JP2005349093A - Polysulfone-based permselective hollow yarn membrane - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a polysulfone-based hollow yarn membrane which is excellent in safety and functional stability, is excellent in module assemblability, has little change on standing during the hemodialysis and is suitable for a highly water-permeable blood purifier for the treatment of chronic renal failure. <P>SOLUTION: The polysulfone-based permselective hollow yarn membrane of this invention is a polysulfone-based hollow yarn membrane which contains a hydrophilic polymer and simultaneously satisfies the conditions that (a) the content of the hydrophilic polymer at the outermost layer of a blood contacting side surface is 1.1 times or more of the content of the hydrophilic polymer at a surface neighboring layer of the blood contacting side surface, (b) the content of the hydrophilic polymer at the outermost layer of an opposite surface of the blood contacting side is 1.1 times or more of the content of the hydrophilic polymer at the outermost layer of the blood contacting side surface, and (c) the above membrane is comprised of an aggregate of agglomerated particles and the hydrophilic polymer is concentrated on a surface of the agglomerated particle. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a highly water-permeable hollow fiber blood purifier for medical use that has high safety and module assemblability, and has high water permeability with little change with time in blood purification treatment in which a large amount of water is removed.

  In blood purification therapy for the treatment of renal failure, natural materials such as cellulose, cellulose diacetate and cellulose triacetate, and synthetic polymers such as polysulfone, Modules such as hemodialyzers, hemofilters or hemodialyzers using dialysis membranes using polymers such as methyl methacrylate and polyacrylonitrile and ultrafiltration membranes as separation materials are widely used. In particular, modules using hollow fiber membranes as separation materials are important in the dialyzer field due to advantages such as reduction of the amount of blood circulating outside the body, high efficiency of removing substances in the blood, and productivity during module production. high.

  Among the above-mentioned membrane materials, polysulfone-based resins having high water permeability are attracting attention as the best match with the progress of dialysis technology. However, when a semi-permeable membrane is made of polysulfone alone, the polysulfone resin is hydrophobic, so it has poor affinity with blood and causes an airlock phenomenon. Can not.

As a method for solving the above-described problems, a method has been proposed in which a hydrophilic polymer is blended with a polysulfone-based resin to form a film, thereby imparting hydrophilicity to the film. For example, a method of blending a polyhydric alcohol such as polyethylene glycol is disclosed (for example, see Patent Documents 1 and 2).
JP-A-61-232860 JP 58-114702 A

Moreover, the method of mix | blending polyvinylpyrrolidone is disclosed (for example, refer patent document 3, 4).
Japanese Patent Publication No. 5-54373 Japanese Examined Patent Publication No. 6-75667

  The above-described problem is solved by the above-described method. However, in the hydrophilization technology by blending a hydrophilic polymer, the membrane performance of the hollow fiber membrane is large due to the concentration of the hydrophilic polymer existing on the inner surface of the membrane in contact with blood and the outer surface of the opposite membrane. Influence and its optimization is important. For example, blood compatibility can be ensured by increasing the hydrophilic polymer concentration on the inner surface of the membrane. However, if the surface concentration becomes too high, the amount of the hydrophilic polymer eluted into the blood increases, and the hydrophilicity to be eluted is increased. Since accumulation of macromolecules causes side effects and complications during long-term dialysis, it is not preferable.

  On the other hand, if the concentration of the hydrophilic polymer present on the outer surface of the opposite surface is too high, the possibility of high hydrophilic endotoxin (endotoxin) contained in the dialysate to enter the blood increases, causing side effects such as fever. New problems such as deterioration of module assemblability, etc. caused by the presence of hydrophilic polymers existing on the outer surface of the membrane when the membrane is dried and the hollow fiber membranes stick together (adhere). It is. Conversely, lowering the hydrophilic high molecular weight present on the outer surface of the membrane is preferable in terms of suppressing the invasion of endotoxin into the blood side, but since the hydrophilicity of the outer surface decreases, When returning a wet bundle of hollow fiber membranes for assembly to a wet state, the priming property, which is the ability to expel air when the wet operation is performed, decreases because the familiarity with the physiological saline used for the wet operation is reduced. This is not preferable because it leads to the generation of a problem.

As a measure for solving the above problems, the concentration of the hydrophilic polymer present in the dense layer on the inner surface of the hollow fiber membrane is within a specific range, and the mass ratio of the hydrophilic polymer present in the dense layer on the inner surface is outside. A method of at least 1.1 times the mass ratio of the hydrophilic polymer present in the surface layer is disclosed (see Patent Document 5). That is, the above technique increases the mass ratio of the hydrophilic polymer present on the dense surface of the inner surface to improve blood compatibility, and conversely lowers the mass ratio of the hydrophilic polymer present on the outer surface to reduce the membrane. This is a technique based on the idea of suppressing the occurrence of sticking between hollow fiber membranes that occurs when dried. In addition to this problem, this technique also improves the problem of endotoxin (endotoxin) contained in dialysate, which is one of the above-mentioned problems, entering the blood side. Since the mass ratio is too low, there remains a problem that leads to generation of the problem that the priming property, which is another problem described above, is lowered, and improvement thereof is necessary.
JP-A-6-165926

In addition, the hollow fiber membrane has a uniform membrane structure, but the hydrophilic polymer content in the inner surface, outer surface, and middle portion of the hydrophilic polymer hollow fiber membrane in the vicinity of the surface determined by the infrared absorption method is A method for improving the problem of endotoxin (endotoxin) contained in the dialysate, which is one of the problems described above, entering the blood side by specifying is disclosed (for example, see Patent Document 6). Although one of the above problems is improved by this technique, for example, as in the above technique, the problem that the priming property is lowered is not solved, and since the hole diameter of the outer surface of the hollow fiber membrane is large, the pressure resistance In particular, there is a concern that the hollow fiber membrane may be damaged when used for a treatment that increases fluid pressure more than conventional methods such as hemodiafiltration.
JP 2001-38170 A

Furthermore, by specifying the surface content of the hydrophilic polymer on the inner surface of the hollow fiber membrane, a method for improving blood compatibility and the amount of hydrophilic polymer eluted into the blood is disclosed (for example, (See Patent Documents 7 to 9).
JP-A-6-296686 JP 11-309355 A JP 2000-157852 A

  None of the above-mentioned techniques has been mentioned at all regarding the abundance ratio of the hydrophilic polymer on the outer surface of the opposite surface of the hollow fiber membrane, and all of the above-mentioned problems due to the abundance ratio of the hydrophilic polymer on the outer surface are described. It has not improved.

Among the above-mentioned problems, with regard to the problem that endotoxin (endotoxin) enters the blood side, a method using the characteristic that endotoxin has a hydrophobic portion in its molecule and is easily adsorbed to hydrophobic materials. Is disclosed (for example, see Patent Document 10). That is, it can be achieved by setting the ratio of the hydrophilic polymer to the hydrophobic polymer on the outer surface of the hollow fiber membrane to 5 to 25%. Certainly, this method is a preferable method for suppressing the invasion of endotoxin into the blood side, but to impart this property, it is necessary to remove the hydrophilic polymer present on the outer surface of the membrane by washing. In addition, this cleaning requires a great deal of processing time and is economically disadvantageous. For example, in the embodiment of the above-mentioned patent, shower cleaning with hot water at 60 ° C. and cleaning with hot water at 110 ° C. are performed for 1 hour each.
JP 2000-254222 A

In addition, it is preferable to reduce the hydrophilic high molecular weight present on the outer surface of the membrane from the viewpoint of suppressing the invasion of endotoxin into the blood side. Therefore, when the dried hollow fiber membrane bundle is returned to a wet state, the familiarity with the physiological saline used for wetting becomes low. This is not preferable because it leads to the problem. As a method for improving this point, for example, a method of blending a hydrophilic compound such as glycerin is disclosed (for example, see Patent Documents 11 and 12). However, in this method, since the hydrophilic compound acts as a foreign substance during dialysis, and the hydrophilic compound is susceptible to degradation such as light degradation, it leads to problems such as adversely affecting the storage stability of the module. In addition, there is also a problem of causing adhesive inhibition of the adhesive when the hollow fiber membrane bundle is fixed to the module in module assembly.
JP 2001-190934 A Japanese Patent No. 3193262

As a method for avoiding the sticking of the hollow fiber membranes, which is another problem described above, a method is disclosed in which the porosity of the outer surface of the membrane is 25% or more (for example, Patent Document 6 and Patents listed above). Reference 13). Certainly, this method is a preferable method for avoiding the sticking, but has a problem that the membrane strength is lowered due to the high porosity, leading to problems such as blood leakage.
JP-A-7-289863

On the other hand, a method has been disclosed in which the porosity and area of the outer surface of the membrane are specified (see, for example, Patent Document 14).
JP 2000-140589 A

In addition, when the membrane pore diameter is enlarged to obtain high water permeability as shown in the present invention, fine particle components such as proteins and lipids in the blood can easily enter the membrane pores, and hollow fibers are used during blood purification treatment. There was a problem that the film was changed with time due to clogging and the like, and the initial solute removal performance could not be maintained. In particular, these problems become apparent in the case of treatment that actively removes water from the blood, such as hemofiltration and hemodiafiltration. Manufactured as a dry-wet spinning method that reduces the clogging of membranes due to the adsorption of proteins, etc. to the membrane surface by increasing the blood flow rate by reducing the inner diameter of the hollow fiber membrane and using gas as the hollow forming agent There is a method of smoothing the inner surface of the film. (For example, refer to Patent Document 15). However, since this method uses a gas as a hollow forming agent, when winding the hollow fiber membrane into a bobbin shape, it does not contain commonly used liquids such as liquid paraffin and isopropyl alcohol. Decrease or collapse, and there is a high possibility that residual blood will occur during clinical use. Furthermore, when the inner diameter of the hollow fiber membrane is reduced, there is a problem that the pressure loss on the blood side increases.
JP-A-8-970

Further, regarding performance degradation, a technology is disclosed that takes into consideration the particle structure forming a membrane and the presence state of the hydrophilic polymer on a permselective membrane composed of a polysulfone-based polymer and a hydrophilic polymer. (For example, refer to Patent Document 16). With this technique, it is possible to stabilize membrane permeability in blood over time by allowing a high concentration of hydrophilic polymer to exist on the surface of particle components that certainly form a membrane. However, this prior art does not show any suggestion or sufficient improvement to the elution of the hydrophilic polymer from the membrane or the assembling property of the module, which is the subject of the present invention. In addition, although there is a suggestion of stabilization of water permeability in the blood, the stability of the permeation performance of the aforementioned protein region size substances such as β2 microglobulin, and the change over time during the treatment procedure especially with a large amount of permeate Is not satisfactory.
JP-A-7-289866

  An object of the present invention is to provide a medical hollow fiber blood purifier having high water permeability that is excellent in safety and module assemblability, and has little change with time in blood purification treatment in which a large amount of water is removed.

As a result of intensive studies to solve the above technical problems, the present invention provides a hollow fiber membrane mainly composed of a polysulfone resin and a hydrophilic polymer.
(A) The content of the hydrophilic polymer in the outermost surface layer on the blood contact side surface in the hollow fiber membrane is 1.1 times or more than the content of the hydrophilic polymer in the surface vicinity layer on the blood contact side surface It is.
(B) The content of the hydrophilic polymer in the outermost layer on the surface opposite to the blood contact side in the hollow fiber membrane is 1.1 times or more than the content of the hydrophilic polymer in the outermost layer on the blood contact side surface It is.
(C) The hollow fiber membrane has at least a dense layer on the blood contact side surface, the dense layer is composed of aggregates of aggregated particles having an average diameter of 20 to 200 nm, and a hydrophilic polymer is formed on the surface of the aggregated particles. By using a concentrated polysulfone-based permselective hollow fiber membrane, the above-mentioned problems can be solved.

