Asymmetric cellulose virus removal filtration membrane and its preparation process

Abstract

This application describes an asymmetric cellulose virus removal filtration membrane comprising a porous body having a liquid supply surface and a liquid discharge surface. The average pore size measured by SEM is larger on the liquid supply surface than on the liquid discharge surface. The area in the filtration membrane where gold colloids with a particle size of x nm are captured is defined as Dx, and D 20 The region is located within 20-99% of the area of ​​the porous body, and the ratio of its thickness to the thickness of the porous body is 15-40%, and D 40 The present application also discloses an asymmetric cellulose virus removal filtration membrane, in which the region is located within 0-80% of the porous body, the ratio of the thickness of the region to the thickness of the porous body is 30-70%, and the interior of the filtration membrane further includes a transition region having a thickness of 20 μm or less, and a preparation process for the filtration membrane. The filtration membrane of the present application uses a cellulose-based raw material with excellent hydrophilicity as the membrane-forming material, has a distance between the 20 nm gold colloid trap region and the drainage surface, the thickness of the 40 nm gold colloid trap region is large, and the thickness of the transition region inside the filtration membrane is small, resulting in a high virus trapping rate, high load capacity, and high protein yield.

Description

[Technical Field]

[0001] The present application relates to the field of membrane separation technology, and in particular to an asymmetric cellulose virus removal filtration membrane and a process for preparing the same. [Background technology]

[0002] Biological products generally refer to substances that are created using modern biotechnology and chemical techniques from various microorganisms (bacteria, bacteriophages, rickettsia, etc.), microbial metabolic products, parasites, human or animal blood or tissues, etc., and that have excellent diagnostic, preventive, and therapeutic effects against specific infectious diseases and immune disorders.

[0003] Biological drugs have developed rapidly over the past few decades due to their unique targeting capabilities, high efficacy, and low side effects. However, with the rapid development of biological drugs, increasing the concentration of active substances (proteins) in biological drugs has become a recognized development direction. High-concentration biological drugs are typically administered by subcutaneous injection, which shortens treatment time and reduces pain for patients compared to intravenous injection. Furthermore, high-concentration biological drugs significantly reduce the filling volume, significantly reducing manufacturing and transportation costs.

[0004] During the manufacturing process of biological products, viruses may be present in raw materials or at each manufacturing step. The newly published "Chinese Pharmacopoeia" and the ICH's Q5A "Viral Safety Assessment of Biological Products" set out clear requirements for the viral safety of biological products. The viral safety assessment and test reports directly affect the outcome of drug declarations and approvals. Therefore, all biopharmaceutical companies must implement viral clearance and / or viral inactivation steps in the manufacturing of all types of biological products to ensure that biological products are free of viral safety issues.

[0005] Membrane separation technology has been widely used in the production of various biological products due to its high separation efficiency, low energy consumption, small installation area, and the fact that it is less likely to cause denaturation of active protein substances, especially when removing viruses. The most central part of membrane separation technology is the filtration membrane, which can filter a wide range of viruses of various sizes in biological products and improve the viral safety of biological products.

[0006] For example, the Chinese invention patent application publication number CN113842792A discloses an asymmetric PES filtration membrane for virus removal, which consists of a prefiltration layer and a virus-trapping separation layer, with one side of the prefiltration layer and the other side of the separation layer transitioning to form a continuous fiber body. While this PES filtration membrane has a good virus removal effect (LRV>4), the PES material itself has low hydrophilicity (still low after hydrophilic modification), which determines its high protein adsorption effect. When the protein concentration in biological products increases, the membrane is easily clogged by the adsorbed proteins, resulting in a rapid decline in permeation flux and a decrease in the concentration of the biological product, potentially affecting the quality of the biological product.

[0007] U.S. Patent US20200238221A1 filed by Sartorius discloses a porous monolayer polymer membrane having at least one major surface with a surface porosity of at least 40%, a total porosity of at least 40% and an asymmetry factor of 1.5 to 10. This filtration membrane is primarily used for filtering viruses, proteins, or polymers. It can only trap large particulate matter of several hundred nanometers. However, its large pore size makes it difficult to trap small proteins and is less susceptible to clogging by viruses, resulting in slower flux decay. However, due to the large pore size of this filtration membrane, it is ineffective at filtering small viruses, such as those with a particle size of 20 nm, and the desired trapping effect for small viruses cannot be achieved.

[0008] A Chinese patent filed by Millipore Corporation and published under publication number CN1759924B discloses a multi-layer composite ultrafiltration membrane comprising: a first filtration membrane layer having a second surface equivalent to the first surface of at least one layer; and a second filtration membrane layer having at least one surface equivalent to the first surface of the first layer, the first layer being connected and overlapping with the second layer, and having a porosity-connected transition region from the equivalent first surface of the second layer to the equivalent second surface of the first layer, wherein at least one of the layers is an asymmetric ultrafiltration membrane.

[0009] The process for preparing the above-mentioned ultrafiltration membrane mainly involves: both solutions cast in layers on the surface of the support have lower critical solution temperatures (LCSTs); the ultrafiltration membrane layer has a high LCST or no LCST or no measurable LCST; the solution for the microporous layer has an LCST; the cast multilayer liquid sheet is heated to a predetermined temperature higher than the LCST of the microporous layer and lower than the LCST of the ultrafiltration membrane layer, and then immersed in a precipitation bath to form a wet multilayer ultrafiltration membrane; or different polymer solutions are supplied to each raw material supply port; the solutions are applied to the surface of the moving support to form a multilayer coating layer on the support; and the solutions are distributed among the multiple layers; before each successive layer is applied, only partial phase separation occurs in the previous layer; and the multiple layers are subjected to a phase separation treatment to complete the phase separation, thereby forming a wet multilayer ultrafiltration membrane. That is, the phase separation mechanism is such that the polymer solution is a homogeneous liquid at low temperatures, and as the temperature rises, it reaches the LCST of the microporous layer, causing the components in the microporous layer to become insoluble in each other and hardening due to phase separation, while the ultrafiltration membrane layer undergoes phase separation hardening through the subsequent precipitation bath. That is, the phase separation mechanism of the ultrafiltration membrane layer is different from that of the microporous layer; the microporous layer quickly reaches a temperature above the LCST due to the action of the support, causing phase separation hardening, while the ultrafiltration membrane layer undergoes phase separation hardening through the action of the precipitation bath after the microporous layer has completely phase separated and hardened.

[0010] This is because the above-mentioned process for preparing an ultrafiltration membrane requires that the microporous layer solution be first cast onto a support and then the ultrafiltration membrane layer solution be cast into the microporous layer solution in order to achieve "phase separation of the microporous layer at high temperatures" and "phase separation of the ultrafiltration membrane layer in a precipitation bath." Otherwise, the ultrafiltration membrane layer solution cannot directly contact the precipitation bath, and the support must first heat the ultrafiltration membrane layer solution to heat the microporous layer solution, making it impossible to form a multi-layer ultrafiltration membrane. This preparation process has many limitations, and during the preparation, the ultrafiltration membrane layer remains in a liquid membrane state for a long time, which can easily reduce the uniformity of the ultrafiltration membrane layer. Furthermore, after the ultrafiltration membrane layer is phase-separated and hardened, the uniformity of the micropore structure may also be reduced, which is another drawback of the preparation process.

[0011] The patent also states that the filtration membrane is less susceptible to trapping at boundaries than a bilayer PVDF membrane (a bilayer PVDF membrane is obtained by casting a liquid PVDF cast solution onto a pre-fabricated solid PVDF membrane and then subjecting it to phase separation and curing). However, although the bilayer cast solution in this patent is in a liquid state, there are distinct phase separation stages, and the phase separation mechanism of the bilayer cast solution is also different, resulting in significant differences in the fiber structure and micropore structure of the two-layer structure after phase separation and curing. This is evident as a clear boundary line in the cross-sectional SEM image of the filtration membrane. While the bilayer membrane with a large difference in fiber structure and micropore structure alleviates the trapping problem of a bilayer PVDF membrane to some extent, the existence of a boundary line between the different layers within the membrane still results in significant trapping at the boundary between the layers, which is another problem that the filtration membrane in this patent still faces.

[0012] In addition, the membrane-forming materials in the examples of the patent are all polyethersulfone (PES), which, due to its characteristics, has poor hydrophilicity even after hydrophilic modification. When poorly hydrophilic membrane-forming materials combine with the small pore structure of the ultrafiltration membrane layer, the protein yield decreases. In the biomedical field, it is difficult to accept low protein yields due to the high price of proteins.

[0013] In various biological preparations whose main active ingredients are proteins, polypeptides, and their derivatives, filtration membranes with high virus-trapping effects tend to have fast flux decay and low protein yields, while filtration membranes with slow flux decay and high protein yields often fail to achieve the desired virus-trapping effect, making it difficult to combine various properties. This is a difficult and urgent problem that needs to be solved in the current virus removal filtration membranes used in the virus filtration processes of various biological preparations. Summary of the Invention [Problem to be solved by the invention]

[0014] The present application provides an asymmetric cellulose virus removal filtration membrane and a process for preparing the same. The filtration membrane of the present application uses a cellulose-based raw material with excellent hydrophilicity as the membrane-forming material, resulting in a high protein yield. When trapping 20 nm gold colloids, there is a distance between the 20 nm gold colloid trapping region and the drainage surface, reducing the risk of virus leakage. When trapping 40 nm gold colloids, the 40 nm gold colloid trapping region has a large thickness, resulting in a high dirt retention capacity. In addition, the transition region inside the filtration membrane has a small thickness, which reduces the impact of impurities 40 nm or larger on the 20 nm gold colloid trapping region and reduces clogging, resulting in a high load capacity. In other words, the filtration membrane of the present application combines a high protein yield, a high virus trapping rate, and a high load capacity. The filtration membrane preparation process of the present application employs a two-layer casting process, and the membrane structure of the filtration membrane can be adjusted relatively easily by adjusting the solid content of each of the two-layer casting solutions and the surface tension of the solvent system. Therefore, the filtration membrane preparation process of the present application is more environmentally friendly and causes less environmental pollution than the cuprammonium fiber preparation process.

[0015] In a first aspect, the present application provides an asymmetric cellulose virus removal filtration membrane using the following technical means:

[0016] In an asymmetric cellulose virus removal filtration membrane including a porous body having an internal non-directional serpentine flow path, one surface of which is a liquid supply surface and the other surface of which is a liquid drainage surface, the average pore diameter measured by SEM is larger on the liquid supply surface than on the liquid drainage surface,

[0017] When trapping gold colloids with the above filtration membrane in a wet state, the area where gold colloids with a particle size of x nm are captured is defined as Dx. In the thickness direction, the liquid supply surface of the porous body is defined as the position where the film thickness is 0%, and the liquid discharge surface of the porous body is defined as the position where the film thickness is 100%.

[0018] D above 20 the region is located within a region of 20 to 99% of the film thickness position of the porous body, and the ratio of the thickness of the region to the thickness of the porous body is 15 to 40%;

[0019] D above 40 the region is located within a region of 0 to 80% of the film thickness position of the porous body, and the ratio of the thickness of the region to the thickness of the porous body is 30 to 70%;

[0020] D above 40 The area is D 20 It is closer to the liquid supply surface than the above D 40 The side of the area closest to the drainage surface and the above D 20 The region between the side closest to the liquid supply surface is a transition region having a thickness of 20 μm or less.

[0021] The filtration membrane provided in the present application adopts the above-mentioned technical means and uses a cellulose-based material as the membrane-forming raw material, which has good hydrophilic effect, weak protein adsorption, and improves protein yield.

