond. Among these, from the viewpoint of the production cost of the carboxylic acid, the saturated aliphatic hydrocarbon group preferably has a straight chain structure or a branched structure, and more preferably has a straight chain structure.

[0018] Examples of monocarboxylic acids in which the hydrocarbon group R is an aromatic hydrocarbon group include benzoic acid and its derivatives, etc. Examples of monocarboxylic acids in which the hydrocarbon group R is a saturated aliphatic hydrocarbon group include acetic acid, propanoic acid, and butyric acid, etc.

[0019] In addition, the hydrocarbon group R may have at least a portion of its hydrogen atoms substituted as desired. However, if the terminal hydrogen atom is substituted with an anionic functional group such as a sulfonic acid group, this may destabilize the protein structure and induce adhesion to the separation membrane surface. Therefore, it is preferable that the terminal hydrogen atom is not substituted with an anionic functional group.

[0020] A smaller number of carbon atoms in the hydrocarbon group R is preferable in terms of lowering the hydrophobicity of the monocarboxylic acid, reducing the hydrophobic interaction with proteins, and preventing adhesion. Therefore, when R is an aliphatic hydrocarbon group or an aromatic hydrocarbon group, the number of carbon atoms is preferably 20 or less, more preferably 9 or less, and even more preferably 5 or less. On the other hand, when R is an aliphatic hydrocarbon group or an aromatic hydrocarbon group, the number of carbon atoms is 1 or more, but from the viewpoint of improving the mobility of the monocarboxylic acid and further suppressing protein adhesion, 2 or more is preferable. When R is a saturated aliphatic hydrocarbon group, the compound with 1 carbon atom is acetic acid, and the compound with 2 carbon atoms is propanoic acid.

[0021] In this specification, the term "unit" refers to a repeating unit in a homopolymer or copolymer obtained by polymerizing a monomer, and the term "vinyl carboxylate unit" refers to a repeating unit obtained by polymerizing a vinyl carboxylate monomer, i.e., a repeating unit represented by "-CH(OCO-R)-CH-" (R is a hydrocarbon group). R is the same as described above for monocarboxylic acid, and preferred examples are also as described above.

[0022] Specific examples of the monocarboxylic acid vinyl ester unit in which the hydrocarbon group R is saturated aliphatic include vinyl propanoate unit, vinyl pivalate unit, vinyl decanoate unit, and vinyl methoxyacetate unit. Since it is preferable that the hydrophobicity is not too strong, preferred examples include vinyl acetate unit (R:CH3), vinyl propanoate unit (R:CH2CH3), vinyl butyrate unit (R:CH2CH2CH3), vinyl pentanoate unit (R:CH2CH2CH2CH3), vinyl pivalate unit (R:C(CH3)3), and vinyl hexanoate unit (R:CH2CH2CH2CH2CH3). Specific examples of the monocarboxylic acid vinyl ester unit in which R is aromatic include vinyl benzoate unit and its substitutes.

[0023] Below, we will show an example of a method for confirming that a biocompatible polymer is supported on a separation membrane, using a case where the biocompatible polymer is a monocarboxylate vinyl ester unit. When other polymers are used as the biocompatible polymer, ion signals derived from the molecular structure specific to the polymer used are measured using a TOF-SIMS device and an appropriate combination of other measurement methods.

[0024] The support of a polymer containing a vinyl monocarboxylate unit can be confirmed by combining measurements using a TOF-SIMS apparatus and X-ray photoelectron spectroscopy (XPS). Specifically, a peak derived from the carboxylate ion of the saturated aliphatic monocarboxylate is detected by measurements using a TOF-SIMS apparatus, and the structure of the monocarboxylate is revealed by analyzing the mass (m / z).

[0025] In measurements using a TOF-SIMS device, pulsed ions (primary ions) are irradiated onto the surface of a sample placed in an ultra-high vacuum, and the ions (secondary ions) emitted from the surface of the sample gain a certain amount of kinetic energy and are guided to a time-of-flight mass spectrometer. Each of the secondary ions accelerated with the same energy passes through the spectrometer at a speed according to its mass, but because the distance to the detector is constant, the time it takes to reach the detector (flight time) is a function of mass, and the mass distribution of the secondary ions, i.e., the mass spectrum, can be obtained by precisely measuring the distribution of this flight time. For example, if Bi3 is used as the primary ion species, ++ When using a fluorine-containing fluorine atom detector to detect secondary positive ions, the peak at m / z=43.01 is C2H3O + That is, it corresponds to acetic acid (aliphatic chain carbon number: 1). Also, the peak at m / z = 57.03 corresponds to C3H5O + , that is, it corresponds to propanoic acid (number of aliphatic chain carbon atoms: 2).

[0026] The conditions for measuring ion signals using a TOF-SIMS device are as follows: the measurement area is 100 μm × 100 μm, the primary ion acceleration voltage is 30 kV, and the pulse width is 7.8 nS. The detection depth in this analysis method is several nm or less. In this case, if the intensity of the ion signal derived from carboxylic acid relative to the total secondary ion intensity is 0.1% or less, it is judged to be noise, and no carboxylate ion is present. More specifically, the measurement is performed according to the method (2) described in the examples below.

[0027] Furthermore, when XPS measurement was performed, the peak of carbon derived from the ester group (COO) was CH x and CC main peak (near 285 eV) at +4.0 to 4.2 eV, it can be seen that the above carboxylic acid forms an ester bond. The measurement angle used for XPS is a value measured at 90°. When measured at a measurement angle of 90°, a region up to a depth of about 10 nm from the surface is detected. The peak of carbon derived from the ester group (COO) can be obtained by peak division of the peak that appears at +4.0 to 4.2 eV from the main peak derived from CH and CC of C1s. More specifically, the C1s peak is mainly CHx It is composed of five components: components derived from CC, C=C, and CS, mainly components derived from CO and CN, components derived from π-π* satellites, components derived from C=O, and components derived from COO. The amount of carbon derived from ester groups (atomic %) can be obtained by dividing the peak into the above five components and calculating the ratio of the peak area derived from ester groups to the total peak area derived from carbon. In this case, if the ratio of the peak area derived from ester groups to the total peak area derived from carbon is 0.4% or less, it is judged to be noise and no ester groups are present. More specifically, it is measured by the method (3) described in the examples below.

[0028] The amount of carbon (atomic percentage) derived from ester groups is preferably greater on the surface on the larger pore size side than on the surface on the smaller pore size side, and the amount of carbon derived from ester groups is preferably 1% or more and 10% or less on both surfaces.

[0029] From the viewpoint of sufficiently suppressing the adhesion of proteins, the coating polymer is preferably supported on the entire separation membrane, and in particular, it is preferable that the coating polymer is supported more in the dense layer to which proteins are likely to adhere after long-term use. Therefore, in the present invention, when the cross section of the porous separation membrane is measured by TOF-SIMS, it is important that the normalized average intensity of the ion signal derived from the biocompatible polymer of the dense layer is 2.0 times or more than the normalized average intensity of the ion signal derived from the biocompatible polymer of the coarse layer. If the ratio of normalized average intensities is less than 2.0, proteins will adhere to the dense layer especially after long-term use, and the protein permeability will decrease. The normalized average intensity of the ion signal derived from the biocompatible polymer of the dense layer is preferably 2.5 times or more, more preferably 3.0 times or more, than the normalized average intensity of the ion signal derived from the biocompatible polymer of the coarse layer. On the other hand, if the amount of biocompatible polymer is too high, there is a concern that the separation performance and permeability of the separation membrane will decrease, so it is preferable that the normalized average intensity of the ion signal derived from the biocompatible polymer in the dense layer is 10 times or less than the normalized average intensity of the ion signal derived from the biocompatible polymer in the coarse layer.