  As a detailed embodiment, the content of the hydrophilic polymer in the outermost surface layer on the blood contact side is usually 5 to 60% by mass, suitably 10 to 50% by mass, more preferably 20 to 40% by mass. . The range of the content of the hydrophilic polymer in the surface vicinity layer adjacent thereto is usually about 2 to 37% by mass, and optimally about 5 to 20% by mass. Furthermore, the hydrophilic polymer content of the outermost surface layer opposite to the blood contact side is 1.1 times or more than the hydrophilic polymer content of the outermost surface layer of the blood contact side. A content of about 25 to 50% by mass on the outer surface of the hollow fiber membrane is sufficient. Appropriate content distribution of each layer can be determined in consideration of the elution of the hydrophilic polymer from the hollow fiber membrane to 10 ppm or less. In addition, the film is formed of an aggregate of aggregated particles having a diameter of 20 to 200 nm, and the hydrophilic polymer is concentrated on the surface of the aggregated particles.

  The hollow fiber blood purifier of the present invention is excellent in safety and module assemblability, and is suitable as a medical hollow fiber blood purifier having high water permeability used for the treatment of chronic renal failure.

Hereinafter, the present invention will be described in detail.
The hollow fiber membrane used in the present invention is characterized in that it is composed of a polysulfone resin containing a hydrophilic polymer. The polysulfone resin in the present invention is a general term for resins having a sulfone bond and is not particularly limited. However, for example, a polysulfone resin or a polyethersulfone resin having a repeating unit represented by Chemical Formula 1 or Chemical Formula 2 is a polysulfone resin. It is preferred because it is widely available on the market and is easily available.

  The hydrophilic polymer in the present invention is a material such as polyethylene glycol, polyvinyl alcohol, polyvinyl pyrrolidone, carboxymethyl cellulose, polypropylene glycol, glycerin, starch and derivatives thereof, but it is preferable to use polyvinyl pyrrolidone from the viewpoint of safety and economy. This is an embodiment. Polyvinyl pyrrolidone is a water-soluble polymer compound obtained by vinyl polymerization of N-vinyl pyrrolidone. There are products of various molecular weights. In general, it is preferable to use a low molecular weight in terms of hydrophilicity imparting efficiency, while it is preferable to use a high molecular weight from the viewpoint of lowering the elution amount, but depending on the required characteristics of the hollow fiber membrane of the final product, it is appropriate. Selected. Those having a single molecular weight may be used, or two or more products having different molecular weights may be mixed and used. Moreover, you may use what refine | purified a commercial product and sharpened molecular weight distribution, for example. As the molecular weight of polyvinylpyrrolidone, those having a mass average molecular weight of 10,000 to 1,500,000 can be used. Specifically, for example, those having a molecular weight of 9,000 (K17) commercially available from BASF, and the same shall apply hereinafter, 45,000 (K30), 450,000 (K60), 900,000 (K80), 1,200 000 (K90) is preferable, and in order to obtain the intended use, characteristics, and structure, they may be used alone or in combination of two or more. In the present invention, it is most preferable to use K90 alone.

  In the present invention, the content of the hydrophilic polymer in the outermost layer on the blood contact side surface in the above (i) polysulfone-based hollow fiber membrane is 1. To make it more than 1 time means that the content of the hydrophilic polymer on the outermost surface is larger than the layer near the surface and that the content of the hydrophilic polymer on the outermost layer is optimal 20 to 40% by mass. The range of the content of the hydrophilic polymer in the layer near the surface adjacent to it must be in the range of about 2 to 37% by mass. It is based on this reason that the appropriate content of the hydrophilic polymer in the surface vicinity layer is actually about 5 to 20% by mass. This means that it is acceptable up to about 10 times, but if the magnification of the difference becomes too large, the diffusion movement of the hydrophilic polymer may reversely move from the outermost layer to the vicinity of the surface. There is also a situation that it is difficult to manufacture a hollow fiber membrane having such a structure. Appropriate content of the hydrophilic polymer in the outermost layer on the blood contact side surface and in the surface vicinity layer is a numerical value based on 5 to 20% by mass, which is an appropriate content of the hydrophilic polymer in the surface vicinity layer. By calculating with a simple multiplier of about 1.1 to 10 times (5 to 20% by mass), the optimal hydrophilic polymer content of the outermost layer on the blood contact side surface is 20 to 40% by mass. % Means that it can be The ratio is usually about 1.1 to 5 times, and in some cases, it is preferable that the hydrophilic polymer exists in a range of about 1.2 to 3 times optimally. Actually, the magnification can be arbitrarily determined in consideration of the performance of the hollow fiber membrane. For example, if the content of the hydrophilic polymer in the surface vicinity layer is 5% by mass of the lowest value, the content of the hydrophilic polymer in the outermost layer is 20 to 40% by mass of an appropriate amount corresponding to 4 to 8 times. It can also be taken within the range.

  In the present invention, the content of the hydrophilic polymer in the outermost layer on the surface opposite to the blood contact side in the (b) polysulfone-based hollow fiber membrane is based on the content of the hydrophilic polymer in the outermost layer on the blood contact side surface. , 1.1 times or more is sufficient if the content of the hydrophilic polymer on the outer surface of the hollow fiber membrane is about 25 to 50% by mass, and the hydrophilic polymer of the outermost layer on the inner surface The content is preferably 1.1 times or more. If the content of the hydrophilic polymer on the outer surface is too small, the amount of protein in the blood adsorbed on the support layer portion of the hollow fiber membrane will increase, and blood compatibility and permeation performance may decrease. Conversely, if the content of the hydrophilic polymer on the outer surface is too high, the endotoxin (endotoxin) contained in the dialysate is likely to enter the blood side, leading to side effects such as fever, When the membrane is dried, a hydrophilic polymer present on the outer surface of the membrane is interposed, and the hollow fiber membranes are fixed to each other, which may cause problems such as deterioration in module assembly.

  In the present invention, (c) the polysulfone-based permselective hollow fiber membrane has a dense layer on at least the blood contact side surface, and the dense layer comprises an aggregate of aggregated particles having an average diameter of 20 to 200 nm. It is preferable that the hydrophilic polymer is concentrated on the surface of the particles. In the permselective membrane comprising the polysulfone resin and the hydrophilic polymer, the membrane has a dense layer on the blood contact surface, and the dense layer is composed of aggregates of aggregated particles. Such a membrane is presumed to have two channels in material permeation. The first permeation channel is mass transfer that permeates through the gaps between the particles in the structure in which the particles are aggregated. In the case of such a film structure, the size of the particles is greatly related to the size of the solute to be permeated or the size of the solute not to be permeated. In the region to be separated according to the present invention, since the Stokes diameter of the low molecular protein to be permeated is 1 nm or less, the pore size of the membrane is suitably several nm. For example, the Stokes diameter of β2 microglobulin to be removed is about 0.3 nm, and albumin to be retained is about 0.8 nm. In the blood purification membrane, fractionation between these two proteins is necessary, and the pore diameter formed by the particle gaps in the dense layer must also be large enough to fractionate this. The gap formed by the particles having an average diameter of 20 to 200 nm of the present invention has a size of around several nm, and forms an optimum pore size for fractionation of a low molecular protein region. More preferably, it is 40-150 nm, More preferably, it is 60-120 nm. Further, when the size of the particles is very polydispersed, voids formed from the particles are very non-uniform. Large voids are undesirable because they cause albumin permeation at the beginning of filtration. Further, as the filtration proceeds, the amount of albumin entering the gap increases, so that clogging occurs and the change with time of the pores increases. At this time, the permeation performance of substances that can permeate sufficiently, such as β2 microglobulin, is also affected over time. On the other hand, if the gap is too small, the initial transmission performance itself is degraded. From these factors, the particle size distribution of the particles is preferably 0.5 times or less of the average value as the standard deviation. More preferably, it is 0.3 times or less. In particular, in the blood purification treatment accompanied by an operation with a large amount of filtration, which is an object of the present invention, an improvement effect for these problems is more exhibited.

  The second permeation channel is a mass transfer region to low molecules such as water and urea having permeability to the aggregated particles themselves forming the dense layer. Aggregated particles are composed of a blend polymer in which both the hydrophilic polymer and the hydrophobic polymer of the present invention are compatible, and the particles are considered to be made of a matrix formed by these polymers. These matrices have a sufficiently large network (portions where no polymer chains are present) to allow small molecules such as urea (Stokes diameter <1 nm) to pass through, and are responsible for the permeation of small molecules and blood purification. Expresses the ability to remove small molecules by diffusion in For the formation of the optimum matrix size for performing low-molecular separation by dialysis, the aforementioned polymer group is excellent and has been selected as a material. The blend material of the hydrophobic polymer and the hydrophilic polymer of the present invention is also a material that can be suitably used. From these points, the second channel does not necessarily have a particle structure, and the focus on stabilization of the transmission performance, which is one of the objects of the present invention, is the first by the above-described aggregated particle structure. In channel optimization technology.

  In the present invention, changes in the removal performance of low molecular weight proteins due to clogging of the membrane during blood purification treatment are suppressed by controlling the size and distribution of aggregated particles that form a dense layer that becomes a membrane active separation active layer. I found out that I can. In other words, the decrease over time in the sieving coefficient of β2-microglobulin using bovine blood is determined by the sieving coefficient of β2-microglobulin after 15 minutes (SCβ2MG (15 minutes)) and β2-microglobulin after 120 minutes. It is preferable that SCβ2MG (120 minutes) / SCβ2MG (15 minutes) ≧ 0.6. When SCβ2MG (120 minutes) / SCβ2MG (15 minutes) is smaller than 0.6, the clogging of the membrane becomes large and the removal amount of β2MG is insufficient, so that there is a possibility that the therapeutic effect cannot be obtained sufficiently. In the hollow fiber membrane of the present invention, the ratio of SCβ2MG (120 minutes) / SCβ2MG (15 minutes) preferably does not exceed 1. When the ratio exceeds 1, the pore diameter of the membrane increases with time during blood purification treatment, and albumin leakage described later may increase. Furthermore, in order to effectively demonstrate the effects of the present invention, it is preferable that the β2 microglobulin removal ability itself is high. In particular, in the embodiment of the present invention, it is preferable that sufficient performance is maintained even 120 minutes after the start of filtration. Therefore, SCβ2MG (120 minutes) ≧ 0.5 is a preferred embodiment, and more preferably 0.6. That's it.

  In addition, it is desirable to improve the removal performance of β2 microglobulin, but if the leakage of albumin, which is a useful protein, increases, the effectiveness in blood purification treatment will be diminished. Therefore, it is necessary to use a membrane that sufficiently suppresses leakage of albumin. In the present invention, the sieving coefficient of albumin using bovine blood is 0.1 or less. In particular, since the leakage amount of albumin decreases with time, it is desirable that the sieving coefficient be a value measured at 15 minutes, which is the initial stage of the filtration operation. That is, it is preferable that SCalb (15 minutes) ≦ 0.1. More preferably, SCalb (15 minutes) ≦ 0.07. More preferably, SCalb (15 minutes) ≦ 0.05.

  The composition ratio in the membrane of the hydrophilic polymer with respect to the polysulfone resin in the present invention is not particularly limited as long as the hydrophilic polymer has a sufficient hydrophilicity and a high water content, and can be arbitrarily set. However, 1-20 mass% is preferable with the mass ratio of the hydrophilic polymer with respect to 80-99 mass% of polysulfone resin, and 3-15 mass% is more preferable. If it is less than 1% by mass, the hydrophilicity-imparting effect of the film may be insufficient. On the other hand, if it exceeds 20% by mass, the effect of imparting hydrophilicity is saturated and the amount of elution from the membrane of the hydrophilic polymer increases, and the amount of elution from the membrane of the hydrophilic polymer described later may exceed 10 ppm. There is.

The preferred embodiment of the present invention will be described in detail based on technical requirements. In the polysulfone-based hollow fiber membrane containing a hydrophilic polymer, the polysulfone-based selectively permeable hollow characterized by satisfying the following characteristics at the same time: It becomes a thread membrane.
(1) Elution of the hydrophilic polymer from the hollow fiber membrane is 10 ppm or less.
(2) The content of the hydrophilic polymer in the outermost layer on the blood contact side surface in the polysulfone-based hollow fiber membrane is 20 to 40% by mass.
(3) The content of the hydrophilic polymer in the surface vicinity layer on the blood contact side surface in the polysulfone-based hollow fiber membrane is 5 to 20% by mass.
(4) The content of the hydrophilic polymer in the outermost layer on the surface opposite to the blood contact side in the polysulfone-based hollow fiber membrane is 25 to 50% by mass, and the content of the hydrophilic polymer in the outermost layer on the inner surface 1.1 times or more.
(5) The hollow fiber membrane has at least a dense layer on the blood contact side surface, the dense layer is composed of aggregates of aggregated particles having an average diameter of 20 to 200 nm, and a hydrophilic polymer is formed on the surface of the aggregated particles. It is concentrated.