[0022] It can be clearly seen that the filtration membrane body structure provided in this application has two surfaces with different pore structures, with the pore structure on one surface having significantly larger pore size and more pores, which serves as the membrane's feed surface, and the pore structure on the other surface having significantly smaller pore size and fewer pores, which serves as the membrane's discharge surface. The interconnected omnidirectional serpentine channels in the porous body are adapted to the larger pore structure on the side closer to the membrane's feed surface to trap large particle impurities in the raw liquid, while the interconnected omnidirectional serpentine channels in the porous body are adapted to the smaller pore structure on the side closer to the membrane's discharge surface to trap viruses in the raw liquid. By trapping particles of different sizes in different regions across the membrane's thickness, the possibility of local clogging of the membrane can be reduced and a higher load capacity can be achieved.

[0023] After conducting a gold colloid trapping experiment using a filtration membrane, the gold colloid distribution in the filtration membrane can be measured according to the virus removal membrane test method in Chinese Patent CN105980038B. Specifically, slices are cut from a virus removal filtration membrane through which a gold colloid solution has been filtered, and the brightness distribution of multiple regions of the cross section of the slice stained with gold colloid is measured using an optical microscope. Because gold colloids absorb light, the brightness variation depends on the amount of gold colloid trapped. If necessary, background noise can be removed from the brightness distribution. Then, the horizontal axis is plotted against the vertical axis, and the brightness variation is determined to determine the region where colloids of a certain particle size are trapped in the membrane thickness direction. In the optical microscope measurement, the brightness variation, calculated by subtracting the measured brightness distribution from the constant (255), indicates that if the absolute value of the brightness variation at a spectral region is less than 10% of the maximum spectral absolute value, the capture of gold colloids in that region can be considered within the error range from the perspective of the virus removal ability of the virus removal membrane.

[0024] Furthermore, although some gold colloids are present in certain regions along the membrane thickness direction, the amount is so small that these regions are not considered to be regions where gold colloids are trapped, and only a portion of the gold colloids remains. Therefore, preferably, the portions of the virus removal membrane that are continuously formed in the membrane thickness direction and capture gold colloids with a diameter of 20 nm should be understood to be regions that actually trap gold colloids of the corresponding particle size.

[0025] It should be understood that in addition to optical microscopy, those skilled in the art can characterize the trapping area and amount of gold colloid by other known techniques. For example, elemental gold analysis of a cross section of a gold colloid trap membrane using energy dispersive X-ray spectroscopy (EDS) can also be performed to obtain a similar spectrum by obtaining a distribution curve of elemental gold across the membrane thickness.

[0026] Generally, to ensure a filtration membrane's effectiveness in trapping small viruses, it is necessary for the membrane to be able to filter viruses approximately 20 nm in size (e.g., the approximately 20 nm Mycovirus of Mice or bacteriophage PP7, a model virus for small viruses according to the relevant regulations of PDATR41). Because 20 nm gold colloids have essentially the same particle size as small viruses, the location and trapping conditions of 20 nm gold colloids in a filtration membrane can essentially characterize the trapping of small viruses by a filtration membrane. Furthermore, because gold colloids are opaque to light, they are easier to characterize than viruses, which are difficult to characterize.

[0027] To ensure that the filtration membrane has a good trapping effect on large-sized impurity particles, it is necessary for the filtration membrane to be able to filter impurities of approximately 40 nm (larger-sized impurities are actually filtered in the previous step). Similarly, the trapping behavior of 40 nm gold colloids in the filtration membrane can fundamentally characterize the trapping of large particle impurities by the filtration membrane.

[0028] where D 20 The area is located close to the drainage surface of the porous body, but does not reach the drainage surface. Its thickness is 15-40% of the thickness of the porous body. 20 It can be seen that the area has a constant thickness, and the 20 nm gold colloids are distributed over a wide area in the film thickness direction (thickness ratio of 15% or more), and are not trapped in a narrow area. In other words, it can be seen that the trapping is not concentrated in a narrow area, which causes clogging of the filtration membrane and reduces the permeation flux. 20 The thickness of the area must not be too large (e.g., greater than 40%), which is 20 If the ratio of the thickness of the region is too large, the filtration membrane will be less effective at trapping the 20 nm gold colloids. As a result, the 20 nm gold colloids will be able to permeate in the direction of the drainage surface of the filtration membrane. Therefore, the 20 nm gold colloids will be widely distributed in the thickness direction of the filtration membrane, resulting in a D 20 However, the pore size distribution of the filtration membrane pore structure is similar to a normal distribution, and since pore structures with small pore sizes generally clog quickly, in this case, the raw liquid is more likely to permeate through the unclogged pore structures with large pore sizes. Furthermore, the risk of virus leakage due to the large pore size pore structure increases significantly, and the LRV decreases. That is, D 20 If the ratio of the thickness of the region is too small, the permeation flux will be easily attenuated rapidly due to the membrane trapping in a narrow area. 20 If the ratio of the thickness of the region is too large, the risk of virus leakage increases due to the large pore size of the filtration membrane, and the LRV is likely to decrease.

[0029] In addition, D 20The thickness distribution of the membrane must not be too close to the feed or drain surface (the trap position should not be less than 20% and not more than 99%). If the membrane is too close to the feed surface, the membrane will usually have a high LRV, but it also means that the membrane's small pore structure is too close to the feed surface. The small pore structure has a high resistance to the feed liquid and often results in a low membrane permeation flux. On the other hand, if the membrane is too close to the drain surface and the thickness ratio is too large, the membrane will usually have a high membrane permeation flux, but it always means that the risk of virus leakage from the membrane is significantly increased.

[0030] Similarly, D 40 The thickness ratio of the region should not be too high or too low. If the thickness ratio is too high, it will often mean that the filtration membrane will not be able to trap large 40nm particle impurities effectively. If large 40nm particle impurities leak into the small pore size region, the small pore size region will be rapidly clogged, the permeation flux will rapidly decrease, and the risk of virus leakage will increase. If the thickness ratio is too low, large 40nm particle impurities will be trapped in a concentrated area, causing localized clogging of the filtration membrane, which will also rapidly decrease the permeation flux of the filtration membrane and further increase the risk of large 40nm particle impurities leaking into the small pore size region.

[0031] In addition, there is a transition region in the filtration membrane that connects the large pore structure with the small pore structure, and D 20 Area and D 40 Between the area or D 20 Area and D 40 It is located in the overlapping region of the regions, and its thickness must not exceed 20 μm. This is because the transition region is D 20 Area and D 40 The overlapping area of ​​the region, D 20 Area and D 40If the overlapping area is large, it means that there is a region in the filtration membrane that traps not only 40 nm gold colloids but also 20 nm gold colloids. In this region, the 40 nm gold colloids and the 20 nm gold colloids clog the pore structures with different pore sizes, causing local clogging of the filtration membrane and a rapid decrease in permeation flux. In addition, the transition region is D 20 Area and D 40 Located between the area D 20 Area and D 40 When the distance from the transition region is large, it is found that there is a region in the filtration membrane that does not trap not only the 40 nm gold colloids but also the 20 nm gold colloids. In this region, the effect of the 40 nm gold colloids on the pore structure with small pore diameters can be significantly reduced. However, if the thickness of the transition region is too large, it will create a large resistance to the feed liquid, often reducing the permeation flux of the filtration membrane.

[0032] D 20 The pore structure with a relatively small pore diameter in the region D 20 By adapting the thickness of the region to a certain value, the pores are stacked in the thickness direction, and the trapping effect becomes smaller than the size of the actual pore structure. For example, the projection surface of the pore structure of the upper layer becomes the fibrous structure of the pores of the lower layer that separates the pore structure of the upper layer, thereby improving the trapping effect of the pore structure for smaller particle sizes. Furthermore, by stacking such pore structures in the film thickness direction, it is possible to ensure a good trapping effect of the pore structure for smaller particle sizes.

[0033] Also, D 20 Since the area does not reach the drainage surface, D 20 Between the area and the drainage surface, D 20It has been found that the membrane has a good trapping effect on viruses leaking from certain regions, significantly reducing the risk of virus leakage, and that some regions have a high LRV. Combined with the use of cellulose-based raw materials with excellent hydrophilicity for the membrane, the membrane not only has a high LRV but also a high protein yield, and the membrane protein yield of the present application can basically be stably maintained at 98% or more. 20 The fact that the area is within 20-99% means that D 20 It is not necessary that the area reaches 99%; it may reach only 80% of the area, or it may reach 50% of the area.

[0034] Also, D 40 The pore structure with a relatively large pore diameter in the region D 40 By adapting to the large thickness of the region, not only can we achieve efficient trapping of 40 nm gold colloids, but also provide a large enough dirt retention space for the filtration membrane to have room to retain large particle impurities even after long-term filtration, thereby improving D 20 Reduce the effect of large particle impurities on the region.

[0035] Therefore, D 20 By limiting the distribution location and distribution thickness of the region, it is possible to ensure that the filtration membrane has a high LRV and a high protein yield. 40 By limiting the distribution location and thickness of the region, it is possible to ensure that the filtration membrane has a high dirt retention capacity and that large particle impurities are less likely to affect the pore structure of the small pore size of the filtration membrane, resulting in a high permeation flux and a high loading capacity.Furthermore, by limiting the location and thickness of the transition region, it is possible to ensure that the filtration membrane has a high loading capacity and a high permeation flux, resulting in a filtration membrane with not only a high LRV but also a high loading capacity and a high protein yield.

[0036] The non-directional serpentine flow path refers to a randomly oriented groove structure and / or a discretely distributed pore structure, and it should be understood that each non-directional serpentine flow path is interconnected, and during filtration, the raw liquid flows through the serpentine pore structure, and impurities in the raw liquid are sieved, adsorbed, or otherwise trapped.

[0037] Although the filtration membrane in this application has a multilayer structure, it is not obtained by stacking multiple layers, and the filtration membrane in this application does not have clear boundaries and is less likely to undergo interlayer separation. Furthermore, in this application, when performing gold colloid trapping tests and virus challenge tests, the pressure is 30 psi. However, the trapping area of ​​the filtration membrane for particles of different sizes may change under different test pressures, and the risk of virus leakage will also differ, so the test pressure must be limited.

[0038] The various surface appearance parameters of filtration membranes (fiber diameter, pore size, etc.) can be measured by characterizing the appearance of the membrane structure using a scanning electron microscope, using computer software (Matlab, NIS-Elements, etc.) or manually, and then performing the corresponding calculations. Furthermore, during the membrane preparation process, various characteristics such as pore size distribution are nearly uniform and essentially constant in the direction perpendicular to the membrane thickness (for flat membranes, this direction is the planar direction; for hollow fiber membranes, this direction is perpendicular to the radial direction). Therefore, measuring the average pore size of a portion of the corresponding plane can reflect the overall average pore size on that plane. In actual measurements, the membrane surface (or cross section) is first characterized using an electron microscope, and the corresponding SEM image is obtained, and the pore size is measured within 1 μm. 2 (1μm×1μm) or 25μm 2A certain area, such as (5 μm×5 μm), is selected, and the specific size of the area varies depending on the actual situation. The pore diameters of all pores within the area are measured using corresponding computer software or manually, and then calculated to determine the average pore diameter of the area. Of course, those skilled in the art can also determine the above parameters using other measurement means, and the above measurement means are for reference only.

[0039] Optionally, D above 20 The region is located within 40 to 99% of the film thickness position of the porous body, and the above D 40 The region is located within a region of 0 to 60% of the film thickness position of the porous body.

[0040] By adopting the above technical means, the D 20 Location of region, D within the porous body 40 By further restricting the location of the region, the problem of reduced permeation flux due to the small pore structure being too close to the liquid feed surface can be further reduced, and the possibility that the 40 nm gold colloids being too close to the liquid discharge surface will affect the small pore structure in the filtration membrane can be reduced, further ensuring that the filtration membrane not only has a high permeation flux but also a high loading capacity.