[0030] In addition, in order to suppress adhesion of proteins that permeate the separation membrane, it is preferable that the coating polymer is uniformly supported on the coarse layer. Therefore, in the present invention, when the membrane cross section is measured by TOF-SIMS, it is important that the minimum value of the normalized intensity of the ion signal derived from the biocompatible polymer of the coarse layer is 0.15 times or more of the maximum value of the normalized intensity of the ion signal derived from the biocompatible polymer of the coarse layer. If the ratio of such normalized average intensities is less than 0.15 times, proteins will adhere to the ion signal after long-term use, and the protein permeability will decrease. The minimum value of the normalized intensity of the ion signal derived from the biocompatible polymer of the coarse layer is more preferably 0.20 times or more of the maximum value of the normalized intensity of the ion signal derived from the biocompatible polymer of the coarse layer. On the other hand, the ratio of such normalized intensities is at most 1.

[0031] From the viewpoint of further improving protein permeability even after long-term use, the normalized average intensity of the ion signal derived from the biocompatible polymer acid in the coarse layer is preferably 0.5 or more, more preferably 1.0 or more, even more preferably 1.3 or more, and particularly preferably 1.5 or more. On the other hand, from the viewpoint of improving the permeability of the separation membrane, the normalized average intensity of the ion signal derived from the biocompatible polymer acid in the coarse layer is preferably 20 or less, more preferably 15 or less, and even more preferably 10 or less.

[0032] The normalized average intensity of the ion signal derived from the biocompatible polymer by the TOF-SIMS device can be measured by the following method. The measurement area is 100 μm × 100 μm, the primary ion acceleration voltage is 30 kV, and the pulse width is 7.8 nS. The normalized intensity is the value obtained by dividing the intensity of the detected ion signal derived from the biocompatible polymer by the intensity of the ion signal derived from the polymer that is the main component of the separation membrane. Furthermore, the normalized average intensity of the dense layer is the average value of the normalized intensity of the ion signal derived from the biocompatible polymer in the region up to 3 μm in the direction from the surface side where the pore size of the separation membrane is smaller toward the coarse layer. The maximum and minimum values ​​of the normalized intensity of the ion signal derived from the biocompatible polymer in the coarse layer are the maximum and minimum values ​​of the normalized intensity in the region equivalent to 80% of the entire thickness of the separation membrane in the direction from the surface side where the pore size of the separation membrane is larger toward the dense layer, respectively. Furthermore, the average value of the normalized intensity of the ion signal derived from the biocompatible polymer in the above region is the normalized average intensity of the coarse layer. More specifically, the measurement is carried out by the method (2) described in the Examples below.

[0033] The weight-average molecular weight of the coating polymer is preferably 1,000 or more, more preferably 5,000 or more, from the viewpoint of sufficiently suppressing protein adhesion and further improving protein permeability. On the other hand, the weight-average molecular weight of the coating polymer is preferably 1,000,000 or less, more preferably 500,000 or less, and even more preferably 100,000 or less, from the viewpoint of the efficiency of introduction into the separation membrane. The weight-average molecular weight of the coating polymer can be measured by gel permeation chromatography (GPC). More specifically, it is measured by the method (1) described in the examples below.

[0034] The coating polymer, particularly the polymer containing a monocarboxylic acid vinyl ester unit, is preferably a copolymer (hereinafter, sometimes simply referred to as "copolymer") consisting of a hydrophilic unit and a hydrophobic unit, and more preferably the hydrophobic unit contains a monocarboxylic acid vinyl ester unit. When the surface of the separation membrane is coated with a hydrophilic polymer such as polyethylene glycol or polyvinyl alcohol, the effect of suppressing adhesion of proteins and the like may be insufficient. This is thought to be because if the hydrophilicity of the separation membrane surface is too strong, the protein structure becomes unstable, and the adhesion of proteins cannot be sufficiently suppressed. In particular, in recent years, attention has been paid to the water around the polymer. A highly hydrophilic polymer has a strong interaction with water, and the motility of the water around the polymer decreases. On the other hand, it is thought that the structure of the protein is stabilized by water called adsorbed water. Therefore, if the motility of the adsorbed water of the protein and the water around the polymer are close, the protein structure is not destabilized, and the adhesion of the protein to the separation membrane surface can be suppressed. It is thought that the copolymer consisting of a hydrophilic unit and a hydrophobic unit can adjust the motility of the water around the polymer by selecting the hydrophilic group, hydrophobic group, and copolymerization ratio used, and can further improve the protein permeability. Here, the hydrophilic unit refers to a polymer of the monomer constituting the unit alone having a weight average molecular weight of 10,000 to 1,000,000 that is soluble in water. "Soluble" refers to a polymer having a solubility of more than 0.1 g in 100 g of water at 20°C.

[0035] As the monomer constituting the hydrophilic unit, a monomer having a solubility of more than 10 g is more preferable. Examples of such monomers include vinyl alcohol monomers, acryloylmorpholine monomers, vinylpyridine-based monomers, vinylimidazole-based monomers, and vinylpyrrolidone monomers. Two or more of these may be used. Among them, monomers having an amide bond, an ether bond, or an ester bond are preferable because they are not too hydrophilic compared to monomers having a carboxy group or a sulfonic acid group, and are easy to balance with hydrophobic monomers. In particular, vinylacetamide monomers, vinylpyrrolidone monomers, and vinylcaprolactam monomers having an amide bond are more preferable. Among these, vinylpyrrolidone monomers are even more preferable because the toxicity of the polymer is low. Therefore, in a preferred embodiment of the present invention, the coating polymer further contains a vinylpyrrolidone unit as a hydrophilic unit.

[0036] Monomers constituting the hydrophobic unit include at least monocarboxylic acid vinyl esters, and other monomers include acrylic acid esters, methacrylic acid esters, vinyl-ε-caprolactam, and the like.

[0037] From the viewpoint of further suppressing protein adhesion, the molar fraction of the hydrophobic units in the entire copolymer consisting of the hydrophilic unit and the hydrophobic unit is preferably 10% to 90%, more preferably 20% to 80%, and even more preferably 30% to 70%. In this case, the hydrophobic unit may be only a monocarboxylic acid vinyl ester unit, or may contain other hydrophobic units. By setting the molar fraction of the hydrophobic units to 90% or less, the increase in hydrophobicity of the entire copolymer can be suppressed, and the adhesion of proteins can be further suppressed. On the other hand, by setting the molar fraction of the hydrophobic units to 10% or more, the increase in hydrophilicity of the entire copolymer can be suppressed, structural instability and denaturation of proteins can be avoided, and adhesion can be further suppressed. The molar fraction can be calculated from the peak area by performing, for example, nuclear magnetic resonance (NMR) measurement. When the molar fraction cannot be calculated by NMR measurement due to peaks overlapping, the molar fraction may be calculated by elemental analysis.

[0038] The coating polymer is particularly preferably a copolymer consisting of a vinyl monocarboxylate unit and a vinylpyrrolidone unit, in which the molar ratio of the vinylpyrrolidone unit to the vinyl monocarboxylate unit is preferably 30:70 to 90:10, more preferably 40:60 to 80:20, and even more preferably 50:50.