  In the present invention, as described above, it is preferable that (1) the elution amount of the hydrophilic polymer from the hollow fiber membrane is 10 ppm or less (Requirement 1). When the elution amount exceeds 10 ppm, side effects and complications due to long-term dialysis due to the eluted hydrophilic polymer may occur. A method of satisfying the characteristics can be achieved, for example, by setting the ratio of the hydrophilic polymer to the hydrophobic polymer within the above range or optimizing the film forming conditions of the hollow fiber membrane.

  In the present invention, as described above, the content of the hydrophilic polymer in the outermost layer on the blood contact side surface in the polysulfone-based hollow fiber membrane is preferably 20 to 40% by mass (requirement 2). First, the content of the hydrophilic polymer in the outermost layer on the blood contact side surface of the polysulfone-based hollow fiber membrane can be arbitrarily set in a wide range such as 10 to 50% by mass, for example, The optimum content for appropriately achieving the effects of the present invention is preferably composed mainly of polysulfone-based resin 60 to 80% by mass and hydrophilic polymer 20 to 40% by mass. If it is less than 20% by mass, the hydrophilicity of the surface of the hollow fiber membrane that comes into contact with blood is low, blood compatibility is deteriorated, blood coagulation tends to occur on the surface of the hollow fiber membrane, and the hollow fiber membrane is blocked by the coagulated thrombus. It may occur and the separation performance of the hollow fiber membrane may decrease, or the residual blood after use for hemodialysis may increase. The content of the hydrophilic polymer in the outermost layer on the inner surface of the hollow fiber membrane is more preferably 21% by weight or more, further preferably 22% by weight or more, and further preferably 23% by weight or more. On the other hand, when the amount exceeds 40% by mass, the hydrophilic polymer eluted in blood increases, and side effects and complications due to long-term dialysis due to the eluted hydrophilic polymer may occur. The content of the hydrophilic polymer in the outermost layer on the inner surface of the hollow fiber membrane is more preferably 39% by mass or less, further preferably 38% by mass or less, and further more preferably 37% by mass or less.

  In the present invention, as described above, it is preferable that the hydrophilic polymer content in the surface vicinity layer on the blood contact side surface in (3) the polysulfone-based hollow fiber membrane is 5 to 20% by mass (Requirement 3). In the polysulfone-based hollow fiber membrane of the present invention, the content of the hydrophilic polymer in the surface vicinity layer on the blood contact side surface is mainly composed of a polysulfone-based resin of 60 to 99% by mass and a hydrophilic polymer of 1 to 40% by mass. However, it is preferable that the content of the appropriate hydrophilic polymer is 5 to 20% by mass. Usually, 7 to 18% by mass is more preferable. As described above, the content of the hydrophilic polymer in the outermost layer on the blood contact side surface of the polysulfone-based hollow fiber membrane is preferably higher than the blood compatibility point, but when the content increases, the hydrophilicity to blood increases. Since this is a contradictory phenomenon in which the amount of molecular elution increases, it is determined to be about 20 to 40% by mass in consideration of the appropriate range.

  On the other hand, the content of the hydrophilic polymer in the inner layer near the surface of the hollow fiber membrane can be in a relatively wide range of 1 to 40% by mass, but more than the outermost layer, for example, 30% by mass of the outermost layer and 35% of the layer near the surface. %, The diffusion and movement of the hydrophilic polymer to the outermost layer becomes active, and the content of the hydrophilic polymer in the outermost layer accumulates more than a predetermined design value, which is not preferable. In short, considering the mechanism of supplying the amount of hydrophilic polymer consumed in the outermost layer by diffusion movement, the hydrophilic polymer content in the surface vicinity layer is relatively lower than that of the outermost layer. It is more preferably no greater than mass%, and even more preferably no greater than 18 mass%. In addition, if the content of the hydrophilic polymer in the vicinity of the inner surface of the hollow fiber membrane is too small, the hydrophilic polymer is not supplied to the outermost layer, so that the solute removal performance and the blood compatibility stability with time decrease. there is a possibility. Therefore, the content of the hydrophilic polymer in the vicinity of the inner surface of the hollow fiber membrane is more preferably 6% by mass or more, and further preferably 7% by mass or more as the optimum amount. The content of the hydrophilic polymer in the vicinity of the surface is slightly higher than the average content of 80 to 99% by mass of the polysulfone polymer constituting the hollow fiber membrane of the present invention and 1 to 20% by mass of the hydrophilic polymer. This is common.

  This requirement 3 is a factor for overcoming the above-mentioned contradictory phenomenon and achieving the optimization of the above phenomenon at a high level that could not be achieved by the prior art, and is one of the novel features of the present invention. That is, the content of the hydrophilic polymer in the outermost layer of the hollow fiber membrane that governs blood compatibility was set to the lowest level at which blood compatibility could be expressed. However, with the content of the outermost layer, the initial blood compatibility is satisfactory, but the hydrophilic polymer present in the outermost layer gradually elutes into the blood after long-term dialysis. The problem that blood compatibility gradually decreases occurs. This persistence of blood compatibility is improved by specifying the hydrophilic polymer content in the surface vicinity layer on the blood contact side surface of the polysulfone-based hollow fiber membrane. By specifying the hydrophilic polymer content in the near-surface layer, blood compatibility by decreasing the content of the hydrophilic polymer in the outermost layer due to elution of the outermost hydrophilic polymer into the blood due to the progress of dialysis It has been completed by the technical idea of ensuring a sustained decrease in blood compatibility in which the property deteriorates with time by transferring the hydrophilic polymer present in the surface vicinity layer to the outermost layer. Therefore, if the content of the hydrophilic polymer in the surface vicinity layer on the blood contact side surface is less than 5% by mass, it may be insufficient to suppress the decrease in blood compatibility. On the other hand, when the amount exceeds 20% by mass, the amount of the hydrophilic polymer eluted in the blood increases, which may cause side effects and complications due to long-term dialysis. There has been no example of elucidating the material content based on the proper content and structure of the hydrophilic polymer in the near-surface layer and near-surface layer of this hollow fiber membrane, and based on the very novel knowledge of the inventors etc. It is.

  In the present invention, as described above, (4) the content of the hydrophilic polymer in the outermost layer on the surface opposite to the blood contact side in the polysulfone-based hollow fiber membrane is 25 to 50% by mass, and the outermost layer on the inner surface is The content is preferably 1.1 times or more of the content of the hydrophilic polymer (Requirement 4). If the content of the hydrophilic polymer on the outer surface is too small, the amount of protein in the blood adsorbed on the support layer portion of the hollow fiber membrane will increase, and blood compatibility and permeation performance may decrease. Although it may be composed mainly of 90 to 40% by mass of a polysulfone resin and 10 to 60% by mass of a hydrophilic polymer, the content of the hydrophilic polymer on the outer surface is actually 27% by mass. The above is more preferable, and 29% by mass or more is more preferable. In the case of a dry film, the priming property may be deteriorated. Conversely, if the content of the hydrophilic polymer on the outer surface is too high, the endotoxin (endotoxin) contained in the dialysate is likely to enter the blood side, leading to side effects such as fever, When the membrane is dried, a hydrophilic polymer present on the outer surface of the membrane is interposed, and the hollow fiber membranes are fixed to each other, which may cause problems such as deterioration in module assembly. The content of the hydrophilic polymer on the outer surface of the hollow fiber membrane is more preferably 43% by mass or less, and further preferably 40% by mass or less.

In addition, as one of the requirements 4, the content of the hydrophilic polymer in the outermost surface outer layer is preferably 1.1 times or more the content of the hydrophilic polymer in the outermost surface layer. The content of the hydrophilic polymer affects the shrinkage rate of the hollow fiber membrane after film formation. That is, the shrinkage rate of the hollow fiber membrane increases as the content of the hydrophilic polymer increases. For example, when the content of the hydrophilic polymer on the innermost outermost layer is higher than the content of the hydrophilic polymer on the outermost outermost layer, the inner surface side may be affected by the difference in shrinkage between the inner surface side and the outer surface side. Micro wrinkles may occur or the hollow fiber membrane may break. If wrinkles enter the inner surface, for example, when used for hemodialysis, blood proteins and the like are likely to be deposited on the membrane surface when blood is flowed, so the permeation performance deteriorates over time. There is a possibility of connection. For these reasons, it is preferable to increase the content of the hydrophilic polymer on the outer surface side. Furthermore, the hollow fiber membrane of the present invention has a dense layer on the inner surface and a structure in which the pore diameter gradually increases toward the outer surface. That is, since the porosity on the outer surface side is higher than that on the inner surface side, the shrinkage rate on the outer surface side is larger. In consideration of the influence, the content of the hydrophilic polymer in the outermost outermost layer is preferably 1.1 times or more the content of the hydrophilic polymer in the outermost outermost layer. More preferably, it is 1.2 times or more, and further preferably 1.3 times or more.
For the above reasons, the content of the hydrophilic polymer on the outermost surface layer is preferably high. However, if the content exceeds 2.0 times, the content of the hydrophilic polymer with respect to the polysulfone polymer becomes too high, resulting in insufficient strength or hollowness. This may cause problems such as sticking of the membranes, reverse influx of endotoxin when using hemodialysis, and elution of hydrophilic polymers. More preferably, it is 1.9 times or less, More preferably, it is 1.8 times or less, More preferably, it is 1.7 times or less.

  Furthermore, it is a preferred embodiment that the hydrophilic polymer is insolubilized by crosslinking. There is no limitation on the crosslinking method, the degree of crosslinking, etc. For example, the crosslinking method includes γ-rays, electron beams, heat, chemical crosslinking, etc. Among them, residues such as initiators do not remain, and in terms of high material permeability, γ-rays and electron beams are used. Crosslinking is preferred.

  Insolubilization in the present invention refers to solubility in dimethylformamide in a crosslinked film. That is, 1.0 g of the crosslinked film is taken, dissolved in 100 ml of dimethylformamide, and the presence or absence of insoluble matter is visually observed and determined. In the case of a module in which a liquid is filled in the module, the filling liquid is first withdrawn, and then pure water is allowed to flow at 500 mL / min for 5 minutes in the dialysate side flow path, and then pure water is similarly applied to the blood side flow path. Flow for 5 minutes at 200 mL / min. Finally, 200 mL / min of pure water is passed through the membrane from the blood side to the dialysate side to finish the washing process. A hollow fiber membrane is taken out from the obtained module and freeze-dried to obtain an insoluble component measurement sample. In the case of a dry hollow fiber membrane module, the same washing treatment is performed to obtain a measurement sample.

  Next, in detail regarding the innermost surface layer and the inner layer near the inner surface of the hollow fiber membrane, the difference between the two layers shows a two-layer structure due to the difference in the concentration of the hydrophilic polymer. Since the pore diameter tends to increase from the dense layer on the surface toward the outer surface, a two-layer structure with a difference in density between the outermost layer portion and the vicinity of the surface may be formed. The thickness of each layer and its boundary line are arbitrarily changed depending on the manufacturing conditions of the hollow fiber membrane, and the structure of the layer also has some influence on the performance. Then, even if we infer from the manufacturing process by coagulation of the hollow fiber membrane, if the outermost layer and the near-surface layer are manufactured almost simultaneously and both layers are manufactured adjacent to each other, two layers are formed. Can be recognized, but the boundary is not something that can be drawn clearly.If you look at the distribution curve of the content of hydrophilic polymer across two layers, it is often connected by a continuous line, hydrophilic polymer There may be a difference in concentration between the two layers due to the difference in the content of. In general, it is technically impossible to assume that two discontinuous layers with different material behaviors are formed because the distribution curve of the hydrophilic polymer content can be broken at the boundary between the two layers. The hydrophilic polymer content of the outermost layer is 20-40% by mass and that of the surface vicinity layer is 5-20% by mass as the optimum range. From the mechanism of diffusion and movement to the outermost layer, for example, a design in which the outermost layer is 40% by mass and the near-surface layer is 5% by mass may not function sufficiently. In short, it is important to design by paying attention to the difference in the content of simple hydrophilic polymers present in the two layers. As an appropriate difference value, for example, 1.1 times or more between the two layers consisting of the outermost surface and the surface vicinity layer, based on the numerical value represented by mass% expressed in the content of both hydrophilic polymers. That is, if calculated by converting the difference in mass% of the hydrophilic polymer content between the two layers, a simple difference in the content (content) of the hydrophilic polymer in the two layers is 1 to 35. If there is a difference in the difference of about 5 to 25% by mass, about 5% by mass, it can be said that the hydrophilic polymer can smoothly diffuse and move from the surface vicinity layer to the outermost layer. For example, when the outermost layer is 32% by mass, the surface vicinity layer is in the range of about 7 to 27% by mass, which satisfies the requirement of about 1.1 to 10 times.