[0041] Optionally, D above 20 In the region, when the distance between the capture peak position of the 20 nm gold colloid and the drainage surface is L1, the ratio of L1 to the thickness of the porous body is 10 to 50%, and the D 40 In this region, when the distance between the capture peak position of 40 nm gold colloid and the liquid supply surface is L2, the ratio of L2 to the thickness of the porous body is 1 to 40%.

[0042] By adopting the above technical means, the gold colloid trapping peak position is the position where the amount of gold colloid trapped corresponding to the particle size is the largest in the film thickness direction, and when observing the spectrum, the brightness offset at that position is the largest (the gold element can be confirmed most in the spectrum by EDS), and the amount of gold colloid trapped gradually decreases from the peak position to both sides. 20 The distance L1 between the capture peak position of the area and the drainage surface characterizes the risk of leakage of 20 nm gold colloids, and D 40 The distance L2 between the capture peak position of the region and the liquid supply surface can characterize the dirt retention capacity of the filtration membrane for large particle bodies, and although the ranges of the two are different, they should be controlled within appropriate ranges.

[0043] Here, if the thickness ratio of L1 is too small (e.g., less than 10%), the capture peak of 20 nm gold colloids is located too close to the drainage surface. This suggests that small pore structures that effectively trap 20 nm gold colloids have a high permeation flux but are close to the drainage surface. However, as mentioned above, the pore size distribution of the filtration membrane is similar to a normal distribution, and even in the small pore structure region, there are also pores with slightly larger pore sizes. Being too close to the drainage surface increases the risk of 20 nm gold colloids leaking through these pores. If the thickness ratio of L1 is too large (e.g., more than 50%), the capture peak of 20 nm gold colloids is located too far from the drainage surface. This means that the filtration membrane begins to trap 20 nm gold colloids too early. While the risk of 20 nm gold colloids leaking can be reduced, this often comes at the cost of a significant decrease in permeation flux, significantly reducing the practical value of the filtration membrane.

[0044] Similarly, if the thickness ratio of L2 is too small (e.g., less than 1%), a large number of 40 nm gold colloids are trapped at the feed surface (rather than being distributed uniformly and dispersedly), which clogs the pore structure near the feed surface of the membrane, resulting in a rapid decline in permeation flux. If the thickness ratio of L2 is too large (e.g., greater than 40%), the membrane structure near the feed surface is weak in trapping 40 nm gold colloids, and the pore size is too large. Large pore sizes often mean poor mechanical properties (such as reduced pressure resistance). Furthermore, if the thickness ratio of L2 is too large, the risk of 40 nm gold colloid leakage increases. If the small pore size of the membrane becomes clogged, the permeation flux rapidly declines, increasing the risk of virus leakage.

[0045] Optionally, when the distance between the capture peak position of the 20 nm gold colloid and the capture peak position of the 40 nm gold colloid is L3, the ratio of L3 to the thickness of the porous body is 20% or more.

[0046] By adopting the above technical means, the distance L3 between the capture peak positions of the 20 nm gold colloid and the 40 nm gold colloid can characterize the distance between the concentrated trapping regions of the 20 nm gold colloid and the 40 nm gold colloid in the thickness direction of the filtration membrane, and D 20 The influence of 40 nm gold colloids on the area can also be characterized to some extent, which may cause local clogging of the filtration membrane. The inventors of this application have found that when the thickness ratio of L3 is greater than 20%, 40 nm gold colloids can cause local clogging of the filtration membrane. 20 It was found that the effect of this on the membrane area, leading to localized clogging of the small pore structure, was significantly reduced, slowing down the decay rate of the permeate flux of the membrane.

[0047] Optionally, D above 20 In the region, the region where the capture amount of 20 nm gold colloids is 0.8 times or more of the peak capture amount is defined as the concentrated capture region, and the above D 20The ratio of the thickness of the concentrated capture area to the thickness of the area is 10 to 45%.

[0048] By adopting the above technical measures, D 20 The area with a size of 0.8 times or more of the peak capture amount within the region is designated as the concentrated capture region, in which many 20 nm gold colloids are trapped. 20 If the ratio to the thickness of the region is too small (e.g., less than 10%), a large number of 20 nm gold colloids are trapped in a small area. 20 Since the pore size of the region is small, the narrow range of concentrated trapping leads to localized clogging of the filtration membrane, resulting in a rapid decay of the permeate flux. 20 If the ratio of the area to the thickness is too large (e.g., greater than 50%), the concentrated capture area will form a platform-like absorption curve, and the 20 nm gold colloid will be D 20 It also means that the small pore size pore structure within the region can pass through more smoothly, greatly increasing the risk of leakage of the 20 nm gold colloid.

[0049] Optionally, D above 40 In the region, the region where the capture amount of 40 nm gold colloid is 0.8 times or more of the peak capture amount is defined as the concentrated dirt retention region, and the above D 40 The ratio of the thickness of the concentrated dirt holding area to the thickness of the area is 60 to 90%.

[0050] By adopting the above technical measures, D 40 The area with a concentration of 0.8 times or more of the peak capture amount within the area is defined as the concentrated dirt retention area, and a large number of 40 nm gold colloids are trapped in the concentrated dirt retention area. 40If the ratio of the thickness of the concentrated dirt retention area to the thickness of the area is too small (e.g., less than 60%), a large number of 40 nm gold colloids will be trapped in a small area. Since the volume of 40 nm gold colloids is much larger than that of 20 nm gold colloids, the concentrated trapping of 40 nm gold colloids is likely to lead to localized clogging of the filtration membrane, resulting in a rapid decline in permeation flux. 40 If the ratio of the thickness of the concentrated stain-retaining area to the thickness of the area is too large (e.g., greater than 90%), the 40 nm gold colloid will not be able to 40 This indicates that a very wide distribution is formed in the region, and there is a very high risk that the 40 nm gold colloids will clog the small pore structure of the filtration membrane.

[0051] In addition, as a virus removal filtration membrane, the risk of virus leakage must be strictly controlled. 20 It is necessary to ensure that the concentration area within the area is not too large. 40 The transition and D regions mainly play a role in pre-filtration and fouling retention. The relatively uniform distribution of large particle impurities is beneficial in improving the fouling capacity of the filtration membrane and significantly reducing the risk of clogging of the filtration membrane by large particle impurities and rapid attenuation of the permeation flux. On the other hand, the transition and D regions trap even small amounts of large particle impurities. 20 There is a region, D 40 Regarding the area, it is necessary to ensure that the concentrated dirt retention area is not too small in order to maximize the dirt retention capacity of the filtration membrane and minimize the attenuation rate of the permeation flux of the filtration membrane. Therefore, for the filtration membrane of the present application, the thickness of the concentrated dirt retention area is not too large, while the thickness of the concentrated dirt retention area is not too small, so as to achieve both a high LRV and a high loading capacity.

[0052] Optionally, D above 40 The side of the area closest to the drainage surface and the above D 20 There is an overlapping area near the feed surface of the area, or

[0053] D above 40 The side of the area closest to the drainage surface and the above D 20 The areas close to the liquid supply surface do not overlap with each other, and a non-overlapping area exists,

[0054] The overlapping region or the non-overlapping region is the transition region, the side of the transition region closer to the liquid supply surface is considered to be 0, the side of the transition region closer to the liquid drain surface is considered to be 1, the region of 0 to 0.5 of the transition region is the upper transition region, and the region of 0.5 to 1 of the transition region is the lower transition region, and the average pore diameter measured by SEM is larger in the upper transition region than in the lower transition region.

[0055] By adopting the above technical means, the transition region is D 40 Area and D 20 The overlapping region or transition region is D 40 Area and D 20 When the transition region is divided equally into an upper transition region and a lower transition region according to thickness, regardless of whether it is a region between the upper and lower transition regions, the average pore size measured by SEM is larger in the upper transition region than in the lower transition region. That is, the average pore size measured by SEM in the transition region gradually decreases from near the liquid supply surface to near the liquid discharge surface. Therefore, the transition region also has a trapping effect for certain large particles, further reducing the risk of large particle impurities affecting the pore structure of small pores in the filtration membrane and causing local clogging of the filtration membrane.

[0056] Optionally, the transition region has an average pore size of 100 to 300 nm as measured by SEM, and the D 40 The average pore size measured by SEM in the region is smaller than the above D 20 This is larger than the average pore size measured by SEM in the area.

[0057] By adopting the above technical means, the average pore size measured by SEM in the transition region should not be too large or too small. If the average pore size measured by SEM in the transition region is too large, D 40This means that the average pore diameter measured by SEM in the transition region is larger, and although stacking the pore structure can trap large particle impurities in a similar manner, the risk of large particle impurities leaking out increases. Also, the average pore diameter measured by SEM in the transition region is similarly large, so large particle impurities are likely to be trapped in the D 20 If the average pore size measured by SEM in the transition region is too small, the D 20 This means that the average pore size measured by SEM in the region is smaller, and the LRV of the filtration membrane increases, but the resistance to the feed liquid increases. As a result, the permeation flux of the filtration membrane decreases, and the dirt retention capacity of the filtration membrane also decreases, which may make it more susceptible to clogging with impurities.

[0058] Optionally, the difference between the average pore size measured by SEM in the 0.3 to 0.5 region of the transition region and the average pore size measured by SEM in the 0.5 to 0.7 region of the transition region is 30 to 100 nm.

[0059] By adopting the above technical measures, the pore structure size in the transition region changes gradiently in the thickness direction. The region of 0.3 to 0.7 mm in thickness of the transition region is a relatively intermediate region of the transition region. The average pore size measured by SEM of the filtration membrane in this region does not change significantly. This indicates that there is no abrupt change in pore size in the intermediate region of the transition region. Naturally, this means that there is less chance of concentrated impurity trapping due to abrupt changes in pore size and reduces the risk of clogging in the transition region causing a rapid decline in the permeation flux of the filtration membrane.

[0060] Optionally, the ratio of the asymmetry coefficient of the upper transition region to the asymmetry coefficient of the lower transition region is 0.8 to 1.2, the asymmetry coefficient of the upper transition region being the ratio of the average pore size measured by SEM in the 0 to 0.1 region of the transition region to the average pore size measured by SEM in the 0.4 to 0.5 region of the transition region, and the asymmetry coefficient of the lower transition region being the ratio of the average pore size measured by SEM in the 0.5 to 0.6 region of the transition region to the average pore size measured by SEM in the 0.9 to 1 region of the transition region.

[0061] By adopting the above technical measures, the difference between the asymmetry coefficients of the upper and lower transition regions is not large, with the ratio between the two being only 0.8 to 1.2. This indicates that the difference in the rate of change of pore size between the upper and lower transition regions is small. Combined with the absence of an area of ​​abrupt pore size change in the middle region of the transition region, the change in pore size is more gradual. This indicates that there is no area of ​​abrupt pore size change in the upper and lower transition regions, which reduces the risk of clogging in the transition region and ensures a high load capacity of the filtration membrane.

[0062] Optionally, the asymmetry factor of said upper transition region is between 1.1 and 1.6, and the asymmetry factor of said lower transition region is between 1.1 and 1.6.

[0063] By adopting the above technical measures, the asymmetry coefficients of the upper and lower transition regions are all small, and coupled with the total thickness of the transition region not exceeding 20 μm and the change in pore size in the middle region of the transition region not exceeding 100 nm, it is suggested that there is no abrupt change in pore size in both the upper and lower transition regions, reducing the risk of clogging in the transition region.

[0064] Optionally, the ratio of the fiber diameter measured by SEM in the upper transition region to the fiber diameter measured by SEM in the lower transition region is 0.8 to 1.2, the fiber diameter measured by SEM in the upper transition region is 20 to 50 nm, and the fiber diameter measured by SEM in the lower transition region is 20 to 45 nm.