[0039] The arrangement of the units in the copolymer may be, for example, a block copolymer, an alternating copolymer, or a random copolymer. Of these, an alternating copolymer or a random copolymer is preferred because the distribution of hydrophilicity and hydrophobicity in the entire copolymer is small. Among these, a random copolymer is more preferred because it is easy to synthesize.

[0040] Although not essential, it is preferable that the coating polymer is immobilized on the separation membrane by chemical bonding in order to prevent the coating polymer from eluting during use. The method of immobilization will be described later.

[0041] The separation membrane of the present invention may be in the form of, for example, a flat membrane or a hollow fiber membrane, with the hollow fiber membrane being preferred from the standpoint of processing efficiency and the like.

[0042] Examples of polymers that can be used as the material for the separation membrane of the present invention include polysulfone-based polymers, polystyrene, polyurethane, polyethylene, polypropylene, polycarbonate, polyvinylidene fluoride, polyacrylonitrile, etc. Two or more of these may be used. Among these, polysulfone-based polymers are preferably used because they are easy to form a separation membrane and are easy to coat a polymer containing a monocarboxylic acid vinyl ester unit.

[0043] The polysulfone-based polymer in the present invention is a polymer having an aromatic ring, a sulfonyl group, and an ether group in the main chain, and examples of such polymers include polysulfone, polyethersulfone, polyarylethersulfone, etc. As the polysulfone-based polymer used in the present invention, a polymer having a repeating unit represented by the following formula (1) or (2) is preferable.

[0044] [ka]

[0045] The polysulfone-based polymer may have other repeating units in addition to the repeating units represented by the above formula (1) or (2) as long as the effects of the present invention are not impaired. In this case, the content of the other repeating units is preferably 10 mass% or less of the total repeating units. In addition, the hydrogen atoms of the hydrocarbon skeleton of the polysulfone-based polymer may be substituted with other atoms such as alkyl groups, functional groups, and halogens, and the polysulfone-based polymer may be a modified product.

[0046] In the present invention, polysulfone-based polymers represented by the following formula (3) or (4) and consisting only of repeating units represented by the above formula (1) or (2) are particularly preferably used.

[0047] [ka]

[0048] In the formulas (3) and (4), n represents an integer of 50 or more, and preferably an integer of 50-200.

[0049] Specific examples of such polysulfone-based polymers include "Udel" (registered trademark) P-1700 and P-3500 (manufactured by SOLVAY), "Ultrason" (registered trademark) S3010 and S6010 (manufactured by BASF), etc. Two or more of these may be used.

[0050] Incidentally, "mainly composed of polysulfone-based polymer" means that the polysulfone-based polymer accounts for 50% by mass or more of the total components constituting the separation membrane. The content of the polysulfone-based polymer is preferably 75% by mass or more, and more preferably 90% by mass or more, of the components constituting the separation membrane.

[0051] The separation membrane of the present invention preferably further contains a hydrophilic polymer. That is, the separation membrane of the present invention is preferably composed of a mixed resin of the above-mentioned polysulfone-based polymer and hydrophilic polymer. The hydrophilic polymer plays a role of adjusting the viscosity of the pore-forming agent and membrane-forming solution when forming a porous separation membrane with the polysulfone-based polymer, and imparting a protein adhesion suppression effect. In the present invention, the hydrophilic polymer means a polymer soluble in water or ethanol, and is preferably a polymer soluble in these at 0.1 g / mL or more at 20°C.

[0052] As the hydrophilic polymer, a good solvent for polysulfone-based polymers and a polymer compatible with polysulfone-based polymers are preferable. Examples of such hydrophilic polymers include polyvinylpyrrolidone, polyethylene glycol, polyvinyl alcohol, and copolymers thereof. Examples of copolymers include copolymers of vinylpyrrolidone with vinyl acetate, vinyl propionate, and vinyl butanoate. Two or more of these may be used. Among them, polyvinylpyrrolidone and its copolymers are preferable from the viewpoint of compatibility with polysulfone-based polymers.

[0053] By using a hydrophilic polymer with a relatively low molecular weight (weight average molecular weight 1,000 to 200,000), the pore-forming effect is strengthened, and the water permeability of the hollow fiber membrane can be improved. On the other hand, when a hydrophilic polymer with a relatively high molecular weight (weight average molecular weight 200,000 to 1,200,000) is used, the molecular chain is long and the interaction with the polysulfone polymer is strong, so that it is likely to remain in the hollow fiber membrane, which contributes to improving the hydrophilicity of the hollow fiber membrane. Therefore, it is more preferable to combine a low molecular weight hydrophilic polymer with a high molecular weight hydrophilic polymer.

[0054] Among biopharmaceuticals, antibodies are expensive, but various separation and purification steps are performed in the manufacturing process, so it is necessary to minimize antibody loss. In particular, separation membranes and the like have a large surface area, so antibodies are easily adsorbed, and the recovery rate is likely to decrease. In particular, when multiple separation membranes are used in succession for treatment, the decrease in antibody recovery rate due to adhesion to the separation membranes becomes a major problem. Therefore, the antibody recovery rate is preferably 80% or more, more preferably 85% or more, and even more preferably 90% or more.

[0055] Furthermore, when antibodies or proteins to be removed adhere to the separation membrane, the treatment speed decreases with an increase in the amount of treatment liquid, leading to an extension of the treatment time and a decrease in the antibody recovery rate. Therefore, it is preferable that the treatment speed does not decrease with the amount of treatment liquid. For example, when a protein-containing treatment liquid is treated on a separation membrane under a constant pressure of 50 to 200 kPa, the ratio of the time required for 195 to 200 mL of filtrate to the time required for the initial 0 to 5 mL of filtrate to permeate, i.e., the permeation speed maintenance rate, is 200 g / m 2 At this point, the level is preferably 50% or more, more preferably 60% or more, and even more preferably 70% or more.

[0056] The manufacturing process of general biopharmaceuticals consists of a cell culture process in which animal cells are used to produce antibodies, a process in which the cells and antibodies are separated, the produced antibodies are collected, a purification process, a virus inactivation process, and a virus removal process. The cell and antibody separation process uses centrifugation or depth filtration. Protein A columns with immobilized Protein A, which specifically adsorbs antibodies, are mainly used to collect the produced antibodies. In the purification process, cation exchange columns and anion exchange columns are used to remove host cell proteins derived from the animal cells used to produce the antibodies. In the virus inactivation process, low pH treatment at a pH of 4 or less is generally used.

[0057] As described above, the separation membrane of the present invention has excellent protein non-adhesion properties and can therefore be used in the separation process between cells and antibodies. It can also be used to remove antibody aggregates that are difficult to remove using a protein A column in the antibody recovery process. Furthermore, it can be used to remove cell-derived proteins by size separation in the antibody purification process. When the separation membrane of the present invention is used in the virus removal process, it is preferable to use it as a virus removal membrane in the final stage of the general biopharmaceutical manufacturing process described above.

[0058] In particular, the current biopharmaceutical manufacturing process has a problem of low efficiency because each step is performed batchwise. Therefore, a method in which a solution containing cells and proteins is continuously treated with an apparatus in which separation membranes with various pore sizes corresponding to the size of the substances to be removed at each stage of the manufacturing process are arranged in series, and the desired cells or proteins are purified and recovered is highly productive and is preferable as a manufacturing process.