  The content of the outermost layer (the surface hydrophilic polymer concentration of the aggregated particles) of the above-described hydrophilic polymer hollow fiber membrane was measured and calculated by the ESCA method as described later. The absolute value of the content of the surface layer portion (depth from several tens to several tens of centimeters from the surface layer) is obtained. Normally, the ESCA method (outermost layer) can measure the hydrophilic polymer (PVP) content up to about 10 nm (100 mm) deep from the blood contact surface. In addition, the hydrophilic polymer content in the surface vicinity layer is an absolute value of the ratio existing in a range up to a depth corresponding to several hundreds of nanometers. In the ATR method (surface vicinity layer), the blood contact surface The hydrophilic polymer content up to a depth of about 1000 to 1500 nm (1 to 1.5 μm) can be measured.

  The content of the hydrophilic polymer on the inner surface and the outer surface may be related to the molecular weight of the hydrophilic polymer. For example, when polyvinyl pyrrolidone with a molecular weight of about 450,000 is lower than when polyvinyl pyrrolidone with a molecular weight of about 1,200,000 is used, the solubility and elution amount of polyvinyl pyrrolidone during coagulation is large, and diffusion transfer Compared to the content of 1 to 20% by mass of the average mass ratio of the hydrophilic polymer to the polysulfone-based polymer, the outermost layer part 20 to 40% by mass and the surface vicinity layer part 5 to 20% by mass % Having a relatively high hydrophilic polymer concentration tends to be produced. For example, a hollow fiber membrane produced by using 80% by mass of a polysulfone resin and 15% by mass of a polyvinyl pyrrolidone having a molecular weight of 900,000 and 5% by mass of a polyvinyl pyrrolidone having a molecular weight of about 45,000 is also used. The content and performance of the polyvinyl pyrrolidone in the layer may be affected, and the design of the hollow fiber membrane from this viewpoint also belongs to the category of the present invention.

  Examples of the method for achieving the above requirements 2 and 3 and 4 in the present invention include setting the ratio of the hydrophilic polymer to the hydrophobic polymer in the above-described range, optimizing the film forming conditions of the hollow fiber membrane, etc. Can be achieved. Specifically, it is preferable that the dense layer formed on the inner surface side of the hollow fiber membrane has a two-layer structure in which there is a density difference between the outermost layer portion and the surface vicinity portion. That is, although the detailed reason is not known, by adjusting the mass ratio of the polysulfone polymer and the hydrophilic polymer in the spinning dope and the concentration and temperature of the internal coagulation liquid within the ranges described later, the innermost surface of the hollow fiber membrane can be obtained. There is a difference in the solidification rate and / or phase separation rate between the surface layer and the vicinity of the surface, and the difference in solubility between the polysulfone polymer and the hydrophilic polymer in the solvent / water has the characteristics as in requirements 2 and 3. I think that it may be expressed. For requirement 4, it is important to optimize the drying conditions. That is, when the wet hollow fiber membrane is dried, the hydrophilic polymer dissolved in water moves from the inside of the hollow membrane to the surface side as the water moves. Here, by using drying conditions as described later, it is possible to give a certain speed to the movement of water and to make the movement speed uniform throughout the hollow fiber membrane, and the hydrophilic polymer inside the hollow fiber membrane. Move quickly to both surfaces without spots. Since the evaporation of water from the membrane surface is greater from the outer surface side than from the inner surface side of the hollow fiber membrane, the amount of the hydrophilic polymer that moves to the outer surface side is increased, and thus the hollow fiber membrane of the present invention is increased. It is presumed that requirement 4 which is a characteristic of the above can be achieved.

  The mass ratio of the hydrophilic polymer to the polysulfone polymer in the dope is preferably 0.1 to 0.6. If the PVP content in the dope is too low, it may be difficult to control the PVP abundance ratio in the film within the range of requirements 2, 3, and 4. Therefore, the hydrophilic polymer / polysulfone polymer in the dope is more preferably 0.15 or more, further preferably 0.2 or more, still more preferably 0.25 or more, and particularly preferably 0.3 or more. Further, if the PVP content in the dope is too large, the amount of PVP in the film also increases, so it is necessary to strengthen the cleaning after film formation, which may lead to an increase in cost. Therefore, the PVP ratio in the dope is more preferably 0.57 or less, and further preferably 0.55 or less.

  As the internal coagulation liquid, a 15 to 70 mass% dimethylacetamide (DMAc) aqueous solution is preferable. If the concentration of the internal coagulation liquid is too low, the solidification rate of the inner surface is increased, which may make it difficult to control the content of the hydrophilic polymer in the vicinity of the inner surface. Therefore, the internal coagulation liquid concentration is more preferably 20% by mass or more, further preferably 25% by mass or more, and further preferably 30% by mass or more. On the other hand, if the concentration of the internal coagulating liquid is too high, the coagulation rate on the inner surface becomes slow, and it may be difficult to control the content of the hydrophilic polymer on the outermost surface. Therefore, the internal coagulation liquid concentration is more preferably 60% by mass or less, further preferably 55% by mass or less, and further preferably 50% by mass or less. Furthermore, it is preferable to control the internal coagulation liquid temperature to -20 to 30 ° C. If the internal coagulating liquid temperature is too low, the outermost surface is solidified immediately after nozzle discharge, and it may be difficult to control the content of the hydrophilic polymer in the vicinity of the inner surface. The internal coagulation liquid temperature is more preferably −10 ° C. or higher, further preferably 0 ° C. or higher, and further preferably 10 ° C. or higher. In addition, if the internal coagulation liquid temperature is too high, the difference in the film structure (dense / dense) between the innermost surface layer and the surface becomes too large, making it difficult to control the content of the hydrophilic polymer near the outermost surface and the surface. Sometimes. The internal coagulation liquid temperature is more preferably 25 ° C. or lower, and further preferably 20 ° C. or lower. In addition, by setting the internal coagulating liquid temperature within the above range, it is possible to suppress the dissolved gas that has dissolved into bubbles when the internal coagulating liquid is discharged from the nozzle. In other words, by suppressing the bubbling of the dissolved gas in the internal coagulation liquid, there is also a secondary effect of suppressing yarn breakage directly below the nozzle and the occurrence of knobs. As a means for controlling the internal coagulating liquid temperature within the above range, it is preferable to provide a heat exchanger in the pipe from the internal coagulating liquid tank to the nozzle.

As a specific example of the method for drying the wet hollow fiber membrane, the wet hollow fiber membrane bundle is put in a microwave dryer and dried by irradiating with a microwave with an output of 0.1 to 20 kW under a reduced pressure of 20 kPa or less. Is a preferred embodiment. In consideration of shortening the drying time, a higher microwave output is preferable.For example, in hollow fiber membranes containing hydrophilic polymers, degradation or decomposition of the hydrophilic polymers may occur due to overdrying or overheating. It is preferable that the output is not increased too much due to problems such as a decrease in wettability. Therefore, the microwave output is more preferably 18 kW or less, further preferably 16 kW or less, and further preferably 14 kW or less. Further, the hollow fiber membrane bundle can be dried even with an output of less than 0.1 kW, but there is a possibility that a problem of reduction in throughput due to an increase in drying time may occur. The output of the microwave is more preferably 0.15 kW or more, and further preferably 0.2 kW or more. The degree of reduced pressure combined with the output is preferably 15 kPa or less, more preferably 10 kPa or less, although it depends on the moisture content of the hollow fiber membrane bundle before drying. A lower degree of vacuum is preferable because the drying speed increases, but it is preferable to set the lower limit to 0.1 kPa in view of cost increase for increasing the degree of sealing of the system. More preferably, it is 0.2 kPa or more, and further preferably 0.3 kPa or more. The optimum value of the combination of the microwave output and the degree of reduced pressure varies depending on the moisture content of the hollow fiber membrane bundle and the number of treatments of the hollow fiber membrane bundle.
For example, as a temporary guideline for carrying out the drying conditions of the present invention, when 20 hollow fiber membrane bundles having a moisture content of 50 g per hollow fiber membrane are dried, the total moisture content is 50 g × 20 = 1,000 g. At this time, the output of the microwave is suitably 1.5 kW and the degree of decompression is suitably 5 kPa.

  The microwave irradiation frequency is preferably 1,000 to 5,000 MHz in consideration of suppression of irradiation spots on the hollow fiber membrane bundle and the effect of extruding water in the pores from the pores. More preferably, it is 1,500-4,500 MHz, More preferably, it is 2,000-4,000 MHz.

  In drying by microwave irradiation, it is important to uniformly heat and dry the hollow fiber membrane bundle. In the above-described microwave drying, nonuniform heating due to the reflected wave that occurs accompanying the generation of the microwave occurs, so it is important to take measures to reduce the nonuniform heating due to the reflected wave. The method is not limited and is arbitrary. For example, a method of providing a reflector in an oven disclosed in Japanese Patent Application Laid-Open No. 2000-340356 to reflect reflected waves and uniformize heating is a preferred embodiment. One.

  With the above combination, the hollow fiber membrane is preferably dried within 5 hours. If the drying time is too long, the movement speed of the water in the hollow fiber membrane is slow, which may affect the movement of the hydrophilic polymer dissolved in the water. As a result, it becomes impossible to move the hydrophilic polymer to the target layer in the hollow fiber membrane, or it becomes easy to cause movement spots, thereby controlling the content of the hydrophilic polymer in each part. It may not be possible. Therefore, the drying time of the hollow fiber membrane is more preferably within 4 hours, and further preferably within 3 hours. A shorter drying time is preferable because the movement of the hydrophilic polymer is less, but a combination of microwave frequency, output, and degree of vacuum is preferred from the viewpoint of suppressing degradation and decomposition of the hydrophilic polymer due to heat generation and reducing dry spots. When selected, it is preferable to take a drying time of 5 minutes or more, more preferably 10 minutes or more, and even more preferably 15 minutes or more.

  Further, the maximum temperature reached by the hollow fiber membrane bundle during drying is preferably 80 ° C. or lower. If the temperature is too high, the hydrophilic polymer may be deteriorated or decomposed. Therefore, the temperature of the hollow fiber membrane during drying is more preferably 75 ° C. or less, and further preferably 70 ° C. or less. However, if the temperature is too low, the drying time becomes long, so that there is a possibility that the hydrophilic high molecular weight of each part of the hollow fiber membrane cannot be controlled as described above. Therefore, the temperature during drying is preferably 20 ° C. or higher, more preferably 30 ° C. or higher, and further preferably 40 ° C. or higher.

  Furthermore, it is preferable that the hollow fiber membrane does not dry out. If it is completely dried, there is a possibility that the wettability decreases during re-wetting at the time of use, and the hydrophilic polymer is less likely to absorb water, so that it can be easily eluted from the hollow fiber membrane. The moisture content of the hollow fiber membrane after drying is preferably 1% by weight or more, and more preferably 1.5% by weight or more. If the water content of the hollow fiber membrane is too high, bacteria may be easily proliferated during storage, and the hollow fiber membrane may be crushed by its own weight. Therefore, the water content of the hollow fiber membrane is 5% by weight. The following is preferable, more preferably 4% by weight or less, and still more preferably 3% by weight or less. The moisture content (%) of the present invention means the mass of the hollow fiber membrane bundle before drying (a) the mass (b) of the hollow fiber membrane bundle after drying, and the moisture content (%) = (a− b) can be easily calculated by / b × 100.