[0065] By adopting the above technical measures, the average pore size measured by SEM in the upper transition region becomes larger than the average pore size measured by SEM in the lower transition region. If the fiber diameter measured by SEM of the fibrous structure is smaller than 20 nm, the pore structure in the upper transition region is likely to be poorly supported. If the fiber diameter measured by SEM of the fibrous structure is larger than 50 nm, the pore structure is well supported, but the fibrous structure as a solid part has too much resistance to the feedstock liquid, which is likely to lead to a decrease in permeation flux. Similarly, for the pore structure in the lower transition region, the fiber diameter measured by SEM of the fibrous structure needs to be controlled to 20 to 45 nm.

[0066] Optionally, the difference between the fiber diameter measured by SEM in the 0.3 to 0.5 region of the transition region and the fiber diameter measured by SEM in the 0.5 to 0.7 region of the transition region is 10 nm or less.

[0067] By adopting the above technical means, the difference in fiber diameter measured by SEM in the middle part of the transition region of the filtration membrane of the present application is small, and the difference between the fiber diameter measured by SEM in the 0.3 to 0.5 region of the transition region and the fiber diameter measured by SEM in the 0.5 to 0.7 region of the transition region is 10 nm or less. This means that in the transition region, not only the pore size of the pore structure but also the diameter of the fiber structure transitions relatively smoothly, which is reflected in the boundary lines and layer separation phenomenon that are clearly visible in the cross section of the SEM image of the filtration membrane.

[0068] When observing an SEM image of a filtration membrane cross section, the surface appearance is the main factor affecting the visual impression of the SEM image. In the case of a pore structure or fibrous structure, if there is a large difference in the pore size or diameter of the fibrous structure between the upper and lower transition regions, a clear boundary line will be shown in the SEM image. The filtration membrane in this application has a small difference in pore size between the upper and lower transition regions, but there is little difference in the fibrous structure. This means that during filtration, the raw liquid does not encounter abnormal resistance from the solid part (i.e., the fibrous structure) that changes suddenly there, the raw liquid flows smoothly, and concentrated traps are less likely to occur in the transition region, which greatly helps to increase the load capacity of the filtration membrane.

[0069] Optionally, D above 40 The average pore size measured by SEM in the region gradually decreases from the liquid supply surface to the liquid drainage surface. 20 The average pore size measured by SEM of the area does not change substantially from the feed surface to the drain surface.

[0070] By adopting the above technical means, the average pore size measured by SEM does not change substantially. 20 Near the liquid supply surface of the area and D 20 The gradient of the average pore size change near the drainage surface of the region measured by SEM was less than 5 nm / μm, which means that the gradient of change is not clear. 20 This does not mean that the average pore size measured by SEM in the region must not vary strictly in the film thickness direction.

[0071] where D 40 The pore structure with gradient change in area is D 40 This means that the side of the region closer to the liquid supply surface has a pore structure with a large pore diameter, which makes it easier for the raw liquid to enter the filtration membrane through the pore structure with a large pore diameter, reducing the resistance to the raw liquid. At the same time, the pore structure with a large pore diameter also has a D 40 To ensure a large dirt holding capacity in the area and to ensure the trapping effect of large particle impurities, 40 The pore structure has a relatively small pore size near the drainage surface of the region, so the D 40 The area not only has a good trapping effect for large particle impurities, but also has a large dirt holding capacity and is difficult to clog quickly.

[0072] where D 20 Since the region has a pore structure in which the pore size does not change substantially, the D region has a pore structure with a small pore size relative to the permeation flux of the filtration membrane. 20 This significantly reduces the effect of the area. 20 When the pore size of the pore structure of the region changes in a large gradient, the smaller the pore size, the higher the trapping effect of 20 nm gold colloids, but the risk of concentrated trapping of 20 nm gold colloids also increases, resulting in D 20This is because the amount of dirt held in the area is reduced and the risk of localized clogging increases. Furthermore, to reduce the risk of leakage of the 20 nm gold colloids, the trapping area of ​​the 20 nm gold colloids must not reach the drainage surface of the filtration membrane. When the pore structure has a large gradient of change in pore size, the pore structure near the drainage surface of the filtration membrane tends to have small pore sizes, and this area creates a large resistance to the raw liquid, resulting in a significant decrease in the permeation flux of the filtration membrane. D, which has a pore structure with substantially unchanged pore size 20 The region does not have an obvious small pore structure, and 20 There is no local sudden increase in resistance to the feed liquid within the region, and the permeation flux of the filtration membrane increases.

[0073] Furthermore, it is generally believed that a pore structure in which the pore size does not change substantially often has a higher risk of virus leakage. 20 The region has a pore structure in which the pore size does not change substantially, and D 20 The thickness of the area is 15-40%, and D 20 It has also been found that, combined with the position of the region not reaching the drainage surface, it is possible to ensure that the filtration membrane has a high LRV and a low risk of leakage.

[0074] Optionally, the liquid supply surface has an average pore size of 350 to 2000 nm as measured by SEM, and the liquid drain surface has an average pore size of 30 to 45 nm as measured by SEM.

[0075] By adopting the above technical means, the porous body near the filtration membrane liquid supply surface is prevented from collapsing under pressure due to insufficient compressive strength, resulting in a loss of the permeation flux and virus separation ability of the filtration membrane. In order to avoid this, the average pore size measured by SEM on the filtration membrane liquid supply surface is not too large (for example, exceeds 2000 nm). Also, the resistance to the raw material liquid increases, resulting in D 40 The average pore size measured by SEM on the membrane feed surface should not be too small (e.g., less than 350 nm) to avoid insufficient foulant retention in the region, which would result in reduced flux and loading capacity.

[0076] The average pore size measured by SEM on the drainage surface of the filtration membrane should likewise not be too large (for example, greater than 45 nm), which is the D of the filtration membrane according to the present application. 20 The region has a pore structure in which the pore size does not change substantially, and D 20 Although the thickness of the region can still form a good trap for 20 nm gold colloids, the risk of leakage of 20 nm gold colloids increases significantly with the extension of the filtration time. The porous body near the drainage surface of the filtration membrane becomes the last barrier of the filtration membrane, and the average pore size measured by SEM at the drainage surface becomes small. 20 This is because it is possible to ensure that small amounts of 20 nm gold colloids that leak from the membrane are efficiently trapped. In addition, the average pore size measured by SEM on the drainage surface of the filtration membrane should not be too small (for example, smaller than 30 nm). This is because a pore structure with a pore size on the drainage surface that is too small would have too much resistance to the feed liquid, resulting in a significant decrease in the permeation flux of the filtration membrane.

[0077] It is now generally recognized both domestically and internationally that a virus removal membrane can essentially filter out all viruses if its virus trapping efficiency reaches an LRV of >6. Based on this, theoretically, improving the membrane's virus trapping efficiency (for example, by further reducing the average pore size measured by SEM on the membrane's outlet surface) can indeed improve the membrane's virus trapping efficiency, but this often comes at the expense of a significant reduction in permeation flux. Furthermore, there is a diminishing return on improving the membrane's virus trapping efficiency, meaning that even if the membrane's virus trapping efficiency is already excellent, further improvement requires a significant reduction in permeation flux. Furthermore, a membrane with an average pore size measured by SEM on the outlet surface of 30-45 nm not only significantly reduces the risk of virus leakage, but also minimizes the reduction in permeation flux. By appropriately adjusting the average pore size measured by SEM on the outlet surface according to actual needs, a higher virus trapping efficiency and higher permeation flux can be achieved.

[0078] Optionally, D above 20 The area between the area and the above-mentioned drainage surface is a leak-proof area, and / or 40 The area between the area and the liquid supply surface is a buffer area.

[0079] By adopting the above technical measures, D 20 The area does not reach the drainage surface, so D 20 Between the area and the drainage surface, there is an area where 20 nm of gold colloid is not trapped but has the ability to trap 20 nm of gold colloid. 20 This area acts as a leak prevention area, trapping small amounts of 20 nm gold colloid that leak out, significantly reducing the risk of leakage during long-term filtration. 40 There may be areas between the membrane and the liquid supply surface that have poor trapping ability for 40 nm gold colloids, and although the trapping effect of 40 nm gold colloids is low, the raw material liquid can be introduced more effectively into the filtration membrane, reducing the resistance to the raw material liquid near the liquid supply surface.

[0080] If the pore size of the membrane supply surface is small, 40 nm gold colloids may be trapped near the supply surface. 40 The region may also include the liquid supply surface, and therefore the buffer region does not necessarily have to be present, and the thickness of the buffer region may be zero.

[0081] Optionally, the ratio of the thickness of the leak-proof region to the thickness of the porous body is 5-30%, and / or the ratio of the thickness of the buffer region to the thickness of the porous body is 1-10%.

[0082] By adopting the above technical measures, it is necessary to control the thickness ratio of the leak-proof area within an appropriate range. If the thickness ratio of the leak-proof area is too large (e.g., greater than 30%), the virus risk of the filtration membrane can be further reduced, but the permeation flux of the filtration membrane tends to be low, resulting in low filtration efficiency. On the other hand, if the thickness ratio of the leak-proof area is too small (e.g., less than 5%), the trapping effect of the 20 nm gold colloid by the leak-proof area will be insufficient, resulting in D 20 The chances that a small amount of 20 nm colloidal gold leaking from the area will pass through the leak-proof area will be greatly increased, resulting in a significantly increased risk of virus leakage.

[0083] Similarly, the thickness of the buffer region must be controlled within an appropriate range. Its primary role is to introduce the raw liquid into the membrane and accommodate larger particle impurities (particle impurities larger than 40 nm may be present in small amounts in the raw liquid after the preliminary large-particle filtration process). Furthermore, if the thickness of the buffer region is too high (e.g., greater than 10%), a large proportion of the region will be less effective at trapping 40 nm gold colloids. This region generally has larger pore sizes, which are believed to better introduce the raw liquid into the membrane and retain larger particle impurities. However, since virus removal membranes typically use dead-end filtration, external pressure is the primary driving force during filtration, and the membrane is subjected to significant pressure from the raw liquid. However, the large pore size of the buffer region, combined with the soft texture of the cellulose material itself, makes the pore structure prone to collapse under pressure from the raw liquid. Once this collapse occurs, both the raw liquid introduction capacity and the membrane's dirt retention capacity are significantly reduced. Therefore, the thickness of the buffer region should not be too large. The thickness ratio of the buffer region is also not too small to avoid excessive resistance to the raw material liquid near the liquid supply surface and clogging of the pore structure near the liquid supply surface by larger particle impurities.

[0084] Furthermore, further increasing the thickness ratio of the leakproof area can indeed further reduce the risk of virus leakage and increase the LRV of the filtration membrane. However, as the thickness ratio of the leakproof area increases, there is a diminishing margin effect on the increase in the LRV of the filtration membrane. That is, as the thickness ratio of the leakproof area increases further, the rate of increase in the LRV of the filtration membrane slows, but as the thickness ratio of the leakproof area increases further, the permeation flux of the filtration membrane decreases significantly.

[0085] Optionally, D above 40 The average pore size measured by SEM in the region is 150 to 500 nm, and the above D 20 The average pore size of the region measured by SEM is 60 to 200 nm.

[0086] By adopting the above technical measures, D 40 The main role of this region is to trap large particles in the raw liquid (for example, particle impurities similar to gold colloids with a particle size of 40 nm), so it is necessary to have a large dirt retention capacity and good trapping effect for large particles. 40 If the average pore size measured by SEM in the region is too large (e.g., larger than 500 nm), D 40 The dirt holding capacity of the area is larger, but the trapping effect of large particles is insufficient, and D 40 If the average pore size measured by SEM in the region is too small (e.g., smaller than 150 nm), D 40 The region has a good trapping effect for large particle impurities, but if the pore size is too small, it means that the dirt retention capacity is insufficient, and clogging by large particle impurities occurs easily, resulting in a rapid decline in permeation flux.