[0059] In addition to the manufacturing process of biopharmaceuticals, it can also be used in the virus removal process of blood products.Furthermore, it can be used in the medical field, such as hemodialysis to remove impurities from blood, plasma separation to separate blood cell components and plasma components in blood, concentrated ascites filtration and reinfusion therapy (CART), food applications such as beverages containing protein components, and water treatment applications.

[0060] As the porous separation membrane of the present invention, an example of a method for producing a hollow fiber membrane mainly composed of the above-mentioned polysulfone polymer is shown below. As a method for producing a hollow fiber membrane, a phase separation method is preferable. As a phase separation method, a method for inducing phase separation with a poor solvent (non-solvent induced separation method, NIPS) or a method for inducing phase separation by cooling a high-temperature membrane-forming solution using a solvent with relatively low solubility (thermally induced phase separation method, TIPS) can be used, but among them, the method for inducing phase separation with a poor solvent is particularly preferable.

[0061] In this membrane-forming process, phase separation proceeds due to contact between the membrane-forming solution and the poor solvent, and the structure of the porous hollow fiber membrane is determined. In particular, when the poor solvent is discharged as a core liquid into the inside of a double-tube nozzle and the membrane-forming solution is flowed on the outside to form a membrane, phase separation begins from the inside (inner surface) of the hollow fiber membrane where the membrane-forming solution and the poor solvent come into contact. After that, the poor solvent diffuses in the membrane thickness direction, and phase separation proceeds continuously. At this time, the pore size of the inner surface of the porous membrane hollow fiber membrane with the highest concentration of the poor solvent is the smallest, and the inner surface side has a dense structure, and the pore size becomes larger toward the inside of the membrane, resulting in a sparse structure. The pore size of the inner surface and the thickness of the dense layer can be adjusted by controlling the phase separation rate described above. Specifically, adjustments can be made to the poor solvent concentration, the discharge temperature, and the concentration of the polysulfone-based polymer in the membrane-forming solution. In particular, changing the poor solvent concentration is effective for adjusting the pore size and the dense layer. By adjusting the concentration of the poor solvent, the diffusion speed of the poor solvent changes, and the pore size of the surface and the thickness of the dense layer can be adjusted. Furthermore, by increasing the concentration of the polysulfone-based polymer in the membrane-forming solution, the polysulfone-based polymer, which is the main component of the hollow fiber membrane, is densely present, and the thickness of the dense layer can be increased.

[0062] By increasing the concentration of the polysulfone-based polymer in the membrane-forming solution, the mechanical strength of the hollow fiber membrane can be increased. For this reason, the concentration of the polysulfone-based polymer in the membrane-forming solution is preferably 10% by mass or more. On the other hand, by decreasing the concentration of the polysulfone-based polymer, the solubility can be improved and an increase in the viscosity of the membrane-forming solution can be suppressed. For this reason, the polymer density on the inner surface of the hollow fiber membrane can be appropriately suppressed, and the water permeability and the molecular weight cutoff can be improved. For this reason, the concentration of the polysulfone-based polymer in the membrane-forming solution is preferably 30% by mass or less.

[0063] When dissolving polysulfone polymers, it is preferable to dissolve them at high temperatures to improve solubility, but there is a concern that the heat may cause denaturation of the polymer or that the composition may change due to evaporation of the solvent. Therefore, the dissolving temperature is preferably 30°C or higher and 120°C or lower. However, the optimal range may vary depending on the type of polysulfone polymer and additives.

[0064] Furthermore, by adding the hydrophilic polymer to the membrane-forming solution, it is possible to expect the effect of improving water permeability as a pore-forming agent as described above, and the effect of suppressing protein adhesion by improving hydrophilicity. In addition, the viscosity of the membrane-forming solution can be adjusted by adding the hydrophilic polymer, and it is possible to suppress the generation of macrovoids that cause a decrease in membrane strength. The optimal amount of hydrophilic polymer to be added to the membrane-forming solution varies depending on the type and the intended performance, but it is preferably 1% by mass or more, and more preferably 20% by mass or less, of the entire membrane-forming solution.

[0065] The liquid (core liquid) discharged from the inner tube of the double tube nozzle is a mixture of a good solvent and a poor solvent for the polysulfone polymer, and the water permeability and molecular weight cutoff, i.e., pore size, of the hollow fiber membrane can be adjusted by adjusting the ratio of the good solvent and the poor solvent. The poor solvent is not particularly limited, but water and alcoholic solvents such as ethanol and isopropyl alcohol are used, and water is most preferably used. The good solvent is not particularly limited, but N-methylpyrrolidone and N,N-dimethylacetamide are preferably used.

[0066] When the membrane-forming solution and the core liquid come into contact with each other, the poor solvent induces phase separation of the membrane-forming solution, and solidification proceeds. The appropriate ratio of the good solvent and the poor solvent in the core liquid varies depending on the type of the good solvent and the poor solvent, but the poor solvent is preferably 10% by mass or more in the mixture of the two solvents, and is preferably 80% by mass or less.

[0067] The temperature of the double-tube nozzle during extrusion can affect the viscosity of the membrane-forming solution, the phase separation behavior, and the diffusion rate of the core liquid into the membrane-forming solution. In general, the higher the temperature of the double-tube nozzle, the faster the diffusion rate of the poor solvent, and the more the phase separation progresses, resulting in a larger pore size of the resulting hollow fiber membrane, and a larger water permeability and molecular weight cutoff. The temperature of the double-tube nozzle is preferably 20°C or higher, while 90°C or lower is preferable.

[0068] In addition, by increasing the draft ratio (= linear extrusion speed of the film-forming solution / pulling speed of the yarn) and stretching before solidification, the major axis of the surface can be made longer than the minor axis. Since the solution is stretched before solidification, problems such as breakage and cracking that are problems with the stretching method do not occur. The draft ratio is 1.5 or more, preferably 2 or more, and more preferably 2.5 or more. On the other hand, if the draft ratio is too high, it will lead to yarn breakage, so the draft ratio needs to be 10 or less, and preferably 9 or less.

[0069] After being discharged from the double-tube die, it is preferable to idle through a specified section called the dry section. In the dry section, the outer surface comes into contact with air, absorbing moisture from the air, which becomes a poor solvent, and phase separation progresses. Therefore, the opening rate of the outer surface can be adjusted by controlling the dew point of the dry section. The dew point of the dry section is preferably 60°C or lower, while 10°C or higher is preferable.

[0070] The dry length of the dry section is preferably 50 mm or more, more preferably 100 mm or more, while the dry length is preferably 600 mm or less.

[0071] After the dry section is idled, it is preferable to subject the solution to a coagulation bath mainly composed of a poor solvent for the polysulfone polymer. Water is preferably used as the poor solvent. When the solution enters the coagulation bath, the solution is coagulated by the large amount of poor solvent in the coagulation bath, and the membrane structure is fixed. In addition, a good solvent may be added to the coagulation bath as necessary. The higher the temperature of the coagulation bath or the higher the concentration of the good solvent, the more coagulation is suppressed and phase separation progresses, resulting in higher water permeability and molecular weight cutoff.