In the present invention, the hole area ratio of the outer surface of the hollow fiber membrane is 8 to 25%, and the average hole area of the hole portion on the outer surface of the hollow fiber membrane is 0.3 to 1.0 μm ² . This is effective for imparting the above-mentioned characteristics and is a preferred embodiment. If the open area ratio is less than 8% or the average pore area is less than 0.3 μm ² , the water permeability may decrease. Further, when the membrane is dried, the hydrophilic polymer existing on the outer surface of the membrane is interposed, and the hollow fiber membranes are fixed to each other, causing problems such as deterioration in module assemblability. Therefore, the open area ratio is more preferably 9% or more, and further preferably 10% or more. The average pore area is more preferably 0.4 μm ² or more, further preferably 0.5 μm ² or more, and further preferably 0.6 μm ² or more. On the contrary, when the open area ratio exceeds 25% or the average pore area exceeds 1.0 μm ² , the burst pressure may decrease. Therefore, the porosity is more preferably 23% or less, further preferably 20% or less, still more preferably 17% or less, and particularly preferably 15% or less. The average pore area is more preferably 0.95 .mu.m ² or less, more preferably 0.90 .mu.m ² or less.

  As a method for bringing the content of the hydrophilic polymer on the outer surface of the hollow fiber membrane into the above-mentioned range and a method for bringing the opening ratio of the outer surface of the hollow fiber membrane into the above-mentioned range, hydrophilicity with respect to the polysulfone-based polymer in the spinning dope described above is used. In addition to adjusting the mass ratio of the conductive polymer and optimizing the drying conditions of the hollow fiber membrane, it is also effective to optimize the cleaning conditions in cleaning the formed hollow fiber membrane. Film forming conditions include temperature and humidity adjustment of the air gap at the nozzle exit, stretching conditions, optimization of the temperature and composition of the external coagulation bath, and cleaning methods include hot water cleaning, alcohol cleaning, and centrifugal cleaning. It is valid. Among these methods, as the film forming conditions, optimization of the humidity of the air gap and the composition ratio of the solvent and the non-solvent in the external coagulating liquid is particularly effective as the cleaning method.

  The air gap portion is preferably surrounded by a member for blocking outside air, and the humidity inside the air gap is preferably adjusted by the spinning solution composition, the nozzle temperature, the air gap length, and the temperature and composition of the external coagulation bath. For example, a spinning stock solution composed of polyethersulfone / polyvinylpyrrolidone / dimethylacetamide / RO water = 10-25 / 0.5 to 12.5 / 52.5 to 89.5 / 0 to 10.0 at 30 to 60 ° C. When discharging from a nozzle, passing through an air gap of 100 to 1000 mm, and leading to an external coagulation bath having a concentration of 0 to 70 mass% and a temperature of 50 to 80 ° C., the absolute humidity of the air gap portion is 0.01 to 0.3 kg / kg dry air. By adjusting the humidity of the air gap part to such a range, it becomes possible to control the outer surface open area ratio, the outer surface average pore area, and the outer surface hydrophilic polymer content to an appropriate range.

  As the external coagulation liquid, it is preferable to use a DMAc aqueous solution of 0 to 50% by mass. If the concentration of the external coagulation solution is too high, the outer surface opening ratio and the outer surface average pore area become too large, which may increase the backflow of endotoxin to the blood side when using dialysis. Therefore, the external coagulation liquid concentration is more preferably 40% by mass or less, further preferably 30% by mass or less, and still more preferably 25% by mass or less. If the concentration of the external coagulation liquid is too low, it is necessary to use a large amount of water to dilute the solvent brought in from the spinning stock solution, and the cost for waste liquid treatment increases. Therefore, the lower limit of the external coagulation liquid concentration is more preferably 5% by mass or more.

  In the production of the hollow fiber membrane of the present invention, it is preferable that stretching is not substantially applied before the hollow fiber membrane structure is completely fixed. “Substantially no stretching” means that the roller speed during the spinning process is controlled so that the spinning dope discharged from the nozzle is not loosened or excessively tensioned. The discharge linear speed / coagulation bath first roller speed ratio (draft ratio) is preferably in the range of 0.7 to 1.8. If the ratio is less than 0.7, the traveling hollow fiber membrane may be loosened, leading to a decrease in productivity, and if it exceeds 1.8, the membrane structure is destroyed such as the dense layer of the hollow fiber membrane is torn. Sometimes. More preferably, it is 0.85-1.7, More preferably, it is 0.9-1.6, Most preferably, it is 1.0-1.5. By adjusting the draft ratio to this range, deformation and destruction of the pores can be prevented, clogging of blood protein into the membrane pores can be prevented, and performance stability over time and sharp fractionation characteristics can be expressed. Is possible.

The hollow fiber membrane that has passed through the water-washing bath is wound into a basket in a wet state to form a bundle of 3,000 to 20,000. Next, the obtained hollow fiber membrane bundle is washed to remove excess solvent and hydrophilic polymer. As a method for cleaning the hollow fiber membrane bundle, in the present invention, it is preferable to treat the hollow fiber membrane bundle by immersing it in hot water at 70 to 130 ° C., or from room temperature to 50 ° C. and 10 to 40 vol% ethanol or isopropanol aqueous solution. .
(1) In the case of hot water washing, the hollow fiber membrane bundle is immersed in excess RO water and treated at 70 to 90 ° C. for 15 to 60 minutes, and then the hollow fiber membrane bundle is taken out and subjected to centrifugal dehydration. This operation is repeated three or four times while updating the RO water to perform the cleaning process.
(2) A method of treating a hollow fiber membrane bundle immersed in excess RO water in a pressurized container at 121 ° C. for about 2 hours can also be employed.
(3) When using an ethanol or isopropanol aqueous solution, it is preferable to repeat the same operation as in (1).
(4) It is also a preferable washing method that the hollow fiber membrane bundle is radially arranged in the centrifugal washer and centrifugal washing is performed for 30 minutes to 5 hours while spraying washing water at 40 ° C. to 90 ° C. from the rotation center in a shower shape. .
Two or more cleaning methods may be combined. In any of the methods, if the processing temperature is too low, it is necessary to increase the number of times of cleaning, which may lead to an increase in cost. On the other hand, when the treatment temperature is too high, the decomposition of the hydrophilic polymer is accelerated, and conversely, the cleaning efficiency may be lowered. By performing the above-described washing, it is possible to optimize the abundance of the hydrophilic polymer on the outer surface, and to suppress sticking and reduce the amount of eluate.

  The formation of aggregated particles in the dense layer is thought to be due to the following process. When the spinning dope is brought into contact with a coagulation solution (such as a hollow forming agent), a decrease in the solvent concentration and an increase in the non-solvent concentration occur rapidly at the interface between the spinning solution and the coagulation solution. At this time, the polymer that has been uniformly dissolved increases the interaction between the polymers, so that a large number of polymer-rich phases are formed in the vicinity of the interface, which becomes the core of the particle. Thereafter, the polymer rapidly precipitates around the nucleus to grow particles. The grown particles come into contact with each other, and an aggregate structure of the particles is formed. That is, (1) particle nucleation (affects the number of particles), (2) particle growth (affects particle size), (3) particle agglomeration (boundary surface formation as individual particles) The influence of the particle gap) determines the aggregate particle structure of the dense layer of the present invention. In particular, in the present application, a high separation property is exhibited by controlling the uniformity of particle size, which is the stage (2).

  The means for forming the particle structure as shown in the present invention is achieved as follows. First, the concentration of the polymer in the film forming solution. When the concentration is high (about 40% or more), the formation of the particles is performed in a polymer-rich system, so that the formation of the boundary surface of the particles is not sufficient, and there is a case where the particle structure does not become clear but becomes a dense structure. In addition, when the concentration is low (about 12% or less), the entanglement and particle growth are remarkably reduced because of the low degree of entanglement between the polymers at the initial stage, and the formation of a dense layer is not substantially caused. Alternatively, the dense layer is not a structure in which particles are aggregated but a network structure. In the present application, it is formed from a blend of a hydrophobic polymer and a hydrophilic polymer, and the total content of both in the spinning dope is 15 to 35% by mass. Preferably, it is 17-30 mass%. As for the content of each polymer, the concentration of the hydrophobic polymer is preferably 12 to 25% by mass, and the concentration of the hydrophilic polymer is preferably 1 to 10% by mass.

  Furthermore, in order to make the dense layer have the optimum particle size and distribution as shown in the present application, optimization by a combination of the following requirements is required. In the dense layer, the formation of the polymer rich portion and the polymer poor portion proceeds rapidly in the process of nucleation to growth. Here, polymer diffusion is the dominant process. The viscosity and temperature of the solution are related as factors for film formation affecting the diffusivity, and the higher the temperature, the greater the diffusibility, the particle growth proceeds, and the particle gap tends to increase. The temperature in this case is the nozzle temperature at which the spinning dope is discharged from the spinning die. In this case, an increase in the sieving coefficient (SCβ2-MG) of β2 microglobulin of the present invention is expected to improve the membrane performance. On the other hand, as described above, there is an upper limit due to the relationship of the sieving coefficient (SCalb) of albumin. However, when the temperature of the discharge stock solution is increased, the particle size distribution tends to increase, which tends to be disadvantageous in stabilizing the film performance over time. It was also possible to increase the particle size and improve the permeability by increasing the concentration of the internal coagulation liquid that is the hollow fiber hollow forming agent. It is presumed that the solidification of the polymer particles is slowed down and the growth of large particles occurs. In this case, the particle size distribution tended to be relatively small. However, only by controlling the particle state by lowering the discharge temperature and increasing the concentration of the internal coagulating liquid within the above-mentioned preferable nozzle temperature and internal coagulating liquid concentration range, the high filtration stability targeted by the present invention is achieved. Was not obtained.

  As a result of intensive studies by the inventors of the present application, in the discharge base (double pipe orifice type), the cross-sectional area of the central hole of the base that discharges the internal coagulating liquid and the cross-sectional area of the hollow portion inner diameter of the hollow fiber to be formed are specified. By making it into the region, the inventors have reached the invention of the hollow fiber membrane of the present invention. When the hollow fiber membrane is produced with the aim of having a perfect inner diameter of 200 μm, the inner diameter of the perfect circle-shaped discharge part of the hollow forming agent of the die is about 100 μm, and the above-mentioned draft ratio is 2 or less, so that the desired hollow fiber membrane Could get. From the examination of the die having various discharge holes, it was found that the ratio of the inner diameter of the internal coagulating liquid to the inner diameter of the hollow fiber membrane is preferably 0.7 or less. Although this mechanism is not clear, when the amount of internal coagulation liquid required to produce the target hollow fiber membrane inner diameter is discharged, the spinning stock solution discharged with positive pressure is applied to the internal liquid side. Inflating work occurs. At this time, the internal coagulating liquid quickly penetrates into the dense layer formed on the inner surface, causing the temperature of the dense layer and the concentration gradient of the coagulating liquid to be uniform, and stabilizing the size and distribution of the particle structure. It is estimated that it will contribute to The ratio of the internal coagulation liquid discharge hole diameter of the die to the inner diameter of the hollow fiber is preferably small, but from the viewpoint of the processing accuracy of the die and the washability of the die when the polymer is clogged in the discharge hole, a very small one is preferred. No. The diameter of the discharge hole inner diameter is preferably 60 μm or more. Therefore, the ratio of the internal coagulation liquid discharge hole diameter of the die to the hollow fiber inner diameter is a preferable aspect of 0.7 to 0.3.

  In the present invention, it is important to satisfy the above requirements 1 to 5 at the same time. By simultaneously achieving the requirements, all the above-mentioned characteristics can be satisfied.

  Since the hollow fiber membrane bundle of the present invention has characteristics as described above, it is a preferred embodiment to be used for a blood purifier.

When used for the above-mentioned blood purifier, the blood purifier is a hollow fiber membrane having a burst pressure of 0.5 MPa or more, and the water permeability of the blood purifier is 150 ml / m ² / hr / mmHg or more. preferable.