[0087] D 20 The main role of this region is to trap small particles (for example, small viruses similar to those in gold colloids with a particle size of 20 nm), and considering that the virus removal filtration membrane must ensure a low risk of virus leakage, D 20 The area must have a good trapping effect for small viruses. 20It is necessary to ensure that the area has a certain dirt retention capacity for small viruses and is not prone to rapid clogging by small viruses. 20 If the average pore size measured by SEM in the region is too large (e.g., greater than 200 nm), the D 20 Since the pore size in the D region does not change substantially, trapping of small viruses is mainly achieved by stacking the pore structure in the thickness direction. 20 A large average pore size measured by SEM in the region often means a high risk of virus leakage, and D 20 If the average pore size measured by SEM in the region is too small (e.g., smaller than 60 nm), the pore structure with small pore size will fit into the layered structure in the thickness direction and obtain an excellent trapping effect for small viruses, but the D 20 The dirt loading of the area tends to be too low, resulting in rapid clogging of the filtration membrane and an excessively rapid rate of flux decay.

[0088] Optionally, the leak-preventing region has an average pore size of 40 to 80 nm as measured by SEM, and the gradient of change in the average pore size as measured by SEM of the leak-preventing region is 2 nm / μm or less.

[0089] By adopting the above technical measures, the leak prevention area needs to have a good trapping effect on small viruses as a structure that further reduces the risk of virus leakage, so the pore diameter of the pore structure of the leak prevention area needs to be appropriately small. However, as the average pore diameter of the leak prevention area measured by SEM becomes smaller, this inevitably leads to a further decrease in the dirt retention capacity of the leak prevention area and the permeation flux. On the other hand, by making the leak prevention area a basic symmetrical structure (the gradient of the change in the average pore diameter measured by SEM is 2 nm / μm or less), it is possible to ensure the trapping effect of small viruses while also ensuring higher dirt retention capacity and permeation flux of the leak prevention area.

[0090] The rate at which the dirt retention capacity decreases is much greater than the rate at which the pore size decreases. For example, if the pore structure is considered a virtual sphere, reducing the pore size by half will reduce the volume of the virtual sphere by one-eighth. This means that the pore size of the pore structure that can accommodate small viruses will be significantly reduced, and the space through which the raw liquid passes will also be significantly reduced (even if only the cross-sectional area of ​​the flow path is considered, it will be reduced to one-quarter). Therefore, even a slight reduction in the pore size of the pore structure will significantly reduce the dirt retention capacity and permeation flux of the pore structure, but the impact on the trapping effect of small viruses will not be significant (i.e., there is a diminishing return effect). Taking this into consideration, the risk of virus leakage can be reduced by appropriately reducing the average pore size measured by SEM in the leak-proof region, but the average pore size measured by SEM in the leak-proof region should not be too small.

[0091] Optionally, the average pore size of the leak-proof region as measured by SEM is D 20 The difference between the average pore size measured by SEM in the leak-prevention region and the average pore size measured by SEM in the leak-prevention region is 130 nm or less. 20 The difference from the gradient of change in average pore size measured by SEM in the region is 1 nm / μm or less.

[0092] By adopting the above technical measures, D 20 The area and the leak-proof area are adjacent areas, and the difference in the average pore size measured by SEM between the two is 130 nm or less, and the difference in the gradient of the average pore size measured by SEM between the two is 1 nm / μm or less. 20 Since both the leak-proof and leak-proof regions have essentially symmetrical structures, the gradient of the average pore size change measured by SEM in both regions is essentially the same (less than 1 nm / μm). 20 The area and leak-proof area means that there is no area of ​​abrupt change in pore size, so D 20 Both the leak prevention area and the leak prevention area are less likely to be clogged by localized concentrated traps, and the attenuation of the permeation flux is slow.

[0093] Optionally, the average pore size of the buffer region as measured by SEM is D40 The average pore size of the buffer region measured by SEM is 300 to 1500 nm, and the average pore size of the buffer region measured by SEM is larger than the average pore size of the buffer region measured by SEM. 40 The difference between the average pore size measured by SEM in the region is 200 to 1000 nm.

[0094] By adopting the above technical measures, the average pore size of the buffer region as measured by SEM must not be too large or too small. If the average pore size of the buffer region as measured by SEM is too small (for example, smaller than 300 nm), not only will the resistance to the feed liquid near the liquid feed surface be too high, but even the buffer region with a low dirt retention capacity will be prone to clogging with large particle impurities, resulting in an excessively rapid attenuation rate of the permeation flux. On the other hand, if the average pore size of the buffer region as measured by SEM is too large (for example, larger than 1500 nm), the membrane forming material of the filtration membrane in this application is cellulose, which has a soft fabric and the pressure of the feed liquid is directly applied to the liquid feed surface of the filtration membrane, so the requirements for the pressure resistance of the buffer region are high. Furthermore, as the average pore size of the buffer region as measured by SEM increases (the volume of the pore structure increases exponentially), the pressure resistance of the buffer region will be significantly reduced and the risk of structural collapse due to external pressure will increase.

[0095] In addition, the average pore size and D 40 The difference between the average pore size measured by SEM in the buffer region and the average pore size measured by SEM in the buffer region is less than 1000 nm. During the membrane production process, the casting liquid near the liquid supply surface comes into direct contact with the external phase separation environment, so it tends to have a pore structure with relatively large changes. 40 If the difference between the average pore size measured by SEM in the buffer region and the average pore size measured by SEM is 1000 nm or less, the filtration membrane can have a good trapping effect on large particle impurities and a sufficient dirt retention effect. 40 The region provides better trapping of larger particle mass due to the change in pore size, reducing the risk of large particle mass leakage.

[0096] Optionally, the flux retention of the filtration membrane is T of 0.4@(10 g / L) or more, where T is calculated by the following formula:

number

[0097] In the above formula, V is the load capacity of the filtration membrane, and when the permeation flux is attenuated by a%, the load capacity is Va, V25 is the load capacity of the filtration membrane when the permeation flux is attenuated by 25%, V50 is the load capacity of the filtration membrane when the permeation flux is attenuated by 50%, and V75 is the load capacity of the filtration membrane when the permeation flux is attenuated by 75%.

[0098] By adopting the above technical means, T of 0.4@(10g / L) or more means that the permeation flux retention (T) of the filtration membrane is 0.4 or more in a system with a protein concentration of 10g / L. Note that when testing using a system with a protein concentration of 10g / L, the load capacity of the filtration membrane is more challenged than in the currently common system with a protein concentration of only 1g / L, and the filtration membrane is more likely to become clogged.

[0099] The inventors of the present application have found that the attenuation rate of the permeation flux of the filtration membrane can be well characterized by the above formula, while ensuring that the filtration membrane has a good trapping effect (high LRV) for small viruses. It is more valuable to characterize the attenuation rate of the permeation flux of the filtration membrane quantitatively than to determine it qualitatively.

[0100] The measurement system for the above data was a protein concentration of 10 g / L, a buffer system of 50 mM acetic acid + 100 mM NaCl, pH 5.0, Condition 30.5 μs / cm, and density 1 g / mL. This system was pre-filtered with a 0.22 μm sterilized membrane, and the protein was IVIG.

[0101] Optionally, the permeation flux of the above filtration membrane is 60 L / (h m 2 )@30psi or more, and the maximum load capacity Vmax of the above filtration membrane is 300L / m2 The logarithmic reduction rate (LRV) of the filtration membrane against bacteriophage PP7 is 5 or more, and the LRV of the filtration membrane after retention in the process is 4 or more.

[0102] By adopting the above technical measures, the filtration membrane has a special structure, which can ensure not only a high permeation flux but also a high loading capacity and LRV. Furthermore, the unique structure of the leak-proof area can ensure that the filtration membrane can achieve an LRV of >4 even after dwelling in the process.

[0103] Process dwell is the point at which, during a virus challenge test, the permeation flux of the filtration membrane drops to 25%, or the point at which the challenge liquid has been filtered until only a small amount remains at the bottom. Then, the external pressure is removed, buffer is added, and the mixture is left for 15 minutes. After the process dwell, pressure is applied for the virus challenge test, and the filtrate obtained in the test is individually collected and the virus titer is measured to calculate the LRV after the process dwell.

[0104] When a conventional virus challenge test is performed, filtration ends when the permeation flux has attenuated to 25%, and the maximum load capacity Vmax of the filtration membrane (when the permeation flux has attenuated to 99% or more) is not measured at that point. Therefore, the maximum load capacity Vmax of the filtration membrane was calculated by fitting the load capacity change data to a curve during the virus challenge test.

[0105] In a second aspect, the present application provides a preparation process for the above filtration membrane using the following technical means:

[0106] The preparation process of the filtration membrane includes the following process steps:

[0107] S1. Casting: A first casting liquid containing 10-30 parts of a first film-forming polymer and 30-150 parts of a first solvent system and a second casting liquid containing 10-20 parts of a second film-forming polymer and 70-300 parts of a second solvent system are sequentially cast onto a support to form a two-layer liquid film, wherein the solids content of the first casting liquid is 17% or more, the solids content of the second casting liquid is 13% or less, and the surface tension of the first solvent system is 30 dyne / cm or less, and / or the surface tension of the second solvent system is 30 dyne / cm or less.

[0108] S2. Phase separation curing: The bilayer liquid membrane is immersed in a coagulation bath of water or ethanol, and the cast liquid is phase separated and cured until the bilayer liquid membrane is completely cured, thereby obtaining a wet membrane.

[0109] S3. Regeneration: The wet membrane is immersed in a regeneration bath to regenerate the membrane, thereby obtaining a regenerated membrane.

[0110] By employing the above technical means, the filtration membrane of the present application employs a two-layer casting process in which a first casting liquid and a second casting liquid with different solid contents are used to form small and large pore structures of the filtration membrane, respectively. Furthermore, the phase separation mechanism in the present application is non-solvent-induced phase separation (NIPS), which requires the interior of the casting liquid to be immersed in a coagulation bath to undergo phase separation and hardening. Therefore, the first casting liquid and the second casting liquid do not begin to separate at the same time. Furthermore, because the first casting liquid and the second casting liquid have roughly the same formulation, they undergo mutual mass transfer. Combined with the difference in the phase separation onset times between the first casting liquid and the second casting liquid described above, when the two liquids are cast sequentially onto a support, they undergo mutual mass transfer before the first casting liquid separates, forming a mixed casting liquid. Furthermore, the mixed casting liquid forms a gradient in solid content from the first casting liquid to the second casting liquid. Although there are differences between the first cast liquid, mixed cast liquid, and second cast liquid, the systems are nearly the same, and the phase separation mechanism is also the same, so the filtration membrane that is ultimately obtained does not have a clear boundary line or layer structure.

[0111] The first casting liquid, after phase separation and hardening, forms small pore regions that are primarily effective in trapping small viruses, and its higher solid content makes it more likely to form a small pore structure. Meanwhile, the second casting liquid, after phase separation and hardening, forms large pore regions that are primarily effective in trapping large particles, and its relatively lower solid content makes it more likely to form a large pore structure. Furthermore, because the second casting liquid comes into contact with the coagulation bath first, the second casting liquid, with its lower solid content, has a lower viscosity and therefore lowers the penetration resistance of the coagulation bath, making it more likely for the coagulation bath to penetrate into the casting liquid and induce phase separation. If the solid content of the second casting liquid is too high, a pore structure with too small a pore size will be formed, which will not only fail to provide a large pore region with a sufficiently large dirt retention capacity, but will also result in excessive resistance to the coagulation bath, potentially resulting in non-uniform phase separation of the casting liquid. If the solid content of the first casting liquid is too low, a pore structure with too large a pore size will be formed, potentially increasing the risk of virus leakage.