[0072] The hollow fiber membrane obtained by coagulation in a coagulation bath contains excess hydrophilic polymer derived from the solvent or the raw solution, and is therefore preferably subjected to further washing. A preferred washing method is to pass the membrane through a solvent having a composition in which the polysulfone polymer does not dissolve but the excess hydrophilic polymer dissolves. Examples of the solvent include water and alcohols such as ethanol, aqueous solutions in which good solvents are mixed to such an extent that the polysulfone polymer does not dissolve. Among these, water is preferred from the viewpoint of ease of handling. In addition, the washing efficiency can be increased by increasing the temperature of the solvent used for washing, and therefore the washing temperature is preferably 50 to 100°C.

[0073] The formed porous membrane may be immersed in an aqueous solution of a non-volatile compound such as glycerin in order to prevent the pore size from changing due to drying.

[0074] The formed porous membrane may be dried. Examples of the drying method include drying with hot air, drying with microwaves, and drying under reduced pressure, and hot air drying is preferably used.

[0075] Furthermore, when the porous membrane is a hollow fiber membrane, it is useful to provide crimps, since this improves the flow of the dialysis fluid when modularized. The crimp pitch is preferably in the range of 5 to 30 mm, and the amplitude is preferably in the range of 0.2 to 3 mm.

[0076] The fiber diameter of the hollow fiber membrane is not particularly limited, but can be measured by the following method. The membrane thickness of 16 randomly selected hollow fiber membranes is measured with a 1000x lens of a microwatcher (VH-Z100; KEYENCE Corporation) to obtain the average value a, and the value is calculated from the following formula. The outer diameter of the hollow fiber membrane refers to the average value obtained by measuring the outer diameter of 16 randomly selected hollow fiber membranes with a laser displacement meter (e.g., LS5040T; KEYENCE Corporation).

[0077] Inner diameter of hollow fiber membrane (μm) = outer diameter of hollow fiber membrane - 2 × membrane thickness Methods for modularizing the porous hollow fiber membrane include a method of fixing the membrane to a case while centrifuging, and a method of forming the hollow fiber membrane into a U-shape and fixing only the opening side of the hollow fiber membrane to the case. Although not particularly limited, an example is as follows. First, the hollow fiber membrane is cut to the required length, and the required number of membranes are bundled and placed in a cylindrical case. Then, temporary caps are placed on both ends, and potting material is placed on both ends of the hollow fiber membrane. In this case, the potting material is placed while rotating the module with a centrifuge, which is a preferred method because the potting material can be filled evenly. After the potting material has solidified, both ends of the hollow fiber membrane are cut so that both ends are open. A hollow fiber membrane module is obtained by attaching treated liquid inlet ports (headers) to both ends of the housing and plugging the headers and the nozzle parts of the housing.

[0078] Methods for supporting a coating polymer on a porous separation membrane include a method of adding the coating polymer to a raw solution or a core liquid during membrane formation, and a method of contacting the surface with a coating polymer solution after membrane formation. Among these, a method of contacting the surface with a coating polymer solution after membrane formation is preferred in that it does not affect the membrane formation conditions. Such a method may be any method, such as a method of immersing a separation membrane in the coating polymer solution, a method of passing the solution through the membrane, or a method of spraying the solution with a spray or the like. Among these, a method of passing the coating polymer solution through the separation membrane is preferred because it is possible to impart the coating polymer to the inside of the separation membrane.

[0079] When the coating polymer solution is passed through the separation membrane, the concentration of the coating polymer in the coating polymer solution is preferably 10 ppm or more, more preferably 100 ppm or more, and even more preferably 300 ppm or more from the viewpoint of more efficiently introducing the coating polymer, while the concentration of the coating polymer in the aqueous solution is preferably 100,000 ppm or less, more preferably 10,000 ppm or less from the viewpoint of suppressing elution from the module.

[0080] The solvent used to prepare the coating polymer solution is preferably water. However, if the coating polymer does not dissolve in water to a predetermined concentration, the coating polymer may be dissolved in an organic solvent that does not dissolve the separation membrane, or a mixed solvent of water and an organic solvent that is compatible with water but does not dissolve the separation membrane. Examples of organic solvents that can be used in the organic solvent or mixed solvent include, but are not limited to, alcohol solvents such as methanol, ethanol, and propanol.

[0081] The direction of passing the coating polymer solution through the separation membrane may be from the dense layer side to the coarse layer side or from the coarse layer side to the dense layer side of the separation membrane, but from the viewpoint of efficiently applying the coating polymer to the inside of the separation membrane, it is preferable to pass the solution from the coarse layer side to the dense layer side. If the size of the coating polymer used is larger than the pore size of the dense layer, passing the solution from the dense layer side will not allow the coating polymer to pass through the pores and will be concentrated on the surface of the dense layer, making it difficult to support the coating polymer in the coarse layer.

[0082] When passing the coating polymer solution, the application of pressure presses the polymer against the separation membrane surface, increasing the coating efficiency, so the pressure during coating is preferably 10 kPa or more, and preferably 50 kPa or more. On the other hand, if the pressure is too high, the amount of polymer coated increases, affecting the permeability of the separation membrane, so the pressure is preferably 300 kPa or less, and preferably 200 kPa or less.

[0083] As described above, the coating polymer is preferably immobilized on the separation membrane by chemical bonding. Methods for immobilization by chemical bonding include, but are not limited to, a method of contacting the coating polymer with radiation and a method of introducing reactive groups such as amino groups or carboxyl groups into the coating polymer and the surface of the separation membrane to be immobilized, followed by condensation.

[0084] Methods for introducing reactive groups onto the surface of a separation membrane include a method in which a monomer having a reactive group is polymerized to obtain a substrate having a reactive group on its surface, and a method in which the reactive group is introduced by ozone treatment or plasma treatment after polymerization.

[0085] The above-mentioned radiation irradiation can be performed using alpha rays, beta rays, gamma rays, X-rays, ultraviolet rays, electron beams, or the like. Radiation is performed when the separation membrane in the separation membrane module is in contact with a solution in which a coating polymer is dissolved, or when the solution in the separation membrane module is removed after the coating polymer is introduced onto the surface, or when the separation membrane is dried. This method is preferable because it can achieve sterilization of the separation membrane module at the same time as immobilizing the coating polymer. In this case, the radiation dose is preferably 15 kGy or more, more preferably 25 kGy or more. On the other hand, if the radiation dose is too high, deterioration and decomposition of the polymer are promoted, so the radiation dose is preferably 100 kGy or less.

[0086] In addition, an antioxidant may be used to suppress the crosslinking reaction of the coating polymer caused by irradiation with radiation. The antioxidant refers to a substance that has the property of easily donating electrons to other molecules, and examples thereof include water-soluble vitamins such as vitamin C, polyphenols, and alcohol solvents such as methanol, ethanol, and propanol, but are not limited to these. These antioxidants may be used alone or in combination of two or more types. When safety needs to be considered, antioxidants with low toxicity such as ethanol and propanol are preferably used.

[0087] When the hollow fiber membrane is irradiated in a dry state, where the moisture content is 20% or less of its own weight, the decomposition reaction proceeds more easily than with cross-linking of coating polymers, etc. Therefore, in order to suppress the decomposition reaction, the oxygen concentration during radiation exposure is preferably 1% or less, and more preferably 0.5% or less.