  The burst pressure in the present invention is an index of the pressure resistance of the hollow fiber membrane after the hollow fiber is made into a module, and the hollow fiber is inside when the inside of the hollow fiber is pressurized with air and the pressure is gradually increased. It is a pressure that bursts without bursting. The higher the burst pressure, the fewer latent defects that lead to the cutting of the hollow fiber membrane and the generation of pinholes during use, so 0.5 MPa or more is preferable, 0.7 MPa or more is more preferable, and 1.0 MPa or more is particularly preferable. If the burst pressure is less than 0.5 MPa, there is a possibility that it is impossible to detect a potential defect that leads to a blood leak as described later. The higher the burst pressure, the better. However, if the focus is on increasing the burst pressure and the film thickness is increased or the porosity is decreased too much, the desired film performance may not be obtained. Therefore, when finished as a hemodialysis membrane, the burst pressure is preferably less than 2.0 MPa. More preferably, it is less than 1.7 MPa, more preferably less than 1.5 MPa, and particularly preferably less than 1.3 MPa.

Further, when the water permeability is less than 150 ml / m ² / hr / mmHg, the solute permeability may be lowered. In order to increase the solute permeability, the pore diameter is increased or the number of pores is increased. However, this tends to cause problems such as a decrease in membrane strength and formation of defects. However, in the hollow fiber membrane of the present invention, the porosity of the support layer portion is optimized by optimizing the pore diameter of the outer surface, and the reduction in solute permeation resistance and the improvement in membrane strength are balanced. A more preferable range of water permeability is 200 ml / m ² / hr / mmHg or more, more preferably 300 ml / m ² / hr / mmHg or more, still more preferably 400 ml / m ² / hr / mmHg or more, particularly preferably 500 ml / m. ² / hr / mmHg or more. In addition, when the water permeability is too high, it becomes difficult to control water removal during hemodialysis, and therefore, 2000 ml / m ² / hr / mmHg or less is preferable. More preferably 1800 ml / m ² / hr / mmHg or less, further preferably 1500 ml / m ² / hr / mmHg or less, even more preferably 1300 ml / m ² / hr / mmHg or less, particularly preferably 1000 ml / m ² / hr / h / g. It is below mmHg.

  Usually, a module used for blood purification is subjected to a leak test in which the inside or outside of the hollow fiber is pressurized with air in order to confirm defects of the hollow fiber or the module at the final stage of production. When a leak is detected by the pressurized air, the module is discarded as a defective product or an operation for repairing the defect is performed. The air pressure in this leak test is often several times the guaranteed pressure resistance (usually 500 mmHg) of the hemodialyzer. However, in the case of a hollow fiber type blood purification membrane having a particularly high water permeability, microscopic scratches, crushing, and tearing of the hollow fiber that cannot be detected by a normal pressure leak test are caused by manufacturing processes (mainly sterilization and (Packing), transportation process, or handling in clinical settings (unpacking, priming, etc.) will lead to the occurrence of hollow fiber cutting and pinholes. It is. The trouble can be avoided by setting the burst pressure to the above characteristic.

  Further, the uneven thickness of the hollow fiber membrane is effective for suppressing the occurrence of the above-described potential defects. It is a preferred embodiment in which the thickness deviation when observing the cross section of 100 hollow fiber membranes in the hollow fiber membrane module is high, and the thickness deviation ratio shown by the ratio between the maximum value and the minimum value is high. The uneven thickness of 100 hollow fiber membranes is preferably 0.6 or more. If at least one hollow fiber membrane includes 100 hollow fiber membranes with an uneven thickness of less than 0.6, the hollow fiber membrane may easily cause a potential defect that causes a leak during clinical use. The thickness deviation of the present invention is not an average value but represents a minimum value of 100. A higher unevenness degree increases the uniformity of the film, suppresses the occurrence of latent defects, and improves the burst pressure. Therefore, it is more preferably 0.65 or more, more preferably 0.7 or more, and still more preferably 0. .75 or more. If it is less than 0.6, latent defects are likely to occur, and the burst pressure becomes low, which may lead to blood leakage.

  In order to make the thickness deviation 0.6 or more, for example, it is preferable to make the slit width of the nozzle that is the discharge port of the film forming solution strictly uniform. The spinning nozzle of the hollow fiber membrane is generally a tube-in-orifice type nozzle having an annular portion for discharging the spinning stock solution and a core liquid discharge hole serving as a hollow forming agent inside thereof. This refers to the width of the annular portion that discharges the spinning dope. By reducing the variation in the slit width, uneven thickness of the spun hollow fiber membrane can be reduced. Specifically, the ratio between the maximum value and the minimum value of the slit width is 1.00 to 1.11, and the difference between the maximum value and the minimum value is preferably 10 μm or less, and more preferably 5 μm or less. In addition, methods such as optimizing the nozzle temperature, reducing discharge spots of the internal liquid during film formation, and optimizing the draw ratio are also effective.

  Further, as a measure for increasing the burst pressure, it is also an effective method to reduce the potential defects by reducing the flaws on the surface of the hollow fiber membrane, the mixing of foreign matter and bubbles. As a method of reducing the occurrence of scratches, the material and surface roughness of rollers and guides in the manufacturing process of the hollow fiber membrane are optimized, the container and the hollow fiber membrane are inserted when the hollow fiber membrane bundle is inserted into the module container at the time of module assembly. It is effective to devise such that contact with the fiber or rubbing between the hollow fiber membranes is reduced. In the present invention, it is preferable to use a roller having a mirror-finished surface in order to prevent the hollow fiber membrane from slipping and scratching the surface of the hollow fiber membrane. Further, it is preferable to use a guide whose surface is textured or knurled in order to avoid contact resistance with the hollow fiber membrane as much as possible. When the hollow fiber membrane bundle is inserted into the module container, the hollow fiber membrane bundle is not directly inserted into the module container. It is preferable to use a method in which a film is inserted into a module container, and after insertion, only the film is extracted from the module container.

  As a method for suppressing the mixing of foreign matter into the hollow fiber membrane, a method using a raw material with less foreign matter, filtering the spinning dope for forming a membrane, and reducing foreign matter is effective. In the present invention, it is preferable to filter the spinning dope using a filter having a pore size smaller than the film thickness of the hollow fiber membrane. Specifically, the pore size provided while the uniformly dissolved spinning dope is led from the dissolution tank to the nozzle. Pass through a 10-50 μm sintered filter. The filtration treatment may be performed at least once. However, when the filtration treatment is performed in several stages, it is preferable to reduce the pore size of the filter as it is in the latter stage in order to extend the filtration efficiency and the filter life. The pore diameter of the filter is more preferably 10 to 45 μm, further preferably 10 to 40 μm. If the filter pore size is too small, the back pressure may increase and the quantitativeness may decrease.

  Further, as a method for suppressing the mixing of bubbles, it is effective to defoam a polymer solution for film formation. Depending on the viscosity of the spinning dope, static defoaming or vacuum defoaming can be used. Specifically, after the pressure in the dissolution tank is reduced to -100 to -760 mmHg, the tank is sealed and allowed to stand for 5 to 30 minutes. This operation is repeated several times to perform defoaming treatment. If the degree of vacuum is too low, the treatment may take a long time because it is necessary to increase the number of defoaming times. On the other hand, when the degree of vacuum is too high, the cost for increasing the degree of sealing of the system may increase. The total treatment time is preferably 5 minutes to 5 hours. If the treatment time is too long, the hydrophilic polymer may deteriorate or decompose due to the effect of reduced pressure. If the treatment time is too short, the defoaming effect may be insufficient.

  Hereinafter, the effectiveness of the present invention will be described with reference to examples, but the present invention is not limited thereto. In addition, the evaluation method of the physical property in the following examples is as follows.

1. Measurement of water permeability The blood outlet circuit of the dialyzer (the outlet side from the pressure measurement point) was sealed with forceps. Purified water kept at 37 ° C is put into a pressurized tank, and the pressure is controlled by a regulator. The pure water is sent to the blood flow path side of the dialyzer kept at 37 ° C constant temperature bath, and the filtrate flowed out from the dialysate side. Was measured. The transmembrane pressure difference (TMP) is
TMP = (Pi + Po) / 2
And Here, Pi is the dialyzer inlet side pressure, and Po is the dialyzer outlet side pressure. Change the TMP at four points, measure the filtration flow rate, and calculate the water permeability (mL / hr / mmHg) from the slope of the relationship. At this time, the correlation coefficient between TMP and filtration flow rate should be 0.999 or more. In order to reduce the pressure loss error due to the circuit, TMP is measured in the range of 100mmHg or less. The water permeability of the hollow fiber membrane is calculated from the membrane area and the water permeability of the dialyzer.
UFR (H) = UFR (D) / A
Where UFR (H) is the water permeability of the hollow fiber membrane (mL / m ² / hr / mmHg), UFR (D) is the water permeability of the dialyzer (mL / hr / mmHg), and A is the membrane area of the dialyzer ( m ² ).

2. Calculation of membrane area The membrane area of the dialyzer is determined as a reference for the inner diameter of the hollow fiber.
A = n × π × d × L
Here, n is the number of hollow fibers in the dialyzer, π is the circumference, d is the inner diameter (m) of the hollow fiber, and L is the effective length (m) of the hollow fiber in the dialyzer.

3. Burst pressure Fill the dialysate side of the module consisting of about 10,000 hollow fiber membranes with water and plug it. Dry air or nitrogen is fed from the blood side at room temperature and pressurized at a rate of 0.5 MPa per minute. The pressure is increased, and the air pressure when the hollow fiber membrane bursts (bursts) with pressurized air and bubbles are generated in the liquid filled on the dialysate side is defined as the burst pressure.

4. Unevenness of thickness The cross section of 100 hollow fibers is observed with a 200 times projector. In one field of view, the thickness of the thickest part and the thinnest part is measured with respect to one yarn cross section having the largest film thickness difference.
Thickness unevenness = thinnest part / thickest part thickening degree = 1, and the film thickness is perfectly uniform.

5. Elution amount of hydrophilic polymer An example of the measurement method when polyvinylpyrrolidone is used as the hydrophilic polymer is shown below.
<Dry hollow fiber membrane module>
Saline was passed through the dialysate side channel of the module at 500 mL / min for 5 minutes, and then passed through the blood side channel at 200 mL / min. Thereafter, the solution was passed for 3 minutes from the blood side to the dialysate side while filtering at 200 mL / min.
<Wet hollow fiber membrane module>
After extracting the module filling liquid, the same processing operation as that of the dry hollow fiber membrane module was performed.
Using the hollow fiber membrane module subjected to the priming treatment, extraction was performed by a method defined in the dialysis artificial kidney device production approval standard, and polyvinylpyrrolidone in the extract was quantified by a colorimetric method.
That is, 100 ml of pure water is added to 1 g of the hollow fiber membrane and extracted at 70 ° C. for 1 hour. The obtained extract (2.5 ml), 0.2 molar aqueous citric acid solution (1.25 ml) and 0.006 normal iodine aqueous solution (0.5 ml) were mixed well and allowed to stand at room temperature for 10 minutes, and the absorbance at 470 nm was measured. The quantification was performed with a calibration curve obtained by measuring according to the above method using a standard polyvinylpyrrolidone.