[0112] The surface tension of both the first and second solvent systems must be kept low (less than 30 dyne / cm). This is because the coagulation bath is the most important driving force for phase separation in the phase separation mechanism used in this application, and the rate of impregnation and penetration of the casting liquid into the interior significantly affects the phase separation rate of the casting liquid. To improve the uniformity of phase separation at each location in the casting liquid, it is necessary to reduce the resistance of the coagulation bath in the casting liquid, improve the uniformity of the dispersion of the coagulation bath at each location in the casting liquid, and reduce the unevenness of the phase separation rate caused by the uneven dispersion of the coagulation bath. Uneven phase separation rates reduce the uniformity of the pore size distribution of the pore structure, increasing the standard deviation of pore sizes and significantly increasing the risk of virus leakage through pore structures with large pore sizes.

[0113] Furthermore, by controlling the surface tension of the first and second solvent systems, it is possible to ensure that the coagulation bath impregnates and penetrates into the support at a faster rate, which results in more uniform phase separation of the first casting liquid and a pore structure that is essentially symmetrical and highly uniform. Furthermore, the proportion of pore structures with very small pore sizes formed by the first casting liquid with a high solid content is low, and the proportion of pore structures with large pore sizes is also low, which not only reduces the risk of virus leakage but also ensures that the filtration membrane has a relatively high permeation flux.

[0114] Optionally, the above step S2 specifically includes:

[0115] S21. Pretreatment: The bilayer liquid membrane is immersed in a pretreatment bath containing 40 to 100% aqueous solvent solution for 0.5 to 10 seconds to obtain a pretreated membrane.

[0116] S22. Curing: The pre-treated membrane is immersed in a coagulation bath of water or ethanol until the pre-treated membrane is completely phase-separated and cured to obtain a wet membrane; Includes:

[0117] By adopting the above-mentioned technical measures, the inventors of the present application discovered that pore formation on the surface of cellulosic materials is difficult, and that when a casting liquid is directly impregnated into a coagulation bath, a dense skin layer structure is easily formed in the second casting liquid, even if the solids content is low. This is related to the cellulosic material's high sensitivity to the coagulation bath, i.e., it undergoes rapid phase separation in a short period of time upon entering the coagulation bath. Generally, the faster the phase separation rate, the smaller the pore size of the formed pore structure, which is thought to be why a skin layer structure is often formed on the surface of the second casting liquid that comes into contact with the coagulation bath. Furthermore, as the coagulation bath passes through the skin layer and penetrates into the casting liquid, the solvent system in the casting liquid dilutes the coagulation bath, reducing its concentration and slowing the phase separation rate of the casting liquid. As a result, a pore structure with large pore size is formed in the casting liquid inside the skin layer structure. Furthermore, since the large pore structure can accommodate a large amount of coagulation bath, the coagulation bath rapidly penetrates into the mixed casting liquid and the first casting liquid, causing rapid phase separation between the mixed casting liquid and the first casting liquid.

[0118] Specifically, the present application discloses that a pretreatment is performed before impregnation in a coagulation bath. The pretreatment bath may be a mixture of solvent and water, or a pure solvent. When the pretreatment bath is a pure solvent, after impregnation of the bilayer liquid membrane, the surface of the second casting liquid is diluted with the pure solvent, resulting in the formation of a certain low-solids content region (lower than the solids content of the second casting liquid) on the surface of the second casting liquid, which significantly reduces the possibility of skin layer formation after impregnation in the coagulation bath. When the pretreatment bath is a mixture of solvent and water, a certain amount of solvent is added to the pretreatment bath compared to the coagulation bath, which significantly reduces the gelation effect. Therefore, the second casting liquid undergoes gel phase separation at a significantly slower rate in the pretreatment bath. Combined with the low solids content of the second casting liquid itself, the dilution effect of the solvent in the pretreatment bath also significantly reduces the possibility of skin layer formation from the second casting liquid.

[0119] The pretreatment bath being a 40-100% aqueous solvent solution means that the volume percentage of the solvent in the pretreatment bath is 40-100%, and it should be understood that the pretreatment bath may be a mixture of solvent and water or a pure solvent. The term "solvent" refers to a substance capable of dissolving a film-forming polymer. In this application, the solvent may be any of acetone, dioxane, dimethylacetamide, N-methylpyrrolidone, acetic acid, propionic acid, butyric acid, and valeric acid.

[0120] Optionally, the surface tension of the first solvent system is less than the surface tension of the second solvent system, and the solids content of the first casting liquid is 5 to 25 wt % higher than the solids content of the second casting liquid.

[0121] By adopting the above technical measures, it is generally believed that for approximately the same system, the first casting liquid with a higher solid content often has a higher viscosity, and that a higher viscosity means not only an increased resistance in the coagulation bath but also an increased resistance to mutual mass transfer between the first casting liquid and the second casting liquid. However, if the mass transfer resistance becomes too large and the coagulation bath in the first cast liquid becomes increasingly non-uniform, the uniformity of the phase separation rate at each location in the first cast liquid decreases, making it difficult to form a uniform and essentially symmetrical pore structure. If the mass transfer resistance of the first cast liquid is too large and it is difficult to form a mixed cast liquid by mutual mass transfer between the first and second cast liquids, the membrane structures formed from the first and second cast liquids will have pore structures with abrupt changes in pore size, which may result in the formation of clear boundaries or interlayer boundaries. Furthermore, such abrupt changes in pore size or boundaries may cause localized clogging due to concentrated trapping of particulate matter, which may rapidly reduce the permeation flux of the filtration membrane. Therefore, the difference in solid content between the first and second cast liquids must not exceed 25 wt %.

[0122] Furthermore, the difference in solids content between the first and second casting liquids should not be too small (for example, less than 5%). This is because a difference between the two that is too small often means that the solids content of the first casting liquid is too low or the solids content of the second casting liquid is too high. If the solids content of the first casting liquid is too low, the pore size of the formed pore structure will be large, making virus leakage more likely to occur. Conversely, if the solids content of the second casting liquid is too high, the pore size of the formed pore structure will be small, resulting in insufficient dirt retention and rapid clogging, which will lead to a rapid decline in the permeation flux of the filtration membrane.

[0123] Optionally, the first solvent system is a mixture of a first good solvent and a first pore-forming agent based on a mass ratio of (1-8):(5-22).

[0124] The second solvent system is a mixture of a second good solvent and a second pore-forming agent based on a mass ratio of (3-13):(4-17).

[0125] The first film-forming polymer is at least one of cellulose diacetate, cellulose triacetate, cellulose propionate, cellulose acetate phthalate, cellulose acetate butyrate, and cellulose acetate propionate.

[0126] The second film-forming polymer is at least one of cellulose diacetate, cellulose triacetate, cellulose propionate, cellulose acetate phthalate, cellulose acetate butyrate, and cellulose acetate propionate.

[0127] By adopting the above-mentioned technical methods, cellulose nitrate, cellulose acetate, cellulose propionate, etc., which are all cellulose ester compounds, can be regenerated into regenerated cellulose by simply hydrolyzing and deesterifying them under certain conditions. These are currently the mainstream cellulose-based film-forming materials, and are more environmentally friendly than cuprammonium fiber.

[0128] Both the first solvent system and the second solvent system are obtained by mixing a good solvent and a pore-forming agent. It should be understood that a good solvent is a solvent that has good solubility for the film-forming polymer, and a pore-forming agent is a component that has poor solubility for the film-forming polymer but promotes the formation of a pore structure and promotes the impregnation and penetration of a coagulation bath into the inside of the casting liquid during the film-forming process.

[0129] The first and second casting liquids have different casting orders and solid contents, and therefore different resistances to the coagulation bath. Furthermore, the first casting liquid, which has a higher solid content and is farther from the coagulation bath, requires more pore-forming agent to ensure rapid and uniform mass transfer between the first and second casting liquids, thereby facilitating mass transfer in the coagulation bath. The inventors of the present application have found that, in the two-layer casting process of the present application, a filtration membrane with a high LRV and a high loading capacity can be obtained when the first solvent system is a mixture of a first good solvent and a first pore-forming agent in a mass ratio of (1-8):(5-22), and the second solvent system is a mixture of a second good solvent and a second pore-forming agent in a mass ratio of (3-13):(4-17).

[0130] Optionally, the first good solvent is at least one of acetone, dioxane, dimethylacetamide, N-methylpyrrolidone, acetic acid, propionic acid, butyric acid, and valeric acid.

[0131] The second good solvent is at least one of acetone, dioxane, dimethylacetamide, N-methylpyrrolidone, acetic acid, propionic acid, butyric acid, and valeric acid.

[0132] The first pore-forming agent is at least one of ethanol, 1-propanol, isopropanol, n-butanol, 1-pentanol, 2-pentanol, hexafluoroisopropanol, or trifluoroethanol.

[0133] The second pore-forming agent is at least one of ethanol, 1-propanol, isopropanol, n-butanol, 1-pentanol, 2-pentanol, hexafluoroisopropanol, or trifluoroethanol.

[0134] Optionally, in the above step S3, the regeneration bath is an aqueous sodium hydroxide solution, the temperature is 20-40°C, and the regeneration time is 30-120 min.

[0135] Optionally, after the above step S3, the recycled membrane is placed in a crosslinking agent, which is at least one of halogenated epoxide, dihaloalkane, and dihalohydrin, to undergo a crosslinking treatment, and after the crosslinking treatment is completed, the membrane is washed to obtain a finished membrane, which is then subjected to further crosslinking in step S4.

[0136] By adopting the above technical means, the present application has decided to employ a pore-forming agent with low surface tension. Compared with the currently common PVP and PEG pore-forming agents, the low surface tension pore-forming agent promotes the penetration of the coagulation bath into the casting liquid, improving the uniformity of phase separation in each region within the casting liquid. Furthermore, pores can be formed after the PVP, PEG, or other pore-forming agents are removed. However, these pore-forming agents have high viscosity and create a large mass transfer resistance in the coagulation bath. Furthermore, given the high solids content and high viscosity of the first casting liquid, the addition of PVP and PEG, which can form pores but have a thickening effect, will significantly affect the uniformity of phase separation in the first casting liquid, potentially resulting in a large gradient and a non-uniform pore structure.

[0137] Furthermore, one of the characteristics of cellulose is its soft texture, which often results in poor mechanical properties for the regenerated fiber membrane obtained after hydrolysis and regeneration. Because virus removal filtration often uses dead-end filtration, the filtration membrane is often subjected to high pressure from the raw liquid, resulting in insufficient pressure resistance and a high risk of the membrane's pore structure collapsing. Pressure-induced collapse of the membrane's pore structure significantly impacts its virus trapping capacity and permeation flux. Therefore, after step S3, a crosslinking treatment with a crosslinking agent is performed to improve the membrane's pressure resistance and reduce the possibility of membrane structure collapse due to pressure. [Effects of the Invention]

[0138] As described above, the present application includes at least one of the following beneficial technical effects.

[0139] 1. The filtration membrane of the present application uses a cellulose-based raw material with excellent hydrophilicity as the membrane forming material, resulting in a high protein yield. When trapping 20 nm gold colloids, there is a distance between the 20 nm gold colloid trapping region and the drainage surface, reducing the risk of virus leakage. When trapping 40 nm gold colloids, the 40 nm gold colloid trapping region has a large thickness, allowing for a large dirt retention capacity. In addition, the transition region inside the filtration membrane is thin, which reduces the impact of impurities larger than 40 nm on the 20 nm gold colloid trapping region and reduces clogging, resulting in a high load capacity. In other words, the filtration membrane of the present application combines a high protein yield, a high virus trapping rate, and a high load capacity.