[0088] From the viewpoint of productivity, the permeation rate of the protein solution through the separation membrane is set at 200 mL / m 2 / h / kPa or more is better, 400mL / m 2 / h / kPa or more is preferable, and 600mL / m 2 / h / kPa or more is preferable. On the other hand, if the permeation rate is too high, the shear stress when in contact with the separation membrane increases, and there is a risk of denaturing the protein. Therefore, a permeation rate of 20,000 mL / m 2 / h / kPa or less is best, 15,000mL / m 2 / h / kPa or less is preferable.

[0089] The thickness of the dense layer can be measured by the following method. The hollow fiber membrane was wetted by soaking in water for 5 minutes, frozen with liquid nitrogen, and quickly folded, or cut with a microtome in a cryostat to expose the cross section, and then freeze-dried to obtain the hollow fiber membrane. The cross section of the hollow fiber membrane was observed at a magnification of 10,000 times using an SEM (S-5500, Hitachi High-Technologies Corporation), and the image was imported into a computer. The size of the image to be imported is not particularly limited, but 640 pixels x 480 pixels is recommended. If the pores of the cross section are blocked when observed with an SEM, the sample is prepared again. The blocking of the pores may occur when the hollow fiber membrane is deformed in the stress direction during the cutting process. The SEM image is cut to a length of 1 μm parallel to the surface of the hollow fiber membrane and an arbitrary length in the membrane thickness direction, and image analysis is performed using image processing software. The length of the analysis range in the membrane direction may be as long as the dense layer fits within it. If the dense layer does not fit within the observation field of the measurement magnification, two or more SEM images may be combined so that the dense layer fits within it. A threshold was determined by binarization processing so that the structural parts were bright and the other parts were dark, and an image was obtained in which the bright parts were white and the dark parts were black. If the structural parts and other parts cannot be separated due to differences in contrast within the image, the image may be cut into parts with the same contrast, binarized separately, and then joined together to return to a single image. If holes are observed twice in the depth direction, the shallower hole is measured. If a part of a hole falls outside the measurement range, that hole is excluded. The image contains noise, and dark parts with five or fewer consecutive pixels cannot be distinguished from noise, so they are treated as bright parts of the structure. To eliminate noise, dark parts with five or fewer consecutive pixels are excluded when measuring the number of pixels. The number of pixels of a scale bar showing a known length in the image is measured, and the length per pixel is calculated. The number of pixels of the hole is counted, and the hole area is calculated by multiplying the number of pixels of the hole by the square of the length per pixel. The diameter of the circle equivalent to the hole area was calculated using the following formula, and this was taken as the hole diameter. Hole diameter = (hole area ÷ pi) 1 / 2 ×2 For example, the thickness of the dense layer in which no pores with a diameter of 130 nm or more are observed can be measured by the following method. The pore area for a pore diameter of 130 nm is 1.3 × 10 4 (nm 2 ). Pores with a diameter of 130 nm or more were identified, and the layer in which no such pores were observed was regarded as the dense layer, and the thickness of the dense layer was measured in the vertical direction from the surface. A perpendicular line was drawn to the surface, and the shortest distance between the surface and the pores with a diameter of 130 nm or more on the perpendicular line (i.e., the distance between the surface and the pores with a diameter of 130 nm or more closest to the surface) was determined. Measurements were taken at five locations in the same image. The same measurement was taken on five more images, and the average value of the total 25 measurement data was calculated, and the value rounded off to two decimal places was regarded as the thickness of the dense layer.

[0090] When identifying pores with a diameter of 130 nm or more, a lower limit for the pore area to be detected can be set during analysis using image processing software. EXAMPLES

[0091] The present invention will be described below with reference to examples, but the present invention is not limited to these examples.

[0092] (1) Weight average molecular weight of coating polymer A 0.1N LiNO3 solution of water / methanol = 50 / 50 (volume ratio) was prepared as a GPC developing solution. The coating polymer used in each Example and Comparative Example was dissolved in this solution to a concentration of 1 mg / mL. 100 μL of this polymer solution was applied to a column (GMPW, manufactured by Tosoh Corporation). XL The sample was injected into a Shimadzu Prominence GPC system connected to a 100-molecular weight column. The flow rate was 0.5 mL / min, and the measurement time was 30 minutes. Detection was performed using a differential refractometer, and the weight-average molecular weight was calculated from the peak derived from the coating polymer that appeared at about 15 minutes of elution. The weight-average molecular weight was calculated by rounding off the 10th digit. A polyethylene oxide standard sample (0.1 kD to 1258 kD) manufactured by Agilent was used to create a calibration curve.

[0093] (2) TOF-SIMS measurement The hollow fiber membranes obtained in each of the Examples and Comparative Examples were wetted by soaking in water for 5 minutes, and then cut into lengths of approximately 1 cm or less using a single-edged blade or the like. The hollow fiber membranes that were kept wet were rapidly frozen to prepare sample blocks. The sample blocks were attached to an ultramicrotome equipped with a cryosystem, sliced ​​at a temperature of -65°C to a thickness of 200 nm, and mounted on slide glasses to prepare measurement samples. The measurement samples were dried at room temperature and normal pressure for 10 hours, and then subjected to measurement. The measurement apparatus and conditions were as follows. Measurement equipment: TOF.SIMS 5 (ION-TOF) Primary ion: Bi3 ++ Primary ion acceleration voltage: 30 kV Pulse width: 7.8ns Secondary ion polarity: positive Number of scans: 16 scans / cycle Measurement range: 100 x 100 μm 2 Mass range (m / z): 0~1,500.

[0094] From the obtained mass m / z spectrum, the presence or absence of carboxylate ions, which are ion signals specific to the biocompatible polymer used in this example, in the hollow fiber membrane was confirmed. However, if the carboxylate ion intensity relative to the total secondary ion intensity is 0.1% or less, it is judged to be noise, and carboxylate is deemed not to be present.

[0095] From the obtained secondary ion image, a line profile was extracted in an arbitrary range in which the hollow fiber membrane cross section was contained in the direction of the membrane thickness and 20 μm parallel to the inner surface of the hollow fiber membrane, and the intensity of the ion signal was calculated. However, the range in which the intensity of the ion signal derived from the polymer that is the main component of the hollow fiber membrane is not judged as noise, for example, when the main component of the hollow fiber membrane is a polysulfone-based polymer, the range in which the intensity of the ion signal derived from polysulfone is 10 or more was taken as the membrane cross-sectional area, and both ends of the range were taken as the membrane surface. When the membrane cross-section does not fit within the measurement range, two or more secondary ion images were synthesized so that the membrane cross-section fits. Figure 1 shows an example of a total secondary ion image obtained by photographing the cross-section of a porous separation membrane by TOF-SIMS. In Figure 1, the rectangular area indicated by the symbol 1 represents the extraction range of the line profile.

[0096] The intensity of the detected ion signal derived from carboxylic acid was divided by the intensity of the ion signal derived from the polymer, which is the main component of the hollow fiber membrane, to calculate the normalized intensity. Furthermore, the average value of the normalized intensity of the ion signal derived from carboxylic acid in the region from the surface side where the pore size of the hollow fiber membrane is small to 3 μm (rounded off to the first decimal place) in the direction of the coarse layer was calculated as the normalized average intensity of the dense layer. In addition, the maximum and minimum values ​​of the normalized intensity of the ion signal derived from carboxylic acid in the coarse layer were respectively the maximum and minimum values ​​of the normalized intensity in the region equivalent to 80% of the entire membrane thickness of the separation membrane from the surface side where the pore size of the separation membrane is larger in the direction of the dense layer (in the case of a membrane thickness of 50 μm, the region is 40 μm). Furthermore, the average value of the normalized intensity of the ion signal derived from carboxylic acid in the above region was calculated as the normalized average intensity of the coarse layer. Note that in each Example and Comparative Example, similar measurements were performed on three different cross sections, and the average value was calculated from the normalized average intensity of the dense layer and the coarse layer of each cross section, and the value was rounded off to the third decimal place. In addition, the average values ​​of the maximum and minimum values ​​of the coarse layer for each cross section were calculated, and the ratio of the minimum and maximum values ​​was calculated from these values ​​and rounded off to the third decimal place.