6. Content of hydrophilic polymer in outermost surface of inner and outer surfaces The content of hydrophilic polymer was determined by X-ray photoelectron spectroscopy (ESCA method). The measurement method when polyvinylpyrrolidone (PVP) is used as the hydrophilic polymer will be exemplified.
One hollow fiber membrane was cut obliquely with a razor so that a part of the inner surface was exposed, and was attached to a sample stage so that the inner and outer surfaces could be measured, and measurement was performed by ESCA. The measurement conditions are as follows.
Measuring device: ULVAC-Phi ESCA5800
Excitation X-ray: MgKα ray X-ray output: 14 kV, 25 mA
Photoelectron escape angle: 45 °
Analysis diameter: 400μmφ
Pass energy: 29.35 eV
Resolution: 0.125 eV / step
Degree of vacuum: about 10 ⁻⁷ Pa or less From the measured value (N) of nitrogen and the measured value (S) of sulfur, the PVP content ratio on the surface was calculated by the following formula.
<In case of PVP-added PES (polyethersulfone) membrane>
PVP content ratio (Hpvp) [%]
= 100 × (N × 111) / (N × 111 + S × 232)
<In the case of PVP-added PSf (polysulfone) membrane>
PVP content ratio (Hpvp) [%]
= 100 × (N × 111) / (N × 111 + S × 442)

7. Measuring method of hydrophilic polymer content rate in the whole hollow fiber membrane The measuring method at the time of using PVP as a hydrophilic polymer is illustrated. The sample was dried at 80 ° C. for 48 hours using a vacuum dryer, and 10 mg of the sample was analyzed with a CHN coder (manufactured by Yanaco Analytical Industrial Co., Ltd., MT-6 type). Calculated with
Mass ratio (mass%) of PVP = nitrogen content (mass%) × 111/14

8. Content of hydrophilic polymer in surface vicinity layer of blood contact surface of hollow fiber membrane An example of a measurement method using PVP as the hydrophilic polymer is illustrated. Measurement was performed by infrared absorption analysis. A measurement sample prepared by the same method as in 6 above was used, the measurement in the vicinity of the surface was performed by the ATR method, and the entire film was measured by the transmission method. In the ATR method, an infrared absorption spectrum was measured by a method using diamond 45 ° as an internal reflection element. For the measurement, IRμs / SIRM manufactured by SPECTRA TECH was used. In the infrared absorption spectrum, the ratio Ap / As of the peak absorption intensity Ap derived from C═O of PVP near 1675 cm ⁻¹ and the peak absorption intensity As derived from polysulfone-based polymer near 1580 cm ⁻¹ was determined. In the ATR method, since the absorption intensity depends on the measured wave number, a ratio υp / υs between the peak position υs of the polysulfone polymer and the peak position υp (wave number) of the PVP was applied to the measured value as a correction value. The content of PVP in the blood contact surface vicinity layer was calculated by the following formula.
Content (mass%) of hydrophilic polymer in the surface vicinity layer = Cav × Ap / As × υp / υs
However, Cav is a PVP mass ratio determined by a method for measuring the hydrophilic polymer content in the entire hollow fiber membrane.

9. Opening ratio of outer surface of hollow fiber membrane The outer surface of the hollow fiber membrane is observed with an electron microscope of 10,000 times and a photograph (SEM photograph) is taken. The image was processed with image analysis processing software to obtain the porosity of the outer surface of the hollow fiber membrane. The image analysis processing software is measured using, for example, Image PRO Plus (Media Cybernetics, Inc.). The emphasis / filtering operation is performed on the captured image so that the hole and the blockage are identified. Thereafter, the holes are counted, and when the lower polymer chain can be seen inside the holes, the holes are combined and counted as one hole. The area (A) of the measurement range and the cumulative area (B) of the pores within the measurement range were determined, and the area ratio (%) = B / A × 100. This was carried out 10 views and the average was obtained. Scale setting is performed as an initial operation, and holes on the measurement range boundary are not excluded during counting.

10. Average pore area of the apertures on the outer surface of the hollow fiber membrane Counted in the same manner as in the previous section, and the area of each pore was determined. In addition, holes on the measurement range boundary were excluded during counting. This was carried out for 10 fields of view and the average of all pore areas was determined.

11. Thickness of hollow fiber membrane The cross section of 100 hollow fibers is observed with a 200 × projector. In one field of view, the thickness of the thickest part and the thinnest part was measured for one yarn cross section having the largest film thickness difference.
Thickness unevenness = thinnest part / thickest part thickening degree = 1, and the film thickness is perfectly uniform.

12. Blood Leakage Test Add 37 ml of cattle blood to which citrate has been added to suppress coagulation to a blood purifier at 200 mL / min, and filter the blood at a rate of 10 mL / min · m ² . At this time, the filtrate is returned to blood to be a circulatory system. After 60 minutes, the filtrate from the blood purifier is collected, and the red color resulting from red blood cell leakage is visually observed. In this blood leak test, 30 blood purifiers were used in each of the examples and comparative examples, and the number of blood leaked modules was examined.

13. Fixing property of hollow fiber membrane About 10,000 hollow fibers were bundled, loaded into a module case of 30 mmφ to 35 mmφ, and sealed with a two-component polyurethane resin to prepare a module. A leak test was conducted for each level of five, and the number of modules that had leaked due to poor urethane resin sealing was counted.

14. Residual blood of hollow fiber membrane Fill the dialysate side of the module with a membrane area of 1.5m ² with physiological saline, fill the blood bag with 200ml of heparinized blood collected from a healthy person, and connect the blood bag and the module with a tube And circulate at 37 ° C. for 1 hour at a blood flow rate of 100 ml / min. Blood samples before the start of circulation and 60 minutes of circulation are sampled, and the white blood cell count and platelet count are measured. The measured value is corrected with the value of hematocrit.
Correction value = measured value (60 minutes) x hematocrit (0 minutes) / hematocrit (60 minutes)
The rate of change of white blood cells and platelets is calculated from the correction value.
Rate of change = correction value (60 minutes) / pre-circulation value x 100
After the circulation for 60 minutes, the blood was returned with physiological saline, and the number of remaining hollow fiber membranes was counted. The evaluation was carried out with 10 or less hollow fiber membranes remaining blood as ◯, 11 or more and 30 or less as Δ, and 31 or more as ×.

15. With the lid on the dialysate side port of the priming property module, distilled water for injection is flowed from the blood side inlet port at 200 mL / min until 10 seconds elapses after the distilled water for injection reaches the outlet port. The module case was detaped by tapping the module case 5 times with forceps, and the number of bubbles passed for 1 minute was visually confirmed. The determination was made according to the following criteria.
10 / min or less: ○
11 / min or more and less than 30 / min: △
30 pieces / minute or more: ×

16. Diameter of aggregated particle, standard deviation After osmium plasma coating of the cross-section obtained by freezing the hollow fiber membrane after freezing in liquid nitrogen, a photograph of a frozen section by a field emission scanning electron microscope (magnification: 4 100 to 70,000 times), 100 aggregated particles in the dense layer were measured, and the average diameter and standard deviation were calculated. The particle diameter was the average value of the major axis and minor axis of each particle. In addition, from the innermost surface toward the outer surface of the membrane, up to about 10 internal parts were measured from the size of the aggregated particles. (For example, when the particle diameter is 50 nm, the dense region inside 500 nm from the film surface is taken as the measurement region.)

17, β2-microglobulin sieve coefficient 1.5m ² hollow fiber membrane module was adjusted to 37 ° C, hematocrit (Ht) 40 ± 2.0%, total protein concentration 6.5 ± 0.5g / dl Sampling blood and filtrate at the inlet and outlet of the module at 15 and 120 minutes after starting blood filtration at a filtration rate of 90 ml / min at 300 ml / min, and then sampling the enzyme immunoassay (Glazyme β2-Microglobulin -EIA Test The concentration of β2-microglobulin (hereinafter abbreviated as β2-MG) was measured by Wako Pure Chemical Industries, Ltd.). In addition, an appropriate amount of human-derived β2-MG was added to the bovine blood flowing into the module in this measurement, and the sampled blood was centrifuged as necessary to be used for β2-MG measurement. The sieving coefficient (SC) of β2-MG was determined from these β2-MG concentration values according to the following formula.
SC = Cfil / ((CI + C0) / 2)
Cfil: β2-MG concentration of the filtrate CI: β2-MG concentration of blood at the inlet of the module blood CO: β2-MG concentration of blood at the outlet of the module blood

18. Sieving coefficient of albumin The experiment similar to the measurement of β2-microglobulin shown above was performed, and after 15 minutes from the start of filtration, blood and filtrate were sampled at the inlet and outlet of the module, and each bromocresol green method was used. The albumin concentration of the liquid was measured. From these concentration values, the sample concentration in the above formula was defined as the albumin concentration, and SC was determined.

(Example 1)
Polyethersulfone (manufactured by Sumika Chemtex Co., Ltd., SUMIKAEXCEL (R) 4800P) 17.9% by mass, polyvinyl pyrrolidone (BASF Koridone (R) K-90) 4.5% by mass, dimethylacetamide (DMAc) 74. Dissolve 7% by mass and 2.9% by mass of RO water uniformly at 50 ° C., and then reduce the pressure to −500 mmHg using a vacuum pump. The system was immediately sealed and left for 15 minutes. This operation was repeated three times to degas the film forming solution. The membrane-forming solution was passed through 30 μm and 15 μm two-stage sintered filters in order, and then discharged from the outer slit of a tube-in orifice nozzle heated to 65 ° C., and at the same time, degassed as an internal coagulating liquid at −700 mmHg for 60 minutes. The treated 45% by mass DMAc aqueous solution at 15 ° C. was discharged from the inner liquid discharge hole, passed through a 450 mm dry part cut off from the outside air by a spinning tube, and then solidified in a 5% DMAc 5% by weight aqueous solution at 60 ° C. I whispered to the wall. The nozzle slit width of the tube-in-orifice nozzle used is an average of 60 μm, the maximum is 61 μm, the minimum is 59 μm, the ratio of the maximum and minimum slit width is 1.03, the center discharge hole diameter of the nozzle is 100 μm, the draft of the film forming solution The ratio was 1.02.

  A polyethylene film whose surface on the hollow fiber bundle side was textured was wound around about 10,000 bundles of the hollow fiber membranes, and then cut into a length of 27 cm to obtain a bundle. This bundle was washed in hot water at 80 ° C. for 30 minutes × 4 times. The obtained wet bundle is centrifuged and drained for 600 rpm × 5 min, the hollow fiber membrane bundle is set in 6 × 4 stages on the rotary table in the drying apparatus, and the reflector is installed in the oven so that it can be heated uniformly. The microwave generator having the structure was irradiated with an initial 1.5 kW microwave, and the inside of the drying apparatus was reduced to 7 kPa by a vacuum pump, and the drying treatment was performed for 28 minutes. Subsequently, a drying process was performed for 12 minutes at a microwave output of 0.5 kW and a reduced pressure of 7 kPa. Further, the microwave output was reduced to 0.2 kW, and the drying process was similarly completed for 8 minutes. At the same time, far infrared irradiation was used in combination. The maximum temperature reached on the surface of the hollow fiber membrane bundle at this time was 65 ° C., and the moisture content of the hollow fiber membrane after drying was an average of 2% by mass. The resulting hollow fiber membrane had an inner diameter of 198.6 μm and a film thickness of 30.5 μm. During the spinning process, the roller contacted with the hollow fiber membrane was a mirror-finished surface, and the guides were all finished with a satin finish.

  As a result of assembling a blood purifier using the hollow fiber membrane thus obtained and conducting a leak test, no adhesion failure due to the sticking of the hollow fibers was found.

  The blood purifier was filled with RO water and irradiated with γ rays at an absorbed dose of 25 kGy for crosslinking treatment. When the hollow fiber membrane was cut out from the blood purifier after γ-irradiation and subjected to the eluate test, the PVP elution amount was 4 ppm, which was a satisfactory level. Further, when the outer surface of the hollow fiber membrane taken out from the blood purifier was observed with a microscope, defects such as scratches were not observed.

  In addition, citrated fresh cow blood was flowed to the blood purifier at a blood flow rate of 200 mL / min and a filtration rate of 10 mL / min, but no blood cell leak was observed. Endotoxin filtered from the outer side of the hollow fiber to the inner side of the hollow fiber was below the detection limit and was at a level with no problem. The other analysis results are shown in Table 1.

  There was a dense layer on the inner surface of the hollow fiber membrane, and when the aggregate diameter was measured, the average was 110 nm, and the standard deviation was 35 μm. The sieving coefficient of β2 microglobulin was SCβ2MG (15 minutes) 0.92, SCβ2MG (120 minutes) 0.76, and SCβ2MG (120 minutes) / SCβ2MG (15 minutes) was 0.83. SCalb (15 minutes) was 0.08.

(Comparative Example 1)
Wet hollow fiber as in Example 1 except that the spinning dope was changed to 2.4% by mass of polyvinylpyrrolidone (Collidon (R) K-90 manufactured by BASF) and 77% by mass of DMAc, and the dry part length was changed to 750 mm. A membrane was obtained. The obtained hollow fiber membrane was washed in the same manner as in Example 1 and dried in a 60 ° C. hot air dryer. The resulting hollow fiber membrane had a moisture content of 3.4% by mass, an inner diameter of 199.9 μm, and a film thickness of 29.7 μm. The properties of the obtained hollow fiber membrane bundle and blood purifier are shown in Table 1. The hollow fiber membrane of Comparative Example 1 had poor residual blood properties, which is presumed to be due to the low PVP content of the inner surface near-surface layer.