[0140] 2. This application discloses a filtration membrane preparation process that employs a two-layer casting process in which a pore structure with a small pore size and a pore structure with a large pore size are formed from casting liquids with different solid contents, respectively. Furthermore, because the two-layer casting liquid systems are nearly identical, mass transfer occurs between them after casting, forming a mixed casting liquid with a gradient in solid content. This significantly reduces the possibility of boundaries in the filtration membrane and reduces the possibility of rapid attenuation of the permeation flux of the filtration membrane due to localized concentrated trapping. [Brief explanation of the drawings]

[0141] [Figure 1] 1 is a general cross-sectional SEM image of the filtration membrane of Example 1 of the present application, the magnification in the figure is 1.5k×. [Figure 2] 1 is an SEM image of the liquid feed surface of the filtration membrane of Example 1 of the present application, the magnification in the figure is 5k×. [Figure 3] 1 is an SEM image of the drainage surface of the filtration membrane of Example 1 of the present application, the magnification in the figure is 20k×. [Figure 4] 1 shows a cross-sectional SEM image of the transition region of the filtration membrane of Example 1 of the present application after trapping 20 nm gold colloids. The magnification in the image is 10kx. Note that the pore structure is not clear in this image because it is filled with 20 nm gold colloids. [Figure 5] FIG. 1 is a cross-sectional SEM image of the D20 region of the filtration membrane of Example 1 of the present application after trapping 20 nm gold colloids, with a magnification of 50k×, showing how the pore structure of the filtration membrane traps the 20 nm gold colloids. [Figure 6] 1 shows a cross-sectional SEM image of the transition region of the filtration membrane of Example 1 of the present application after trapping 40 nm gold colloids. The magnification in the image is 20kx. Note that the pore structure is not clear in this image because it has been filled with 40 nm gold colloids. [Figure 7] FIG. 1 is a cross-sectional SEM image of the D40 region of the filtration membrane of Example 1 of the present application after trapping 40 nm gold colloids, with a magnification of 50k×, showing how the pore structure of the filtration membrane traps 40 nm gold colloids. [Figure 8] 1 is a graph showing the distribution of 40 nm gold colloids in the membrane thickness direction after trapping the 40 nm gold colloids with the filtration membrane of Example 4 of the present application. [Figure 9] 1 is a graph showing the distribution of 20 nm gold colloids in the membrane thickness direction after trapping the 20 nm gold colloids with the filtration membrane of Example 4 of the present application. [Figure 10]1 is a graph showing the distribution of 40 nm gold colloids in the membrane thickness direction after trapping the 40 nm gold colloids with the filtration membrane of Example 7 of the present application. [Figure 11] 1 is a graph showing the distribution of 20 nm gold colloids in the membrane thickness direction after trapping the 20 nm gold colloids in the filtration membrane in Example 7 of the present application. DETAILED DESCRIPTION OF THE INVENTION

[0142] The present application will now be further described in conjunction with the accompanying Figures 1 to 11. Example 1

[0143] In the examples of the present application, a process for preparing an asymmetric cellulose virus removal filtration membrane was disclosed, which included the following steps:

[0144] S1. Casting: The raw materials were mixed according to the composition ratios in Table 1 to prepare a first casting liquid and a second casting liquid. The first casting liquid located on the support side and the second casting liquid located above the first casting liquid were sequentially cast onto the support. After being cast onto the support, a two-layer liquid film was formed. In this example, the first film-forming polymer and the second film-forming polymer were both diacetate fiber, the first good solvent and the second good solvent were both dimethylacetamide, and the first pore-forming agent and the second pore-forming agent were both ethanol.

[0145] S2. Phase separation curing: Specifically, it includes the following steps:

[0146] S21. Pretreatment: The bilayer liquid membrane was immersed in a pretreatment bath containing 100% acetone for 6 seconds, and after the pretreatment, a pretreated membrane was obtained.

[0147] S22. Curing: The pretreated membrane obtained after the pretreatment was immersed in a coagulation bath until the pretreated membrane was completely phase-separated and cured to obtain a wet membrane. In this example, the coagulation bath was water.

[0148] S3. Regeneration: After sufficient hardening, the wet membrane obtained was immersed in a regeneration bath at 30°C for 80 minutes to hydrolyze the cellulose acetate into regenerated cellulose. The hydrolyzed membrane was then removed and washed with water until the pH was neutral, yielding a regenerated membrane. In this example, the regeneration bath was a 0.05 mol / L aqueous sodium hydroxide solution at 30°C.

[0149] S4. Crosslinking: The recycled membrane was placed in a crosslinking agent to undergo crosslinking. After the crosslinking process was completed, the finished membrane was obtained. In this example, the crosslinking agent was an aqueous solution containing 10 wt% epichlorohydrin, the crosslinking time was 25 min, and the crosslinking temperature was 50°C. Examples 2 to 6

[0150] The differences between Examples 2 to 6 and Example 1 are mainly in the process parameters for each step and the compositional blending of the casting liquid, details of which are shown in Table 1. Example 7

[0151] The main difference between Example 7 and Example 1 is that in Example 7, the two-layer structure was not pretreated using a pretreatment bath, but the two-layer liquid membrane was directly immersed in a coagulation bath to phase-separate and harden. The process parameters for each step and the composition of the casting liquid are detailed in Table 1. Step S2 specifically includes the following steps:

[0152] S2. Phase separation hardening: The two-layer liquid membrane was immersed in a coagulation bath, and the casting liquid was phase-separated and hardened until the two-layer liquid membrane was completely hardened, thereby obtaining a wet membrane. In this example, the coagulation bath was water. Example 8

[0153] The main difference between Example 8 and Example 1 is that the regenerated membrane was not subjected to a crosslinking treatment in Example 8. Other than that, the process parameters for each step and the composition of the casting liquid are shown in Table 1. Comparative Example

[0154] Comparative Example 1 The main difference between Comparative Example 1 and Example 1 is that in Comparative Example 1, the filtration membrane was prepared by a single-layer casting process, i.e., in Comparative Example 1, only the first casting liquid was cast onto the support, and the second casting liquid was not further cast above the first casting liquid. Other details of the process parameters for each step and the compositional formulation of the casting liquid are shown in Table 1. The preparation process of the filtration membrane in Comparative Example 1 included the following steps.

[0155] S1. Casting: A first casting liquid was prepared by mixing the raw materials according to the blending ratios in Table 1, and the resulting first casting liquid was cast onto a support to form a single-layer liquid film. In this comparative example, the first film-forming polymer was diacetate fiber, N-methylpyrrolidone was used as the first good solvent, and 1-propanol was used as the first pore-forming agent.

[0156] S2. Phase separation curing: Specifically, it includes the following steps: S21. Pretreatment: The single-layer liquid membrane was immersed in a pretreatment bath containing 80% aqueous acetone for 5 seconds. After the pretreatment, a pretreated membrane was obtained.

[0157] S22. Curing: The pretreated membrane obtained after the pretreatment was immersed in a coagulation bath until the pretreated membrane was completely phase-separated and cured to obtain a wet membrane. In this comparative example, the coagulation bath was water.

[0158] S3. Regeneration: After sufficient hardening, the wet membrane obtained was immersed in a regeneration bath at 30°C for 60 minutes to hydrolyze the cellulose acetate into regenerated cellulose. The hydrolyzed membrane was removed and washed with water to neutralize the pH, yielding a regenerated membrane. In this comparative example, the regeneration bath was a 0.05 mol / L aqueous sodium hydroxide solution at 30°C.

[0159] S4. Crosslinking: The regenerated membrane was placed in a crosslinking agent to undergo crosslinking treatment. After the crosslinking treatment was completed, the final membrane was obtained. In this comparative example, the crosslinking agent was an aqueous solution of epichlorohydrin with a concentration of 10 wt%, the crosslinking time was 25 min, and the crosslinking temperature was 50°C.

[0160] Comparative Example 2 The main difference between Comparative Example 2 and Example 1 is that in Comparative Example 2, a filtration membrane was prepared by a two-layer casting process. In Comparative Example 2, the pore-forming agent in both the first and second solvent systems was PEG-400. The process parameters for each step and the composition of the casting liquid are shown in Table 1.

[0161] Table 1: Composition and process parameters of the casting liquids in Examples 1 to 8 [Table 1]

[0162] Characteristic inspection and characteristic data

[0163] A. Viral challenge test The virus challenge test was performed according to the PDA TR41 standard. The model virus was bacteriophage PP7, the model protein was IVIG, and the buffer was PBS. During the test, the permeation flux and loading capacity were recorded over time, and the LRV, permeation flux, and loading capacity of the filtration membrane were obtained. When the permeation flux of the filtration membrane decreased by 25% or when only a small amount of challenge solution remained at the bottom, the external pressure was removed, buffer was added, and the membrane was left to stand for 15 minutes. After the process dwell, the membrane was pressurized to 30 psi for the virus challenge test. The filtrate obtained from the test was individually collected and the virus titer was measured, and the LRV after the process dwell was calculated.

[0164] Table 2 shows data on trapping of gold colloid in each of the examples and comparative examples.

[0165] Table 2: Data related to trapping of gold colloids in each example and comparative example [Table 2]

[0166] Regarding the gold colloid trapping test, the LRV of the filtration membrane measured in Comparative Example 2 was low (LRV<4), and the desired virus trapping effect was not obtained, so further testing was not performed.

[0167] Table 3 shows data relating to surface appearance parameters for each example and comparative example.

[0168] Table 3: Surface appearance parameters of each example and comparative example [Table 3]

[0169] Table 4 shows data on virus trapping by the filtration membranes of the examples and comparative examples.

[0170] Table 4: Data on virus traps in each example and comparative example [Table 4]

[0171] conclusion Comparing the technical solutions of Examples 1-6 with the data in Tables 2-4, it is easy to see that Examples 1-6 all have high permeation flux and virus trapping effect. Because no pretreatment was performed in Example 7, the pore structure on the liquid feed surface had significantly smaller pore sizes, resulting in smaller total load capacity and T. However, because no crosslinking was performed in Example 8, the filtration membrane had a buffer region structure with larger pore sizes, but its load capacity and T were small. This is because the uncrosslinked fiber structure had poor pressure resistance, causing the pore structure to collapse under pressure.

[0172] Comparing the technical means of Example 1 and Comparative Example 1 and the data in Tables 2-4, it is easy to see that Comparative Example 1 forms a three-layer structure in which the pore size changes from large to small and then to large (the pore size is large in the drainage surface and leak prevention area) through the preparation process of the single-layer cast liquid. Furthermore, the filtration membrane of Comparative Example 1 exhibited a high virus trapping effect, but the permeation flux, loading capacity, and T were all small. This is because a large pore size change gradient was formed overall during the membrane production, and the D 20 Area and D 40 This is because the thickness ratio is low in both regions, and this indicates that for both the 20 nm and 40 nm gold colloids, concentrated traps occur within a narrow range, making clogging more likely to occur.

[0173] Comparing the technical solutions of Example 1 and Comparative Example 2 and the data in Tables 2-4, it is easy to see that the poor virus-trapping effect of the filtration membrane of Comparative Example 2 is due to the fact that the pore-forming agent added to the casting liquid is not a low-surface tension substance, which increases the surface tension of the solvent system; and because PEG has a high viscosity, adding PEG may further increase the viscosity of the system. The coagulation bath does not easily penetrate into the casting liquid, resulting in uneven dispersion of the coagulation bath within the casting liquid, and the amount is small, which results in a slow and uneven phase separation rate within the casting liquid.

[0174] The specific examples are merely for explanation of the present application and do not limit the present application. After reading this specification, a person skilled in the art may make modifications without inventiveness as necessary, but as long as they fall within the scope of the claims of the present application, they will be protected by patent law.

Claims

1. An asymmetric cellulose virus removal filtration membrane including a porous body having an internal non-directional serpentine flow path, one surface of which is a liquid supply surface and the other surface of which is a liquid discharge surface, The average pore size measured by SEM is larger on the liquid supply surface than on the liquid drainage surface; When trapping gold colloid with the filtration membrane in a wet state, the region where gold colloid with a particle size of x nm is captured is defined as Dx, and in the thickness direction, the liquid supply surface of the porous body is defined as the position where the film thickness is 0%, and the liquid discharge surface of the porous body is defined as the position where the film thickness is 100%. The above D ２０ the region is located within a region of 20 to 99% of the thickness position of the porous body, and the ratio of the thickness of the region to the thickness of the porous body is 15 to 40%; The above D ４０ the region is located within a region of 0 to 80% of the thickness position of the porous body, and the ratio of the thickness of the region to the thickness of the porous body is 30 to 70%; The above D ４０ The region is the D ２０ It is closer to the liquid supply surface than the D ４０ The side of the region close to the drainage surface and the D ２０ An asymmetric cellulose virus removal filtration membrane, characterized in that the region between the side closer to the liquid supply surface and the region closer to the liquid supply surface is a transition region having a thickness of 20 μm or less.