[0097] (3) X-ray photoelectron spectroscopy (XPS) measurement The hollow fiber membranes obtained in each Example and Comparative Example were cut into semi-cylindrical pieces with a single-edged blade, and the inner or outer surfaces of the hollow fiber membranes were measured. The measurement samples were rinsed with ultrapure water, dried at room temperature and 0.5 Torr for 10 hours, and then subjected to measurement. The measurement apparatus and conditions were as follows. Measurement device: ESCALAB220iXL (VG) Excitation X-ray: monochromatic Al Kα1,2 line (1486.6 eV) X-ray diameter: 0.15mm Photoelectron escape angle: 90° (detector tilt with respect to the sample surface).

[0098] The presence or absence of ester groups (COO) in the separation membrane was confirmed from the obtained photoelectron spectrum. The C1s peak is composed of five components: components derived mainly from CHx, CC, C=C, and CS, components derived mainly from CO and CN, components derived from π-π* satellites, components derived from C=O, and components derived from COO. The peak was divided into the above five components. The components derived from COO are peaks that appear at +4.0 to 4.2 eV from the main peaks of CHx and CC (around 285 eV). The peak area ratios of each component were calculated by rounding off to the first decimal place. If the ratio of the peak area derived from ester groups to the total peak area derived from carbon as a result of the peak division is 0.4% or less, it is judged to be noise and that no ester groups are present. Note that measurements were performed at two different points on the separation membrane, and the average value of the values ​​at the two points was used.

[0099] (4) Measurement of protein permeability A phosphate buffer solution containing 2.0 g / L of albumin (derived from bovine serum, Wako Pure Chemical Industries) or 2.0 g / L of IgG (derived from human serum, Oriental Yeast Co., Ltd.) was prepared as a stock solution. 200 mL of the resulting stock solution was passed through the hollow fiber membrane module obtained in each Example and Comparative Example in the direction from the outer surface of the hollow fiber membrane to the inner surface at an applied pressure of 200 kPa. At this time, the filtrate was sampled at 5 mL intervals. The protein permeation rate retention rate was calculated as the time it took for the first 0 to 5 mL of filtrate to permeate (F 5mL ) and the time it takes for 95-100 mL and 195-200 mL of filtrate to pass through (F 100mL or F200mL ) was calculated using the following formula: Protein permeation rate maintenance rate (%) = (F 100mL or F 200mL ) / (F 5mL ) x 100. The absorbance at 280 nm of the resulting solution (A 5mL , A 100mL , A 200mL ) and calculated the protein permeability using the following formula, rounded off to the nearest integer. Protein transmittance (%) = (A 100mL Or A 200mL ) / (A 5mL ) x 100.

[0100] (5) Measurement of the permeation rate of protein solution The value obtained in (4) above was used to calculate the following formula, rounding off to the nearest 1. Protein permeation rate (mL / m 2 / hr / kPa) = 5mL / A / F 5mL / 200kPa A: Effective membrane area of ​​the separation membrane (m 2 ) [Example 1] 20 parts by weight of polysulfone (SOLVAY's "Udel" (registered trademark) P-3500), 6 parts by weight of polyvinylpyrrolidone (ASHLAND LCC's Povidone (PLASDONE) K29 / K32), and 3 parts by weight of polyvinylpyrrolidone (ASHLAND LCC's Povidone (PLASDONE) K90) were added to a solution consisting of 70 parts by weight of N,N-dimethylacetamide and 1 part by weight of water, and the mixture was heated and dissolved at 90 ° C for 14 hours to obtain a membrane-forming stock solution. This membrane-forming stock solution was discharged from an orifice-type double cylindrical nozzle adjusted to 40 ° C., and at the same time, a solution consisting of 72 parts by weight of N,N-dimethylacetamide and 28 parts by weight of water was discharged from the inner tube as a core liquid. The discharged liquid was passed through a space with a dry length of 350 mm, and then introduced into a coagulation bath at 40 ° C. containing water to obtain an asymmetric hollow fiber membrane having a dense layer on the inner surface side and a coarse layer on the outer surface side. The obtained hollow fiber membrane had an inner diameter of 280 μm and a membrane thickness of 50 μm. The thickness of the dense layer in which no pores with a diameter of 130 nm or more were observed was 5.27 μm.

[0101] Ten of the obtained hollow fiber membranes were packed into a housing with a diameter of about 5 mm and a length of about 17 cm, and both ends were potted using an epoxy resin-based chemical reaction adhesive "Quick Mender" manufactured by Konishi Co., Ltd., and cut to open, to produce a hollow fiber membrane module. The hollow fiber membrane of the obtained module and the inside of the module were washed with distilled water for 30 minutes, and then 10 mL of an aqueous solution in which a vinylpyrrolidone / vinyl propanoate random copolymer (molar fraction of vinyl propanoate units 40%, weight average molecular weight 68,000) was dissolved at a concentration of 100 ppm and ethanol at a concentration of 1,000 ppm was passed from the outside to the inside of the hollow fiber membrane at an applied pressure of 100 kPa, coating the entire membrane. A hollow fiber membrane module 1 was obtained by irradiating with 25 kGy of gamma rays.

[0102] [Example 2] A hollow fiber membrane module 2 was obtained in the same manner as in Example 1, except that the amount of the coating solution, an aqueous solution of vinylpyrrolidone / vinyl propanoate random copolymer (molar fraction of vinyl propanoate units: 40%, weight average molecular weight: 68,000), passed through was 15 mL.

[0103] [Example 3] A hollow fiber membrane module 3 was obtained in the same manner as in Example 1, except that the polymer used in the coating solution was a vinylpyrrolidone / vinyl acetate random copolymer (BASF "KOLLIDON" (registered trademark) VA64).

[0104] [Example 4] A hollow fiber membrane module 4 was obtained in the same manner as in Example 1, except that the polymer used in the coating solution was a vinylpyrrolidone / vinyl pivalate random copolymer (molar fraction of vinyl pivalate units: 50%, weight average molecular weight: 7,700).

[0105] [Example 5] A hollow fiber membrane module 5 was obtained in the same manner as in Example 1, except that the polymer used in the coating solution was a vinylpyrrolidone / vinyl hexanoate random copolymer (molar fraction of vinyl hexanoate units: 40%, weight average molecular weight: 2,200).

[0106] [Example 6] A hollow fiber membrane module 6 was obtained in the same manner as in Example 1, except that the polymer used in the coating solution was a vinylpyrrolidone / vinyl propanoate random copolymer (molar fraction of vinyl propanoate units: 40%, weight average molecular weight: 100,000).

[0107] [Example 7] A hollow fiber membrane module 7 was obtained in the same manner as in Example 1, except that the amount of the coating solution, an aqueous solution of vinylpyrrolidone / vinyl propanoate random copolymer (molar fraction of vinyl propanoate units: 40%, weight average molecular weight: 68,000), passed through was 2.5 mL.