(Comparative Example 2)
The membrane-forming solution prepared in the same manner as in Example 1 was passed through a 100 μm filter and then degassed in advance at −700 mmHg for 2 hours as a hollow forming agent from a tube-in orifice nozzle heated to 50 ° C. The solution was simultaneously discharged using an aqueous solution, passed through a 600 mm dry section cut off from the outside air by a spinning tube, and then solidified in a DMAc aqueous solution having a concentration of 10% by mass and 60 ° C. The nozzle slit width of the tube-in-orifice nozzle used is an average of 100 μm, the maximum is 110 μm, the minimum is 90 μm, the ratio of the maximum and minimum slit width is 1.22, the center discharge hole diameter of the nozzle is 180 μm, and the draft ratio is 2.45. It was. The obtained hollow fiber membrane was passed through a water washing tank at 40 ° C. for 45 seconds to remove the solvent and excess hydrophilic polymer, and then wound up in a wet state and dried in air at 50 ° C. The resulting hollow fiber membrane had an inner diameter of 198.9 μm and a film thickness of 29.7 μm. It was 7.4 mass% when the mass ratio of the hydrophilic polymer in a hollow fiber membrane was measured. The properties of the obtained hollow fiber membrane bundle and blood purifier are shown in Table 1. The hollow fiber membrane obtained in this comparative example had a high PVP content in the outermost layer on the inner surface and a high PVP elution amount. Further, since the hydrophilic polymer content on the outer surface of the hollow fiber membrane is large, permeation of endotoxin to the blood side was observed.

(Comparative Example 3)
In Comparative Example 2, a hollow fiber membrane bundle and a blood purifier were obtained in the same manner as in Comparative Example 2, except that the number of hot water washings was changed to 6 hours. The properties of the obtained hollow fiber membrane bundle and blood purifier are shown in Table 1. The hollow fiber membrane bundle obtained in this comparative example had poor priming properties because the outermost surface PVP content on the outer surface was low and the hydrophilicity on the outer surface was low.

(Comparative Example 4)
In Example 1, 30.0% by mass of polyethersulfone (manufactured by Sumika Chemtex, Sumika Excel (R) 4800P), 7.0% by mass of polyvinylpyrrolidone (Collidon (R) K-90, manufactured by BASF), dimethylacetamide (DMAc) A hollow fiber membrane and a blood purifier were obtained in the same manner as in Example 1 except that 60.0 mass% and RO water were 3.0 mass%. The properties of the obtained hollow fiber membrane bundle and blood purifier are shown in Table 1. Agglomerated particles were not observed in this film.

(Example 2)
Polyethersulfone (manufactured by Sumika Chemtex, Sumika Excel (R) 4800P) 18.8% by mass, polyvinyl pyrrolidone (BASF Koridone (R) K-90) 5.2% by mass, DMAc 72.0% by mass, water Dissolve 4% by mass at 50 ° C, and then reduce the pressure in the system to -700 mmHg using a vacuum pump. Immediately seal the system and leave it for 10 minutes to prevent the solvent from evaporating and changing the composition of the film forming solution. did. This operation was repeated three times to degas the film forming solution. The obtained film-forming solution was passed through a two-stage filter of 15 μm and 15 μm, and then discharged from the outer slit of a tube-in orifice nozzle heated to 70 ° C., and at the same time degassed as an internal coagulation liquid at −700 mmHg for 2 hours. The treated 55 mass% DMAc aqueous solution at 10 ° C. was discharged from the inner liquid discharge hole, passed through a 330 mm air gap portion blocked from the outside air by a spinning tube, and then solidified in water at 60 ° C. The nozzle slit width of the tube-in-orifice nozzle used is an average of 45 μm, the maximum is 45.5 μm, the minimum is 44.5 μm, the ratio of the maximum value and the minimum value of the slit width is 1.02, the center discharge hole diameter of the nozzle is 100 μm, The draft ratio was 1.06, and the absolute humidity of the dry section was 0.12 kg / kg dry air. The hollow fiber membrane pulled up from the coagulation bath was passed through a 90 ° C. water washing tank for 30 seconds to remove the solvent and excess hydrophilic polymer, and then wound up. A polyethylene film similar to that of Example 1 was wrapped around a bundle of about 10,000 hollow fiber membranes, and then immersed in a 40 vol% isopropanol aqueous solution at 30 ° C. for 30 minutes × twice and replaced with water. did.

  The resulting wet hollow fiber membrane bundle is centrifuged and drained for 600 rpm × 5 min, set on a rotary table in the drying device in 48 × 2 stages, irradiated with an initial 7 kW of microwaves, and inside the drying device. The pressure was reduced to 5 kPa and the drying process was performed for 65 minutes. Subsequently, a drying process was performed for 50 minutes at a microwave output of 3.5 kW and a degree of vacuum of 5 kPa. Further, the microwave output was reduced to 2.5 kW, and similarly a drying treatment was performed for 10 minutes to complete the process. The maximum temperature reached on the surface of the hollow fiber membrane bundle during the drying treatment was 65 ° C., and the water content averaged 2.8% by mass. The roller for changing the yarn path during the spinning process was a mirror-finished surface, and the fixing guide was a satin-finished surface. The hollow fiber membrane obtained had an inner diameter of 200.2 μm and a film thickness of 29.1 μm.

  As a result of assembling a blood purifier using the hollow fiber membrane thus obtained and conducting a leak test, no adhesion failure due to the sticking of the hollow fibers was found.

  The blood purifier was subjected to the subsequent analysis without performing the crosslinking treatment of the hydrophilic polymer. When the hollow fiber membrane was cut out from the blood purifier not irradiated with γ-rays and subjected to the eluate test, the elution amount of PVP was as good as 7 ppm. Further, when the outer surface was observed with a microscope, no defects such as scratches were observed.

  No blood cell leak was found in the blood leak test using bovine blood. In addition, as a result of the endotoxin permeation test, endotoxin filtered from the outside of the hollow fiber to the inside of the hollow fiber was below the detection limit and was at a level with no problem. The other analysis results are shown in Table 1.

(Comparative Example 5)
23% by mass of polyethersulfone (Sumika Chemtex, Sumika Excel (R) 7800P), 7% by mass of PVP (Collidon (R) K-30, BASF), 67.5% by mass of DMAc, 2.5% by mass of water The solution was melted at 50 ° C., and the pressure in the system was reduced to −450 mmHg using a vacuum pump. The system was immediately sealed and allowed to stand for 30 minutes so that the solvent and the like were volatilized and the film forming solution composition did not change. This operation was repeated twice to degas the film forming solution. The obtained film-forming solution is passed through a two-stage filter of 30 μm and 30 μm, and then discharged from the outer slit of the tube-in orifice nozzle heated to 50 ° C., and at the same time, the inner solidification from the inner liquid discharge hole of the tube-in orifice nozzle A 50% by mass DMAc aqueous solution at 50 ° C. that had been degassed under reduced pressure in advance was discharged as a liquid, and after passing through a 350 mm air gap portion that was blocked from outside air by a spinning tube, the solution was solidified in 50 ° C. water. A 50% by mass DMAc aqueous solution at 50 ° C. that had been degassed in advance as an internal coagulating liquid was discharged from the internal liquid discharge hole of the tube-in orifice nozzle, passed through a 350 mm air gap portion that was blocked from the outside air by a spinning tube, and then 50 ° C. Coagulated in water. The nozzle slit width of the used tube-in-orifice nozzle is an average of 45 μm, the maximum is 45.5 μm, the minimum is 44.5 μm, the ratio of the maximum value and the minimum value of the slit width is 1.02, the center discharge hole diameter of the nozzle is 100 μm, The draft ratio was 1.06, and the absolute humidity of the dry section was 0.07 kg / kg dry air. The hollow fiber membrane pulled up from the coagulation bath was passed through a water washing tank at 85 ° C. for 45 seconds to remove the solvent and excess hydrophilic polymer, and then wound up. The resulting 10,000 hollow fiber membrane bundles were not washed and dried at 60 ° C. for 20 hours. Adherence was observed in the hollow fiber membrane bundle after drying, and when assembling the blood purifier, the end portion adhesive resin did not enter well between the hollow fiber membranes, and the blood purifier could not be assembled. The analysis results are shown in Table 1.

  The polysulfone-based hollow fiber membrane of the present invention has high safety and stability of performance, is excellent in module assembly, has little change over time during hemodialysis, and has high water permeability for use in the treatment of chronic renal failure. It is suitable for a hollow fiber blood purifier having a large contribution to the industry.

Claims (11)

A hollow fiber membrane comprising a polysulfone resin and a hydrophilic polymer as main components, wherein the following (i) to (c) are simultaneously satisfied:
(A) The content of the hydrophilic polymer in the outermost surface layer on the blood contact side surface in the hollow fiber membrane is 1.1 times or more than the content of the hydrophilic polymer in the surface vicinity layer on the blood contact side surface It is.
(B) The content of the hydrophilic polymer in the outermost layer on the surface opposite to the blood contact side in the hollow fiber membrane is 1.1 times or more than the content of the hydrophilic polymer in the outermost layer on the blood contact side surface It is.
(C) The hollow fiber membrane has at least a dense layer on the blood contact side surface, the dense layer is composed of aggregates of aggregated particles having an average diameter of 20 to 200 nm, and a hydrophilic polymer is formed on the surface of the aggregated particles. It is concentrated.   In the polysulfone-based hollow fiber membrane, the outermost layer on the blood contact side surface is a layer having a depth of 10 nm from the blood contact surface, and the near surface layer is 1000 to 1500 nm (1 to 1.5 μm) in depth from the blood contact surface. The polysulfone-based permselective hollow fiber membrane according to claim 1, wherein the layer is a layer up to.   The content of the hydrophilic polymer in the polysulfone-based hollow fiber membrane is 20 to 40% by mass in the outermost layer on the blood contact side surface, 5 to 20% by mass in the surface vicinity layer, and 25 in the outermost layer on the surface opposite to the blood contact side. The polysulfone-based permselective hollow fiber membrane according to claim 1 or 2, wherein the content is -50 mass%.   The polysulfone-based selectively permeable hollow fiber membrane according to any one of claims 1 to 3, comprising 99 to 80 mass% of a polysulfone resin and 1 to 20 mass% of a hydrophilic polymer as main components.   The polysulfone-based permselective hollow fiber membrane according to any one of claims 1 to 4, wherein the hydrophilic polymer is polyvinylpyrrolidone.   The polysulfone-based permselective hollow fiber membrane according to any one of claims 1 to 5, wherein an elution amount of the hydrophilic polymer from the hollow fiber membrane is 10 ppm or less.   The polysulfone-based permselective hollow fiber membrane according to any one of claims 1 to 6, wherein the porosity of the outer surface of the hollow fiber membrane is 8% or more and less than 25%.   The polysulfone-based permselective hollow fiber membrane according to any one of claims 1 to 7, wherein the hydrophilic polymer is crosslinked and insoluble in water.   9. The polysulfone-based permselective hollow fiber membrane according to claim 1, wherein the standard deviation value is less than 0.5 times the average value in the particle size distribution of the aggregated particles forming the dense layer. . To a blood purifier having a membrane area of 1.5 m ² housing the hollow fiber membrane, 300 ml / min of bovine blood adjusted to a hematocrit (Ht) of 40 ± 2.0% and a total protein concentration of 6.5 ± 0.5 g / dl. SCβ2MG (120 minutes) when the sieving coefficients of β2-microglobulin 15 minutes and 120 minutes after the start of blood filtration at the filtration rate of 90 ml / min are SCβ2MG (15 minutes) and SCβ2MG (120 minutes), respectively. The polysulfone-based permselective hollow fiber membrane according to claim 1, wherein / SCβ2MG (15 minutes) ≧ 0.6. To a blood purifier having a membrane area of 1.5 m ² housing the hollow fiber membrane, 300 ml / min of bovine blood adjusted to a hematocrit (Ht) of 40 ± 2.0% and a total protein concentration of 6.5 ± 0.5 g / dl. The polysulfone-based permselectivity according to any one of claims 1 to 10, wherein the sieving coefficient of bovine serum albumin at 15 minutes after the start of blood filtration at a filtration rate of 90 ml / min is 0.1 or less. Hollow fiber membrane.