2. The above D ２０ The region is located within a region of 40 to 99% of the thickness of the porous body, and the D ４０ The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein the region is located within a region of 0 to 60% of the membrane thickness position of the porous body.

3. The above D ２０ In the region, when the distance between the capture peak position of 20 nm gold colloid and the drainage surface is L1, the ratio of L1 to the thickness of the porous body is 10 to 50%, and ４０ 2. The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein, in the region, when the distance L2 is the distance between the capture peak position of 40 nm gold colloid and the liquid supply surface, the ratio of L2 to the thickness of the porous body is 1 to 40%.

4. 2. The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein the ratio of L3 to the thickness of the porous body is 20% or more, where L3 is the distance between the capture peak position of a 20 nm gold colloid and the capture peak position of a 40 nm gold colloid.

5. The above D ２０ In the region, the region where the amount of captured 20 nm gold colloid is 0.8 times or more the peak captured amount is defined as the concentrated capture region, and ２０ 2. The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein the ratio of the thickness of the concentrated capture region to the thickness of the region is 10 to 45%.

6. The above D ４０ In the region, the region where the amount of 40 nm gold colloid captured is 0.8 times or more the peak captured amount is defined as the concentrated dirt retention region, and ４０ 2. The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein the ratio of the thickness of the concentrated capture region to the thickness of the region is 60 to 90%.

7. The above D ４０ The side of the region close to the drainage surface and the D ２０ There is an overlap area with the side of the area closer to the liquid supply surface, or The above D ４０ The side of the region close to the drainage surface and the D ２０ The areas close to the liquid supply surface do not overlap with each other, and a non-overlapping area exists, The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein the overlapping region or the non-overlapping region is the transition region, the side of the transition region closer to the liquid supply surface is regarded as 0 and the side of the transition region closer to the liquid discharge surface is regarded as 1, the region of the transition region between 0 and 0.5 is regarded as an upper transition region, and the region of the transition region between 0.5 and 1 is regarded as a lower transition region, and the average pore size measured by SEM is larger in the upper transition region than in the lower transition region.

8. The average pore diameter of the transition region measured by SEM is 100 to 300 nm, and the D ４０ The average pore size of the region measured by SEM is smaller than the D ２０ The asymmetric cellulose virus removal filtration membrane according to claim 7, characterized in that the pore size is larger than the average pore size measured by SEM in the region.

9. The asymmetric cellulose virus removal filtration membrane according to claim 7, wherein the difference between the average pore size of the transition region measured by SEM in the 0.3 to 0.5 region and the average pore size of the transition region measured by SEM in the 0.5 to 0.7 region is 30 to 100 nm.

10. The asymmetric cellulose virus removal filtration membrane according to claim 7, wherein the ratio of the asymmetry coefficient of the upper transition region to the asymmetry coefficient of the lower transition region is 0.8 to 1.2, the asymmetry coefficient of the upper transition region is the ratio of the average pore size of the transition region measured by SEM in a region of 0 to 0.1 to the average pore size of the transition region measured by SEM in a region of 0.4 to 0.5, and the asymmetry coefficient of the lower transition region is the ratio of the average pore size of the transition region measured by SEM in a region of 0.5 to 0.6 to the average pore size of the transition region measured by SEM in a region of 0.9 to 1.

11. The asymmetric cellulose virus removal filtration membrane according to claim 10, wherein the asymmetry coefficient of the upper transition region is 1.1 to 1.6, and the asymmetry coefficient of the lower transition region is 1.1 to 1.

6.

12. The asymmetric cellulose virus removal filtration membrane according to claim 7, wherein the ratio of the fiber diameter of the upper transition region measured by SEM to the fiber diameter of the lower transition region measured by SEM is 0.8 to 1.2, the fiber diameter of the upper transition region measured by SEM is 20 to 50 nm, and the fiber diameter of the lower transition region measured by SEM is 20 to 45 nm.

13. The asymmetric cellulose virus removal filtration membrane according to claim 7, wherein the difference between the fiber diameter measured by SEM in the 0.3 to 0.5 region of the transition region and the fiber diameter measured by SEM in the 0.5 to 0.7 region of the transition region is 10 nm or less.

14. The above D ４０ The average pore size measured by SEM in the region gradually decreases from the liquid supply surface to the liquid discharge surface. ２０ 2. The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein the average pore size of the region measured by SEM does not change substantially from the liquid supply surface to the liquid discharge surface.

15. The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein the liquid supply surface has an average pore diameter measured by SEM of 350 to 2000 nm, and the liquid discharge surface has an average pore diameter measured by SEM of 30 to 45 nm.

16. The above D ２０ The area between the drainage surface and the drainage area is a leak-proof area, and / or ４０ The asymmetric cellulose virus removal filtration membrane according to claim 1, wherein the region between the membrane and the liquid supply surface is a buffer region.

17. The asymmetric cellulose virus removal filtration membrane according to claim 16, wherein the ratio of the thickness of the leak prevention region to the thickness of the porous body is 5 to 30%, and / or the ratio of the thickness of the buffer region to the thickness of the porous body is 1 to 10%.

18. The above D ４０ The average pore diameter of the region measured by SEM is 150 to 500 nm, and ２０ The asymmetric cellulose virus removal filtration membrane according to claim 16, characterized in that the average pore size of the region measured by SEM is 60 to 200 nm.

19. The asymmetric cellulose virus removal filtration membrane according to claim 16, wherein the leak prevention region has an average pore size of 40 to 80 nm as measured by SEM, and the change gradient of the average pore size as measured by SEM in the leak prevention region is 2 nm / μm or less.

20. The average pore diameter of the leak-preventing region measured by SEM is ２０ the difference between the average pore size measured by SEM in the leak prevention region and the D ２０ The asymmetric cellulose virus removal filtration membrane according to claim 16, wherein the difference between the gradient of change in average pore size in the region measured by SEM and the gradient of change in average pore size in the region is 1 nm / μm or less.

21. The average pore size of the buffer region measured by SEM is ４０ the average pore diameter of the buffer region measured by SEM is larger than the average pore diameter of the buffer region measured by SEM, the average pore diameter of the buffer region measured by SEM is 300 to 1500 nm, and the average pore diameter of the buffer region measured by SEM is larger than the average pore diameter of the buffer region measured by SEM, ４０ The asymmetric cellulose virus removal filtration membrane according to claim 16, characterized in that the difference between the average pore size of the region measured by SEM is 200 to 1000 nm.

22. The filtration membrane has a flux retention of 0.4@(10 g / L) or more, The T is calculated by the following formula: [Equation 1] The asymmetric cellulose virus removal filtration membrane according to any one of claims 1 to 21, characterized in that, in the above formula, V is the load capacity of the filtration membrane, when the permeation flux is attenuated by a %, the load capacity is Va, V25 is the load capacity of the filtration membrane when the permeation flux is attenuated by 25%, V50 is the load capacity of the filtration membrane when the permeation flux is attenuated by 50%, and V75 is the load capacity of the filtration membrane when the permeation flux is attenuated by 75%.

23. The permeation flux of the filtration membrane is 60 L / (h m ２ ) @ 30 psi or more, and the maximum load capacity Vmax of the filtration membrane is 300 L / m ２ The asymmetric cellulose virus removal filtration membrane according to any one of claims 1 to 21, characterized in that the log reduction rate (LRV) of the filtration membrane against bacteriophage PP7 is 5 or more, and the LRV of the filtration membrane after process retention is 4 or more.

24. A process for preparing a filtration membrane according to any one of claims 1 to 23, comprising the steps of: S1. Casting: A process step of sequentially casting a first casting liquid containing 10-30 parts of a first film-forming polymer and 30-150 parts of a first solvent system onto a support, and a second casting liquid containing 10-20 parts of a second film-forming polymer and 70-300 parts of a second solvent system onto a support to form a two-layer liquid film, wherein the solid content of the first casting liquid is greater than that of the second casting liquid, and the difference between the solid contents is 3% or more, and the surface tension of the first solvent system is 30 dyne / cm or less, and / or the surface tension of the second solvent system is 30 dyne / cm or less; S2. Phase separation hardening: the bilayer liquid membrane is immersed in a coagulation bath of water or ethanol, and the casting liquid is phase separated and hardened until the bilayer liquid membrane is completely hardened, thereby obtaining a wet membrane; S3. Regeneration: A process step of immersing the wet membrane in a regeneration bath to regenerate the membrane, thereby obtaining a regenerated membrane; A process for preparing a filtration membrane, comprising:

25. Specifically, step S2 is S21. Pretreatment: The bilayer liquid membrane is immersed in a pretreatment bath containing 40 to 100% aqueous solvent solution for 0.5 to 10 seconds to obtain a pretreated membrane. S22. Curing: The pre-treated membrane is immersed in a coagulation bath of water or ethanol until the pre-treated membrane is completely phase-separated and cured to obtain a wet membrane; 25. A process for preparing a filtration membrane according to claim 24, comprising:

26. 26. The process for preparing a filtration membrane according to claim 24 or 25, wherein the surface tension of the first solvent system is lower than the surface tension of the second solvent system, the first casting liquid has a solids content of 17% or more, the second casting liquid has a solids content of 13% or less, and the first casting liquid has a solids content higher than the second casting liquid by 5 to 25 wt %.

27. the first solvent system is a mixture of a first good solvent and a first pore-forming agent in a mass ratio of (1-8):(5-22); the second solvent system is a mixture of a second good solvent and a second pore-forming agent in a mass ratio of (3-13):(4-17); the first film-forming polymer is at least one of cellulose diacetate, cellulose triacetate, cellulose propionate, cellulose acetate phthalate, cellulose acetate butyrate, and cellulose acetate propionate; 26. The process for preparing a filtration membrane according to claim 24 or 25, wherein the second film-forming polymer is at least one of cellulose diacetate, cellulose triacetate, cellulose propionate, cellulose acetate phthalate, cellulose acetate butyrate, and cellulose acetate propionate.

28. the first good solvent is at least one of acetone, dioxane, dimethylacetamide, N-methylpyrrolidone, acetic acid, propionic acid, butyric acid, and valeric acid; the second good solvent is at least one of acetone, dioxane, dimethylacetamide, N-methylpyrrolidone, acetic acid, propionic acid, butyric acid, and valeric acid; the first pore-forming agent is at least one of ethanol, 1-propanol, isopropanol, n-butanol, 1-pentanol, 2-pentanol, hexafluoroisopropanol, or trifluoroethanol; 28. The process for preparing a filtration membrane according to claim 27, wherein the second pore-forming agent is at least one of ethanol, 1-propanol, isopropanol, n-butanol, 1-pentanol, 2-pentanol, hexafluoroisopropanol, or trifluoroethanol.

29. The process for preparing a filtration membrane according to claim 24 or 25, wherein in step S3, the regeneration bath is an aqueous sodium hydroxide solution, the temperature is 20 to 40 ° C, and the regeneration time is 30 to 120 min.

30. 26. The process for preparing a filtration membrane according to claim 24 or 25, characterized in that after step S3, the regenerated membrane is placed in a crosslinking agent which is at least one of a halogenated epoxide, a dihaloalkane, and a dihalohydrin to undergo a crosslinking treatment, and after the crosslinking treatment is completed, the regenerated membrane is washed to obtain a product membrane, which is further crosslinked in step S4.

Images