[0108] [Example 8] A hollow fiber membrane module 8 was obtained in the same manner as in Example 1, except that the applied pressure during coating was 50 kPa.

[0109] [Comparative Example 1] A hollow fiber membrane module 9 was obtained in the same manner as in Example 1, except that an aqueous solution of vinylpyrrolidone / vinyl propanoate random copolymer (vinyl propanoate unit molar fraction 40%, weight average molecular weight 68,000) was not passed through.

[0110] [Comparative Example 2] 16 parts by weight of polysulfone (SOLVAY's "Udel" (registered trademark) P-3500), 4 parts by weight of polyvinylpyrrolidone (ASHLAND LCC's Povidone (PLASDONE) K29 / K32), and 2 parts by weight of polyvinylpyrrolidone (ASHLAND LCC's Povidone (PLASDONE) K90) were added to a solution consisting of 77 parts by weight of N,N-dimethylacetamide and 1 part by weight of water, and the mixture was heated and dissolved at 90 ° C for 14 hours to obtain a membrane-forming stock solution. This membrane-forming stock solution was discharged from an orifice-type double cylindrical nozzle adjusted to 40 ° C., and at the same time, a solution consisting of 64 parts by weight of N,N-dimethylacetamide and 36 parts by weight of water was discharged from the inner tube as a core liquid. The discharged liquid was passed through a space with a dry length of 350 mm, and then introduced into a coagulation bath at 40 ° C. containing water to obtain an asymmetric hollow fiber membrane having a dense layer on the inner surface side and a coarse layer on the outer surface side. The obtained hollow fiber membrane had an inner diameter of 200 μm and a membrane thickness of 40 μm. The thickness of the dense layer in which no pores with a diameter of 130 nm or more were observed was 0.78 μm.

[0111] Ten of the obtained hollow fiber membranes were packed into a housing with a diameter of about 5 mm and a length of about 17 cm, and both ends were potted using an epoxy resin-based chemical reaction adhesive "Quick Mender" manufactured by Konishi Co., Ltd., and cut to open, to produce a hollow fiber membrane module. The hollow fiber membrane of the obtained module and the inside of the module were washed with distilled water for 30 minutes, and then 5 mL of an aqueous solution in which a vinylpyrrolidone / vinyl propanoate random copolymer (molar fraction of vinyl propanoate units 40%, weight average molecular weight 68,000) was dissolved at a concentration of 300 ppm and ethanol at a concentration of 1,000 ppm was passed from the inside to the outside of the hollow fiber membrane, coating the entire membrane. A hollow fiber membrane module 10 was obtained by irradiating with 25 kGy of gamma rays.

[0112] [Comparative Example 3] 18 parts by weight of polysulfone (SOLVAY's "Udel" (registered trademark) P-3500) and 9 parts by weight of vinylpyrrolidone / vinyl acetate random copolymer (BASF's "KOLLIDON" (registered trademark) VA64) were added to a solution consisting of 67 parts by weight of N,N-dimethylacetamide and 1 part by weight of water, and dissolved by heating at 90 ° C for 14 hours to obtain a membrane-forming stock solution. This membrane-forming stock solution was discharged from an orifice-type double cylindrical nozzle adjusted to 40 ° C., and at the same time, a solution consisting of 72 parts by weight of N,N-dimethylacetamide and 28 parts by weight of water was discharged from the inner tube as a core liquid. The discharged liquid was passed through a space with a dry length of 350 mm, and then introduced into a coagulation bath at 40 ° C. containing water to obtain a hollow fiber membrane with an asymmetric structure having a dense layer on the inner surface side and a coarse layer on the outer surface side. The inner diameter of the obtained hollow fiber membrane was 200 μm and the membrane thickness was 41 μm. The thickness of the dense layer in which no pores with a diameter of 130 nm or more were observed was 3.11 μm.

[0113] Ten of the obtained hollow fiber membranes were packed into a housing having a diameter of approximately 5 mm and a length of approximately 17 cm, and both ends were potted using an epoxy resin-based chemical reaction adhesive "Quick Mender" manufactured by Konishi Co., Ltd., and cut to create an opening to produce a hollow fiber membrane module. After that, the module was irradiated with 25 kGy of gamma rays to obtain hollow fiber membrane module 11.

[0114] Table 1 shows the main configurations and evaluation results of each of the examples and comparative examples.

[0115] [Comparative Example 4] A hollow fiber membrane module 12 was obtained in the same manner as in Example 1, except that an aqueous solution of vinylpyrrolidone / vinyl propanoate random copolymer (molar fraction of vinyl propanoate units: 40%, weight average molecular weight: 68,000) was passed from the inside to the outside of the hollow fiber membrane.

[0116] [Table 1] [Explanation of symbols]

[0117] 1. Line profile extraction range

Claims

1. The asymmetric structure has one surface being a dense layer that does not have pores with a pore size of 130 nm or more, and the other surface being a coarse layer, A porous separation membrane carrying a biocompatible polymer selected from the group consisting of 2-methacryloyloxyethyl phosphorylcholine (MPC), 2-methoxyethyl acrylate (PMEA), and a polymer containing a monocarboxylic acid vinyl ester unit, The amount of carbon (atomic percentage) derived from ester groups on the surface, as determined by X-ray photoelectron spectroscopy, is greater on the coarse layer side than on the dense layer side, and A porous separation membrane which satisfies the following (1) and (2) in a surface analysis by TOF-SIMS of a cross section including a dense layer and a coarse layer: (1) The minimum normalized intensity of the ion signal derived from the biocompatible polymer in the coarse layer is 0.15 times or more the maximum. (2) The normalized average intensity of the ion signal derived from the biocompatible polymer in the dense layer is 2.0 times or more the normalized average intensity of the ion signal derived from the biocompatible polymer in the coarse layer.

2. The porous separation membrane described in claim 1, wherein the biocompatible polymer further contains a biocompatible polymer selected from the group consisting of polyethylene glycol, polyethyleneimine, polyvinyl alcohol, and polyvinylpyrrolidone.

3. 3. The porous separation membrane according to claim 1, wherein the normalized average intensity of an ion signal derived from the biocompatible polymer in the coarse layer is 0.5 or more.

4. The porous separation membrane according to any one of claims 1 to 3, wherein the ion signal derived from the biocompatible polymer is a carboxylate ion signal.

5. 5. The porous separation membrane according to claim 1, wherein the biocompatible polymer is a polymer containing a monocarboxylate vinyl ester unit.

6. The porous separation membrane according to claim 5 , wherein the polymer containing a monocarboxylic acid vinyl ester unit is a copolymer comprising a hydrophobic unit containing a monocarboxylic acid vinyl ester unit and a hydrophilic unit.

7. The porous separation membrane according to claim 6 , wherein the hydrophilic unit is a vinylpyrrolidone unit.

8. The porous separation membrane according to any one of claims 5 to 7, wherein the weight average molecular weight of the polymer containing the monocarboxylic acid vinyl ester unit is 1,000 or more and 1,000,000 or less.

9. The porous separation membrane according to any one of claims 1 to 8, which is mainly composed of a polysulfone-based polymer.

10. The porous separation membrane according to any one of claims 1 to 9, which is a hollow fiber membrane.

11. The porous separation membrane according to claim 10 , wherein the inner surface is a dense layer.

Images